import PortalTransforms

git-svn-id: https://svn.erp5.org/repos/public/erp5/trunk@24294 20353a03-c40f-0410-a6d1-a30d3c3de9de

product/PortalTransforms/ChangeLog

+DONT USE ChangeLog USE HISTORY.txt instead.
+
+2004-07-24  Christian Heimes <heimes@faho.rwth-aachen.de>
+
+    * Changed version to stick to Archetypes version.
+
+2004-05-25  Christian Heimes <heimes@faho.rwth-aachen.de>
+
+    * Seperate MimetypesRegistry to a new product
+
+2004-04-20  Christian Heimes <heimes@faho.rwth-aachen.de>
+
+    * transforms/rest.py: rest transform is now using the zope implementation if
+      available
+
+2004-04-07  Christian Heimes <heimes@faho.rwth-aachen.de>
+
+    * transforms/text_pre_to_html.py: new transform for preformatted plain text
+    * transforms/text_to_html.py: changed <br/> to <br />
+
+2004-03-17  Christian Heimes <heimes@faho.rwth-aachen.de>
+
+    * transforms/pdf_to_text.py: return text utf-8 encoded
+
+2004-02-04  Sylvain Thénault  <syt@logilab.fr>
+
+	* transforms/office_com.py: fix wrong import
+	
+2003-12-03  Sidnei da Silva  <sidnei@awkly.org>
+
+	* mime_types/magic.py (guessMime): Don't try to be so magic :)
+
+2003-11-18  Andreas Jung <andreas@andreas-jung.com)
+	
+	* commandtransform.py: fixed sick cleanDir() implementation
+
+2003-11-17  Andreas Jung <andreas@andreas-jung.com)
+
+	* added rtf_to_html.py converter
+    
+        * added rtf to as mimetypes to mime_types/__init__.py
+
+    * added rtf_to_xml.py converter
+     
+    * added pdf_to_text.py converter
+
+    * removed dependency from CMFDefault.utils for misc converters
+      (integrated code into libtransforms/utils.py)
+
+2003-11-14  Sidnei da Silva  <sidnei@plone.org>
+
+	* MimeTypesRegistry.py (MimeTypesRegistry.classify): If no results
+	this far, use magic.py module, written by Jason Petrone, and
+	updated by Gabriel Wicke with the data from gnome-vfs-mime-magic.
+
+2003-11-07  Sylvain Thénault  <syt@logilab.fr>
+
+	* use the same license as Archetypes (BSD like instead of GPL)
+
+        * www/tr_widgets.zpt: fix bug in the list widget (space before the
+	parameter's name, making it unrecognized)
+
+	* zope/Transform.py: fix set parameters to correctly remap
+	transform if editable inputs or output. (fix #837244)
+
+	* TransformEngine.py: better error messages, a few lines wrapping
+
+	* zope/__init__.py: use pt_globals instead of globals for variable
+	handling the product globals, making it reloadable
+
+	* Extensions/Install.py: use pt_globals
+
+	* www/listMimeTypes.zpt: use mt/normalized as id instead of mt/name
+
+2003-11-05  Sylvain Thénault  <syt@logilab.fr>
+
+	* unsafe_tranforms/command.py: added dummy output mime type to avoid
+	error when added via the ZMI (fix #837252)
+
+2003-10-30 Sylvain Thénault  <syt@logilab.fr>
+
+	* fixed addMimeType, editMimeType and tr_widget templates (fix #832958)
+
+2003-10-03  Sidnei da Silva  <sidnei@dreamcatcher.homeunix.org>
+
+	* utils.py (TransformException.getToolByName): Modified
+	getToolByName to have a fallback mimetypes_registry, so we can
+	simplify BaseUnit.
+
+2003-09-23  Sylvain Thénault <syt@logilab.fr>
+
+	* MimesTypesRegistry.py: make unicode error handling configurable
+
+	* zope/MimesTypesTool.py: add a property for unicode error handling
+
+	* zope/Transform.py: make tests working
+
+2003-08-19  Sylvain Thénault  <syt@logilab.fr>
+
+	* transforms/rest.py: override "traceback" setting to avoid
+	sys.exit !
+
+	* transforms/text_to_html.py: use html_quote
+
+
+2003-08-12  Sylvain Thénault  <syt@logilab.fr>
+
+	* TransformEngine.py: set "encoding" in datastream metadata if
+	tranform provides a "output_encoding" attribute. Fix access to
+	"id" instead of "name()"
+
+	* zope/Transform.py: add some code to handle output encoding...
+
+2003-08-08  Sylvain Thénault  <syt@logilab.fr>
+
+	* MimeTypesRegistry.py: use suffix map has the standard mime types
+	module, hopefully correct behaviour of classify
+
+	* unsafe_transforms/build_transforms.py: fix inputs and output
+	mime type of ps_to_text transform
+
+2003-08-07  Sylvain Thenault  <sylvain.thenault@logilab.fr>
+
+	* encoding.py: new module which aims to detect encoding of text
+	files
+
+	* MimeTypesRegistry.py: use the encoding module in iadapter
+
+2003-08-06  Sylvain Thenault  <sylvain.thenault@logilab.fr>
+
+	* MimeTypesRegistry.py (classify): return
+	'application/octet-stream' instead of None
+
+	* transforms/text_to_html.py: replace '\n' with <br/> instead of
+	<pre> wrapping
+
+	* unsafe_transforms/build_transforms.py: create a ps_to_text
+	transform if ps2ascii is available
+
+	* tests/test_transforms.py: handle name of transforms to test on
+	command line
+
+	* transforms/__init__.py: do not print traceback on missing binary
+	exception
+
+2003-08-01  Sylvain Thenault  <sylvain.thenault@logilab.fr>
+
+	* transforms/text_to_html.py: new transform to wrap plain text in
+	<pre> for html
+
+	* transforms/test_transforms.py: add test for text_to_html
+
+2003-07-28  Sylvain Thenault  <sylvain.thenault@logilab.fr>
+
+	* zope/TransformsChain.py: fixes to make it works within Zope.
+
+	* www/editTransformsChain.zpt: add inputs / output information.
+
+2003-07-28  Sylvain Thenault  <sylvain.thenault@logilab.fr>
+
+	* transforms/rest.py: remove class="document"
+
+	* tests/test_transforms.py: added missing output for the identity
+	transform's test, fix initialize method.
+
+2003-07-21  Sylvain Thenault  <sylvain.thenault@logilab.fr>
+
+	* transforms/identity.py: added identity transform (used for instance
+	to convert text/x-rest to text/plain).
+
+	* tests/test_transforms.py: added test for the identity transform.
+
+2003-07-11  Sylvain Thenault  <sylvain.thenault@logilab.fr>
+
+	* unsafe_transforms/xml.py: just make it working.
+
+	* unsafe_transforms/command.py: add missing "name" argument to the
+	constructor. Use popen3 instead of popen4.
+
+	* unsafe_transforms/build_transforms.py: create an xml_to_html
+	transform if an xslt processor is available (however this transform
+	is not configured for any doctypes / dtds). Create tidy_html
+	transform if the tidy command is available.
+
+	* tests/test_transforms.py: add test cases for the xml and
+	html_tidy transform.
+
+	* transform.py: added transform_customize hook.
+
+	* docs/user_manual.rst: explain difference between python distro
+	and zope product. Added notice about archetypes integration.
+
+	* docs/dev_manual.rst: minor fixes.
+
+003-07-10  Sylvain Thenault  <sylvain.thenault@logilab.fr>
+
+	* refactoring to permit use of this package outside zope :)
+	Zope mode is triggered when "import Zope" doesn't fail
+
+	* fix bug in word_to_html / office_wvware transform
+
+	* add a generic test for transforms. It's much more easier now to
+	add a test for a transform :)
+
+	* add licensing information
+
+	* interfaces.py: complete / cleanup interfaces
+
+	* bin/tranform: add command line tool
+
+	* unsafe_transforms/command.py: bug fix
+
+	* addTransformsChain.zpt: fix typo
+
+	* fix #768927
+
+2003-07-09  Sylvain Thenault  <sylvain.thenault@logilab.fr>
+
+	* code cleaning:
+	  - moved Transform and TransformsChain in their own files
+	  - removed no more used bindingmixin and sourceAdapter
+	  - merged transform and chain classes together
+	  - generic cache and misc utilities in the new utils.py.
+
+	* ready for 1.0 alpha1 :)
+
+2003-07-05  Sylvain Thenault  <sylvain.thenault@logilab.fr>
+
+	* make the PortalTransforms product from the original transform
+	package and the mimetypes / transforms tools originaly defined in
+	Archetypes.
+
+	* drop the ability to use it as a standalone python package, since
+	there was too much duplicated code to make it works.
+
+	* some works on tests to make them succeed :)
+
+	* MimeTypesTool.py (MimeTypesTool.lookup): return an empty list
+	instead of None when no matching mime types is found.
+
+2003-05-14  Sidnei da Silva  <sidnei@x3ng.com>
+
+	* interface.py: Trying to normalize the way interfaces are
+	imported in different versions of Zope.
+
+2003-04-21  Sidnei da Silva  <sidnei@x3ng.com>
+
+	* __init__.py: Fixed lots of things here and there to make it work
+	with the new BaseUnit in Archetypes.
+
+2003-04-20  Sidnei da Silva  <sidnei@x3ng.com>
+
+	* tests/output/rest3.out: Fixed subtitle and added a test.
+
+2003-04-19  Sidnei da Silva  <sidnei@x3ng.com>
+
+	* tests/test_rest.py (BaseReSTFileTest.testSame): Added tests
+	based on input/output dirs to make it easy to add new tests for reST.
+
+	* transforms/rest.py (rest.convert): Rendering of
+	reST was broken. It was not rendering references the right way,
+	and it didnt seem like it was doing the right thing with
+	titles. Updated to use docutils.core.publish_string.
+
+	* tests/test_all.py (test_suite): Added lynx_dump to transform
+	html -> text. With tests.
+
+
+2003-04-18  Sidnei da Silva  <sidnei@x3ng.com>
+
+	* tests/test_all.py (test_suite): Removed dependencies from
+	CMFCore on testsuite.
+
+	* __init__.py: Made it work without being inside Products. We
+	eventually need to make a distutils setup, and then this can be
+	removed. If someone knows a better way to do this, please do.
+

product/PortalTransforms/DEPENDS

+lynx
+pdftohtml
+python-docutils

product/PortalTransforms/Extensions/Install.py

+import os
+from Products.CMFCore.DirectoryView import addDirectoryViews
+from Products.CMFCore.DirectoryView import registerDirectory
+from Products.CMFCore.DirectoryView import createDirectoryView
+from Products.CMFCore.DirectoryView import manage_listAvailableDirectories
+from Products.CMFCore.utils import getToolByName
+from Products.CMFCore.utils import minimalpath
+from Globals import package_home
+from Acquisition import aq_base
+from OFS.ObjectManager import BadRequestException
+
+from Products.PortalTransforms import GLOBALS
+from Products.PortalTransforms import skins_dir
+
+from StringIO import StringIO
+
+def install(self):
+    out = StringIO()
+    qi=getToolByName(self, 'portal_quickinstaller')
+    qi.installProduct('MimetypesRegistry',)
+    id = 'portal_transforms'
+    if hasattr(aq_base(self), id):
+        pt = getattr(self, id)
+        if not getattr(aq_base(pt), '_new_style_pt', None) == 1:
+            print >>out, 'Removing old portal transforms tool'
+            self.manage_delObjects([id,])
+
+    if not hasattr(aq_base(self), id):
+        addTool = self.manage_addProduct['PortalTransforms'].manage_addTool
+        addTool('Portal Transforms')
+        print >>out, 'Installing portal transforms tool'
+
+    updateSafeHtml(self, out)
+
+    correctMapping(self, out)
+    # not required right now
+    # installSkin(self)
+    
+    return out.getvalue()
+
+def correctMapping(self, out):
+
+    pt = getToolByName(self, 'portal_transforms')
+    pt_ids = pt.objectIds()
+
+    for m_in, m_out_dict in pt._mtmap.items():
+        for m_out, transforms in m_out_dict.items():
+            for transform in transforms:
+                if transform.id not in pt_ids:
+                    #error, mapped transform is no object in portal_transforms. correct it!
+                    print >>out, "have to unmap transform (%s) cause its not in portal_transforms ..." % transform.id
+                    try:
+                        pt._unmapTransform(transform)
+                    except:
+                        raise
+                    else:
+                        print >>out, "...ok"
+
+def updateSafeHtml(self, out):
+    print >>out, 'Update safe_html...'
+    safe_html_id = 'safe_html'
+    safe_html_module = "Products.PortalTransforms.transforms.safe_html"
+    pt = getToolByName(self, 'portal_transforms')
+    for id in pt.objectIds():
+        transform = getattr(pt, id)
+        if transform.id == safe_html_id and transform.module == safe_html_module:
+            
+            try:
+                disable_transform = transform.get_parameter_value('disable_transform')
+            except KeyError:
+                print >>out, '  replace safe_html (%s, %s) ...' % (transform.name(), transform.module)
+                try:
+                    pt.unregisterTransform(id)
+                    pt.manage_addTransform(id, safe_html_module)
+                except:
+                    raise
+                else:
+                    print >>out, '  ...done'
+    
+    print >>out, '...done'
+
+def installSkin(self):
+    skinstool=getToolByName(self, 'portal_skins')
+
+    fullProductSkinsPath = os.path.join(package_home(GLOBALS), skins_dir)
+    productSkinsPath = minimalpath(fullProductSkinsPath)
+    registered_directories = manage_listAvailableDirectories()
+    if productSkinsPath not in registered_directories:
+        registerDirectory(skins_dir, GLOBALS)
+    try:
+        addDirectoryViews(skinstool, skins_dir, GLOBALS)
+    except BadRequestException, e:
+        pass  # directory view has already been added
+
+    files = os.listdir(fullProductSkinsPath)
+    for productSkinName in files:
+        if os.path.isdir(os.path.join(fullProductSkinsPath, productSkinName)) \
+               and productSkinName != 'CVS':
+            for skinName in skinstool.getSkinSelections():
+                path = skinstool.getSkinPath(skinName)
+                path = [i.strip() for i in  path.split(',')]
+                try:
+                    if productSkinName not in path:
+                        path.insert(path.index('custom') +1, productSkinName)
+                except ValueError:
+                    if productSkinName not in path:
+                        path.append(productSkinName)
+                path = ','.join(path)
+                skinstool.addSkinSelection(skinName, path)

product/PortalTransforms/Extensions/__init__.py

+#poof, you're a product

product/PortalTransforms/HISTORY.txt

+1.4.0-final - 2006-06-16
+========================
+
+  * Shut down a noisy logging message to DEBUG level.
+    [hannosch]
+
+  * Converted logging infrastructure from zLOG usage to Python's logging module.
+    [hannosch]
+
+  * Avoid DeprecationWarning for manageAddDelete.
+    [hannosch]
+
+  * Spring-cleaning of tests infrastructure.
+    [hannosch]
+
+1.4.0-beta1 - 2006-03-26
+========================
+
+  * removed odd archetypes 1.3 style version checking
+    [jensens]
+
+  * Removed BBB code for CMFCorePermissions import location.
+    [hannosch]
+
+  * removed deprecation-warning for ToolInit
+    [jensens]
+
+1.3.9-final02 - 2006-01-15
+==========================
+
+  * nothing - the odd version checking needs a version change to stick to 
+    Archetypes version.
+    [yenzenz]
+
+1.3.9-RC1 - 2005-12-29
+======================
+
+  * Fixed [ 1293684 ], unregistered Transforms are not unmaped,
+    Transformation was deleted from portal_transforms, but remained 
+    active. 
+    http://sourceforge.net/tracker/index.php?func=detail&aid=1293684&group_id=75272&atid=543430
+    Added a cleanup that unmaps deleted transforms on reinstall
+    [csenger]
+
+  * Replaced the safe_html transformation with a configurable version
+    with the same functionality. Migration is handled on reinstall.
+    http://trac.plone.org/plone/ticket/4538
+    [csenger] [dreamcatcher]
+
+  * Removed CoUnInitialize call. According to Mark Hammond: The
+    right thing to do is call that function, although almost noone
+    does (including pywin32 itself, which does CoInitialize the main
+    thread) and I've never heard of problem caused by this
+    omission.
+    [sidnei]
+
+  * Fix a long outstanding issue with improper COM thread model
+    initialization. Initialize COM for multi-threading, ignoring any
+    errors when someone else has already initialized differently.
+    https://trac.plone.org/plone/ticket/4712
+    [sidnei]
+
+  * Correct some wrong security settings.
+    [hannosch]
+
+  * Fixed the requirements look-up from the policy
+    (#1358085)
+
+
+1.3.8-final02 - 2005-10-11
+==========================
+
+  * nothing - the odd version checking needs a version change to stick to 
+    Archetypes version.
+    [yenzenz]
+
+1.3.7-final01 - 2005-08-30
+==========================
+
+  * nothing - the odd version checking needs a version change to stick to 
+    Archetypes version.
+    [yenzenz]
+
+1.3.6-final02 - 2005-08-07
+==========================
+
+  * nothing - the odd version checking needs a version change to stick to 
+    Archetypes version.
+    [yenzenz]
+
+1.3.6-final - 2005-08-01
+========================
+
+  * Added q to the list of valid and safe html tags by limi's request.
+    Wrote test for safe_html parsing.
+    [hannosch]
+
+  * Added ins and del to the list of valid and safe html tags.
+    [ 1199917 ] XHTML DEL tag is removed during the safe_html conversion
+    [tiran]
+
+1.3.5-final02 - 2005-07-17
+==========================
+
+  * changed version to stick to appropiate Archetypes Version.
+    [yenzenz]
+
+1.3.5-final - 2005-07-06
+========================
+
+  * pdf_to_html can show images now. Revert it to command transformer and
+    make it work under windows.
+    [panjunyong]
+
+  * refined command based unsafe transform to make it work with windows.
+    [panjunyong]
+
+  * Disabled office_uno by default because it doesn't support multithread yet
+    [panjunyong]
+
+  * Rewrote office_uno to make it work for the recent PyUNO.
+    [panjunyong]
+
+1.3.4-final01 - 2005-05-20
+==========================
+
+  * nothing (I hate to write this. But the odd version checking needs it).
+    [yenzenz]
+
+1.3.4-rc1 - 2005-03-25
+======================
+
+ * Better error handling for safe html transformation
+   [tiran]
+
+1.3.3-final - 2005-03-05
+========================
+
+ * Updated link to rtf converter to http://freshmeat.net/projects/rtfconverter/
+   [tiran]
+
+ * Small fix for the com office converter. COM could crash if word is 
+   invisible. Also a pop up might appeare when quitting word.
+   [gogo]
+
+ * Fixed [ 1053846 ] Charset problem with wvware word_to_html conversion
+   [flacoste]
+
+ * Fixed python and test pre transforms to use html quote special characters.
+   Thx to stain. [ 1091670 ] Python source code does not escape HTML.
+   [tiran]
+
+ * Fixed [ 1121812 ] fix PortalTransforms unregisterTransformation()
+   unregisterTransformation() misses to remove from the zodb the persistance
+   wrapper added to the trasformation
+   [dan_t]
+
+ * Fixed [ 1118739 ] popentransform does not work on windows
+   [duncanb]
+
+ * Fixed [ 1122175 ] extra indnt sytax error in office_uno.py
+   [ryuuguu]
+
+ * fixed bug with some transformers' temp filename: it tried to use original filename
+   which is encoded in utf8 and may contrain invalid charset for my Windows server. 
+   Just use filename as: unknown.suffix
+   [panjunyong]
+
+ * STX header level is set to 2 instead of using zope.conf. Limi forced me to
+   change it.
+   [tiran]
+
+ * fixed bug: word_to_html uses office_com under windows
+
+1.3.2-5 - 2004-10-17
+====================
+
+ * Fixed [ 1041637 ] RichWidget: STX level should be set to 3 instead 1. The
+   structured text transform is now using the zope.conf option or has an
+   optional level paramenter in the convert method.
+   [tiran]
+
+ * Added win32api.GetShortPathName to libtransforms/commandtransform
+   so binaries found in directories which have spaces in their names
+   will work as expected
+   [runyaga]
+
+1.3.2-4 - 2004-09-30
+====================
+
+ * nothing changed
+
+1.3.2-3 - 2004-09-25
+====================
+
+ * Fixed more unit tests
+   [tiran]
+
+1.3.2-2 - 2004-09-17
+====================
+
+ * Fixed [ 1025066 ] Serious persistency bug
+   [dmaurer]
+
+ * Fixed some unit tests failurs. Some unit tests did fail because the reST
+   and STX output has changed slightly.
+   [tiran]
+
+ * Don't include the first three lines of the lynx output which are url,
+   title and a blank line. This fixed also a unit test because the url
+   which was a file in the fs did change every time.
+   [tiran]
+
+ * Fixed a bug in make_unpersistent. It seemed that this method touched values
+   inside the mapping.
+   [dreamcatcher]
+
+1.3.2-1 - 2004-09-04
+====================
+
+ * Disabled filters that were introduced in 1.3.1-1. The currently used 
+   transform path algo is broken took too long to find a path.
+   [tiran]
+
+ * Cleaned up major parts of PT by removing the python only implementation which
+   was broken anyway
+
+ * Fixed [ 1019632 ] current svn bundle (rev 2942) broken
+
+1.3.1-1 - 2004-08-16
+====================
+
+ * Introduce the concept of filters (one-hop transforms where the source and
+   destination are the same mimetype).
+   [dreamcatcher]
+
+ * Add a html filter to extract the content of the body tag (so we don't get a 
+   double <body> when uploading full html files).
+   [dreamcatcher]
+
+ * Change base class for Transform to SimpleItem which is equivalent to the
+   previous base classes and provides a nice __repr__.
+   [dreamcatcher]
+
+ * Lower log levels.
+   [dreamcatcher]
+
+ * cache.py: Added purgeCache, fixed has cache test.
+   [tiran]
+
+ * Fixed non critical typo in error message: Unvalid -> Invalid
+   [tirna]
+
+1.3.0-3 - 2004-08-06
+====================
+
+ * Added context to the convert, convertTo and __call__ methods. The context is
+   the object on which the transform was called.
+   [tiran]
+
+ * Added isCacheable flag and setCacheable to idatastream (data.py). Now you can
+   disable the caching of the result of a transformation.
+   [tiran]
+
+ * Added __setstate__ to load new transformations from the file system.
+   [tiran]
+
+ * Fixed [ 1002014 ] Add policy screen doesn't accept single entry
+   [tiran]
+   
+1.3.0-2 - 2004-07-29
+====================
+
+ * Added workaround for [ 997998 ] PT breaks ZMI/Find [tiran]

product/PortalTransforms/LICENSE.txt

+Copyright (c) 2002-2003, Benjamin Saller <bcsaller@ideasuite.com>, and 
+	                the respective authors.
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are
+met:
+
+    * Redistributions of source code must retain the above copyright
+      notice, this list of conditions and the following disclaimer.	
+    * Redistributions in binary form must reproduce the above
+      copyright notice, this list of conditions and the following disclaimer
+      in the documentation and/or other materials provided with the
+      distribution. 
+    * Neither the name of Archetypes nor the names of its contributors
+      may be used to endorse or promote products derived from this software
+      without specific prior written permission. 
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+

product/PortalTransforms/MANIFEST.in

+include ChangeLog
+include README
+include TODO
+include version.txt
+include bin/transform
+include bin/transform.bat
+recursive-include docs *.txt
+recursive-include docs *.rst
+recursive-include docs *.html
+recursive-include tests/input *
+recursive-include tests/output *

product/PortalTransforms/Makefile

+NAME=PortalTransforms
+MAJOR_VERSION=1.0
+MINOR_VERSION=4
+RELEASE_TAG=
+PACKAGE_NAME=${NAME}-${MAJOR_VERSION}.${MINOR_VERSION}${RELEASE_TAG}
+
+PYTHON="/usr/bin/python"
+TMPDIR=~/tmp
+
+CURDIR=~/src/archetypes/head/PortalTransforms
+BASE_DIR=${CURDIR}/..
+SOFTWARE_HOME=~/src/zope/2_7/lib/python
+INSTANCE_HOME=~/src/instance/shellex
+PACKAGES=PortalTransforms
+
+
+RM=rm -f
+RMRF=rm -rf
+FIND=find
+XARGS=xargs
+CD=cd
+LN=ln -sfn
+CP=cp
+TAR=tar
+MKDIR=mkdir -p
+
+.PHONY : clean test reindent reindent_clean sdist
+.PHONY : default
+
+# default:     The default step (invoked when make is called without a target)
+default: clean test
+
+clean :
+	find . \( -name '*~' -o -name '*.py[co]' -o -name '*.bak' \) -exec rm {} \; -print
+
+reindent :
+	~/src/reindent.py -r -v .
+
+test :
+	export INSTANCE_HOME=${INSTANCE_HOME}; export SOFTWARE_HOME=${SOFTWARE_HOME}; \
+	cd ${CURDIR}/tests && ${PYTHON} runalltests.py
+
+
+# sdist:       Create a source distribution file (implies clean).
+#
+sdist: reindent clean sdist_tgz
+
+# sdist_tgz:   Create a tgz archive file as a source distribution.
+#
+sdist_tgz:
+	echo -n "${MAJOR_VERSION}.${MINOR_VERSION}${RELEASE_TAG}" >\
+          ${CURDIR}/version.txt
+	${MKDIR} ${TMPDIR}/${PACKAGE_NAME}
+	${CD} ${TMPDIR}/${PACKAGE_NAME} && \
+          for package in ${PACKAGES}; do ${LN} ${BASE_DIR}/$$package .; done && \
+          ${CD} ${TMPDIR} && ${TAR} czfh ${BASE_DIR}/${PACKAGE_NAME}.tgz ${PACKAGE_NAME} \
+           --exclude=${PACKAGE_NAME}.tgz\
+           --exclude=CVS \
+           --exclude=.cvsignore \
+           --exclude=makefile \
+           --exclude=Makefile \
+           --exclude=*.pyc \
+           --exclude=TAGS \
+           --exclude=*~ \
+           --exclude=.#*
+	${RMRF} ${TMPDIR}/${PACKAGE_NAME}

product/PortalTransforms/README

+Portal Transforms
+=================
+
+This Zope product provides two new tools for the CMF in order to make MIME
+types based transformations on the portal contents, and so an easy to way to
+plugin some new transformations for previously unsupported content types. The
+provided tools are : 
+
+ * portal_transform (the transform tool) : handle transformation of data from a
+ mime type to another
+
+A bunch of ready to use transformations are also provided. Look at the
+documentation for more information.
+
+Notice this package can also be used as a standalone Python package. If
+you've downloaded the Python distribution, you can't make it a Zope
+product since Zope files have been removed from this distribution.
+
+This product is an off-spring of the Archetypes project. 
+
+
+Installation
+------------
+
+WARNING : The two installation methods may conflict, choose the one adapted to
+          your need.
+
+Zope
+````
+
+  * Put this package in your Zope's Products directory and restart Zope
+
+  * either use the QuickInstaller to add this product to your CMF site or add an 
+  external method to the root of your CMF site with the following information :
+
+    :module: PortalTransforms.Install
+
+    :method: install
+
+  and click the test tab to run it.
+
+Python
+``````
+
+  * Extract the tarball 
+
+  * Run "python setup.py install". See "python setup.py install  --help" for
+  installation options. 
+
+  * That's it, you should have the library and the *transform* command line tool
+  installed.
+
+
+Documentation
+-------------
+See the *docs* directory in this package.
+
+
+Mailing-list
+------------
+Discussion about this products occurs to the archetypes mailing list :
+http://sourceforge.net/mail/?group_id=75272
+or on the #plone channel of irc.freenode.net.
+
+
+Authors
+-------
+
+Benjamin Saller <bcsaller@yahoo.com>
+Sidnei da Silva  <sidnei@x3ng.com>
+Sylvain Thénault <sylvain.thenault@logilab.fr>

product/PortalTransforms/SUGGESTS

+wv
+xsltproc
+tidy
+unrtf
+ppthtml
+xlhtml
+gs-common
+

product/PortalTransforms/TODO

+TODO list for the Portal Transforms product
+-------------------------------------------
+
+* enhance unsafe_transforms/build_transforms to provide a bunch of
+  transformations using command/xml with various configuration
+
+* iencoding_classifier ?
+
+* make more transforms :)
+

product/PortalTransforms/Transform.py

+from logging import ERROR
+from UserDict import UserDict
+
+from Products.PageTemplates.PageTemplateFile import PageTemplateFile
+from Globals import InitializeClass
+from Globals import PersistentMapping
+try:
+    from ZODB.PersistentList import PersistentList
+except ImportError:
+    from persistent.list import PersistentList
+from OFS.SimpleItem import SimpleItem
+from AccessControl import ClassSecurityInfo
+
+from Products.CMFCore.permissions import ManagePortal
+from Products.CMFCore.utils import getToolByName
+
+from Products.PortalTransforms.interfaces import itransform
+from Products.PortalTransforms.utils import TransformException, log, _www
+from Products.PortalTransforms.transforms.broken import BrokenTransform
+
+__revision__ = '$Id: Transform.py 6255 2006-04-11 15:29:29Z hannosch $'
+
+def import_from_name(module_name):
+    """ import and return a module by its name """
+    __traceback_info__ = (module_name, )
+    m = __import__(module_name)
+    try:
+        for sub in module_name.split('.')[1:]:
+            m = getattr(m, sub)
+    except AttributeError, e:
+        raise ImportError(str(e))
+    return m
+
+def make_config_persistent(kwargs):
+    """ iterates on the given dictionnary and replace list by persistent list,
+    dictionary by persistent mapping.
+    """
+    for key, value in kwargs.items():
+        if type(value) == type({}):
+            p_value = PersistentMapping(value)
+            kwargs[key] = p_value
+        elif type(value) in (type(()), type([])):
+            p_value = PersistentList(value)
+            kwargs[key] = p_value
+
+def make_config_nonpersistent(kwargs):
+    """ iterates on the given dictionary and replace ListClass by python List,
+        and DictClass by python Dict
+    """
+    for key, value in kwargs.items():
+        if isinstance(value, PersistentMapping):
+            p_value = dict(value)
+            kwargs[key] = p_value
+        elif isinstance(value, PersistentList):
+            p_value = list(value)
+            kwargs[key] = p_value
+
+VALIDATORS = {
+    'int' : int,
+    'string' : str,
+    'list' : PersistentList,
+    'dict' : PersistentMapping,
+    }
+
+class Transform(SimpleItem):
+    """A transform is an external method with
+    additional configuration information
+    """
+
+    __implements__ = itransform
+
+    meta_type = 'Transform'
+
+    meta_types = all_meta_types = ()
+
+    manage_options = (
+                      ({'label':'Configure',
+                       'action':'manage_main'},
+                       {'label':'Reload',
+                       'action':'manage_reloadTransform'},) +
+                      SimpleItem.manage_options
+                      )
+
+    manage_main = PageTemplateFile('configureTransform', _www)
+    manage_reloadTransform = PageTemplateFile('reloadTransform', _www)
+    tr_widgets = PageTemplateFile('tr_widgets', _www)
+
+    security = ClassSecurityInfo()
+    __allow_access_to_unprotected_subobjects__ = 1
+
+    def __init__(self, id, module, transform=None):
+        self.id = id
+        self.module = module
+        # DM 2004-09-09: 'Transform' instances are stored as
+        #  part of a module level configuration structure
+        #  Therefore, they must not contain persistent objects
+        self._config = UserDict()
+        self._config.__allow_access_to_unprotected_subobjects__ = 1
+        self._config_metadata = UserDict()
+        self._tr_init(1, transform)
+
+    def __setstate__(self, state):
+        """ __setstate__ is called whenever the instance is loaded
+            from the ZODB, like when Zope is restarted.
+
+            We should reload the wrapped transform at this time
+        """
+        Transform.inheritedAttribute('__setstate__')(self, state)
+        self._tr_init()
+
+    def _tr_init(self, set_conf=0, transform=None):
+        """ initialize the zope transform by loading the wrapped transform """
+        __traceback_info__ = (self.module, )
+        if transform is None:
+            transform = self._load_transform()
+        else:
+            self._v_transform = transform
+        # check this is a valid transform
+        if not hasattr(transform, '__class__'):
+            raise TransformException('Invalid transform : transform is not a class')
+        if not itransform.isImplementedBy(transform):
+            raise TransformException('Invalid transform : itransform is not implemented by %s' % transform.__class__)
+        if not hasattr(transform, 'inputs'):
+            raise TransformException('Invalid transform : missing required "inputs" attribute')
+        if not hasattr(transform, 'output'):
+            raise TransformException('Invalid transform : missing required "output" attribute')
+        # manage configuration
+        if set_conf and hasattr(transform, 'config'):
+            conf = dict(transform.config)
+            self._config.update(conf)
+            make_config_persistent(self._config)
+            if hasattr(transform, 'config_metadata'):
+                conf = dict(transform.config_metadata)
+                self._config_metadata.update(conf)
+                make_config_persistent(self._config_metadata)
+        transform.config = dict(self._config)
+        make_config_nonpersistent(transform.config)
+        transform.config_metadata = dict(self._config_metadata)
+        make_config_nonpersistent(transform.config_metadata)
+
+        self.inputs = transform.inputs
+        self.output = transform.output
+        self.output_encoding = getattr(transform, 'output_encoding', None)
+        return transform
+
+    def _load_transform(self):
+        m = import_from_name(self.module)
+        if not hasattr(m, 'register'):
+            msg = 'Invalid transform module %s: no register function defined' % self.module
+            raise TransformException(msg)
+        try:
+            transform = m.register()
+        except Exception, err:
+            transform = BrokenTransform(self.id, self.module, err)
+            msg = "Cannot register transform %s, using BrokenTransform: Error\n %s" % (self.id, err)
+            self.title = 'BROKEN'
+            log(msg, severity=ERROR)
+        else:
+            self.title = ''
+        self._v_transform = transform
+        return transform
+
+    security.declarePrivate('manage_beforeDelete')
+    def manage_beforeDelete(self, item, container):
+        SimpleItem.manage_beforeDelete(self, item, container)
+        if self is item:
+            # unregister self from catalog on deletion
+            # While building business template, transform tool will be
+            # copied and this transform will be removed from copied one
+            # so that this must not use getToolByName to retrive the tool.
+            tr_tool = self.aq_parent
+            tr_tool._unmapTransform(self)
+
+    security.declarePublic('get_documentation')
+    def get_documentation(self):
+        """ return transform documentation """
+        if not hasattr(self, '_v_transform'):
+            self._load_transform()
+        return self._v_transform.__doc__
+
+    security.declarePublic('get_documentation')
+    def convert(self, *args, **kwargs):
+        """ return apply the transform and return the result """
+        if not hasattr(self, '_v_transform'):
+            self._load_transform()
+        return self._v_transform.convert(*args, **kwargs)
+
+    security.declarePublic('name')
+    def name(self):
+        """return the name of the transform instance"""
+        return self.id
+
+    security.declareProtected(ManagePortal, 'get_parameters')
+    def get_parameters(self):
+        """ get transform's parameters names """
+        if not hasattr(self, '_v_transform'):
+            self._load_transform()
+        keys = self._v_transform.config.keys()
+        keys.sort()
+        return keys
+
+    security.declareProtected(ManagePortal, 'get_parameter_value')
+    def get_parameter_value(self, key):
+        """ get value of a transform's parameter """
+        value = self._config[key]
+        type = self.get_parameter_infos(key)[0]
+        if type == 'dict':
+            result = {}
+            for key, val in value.items():
+                result[key] = val
+        elif type == 'list':
+            result = list(value)
+        else:
+            result = value
+        return result
+
+    security.declareProtected(ManagePortal, 'get_parameter_infos')
+    def get_parameter_infos(self, key):
+        """ get informations about a parameter
+
+        return a tuple (type, label, description [, type specific data])
+        where type in (string, int, list, dict)
+              label and description are two string describing the field
+        there may be some additional elements specific to the type :
+             (key label, value label) for the dict type
+        """
+        try:
+            return tuple(self._config_metadata[key])
+        except KeyError:
+            return 'string', '', ''
+
+    security.declareProtected(ManagePortal, 'set_parameters')
+    def set_parameters(self, REQUEST=None, **kwargs):
+        """ set transform's parameters """
+        if not kwargs:
+            kwargs = REQUEST.form
+        self.preprocess_param(kwargs)
+        for param, value in kwargs.items():
+            try:
+                self.get_parameter_value(param)
+            except KeyError:
+                log('Warning: ignored parameter %r' % param)
+                continue
+            meta = self.get_parameter_infos(param)
+            self._config[param] = VALIDATORS[meta[0]](value)
+
+        tr_tool = getToolByName(self, 'portal_transforms')
+        # need to remap transform if necessary (i.e. configurable inputs / output)
+        if kwargs.has_key('inputs') or kwargs.has_key('output'):
+            tr_tool._unmapTransform(self)
+            if not hasattr(self, '_v_transform'):
+                self._load_transform()
+            self.inputs = kwargs.get('inputs', self._v_transform.inputs)
+            self.output = kwargs.get('output', self._v_transform.output)
+            tr_tool._mapTransform(self)
+        # track output encoding
+        if kwargs.has_key('output_encoding'):
+            self.output_encoding = kwargs['output_encoding']
+        if REQUEST is not None:
+            REQUEST['RESPONSE'].redirect(tr_tool.absolute_url()+'/manage_main')
+
+
+    security.declareProtected(ManagePortal, 'reload')
+    def reload(self):
+        """ reload the module where the transformation class is defined """
+        log('Reloading transform %s' % self.module)
+        m = import_from_name(self.module)
+        reload(m)
+        self._tr_init()
+
+    def preprocess_param(self, kwargs):
+        """ preprocess param fetched from an http post to handle optional dictionary
+        """
+        for param in self.get_parameters():
+            if self.get_parameter_infos(param)[0] == 'dict':
+                try:
+                    keys = kwargs[param + '_key']
+                    del kwargs[param + '_key']
+                except:
+                    keys = ()
+                try:
+                    values = kwargs[param + '_value']
+                    del kwargs[param + '_value']
+                except:
+                    values = ()
+                kwargs[param] = dict = {}
+                for key, value in zip(keys, values):
+                    key = key.strip()
+                    if key:
+                        value = value.strip()
+                        if value:
+                            dict[key] = value
+
+InitializeClass(Transform)
+

product/PortalTransforms/TransformEngine.py

+from AccessControl.Role import RoleManager
+from AccessControl import ClassSecurityInfo
+from Acquisition import Implicit
+from Acquisition import aq_parent
+from Acquisition import aq_base
+from Globals import Persistent
+from Globals import InitializeClass
+from Globals import PersistentMapping
+try:
+    from ZODB.PersistentList import PersistentList
+except ImportError:
+    from persistent.list import PersistentList
+from OFS.Folder import Folder
+from OFS.SimpleItem import Item
+
+from Products.PageTemplates.PageTemplateFile import PageTemplateFile
+
+from Products.CMFCore.ActionProviderBase import ActionProviderBase
+from Products.CMFCore.permissions import ManagePortal, View
+from Products.CMFCore.utils import UniqueObject
+from Products.CMFCore.utils import getToolByName
+
+from Products.PortalTransforms.libtransforms.utils import MissingBinary
+from Products.PortalTransforms import transforms
+from Products.PortalTransforms.interfaces import iengine
+from Products.PortalTransforms.interfaces import idatastream
+from Products.PortalTransforms.interfaces import itransform
+from Products.PortalTransforms.data import datastream
+from Products.PortalTransforms.chain import TransformsChain
+from Products.PortalTransforms.chain import chain
+from Products.PortalTransforms.cache import Cache
+from Products.PortalTransforms.Transform import Transform
+from Products.PortalTransforms.utils import log
+from Products.PortalTransforms.utils import TransformException
+from Products.PortalTransforms.utils import BadRequest
+from Products.PortalTransforms.utils import _www
+
+__revision__ = '$Id: TransformEngine.py 6255 2006-04-11 15:29:29Z hannosch $'
+
+from logging import DEBUG
+
+class TransformTool(UniqueObject, ActionProviderBase, Folder):
+
+    id        = 'portal_transforms'
+    meta_type = id.title().replace('_', ' ')
+    isPrincipiaFolderish = 1 # Show up in the ZMI
+
+    __implements__ = iengine
+
+    meta_types = all_meta_types = (
+        { 'name'   : 'Transform',
+          'action' : 'manage_addTransformForm'},
+        { 'name'   : 'TransformsChain',
+          'action' : 'manage_addTransformsChainForm'},
+        )
+
+    manage_addTransformForm = PageTemplateFile('addTransform', _www)
+    manage_addTransformsChainForm = PageTemplateFile('addTransformsChain', _www)
+    manage_cacheForm = PageTemplateFile('setCacheTime', _www)
+    manage_editTransformationPolicyForm = PageTemplateFile('editTransformationPolicy', _www)
+    manage_reloadAllTransforms = PageTemplateFile('reloadAllTransforms', _www)
+
+    manage_options = ((Folder.manage_options[0],) + Folder.manage_options[2:] +
+                      (
+        { 'label'   : 'Caches',
+          'action' : 'manage_cacheForm'},
+        { 'label'   : 'Policy',
+          'action' : 'manage_editTransformationPolicyForm'},
+        { 'label'   : 'Reload transforms',
+          'action' : 'manage_reloadAllTransforms'},
+        )
+                      )
+
+    security = ClassSecurityInfo()
+
+    def __init__(self, policies=None, max_sec_in_cache=3600):
+        self._mtmap = PersistentMapping()
+        self._policies = policies or PersistentMapping()
+        self.max_sec_in_cache = max_sec_in_cache
+        self._new_style_pt = 1
+
+    # mimetype oriented conversions (iengine interface) ########################
+
+    def unregisterTransform(self, name):
+        """ unregister a transform
+        name is the name of a registered transform
+        """
+        self._unmapTransform(getattr(self, name))
+        if name in self.objectIds():
+            self._delObject(name)
+
+
+    def convertTo(self, target_mimetype, orig, data=None, object=None,
+                  usedby=None, context=None, **kwargs):
+        """Convert orig to a given mimetype
+
+        * orig is an encoded string
+
+        * data an optional idatastream object. If None a new datastream will be
+        created and returned
+
+        * optional object argument is the object on which is bound the data.
+        If present that object will be used by the engine to bound cached data.
+
+        * additional arguments (kwargs) will be passed to the transformations.
+        Some usual arguments are : filename, mimetype, encoding
+
+        return an object implementing idatastream or None if no path has been
+        found.
+        """
+        target_mimetype = str(target_mimetype)
+
+        if object is not None:
+            cache = Cache(object)
+            data = cache.getCache(target_mimetype)
+            if data is not None:
+                time, data = data
+                if self.max_sec_in_cache == 0 or time < self.max_sec_in_cache:
+                    return data
+
+        if data is None:
+            data = self._wrap(target_mimetype)
+
+        registry = getToolByName(self, 'mimetypes_registry')
+
+        if not getattr(aq_base(registry), 'classify', None):
+            # avoid problems when importing a site with an old mimetype registry
+            # XXX return None or orig?
+            return None
+
+        orig_mt = registry.classify(orig,
+                                    mimetype=kwargs.get('mimetype'),
+                                    filename=kwargs.get('filename'))
+        orig_mt = str(orig_mt)
+        if not orig_mt:
+            log('Unable to guess input mime type (filename=%s, mimetype=%s)' %(
+                kwargs.get('mimetype'), kwargs.get('filename')), severity=DEBUG)
+            return None
+
+        target_mt = registry.lookup(target_mimetype)
+        if target_mt:
+            target_mt = target_mt[0]
+        else:
+            log('Unable to match target mime type %s'% str(target_mimetype),
+                severity=DEBUG)
+            return None
+
+        ## fastpath
+        # If orig_mt and target_mt are the same, we only allow
+        # a one-hop transform, a.k.a. filter.
+        # XXX disabled filtering for now
+        filter_only = False
+        if orig_mt == str(target_mt):
+            filter_only = True
+            data.setData(orig)
+            md = data.getMetadata()
+            md['mimetype'] = str(orig_mt)
+            if object is not None:
+                cache.setCache(str(target_mimetype), data)
+            return data
+
+        ## get a path to output mime type
+        requirements = self._policies.get(str(target_mt), [])
+        path = self._findPath(orig_mt, target_mt, list(requirements))
+        if not path and requirements:
+            log('Unable to satisfy requirements %s' % ', '.join(requirements),
+                severity=DEBUG)
+            path = self._findPath(orig_mt, target_mt)
+
+        if not path:
+            log('NO PATH FROM %s TO %s : %s' % (orig_mt, target_mimetype, path),
+                severity=DEBUG)
+            return None #XXX raise TransformError
+
+        if len(path) > 1:
+            ## create a chain on the fly (sly)
+            transform = chain()
+            for t in path:
+                transform.registerTransform(t)
+        else:
+            transform = path[0]
+
+        result = transform.convert(orig, data, context=context, usedby=usedby, **kwargs)
+        assert(idatastream.isImplementedBy(result),
+               'result doesn\'t implemented idatastream')
+        self._setMetaData(result, transform)
+
+        # set cache if possible
+        if object is not None and result.isCacheable():
+            cache.setCache(str(target_mimetype), result)
+
+        # return idatastream object
+        return result
+
+    security.declarePublic('convertToData')
+    def convertToData(self, target_mimetype, orig, data=None, object=None,
+                      usedby=None, context=None, **kwargs):
+        """Convert to a given mimetype and return the raw data
+        ignoring subobjects. see convertTo for more information
+        """
+        data =self.convertTo(target_mimetype, orig, data, object, usedby,
+                       context, **kwargs)
+        if data:
+            return data.getData()
+        return None
+
+    security.declarePublic('convert')
+    def convert(self, name, orig, data=None, context=None, **kwargs):
+        """run a tranform of a given name on data
+
+        * name is the name of a registered transform
+
+        see convertTo docstring for more info
+        """
+        if not data:
+            data = self._wrap(name)
+        try:
+            transform = getattr(self, name)
+        except AttributeError:
+            raise Exception('No such transform "%s"' % name)
+        data = transform.convert(orig, data, context=context, **kwargs)
+        self._setMetaData(data, transform)
+        return data
+
+
+    def __call__(self, name, orig, data=None, context=None, **kwargs):
+        """run a transform by its name, returning the raw data product
+
+        * name is the name of a registered transform.
+
+        return an encoded string.
+        see convert docstring for more info on additional arguments.
+        """
+        data = self.convert(name, orig, data, context, **kwargs)
+        return data.getData()
+
+
+    # utilities ###############################################################
+
+    def _setMetaData(self, datastream, transform):
+        """set metadata on datastream according to the given transform
+        (mime type and optionaly encoding)
+        """
+        md = datastream.getMetadata()
+        if hasattr(transform, 'output_encoding'):
+            md['encoding'] = transform.output_encoding
+        md['mimetype'] = transform.output
+
+    def _wrap(self, name):
+        """wrap a data object in an icache"""
+        return datastream(name)
+
+    def _unwrap(self, data):
+        """unwrap data from an icache"""
+        if idatastream.isImplementedBy(data):
+            data = data.getData()
+        return data
+
+    def _mapTransform(self, transform):
+        """map transform to internal structures"""
+        registry = getToolByName(self, 'mimetypes_registry')
+        inputs = getattr(transform, 'inputs', None)
+        if not inputs:
+            raise TransformException('Bad transform %s : no input MIME type' %
+                                     (transform))
+        for i in inputs:
+            mts = registry.lookup(i)
+            if not mts:
+                msg = 'Input MIME type %r for transform %s is not registered '\
+                      'in the MIME types registry' % (i, transform.name())
+                raise TransformException(msg)
+            for mti in mts:
+                for mt in mti.mimetypes:
+                    mt_in = self._mtmap.setdefault(mt, PersistentMapping())
+                    output = getattr(transform, 'output', None)
+                    if not output:
+                        msg = 'Bad transform %s : no output MIME type'
+                        raise TransformException(msg % transform.name())
+                    mto = registry.lookup(output)
+                    if not mto:
+                        msg = 'Output MIME type %r for transform %s is not '\
+                              'registered in the MIME types registry' % \
+                              (output, transform.name())
+                        raise TransformException(msg)
+                    if len(mto) > 1:
+                        msg = 'Wildcarding not allowed in transform\'s output '\
+                              'MIME type'
+                        raise TransformException(msg)
+
+                    for mt2 in mto[0].mimetypes:
+                        try:
+                            if not transform in mt_in[mt2]:
+                                mt_in[mt2].append(transform)
+                        except KeyError:
+                            mt_in[mt2] = PersistentList([transform])
+
+    def _unmapTransform(self, transform):
+        """unmap transform from internal structures"""
+        registry = getToolByName(self, 'mimetypes_registry')
+        for i in transform.inputs:
+            for mti in registry.lookup(i):
+                for mt in mti.mimetypes:
+                    mt_in = self._mtmap.get(mt, {})
+                    output = transform.output
+                    mto = registry.lookup(output)
+                    for mt2 in mto[0].mimetypes:
+                        l = mt_in[mt2]
+                        for i in range(len(l)):
+                            if transform.name() == l[i].name():
+                                l.pop(i)
+                                break
+                        else:
+                            log('Can\'t find transform %s from %s to %s' % (
+                                transform.name(), mti, mt),
+                                severity=DEBUG)
+
+    def _findPath(self, orig, target, required_transforms=()):
+        """return the shortest path for transformation from orig mimetype to
+        target mimetype
+        """
+        path = []
+
+        if not self._mtmap:
+            return None
+
+        # naive algorithm :
+        #  find all possible paths with required transforms
+        #  take the shortest
+        #
+        # it should be enough since we should not have so much possible paths
+        shortest, winner = 9999, None
+        for path in self._getPaths(str(orig), str(target), required_transforms):
+            if len(path) < shortest:
+                winner = path
+                shortest = len(path)
+
+        return winner
+
+    def _getPaths(self, orig, target, requirements, path=None, result=None):
+        """return a all path for transformation from orig mimetype to
+        target mimetype
+        """
+        if path is None:
+            result = []
+            path = []
+            requirements = list(requirements)
+        outputs = self._mtmap.get(orig)
+        if outputs is None:
+            return result
+        path.append(None)
+        for o_mt, transforms in outputs.items():
+            for transform in transforms:
+                required = 0
+                name = transform.name()
+                if name in requirements:
+                    requirements.remove(name)
+                    required = 1
+                if transform in path:
+                    # avoid infinite loop...
+                    continue
+                path[-1] = transform
+                if o_mt == target:
+                    if not requirements:
+                        result.append(path[:])
+                else:
+                    self._getPaths(o_mt, target, requirements, path, result)
+                if required:
+                    requirements.append(name)
+        path.pop()
+
+        return result
+
+    security.declarePrivate('manage_afterAdd')
+    def manage_afterAdd(self, item, container):
+        """ overload manage_afterAdd to finish initialization when the
+        transform tool is added
+        """
+        Folder.manage_afterAdd(self, item, container)
+        transforms.initialize(self)
+        # XXX required?
+        #try:
+        #    # first initialization
+        #    transforms.initialize(self)
+        #except:
+        #    # may fail on copy
+        #    pass
+
+    security.declareProtected(ManagePortal, 'manage_addTransform')
+    def manage_addTransform(self, id, module, REQUEST=None):
+        """ add a new transform to the tool """
+        transform = Transform(id, module)
+        self._setObject(id, transform)
+        self._mapTransform(transform)
+        if REQUEST is not None:
+            REQUEST['RESPONSE'].redirect(self.absolute_url()+'/manage_main')
+
+    security.declareProtected(ManagePortal, 'manage_addTransform')
+    def manage_addTransformsChain(self, id, description, REQUEST=None):
+        """ add a new transform to the tool """
+        transform = TransformsChain(id, description)
+        self._setObject(id, transform)
+        self._mapTransform(transform)
+        if REQUEST is not None:
+            REQUEST['RESPONSE'].redirect(self.absolute_url()+'/manage_main')
+
+    security.declareProtected(ManagePortal, 'manage_addTransform')
+    def manage_setCacheValidityTime(self, seconds, REQUEST=None):
+        """set  the lifetime of cached data in seconds"""
+        self.max_sec_in_cache = int(seconds)
+        if REQUEST is not None:
+            REQUEST['RESPONSE'].redirect(self.absolute_url()+'/manage_main')
+
+    security.declareProtected(ManagePortal, 'reloadTransforms')
+    def reloadTransforms(self, ids=()):
+        """ reload transforms with the given ids
+        if no ids, reload all registered transforms
+
+        return a list of (transform_id, transform_module) describing reloaded
+        transforms
+        """
+        if not ids:
+            ids = self.objectIds()
+        reloaded = []
+        for id in ids:
+            o = getattr(self, id)
+            o.reload()
+            reloaded.append((id, o.module))
+        return reloaded
+
+    # Policy handling methods #################################################
+
+    def manage_addPolicy(self, output_mimetype, required_transforms, REQUEST=None):
+        """ add a policy for a given output mime types"""
+        registry = getToolByName(self, 'mimetypes_registry')
+        if not registry.lookup(output_mimetype):
+            raise TransformException('Unknown MIME type')
+        if self._policies.has_key(output_mimetype):
+            msg = 'A policy for output %s is yet defined' % output_mimetype
+            raise TransformException(msg)
+
+        required_transforms = tuple(required_transforms)
+        self._policies[output_mimetype] = required_transforms
+        if REQUEST is not None:
+            REQUEST['RESPONSE'].redirect(self.absolute_url()+'/manage_editTransformationPolicyForm')
+
+    def manage_delPolicies(self, outputs, REQUEST=None):
+        """ remove policies for given output mime types"""
+        for mimetype in outputs:
+            del self._policies[mimetype]
+        if REQUEST is not None:
+            REQUEST['RESPONSE'].redirect(self.absolute_url()+'/manage_editTransformationPolicyForm')
+
+    def listPolicies(self):
+        """ return the list of defined policies
+
+        a policy is a 2-uple (output_mime_type, [list of required transforms])
+        """
+        # XXXFIXME: backward compat, should be removed latter
+        if not hasattr(self, '_policies'):
+            self._policies = PersistentMapping()
+        return self._policies.items()
+
+    # mimetype oriented conversions (iengine interface) ########################
+
+    def registerTransform(self, transform):
+        """register a new transform
+
+        transform isn't a Zope Transform (the wrapper) but the wrapped transform
+        the persistence wrapper will be created here
+        """
+        # needed when call from transform.transforms.initialize which
+        # register non zope transform
+        module = str(transform.__module__)
+        transform = Transform(transform.name(), module, transform)
+        if not itransform.isImplementedBy(transform):
+            raise TransformException('%s does not implement itransform' % transform)
+        name = transform.name()
+        __traceback_info__ = (name, transform)
+        if name not in self.objectIds():
+            self._setObject(name, transform)
+            self._mapTransform(transform)
+
+    security.declareProtected(ManagePortal, 'ZopeFind')
+    def ZopeFind(self, *args, **kwargs):
+        """Don't break ZopeFind feature when a transform can't be loaded
+        """
+        try:
+            return Folder.ZopeFind(self, *args, **kwargs)
+        except MissingBinary:
+            log('ZopeFind: catched MissingBinary exception')
+
+    security.declareProtected(View, 'objectItems')
+    def objectItems(self, *args, **kwargs):
+        """Don't break ZopeFind feature when a transform can't be loaded
+        """
+        try:
+            return Folder.objectItems(self, *args, **kwargs)
+        except MissingBinary:
+            log('objectItems: catched MissingBinary exception')
+            return []
+
+InitializeClass(TransformTool)

product/PortalTransforms/TransformTool.py

+from Products.PortalTransforms.TransformEngine import TransformTool

product/PortalTransforms/TransformsChain.py

+"""FIXME: backward compat, remove later
+"""
+from Products.PortalTransforms.chain import Chain as TransformsChain
+

product/PortalTransforms/__init__.py

+import os.path
+__version__ = open(os.path.join(__path__[0], 'version.txt')).read().strip()
+
+from Products.PortalTransforms.utils import skins_dir
+from Products.PortalTransforms.TransformEngine import TransformTool
+
+GLOBALS = globals()
+PKG_NAME = 'PortalTransforms'
+
+tools = (
+    TransformTool,
+    )
+
+# XXX backward compatibility tricks to make old PortalTransform based Mimetypes
+# running (required)
+import sys
+this_module = sys.modules[__name__]
+
+from Products.MimetypesRegistry import mime_types
+setattr(this_module, 'mime_types', mime_types)
+
+from Products.MimetypesRegistry import MimeTypeItem
+setattr(this_module, 'MimeTypeItem', MimeTypeItem)
+
+from Products.MimetypesRegistry import MimeTypeItem
+sys.modules['Products.PortalTransforms.zope.MimeTypeItem'] = MimeTypeItem
+
+def initialize(context):
+    from Products.CMFCore.DirectoryView import registerDirectory
+    #registerDirectory(skins_dir, GLOBALS)
+
+    from Products.CMFCore import utils
+    utils.ToolInit("%s Tool" % PKG_NAME, 
+                   tools=tools,
+                   icon="tool.gif",
+                   ).initialize(context)

product/PortalTransforms/__pkginfo__.py

+from Products import PortalTransforms as PRODUCT
+import os.path
+
+version=PRODUCT.__version__
+modname=PRODUCT.__name__
+
+# (major, minor, patchlevel, release info) where release info is:
+# -99 for alpha, -49 for beta, -19 for rc and 0 for final
+# increment the release info number by one e.g. -98 for alpha2
+
+major, minor, bugfix =  version.split('.')[:3]
+bugfix, release = bugfix.split('-')[:2]
+
+relinfo=-99 #alpha
+if 'beta' in release:
+    relinfo=-49
+if 'rc' in release:
+    relinfo=-19
+if 'final' in release:
+    relinfo=0
+
+numversion = (int(major), int(minor), int(bugfix), relinfo)
+
+license = 'BSD like'
+license_text = open(os.path.join(PRODUCT.__path__[0], 'LICENSE.txt')).read()
+copyright = '''Copyright (c) 2003 LOGILAB S.A. (Paris, FRANCE)'''
+
+author = "Archetypes developement team"
+author_email = "archetypes-devel@lists.sourceforge.net"
+
+short_desc = "MIME types based transformations for the CMF"
+long_desc = """This package provides two new CMF tools in order to
+make MIME types based transformations on the portal contents and so an
+easy to way to plugin some new transformations for previously
+unsupported content types. You will find more info in the package's
+README and docs directory.
+.
+It's part of the Archetypes project, but the only requirement to use it
+is to have a CMF based site. If you are using Archetypes, this package
+replaces the transform package.
+.
+Notice this package can also be used as a standalone Python package. If
+you've downloaded the Python distribution, you can't make it a Zope
+product since Zope files have been removed from this distribution.
+"""
+
+web = "http://plone.org/products/archetypes"
+ftp = ""
+mailing_list = "archetypes-devel@lists.sourceforge.net"
+
+debian_name = "zope-cmftransforms"
+debian_maintainer = "Sylvain Thenault"
+debian_maintainer_email = "sylvain.thenault@logilab.fr"
+debian_handler = "zope"

product/PortalTransforms/bridge.zcml

+<configure
+    xmlns="http://namespaces.zope.org/five"
+    >
+ 
+  <bridge
+      zope2=".interfaces.idatastream"
+      package=".z3.interfaces"
+      name="IDataStream"
+      /> 
+
+  <bridge
+      zope2=".interfaces.itransform"
+      package=".z3.interfaces"
+      name="ITransform"
+      /> 
+
+  <bridge
+      zope2=".interfaces.ichain"
+      package=".z3.interfaces"
+      name="IChain"
+      /> 
+
+  <bridge
+      zope2=".interfaces.iengine"
+      package=".z3.interfaces"
+      name="IEngine"
+      />
+
+</configure>

product/PortalTransforms/cache.py

+"""Cache
+"""
+from time import time
+from Acquisition import aq_base
+
+class Cache:
+
+    def __init__(self, context, _id='_v_transform_cache'):
+        self.context = context
+        self._id =_id
+
+    def _genCacheKey(self, identifier, *args):
+        key = identifier
+        for arg in args:
+            key = '%s_%s' % (key, arg)
+        key = key.replace('/', '_')
+        key = key.replace('+', '_')
+        key = key.replace('-', '_')
+        key = key.replace(' ', '_')
+        return key
+
+    def setCache(self, key, value):
+        """cache a value indexed by key"""
+        if not value.isCacheable():
+            return
+        context = self.context
+        key = self._genCacheKey(key)
+        if getattr(aq_base(context), self._id, None) is None:
+            setattr(context, self._id, {})
+        getattr(context, self._id)[key] = (time(), value)
+        return key
+
+    def getCache(self, key):
+        """try to get a cached value for key
+
+        return None if not present
+        else return a tuple (time spent in cache, value)
+        """
+        context = self.context
+        key = self._genCacheKey(key)
+        dict = getattr(context, self._id, None)
+        if dict is None :
+            return None
+        try:
+            orig_time, value = dict.get(key, None)
+            return time() - orig_time, value
+        except TypeError:
+            return None
+        
+    def purgeCache(self, key=None):
+        """Remove cache
+        """
+        context = self.context
+        id = self._id
+        if not shasattr(context, id):
+            return
+        if key is None:
+            delattr(context, id)
+        else:
+            cache = getattr(context, id)
+            key = self._genCacheKey(key)
+            if cache.has_key(key):
+                del cache[key]

product/PortalTransforms/chain.py

+from Products.PageTemplates.PageTemplateFile import PageTemplateFile
+from Globals import Persistent
+from Globals import InitializeClass
+from Acquisition import Implicit
+from OFS.SimpleItem import Item
+from AccessControl.Role import RoleManager
+from AccessControl import ClassSecurityInfo
+
+from Products.CMFCore.permissions import ManagePortal, ManageProperties
+from Products.CMFCore.utils import getToolByName
+
+from Products.PortalTransforms.utils import TransformException, _www
+from  Products.PortalTransforms.interfaces import ichain
+from  Products.PortalTransforms.interfaces import itransform
+
+from UserList import UserList
+
+class chain(UserList):
+    """A chain of transforms used to transform data"""
+
+    __implements__ = (ichain, itransform)
+
+    def __init__(self, name='',*args):
+        UserList.__init__(self, *args)
+        self.__name__ = name
+        if args:
+            self._update()
+
+    def name(self):
+        return self.__name__
+
+    def registerTransform(self, transform):
+        self.append(transform)
+
+    def unregisterTransform(self, name):
+        for i in range(len(self)):
+            tr = self[i]
+            if tr.name() == name:
+                self.pop(i)
+                break
+        else:
+            raise Exception('No transform named %s registered' % name)
+
+    def convert(self, orig, data, **kwargs):
+        for transform in self:
+            data = transform.convert(orig, data, **kwargs)
+            orig = data.getData()
+        md = data.getMetadata()
+        md['mimetype'] = self.output
+        return data
+
+    def __setitem__(self, key, value):
+        UserList.__setitem__(self, key, value)
+        self._update()
+
+    def append(self, value):
+        UserList.append(self, value)
+        self._update()
+
+    def insert(self, *args):
+        UserList.insert(*args)
+        self._update()
+
+    def remove(self, *args):
+        UserList.remove(*args)
+        self._update()
+
+    def pop(self, *args):
+        UserList.pop(*args)
+        self._update()
+
+    def _update(self):
+        self.inputs = self[0].inputs
+        self.output = self[-1].output
+        for i in range(len(self)):
+            if hasattr(self[-i-1], 'output_encoding'):
+                self.output_encoding = self[-i-1].output_encoding
+                break
+        else:
+            try:
+                del self.output_encoding
+            except:
+                pass
+
+class TransformsChain(Implicit, Item, RoleManager, Persistent):
+    """ a transforms chain is suite of transforms to apply in order.
+    It follows the transform API so that a chain is itself a transform.
+    """
+
+    meta_type = 'TransformsChain'
+
+    meta_types = all_meta_types = ()
+
+    manage_options = (
+                      ({'label':'Configure',
+                       'action':'manage_main'},
+                       {'label':'Reload',
+                       'action':'manage_reloadTransform'},) +
+                      Item.manage_options
+                      )
+
+    manage_main = PageTemplateFile('editTransformsChain', _www)
+    manage_reloadTransform = PageTemplateFile('reloadTransform', _www)
+
+    security = ClassSecurityInfo()
+
+    def __init__(self, id, description, ids=()):
+        self.id = id
+        self.description = description
+        self._object_ids = list(ids)
+        self.inputs = ('application/octet-stream',)
+        self.output = 'application/octet-stream'
+        self._chain = None
+
+    def __setstate__(self, state):
+        """ __setstate__ is called whenever the instance is loaded
+            from the ZODB, like when Zope is restarted.
+
+            We should rebuild the chain at this time
+        """
+        TransformsChain.inheritedAttribute('__setstate__')(self, state)
+        self._chain = None
+
+    def _chain_init(self):
+        """ build the transforms chain """
+        tr_tool = getToolByName(self, 'portal_transforms')
+        self._chain = c = chain()
+        for id in self._object_ids:
+            object = getattr(tr_tool, id)
+            c.registerTransform(object)
+        self.inputs = c.inputs or ('application/octet-stream',)
+        self.output = c.output or 'application/octet-stream'
+
+    security.declarePublic('convert')
+    def convert(self, *args, **kwargs):
+        """ return apply the transform and return the result """
+        if self._chain is None:
+            self._chain_init()
+        return self._chain.convert(*args, **kwargs)
+
+    security.declarePublic('name')
+    def name(self):
+        """return the name of the transform instance"""
+        return self.id
+
+    security.declarePrivate('manage_beforeDelete')
+    def manage_beforeDelete(self, item, container):
+        Item.manage_beforeDelete(self, item, container)
+        if self is item:
+            # unregister self from catalog on deletion
+            tr_tool = getToolByName(self, 'portal_transforms')
+            tr_tool.unregisterTransform(self.id)
+
+    security.declareProtected(ManagePortal, 'manage_addObject')
+    def manage_addObject(self, id, REQUEST=None):
+        """ add a new transform or chain to the chain """
+        assert id not in self._object_ids
+        self._object_ids.append(id)
+        self._chain_init()
+        if REQUEST is not None:
+            REQUEST['RESPONSE'].redirect(self.absolute_url()+'/manage_main')
+
+    security.declareProtected(ManagePortal, 'manage_delObjects')
+    def manage_delObjects(self, ids, REQUEST=None):
+        """ delete the selected mime types """
+        for id in ids:
+            self._object_ids.remove(id)
+        self._chain_init()
+        if REQUEST is not None:
+            REQUEST['RESPONSE'].redirect(self.absolute_url()+'/manage_main')
+
+
+    # transforms order handling #
+
+    security.declareProtected(ManagePortal, 'move_object_to_position')
+    def move_object_to_position(self, id, newpos):
+        """ overriden from OrderedFolder to store id instead of objects
+        """
+        oldpos = self._object_ids.index(id)
+        if (newpos < 0 or newpos == oldpos or newpos >= len(self._object_ids)):
+            return 0
+        self._object_ids.pop(oldpos)
+        self._object_ids.insert(newpos, id)
+        self._chain_init()
+        return 1
+
+    security.declareProtected(ManageProperties, 'move_object_up')
+    def move_object_up(self, id, REQUEST=None):
+        """  move object with the given id up in the list """
+        newpos = self._object_ids.index(id) - 1
+        self.move_object_to_position(id, newpos)
+        if REQUEST is not None:
+            REQUEST['RESPONSE'].redirect(self.absolute_url()+'/manage_main')
+
+    security.declareProtected(ManageProperties, 'move_object_down')
+    def move_object_down(self, id, REQUEST=None):
+        """  move object with the given id down in the list """
+        newpos = self._object_ids.index(id) + 1
+        self.move_object_to_position(id, newpos)
+        if REQUEST is not None:
+            REQUEST['RESPONSE'].redirect(self.absolute_url()+'/manage_main')
+
+
+    # Z transform interface #
+
+    security.declareProtected(ManagePortal, 'reload')
+    def reload(self):
+        """ reload the module where the transformation class is defined """
+        for tr in self.objectValues():
+            tr.reload()
+
+
+    # utilities #
+
+    security.declareProtected(ManagePortal, 'listAddableObjectIds')
+    def listAddableObjectIds(self):
+        """ return a list of addable transform """
+        tr_tool = getToolByName(self, 'portal_transforms')
+        return [id for id in tr_tool.objectIds() if not (id == self.id or id in self._object_ids)]
+
+    security.declareProtected(ManagePortal, 'objectIds')
+    def objectIds(self):
+        """ return a list of addable transform """
+        return tuple(self._object_ids)
+
+    security.declareProtected(ManagePortal, 'objectValues')
+    def objectValues(self):
+        """ return a list of addable transform """
+        tr_tool = getToolByName(self, 'portal_transforms')
+        return [getattr(tr_tool, id) for id in self.objectIds()]
+
+InitializeClass(TransformsChain)

product/PortalTransforms/data.py

+from Products.PortalTransforms.interfaces import idatastream
+
+class datastream:
+    """A transformation datastream packet"""
+    __implements__ = idatastream
+    __slots__ = ('name', '_data', '_metadata')
+
+    def __init__(self, name):
+        self.__name__ = name
+        self._data = ''
+        self._metadata = {}
+        self._objects = {}
+        self._cacheable = 1
+
+    def __str__(self):
+        return self.getData()
+
+    def name(self):
+        return self.__name__
+
+    def setData(self, value):
+        """set the main data produced by a transform, i.e. usually a string"""
+        self._data = value
+
+    def getData(self):
+        """provide access to the transformed data object, i.e. usually a string.
+        This data may references subobjects.
+        """
+        if callable(self._data):
+            data = self._data()
+        else:
+            data = self._data
+        return data
+
+    def setSubObjects(self, objects):
+        """set a dict-like object containing subobjects.
+        keys should be object's identifier (e.g. usually a filename) and
+        values object's content.
+        """
+        self._objects = objects
+
+    def getSubObjects(self):
+        """return a dict-like object with any optional subobjects associated
+        with the data"""
+        return self._objects
+
+    def getMetadata(self):
+        """return a dict-like object with any optional metadata from
+        the transform"""
+        return self._metadata
+    
+    def isCacheable(self):
+        """Return a bool which indicates wether the result should be cached
+        
+        Default is true
+        """
+        return self._cacheable
+    
+    def setCacheable(self, value):
+        """Set cacheable flag to yes or no
+        """
+        self._cacheable = not not value
+
+    #data = property('getData', 'setData', None, """idata.data""")
+
+    #metadata = property('getMetadata', 'setMetadata', None,
+    #"""idata.metadata""")

product/PortalTransforms/docs/dev_manual.rst

+===================================
+Portal Transforms'Developper manual
+===================================
+
+:Author: Sylvain Thenault
+:Contact: syt@logilab.fr
+:Date: $Date: 2005-08-19 23:43:41 +0200 (Fre, 19 Aug 2005) $
+:Version: $Revision: 1.5 $
+:Web site: http://sourceforge.net/projects/archetypes
+
+.. contents::
+
+
+Tools interfaces
+----------------
+
+The MIME types registry
+```````````````````````
+
+    class isourceAdapter(Interface):
+
+	def __call__(data, \**kwargs):
+	    """convert data to unicode, may take optional kwargs to aid in conversion"""
+
+    class imimetypes_registry(isourceAdapter):
+
+	def classify(data, mimetype=None, filename=None):
+	    """return a content type for this data or None
+	    None should rarely be returned as application/octet can be
+	    used to represent most types.
+	    """
+
+	def lookup(mimetypestring):
+	    """Lookup for imimetypes object matching mimetypestring.
+	    mimetypestring may have an empty minor part or containing a wildcard (*).
+	    mimetypestring may be an imimetype object (in this case it will be
+	    returned unchanged, else it should be a RFC-2046 name.
+	    Return a list of mimetypes objects associated with the RFC-2046 name.
+	    Return an empty list if no one is known.
+	    """
+
+	def lookupExtension(filename):
+	    """ return the mimetypes object associated with the file's extension
+	    or None if it is not known.
+	    filename maybe a file name like 'content.txt' or an extension like 'rest'
+	    """
+
+	def mimetypes():
+	    """return all defined mime types, each one implements at least imimetype
+	    """
+
+	def list_mimetypes():
+	    """return all defined mime types, as string"""
+
+
+
+The tranformation tool
+``````````````````````
+
+    class iengine(Interface):
+
+	def registerTransform(transform):
+	    """register a transform
+	    transform must implements itransform
+	    """
+
+	def unregisterTransform(name):
+	    """ unregister a transform
+	    name is the name of a registered transform
+	    """
+
+	def convertTo(mimetype, orig, idata=None, \**kwargs):
+	    """Convert orig to a given mimetype
+	    return an object implementing idatastream or None if not path has been
+	    found
+	    """
+
+	def convert(name, orig, idata=None, \**kwargs):
+	    """run a tranform of a given name on data
+	    name is the name of a registered transform
+	    return an object implementing idatastream
+	    """
+
+	def __call__(name, orig, idata=None, \**kwargs):
+	    """run a transform returning the raw data product
+	    name is the name of a registered transform
+	    return an object implementing idatastream
+	    """
+
+
+
+Writing a new transformation
+----------------------------
+
+Writing a new transform should be an easy task. You only have to follow a
+simple interface to do it, but knowing some advanced features and provided
+utilities may help to do it quicker... 
+
+
+Related interfaces
+``````````````````
+
+    class itransform(Interface):
+	"""A transformation plugin -- tranform data somehow must be threadsafe and stateless"""
+
+        inputs = Attribute("""list of imimetypes (or registered rfc-2046
+                              names) this transform accepts as inputs""")
+
+        output = Attribute("output imimetype as instance or rfc-2046 name"")
+
+	def name(self):
+	    """return the name of the transform instance"""
+
+	def convert(data, idata, \**kwargs):
+	    """convert the data, store the result in idata and return that"""
+
+    class idatastream(Interface):
+	"""data stream, is the result of a transform"""
+
+	def setData(self, value):
+	    """set the main data produced by a transform, i.e. usually a string"""
+
+	def getData():
+	    """provide access to the transformed data object, i.e. usually a string.
+	    This data may references subobjects.
+	    """
+
+	def setSubObjects(self, objects):
+	    """set a dict-like object containing subobjects.
+            keys should be object's identifier (e.g. usually a filename) and
+	    values object's content.
+	    """
+
+	def getSubObjects(self):
+	    """return a dict-like object with any optional subobjects associated
+	    with the data"""
+
+	def getMetadata():
+	    """return a dict-like object with any optional metadata from
+	    the transform"""
+
+
+Important note about encoding
+`````````````````````````````
+
+A transform receive data as an encoded string. A priori, no assumption can be
+made about the used encoding. Data returned by a transform must use the same
+encoding as received data, unless the transform provides a *output_encoding*
+attribute indicating the output encoding (for instance this may be usefull for
+XSLT based transforms).
+
+
+Configurable transformation
+```````````````````````````
+
+You can make your transformation configurable through the ZMI by setting a
+*config* dictionnary on your transform instance or class. Keys are parameter's
+name and values parameter's value. Another dictionnary *config_metadata*
+describes each parameter. In this mapping, keys are also parameter's name but
+values are a tree-uple : (<parameter's type>, <parameter's label>, <parameter's
+description>).
+
+Possible types for parameters are :
+
+  :int: field is an integer
+
+  :string: field is a string
+
+  :list: field is a list
+
+  :dict: field is a dictionnary
+
+You can look at the **command** and **xml** transforms for an example of
+configurable transform.
+
+
+Images / sub-objects  management
+````````````````````````````````
+
+A transformation may produce some sub-objects, for instance when you convert a
+PDF document to HTML. That's the purpose of the setObjects method of
+the idatastream interface.
+
+
+Some utilities
+``````````````
+
+Transform utilities may be found in the libtransforms subpackage. You'll find
+there the following modules :
+
+*commandtransform*
+  provides a base class for external command based transforms. 
+
+*retransform*
+  provides a base class for regular expression based transforms. 
+
+*html4zope*
+  provides a docutils HTML writer for Zope.
+
+*utils*
+  provides some utilities functions. 
+
+
+Write a test your transform !
+`````````````````````````````
+
+Every transform should have its test... And it's easy to write a test for your
+transform ! Imagine you have made a transform named "colabeer" which transforms
+cola into beer (I let you find MIME type for these content types ;). Basically,
+your test file should be :
+
+    from test_transforms import make_tests
+    
+    tests =('Products.MyTransforms.colabeer', "drink.cola", "drink.beer", None, 0)
+
+    def test_suite():
+        return TestSuite([makeSuite(test) for test in make_tests()])
+
+    if __name__=='__main__':
+        main(defaultTest='test_suite')
+
+
+
+In this example :
+
+- "Products.MyTransforms.colabeer" is the module defining your transform (you
+  can also give directly the transform instance).  
+
+- "drink.cola" is the name of the file containing data to give to your transform
+  as input. 
+
+- "drink.beer" is the file containing expected transform result (what the getData
+  method of idatastream will return).
+
+- Additional arguments (*None* and *0*) are respectivly an optional normalizing
+  function to apply to both the transform result and the output file content, and
+  the number of subobjects that the transform is expected to produce.
+
+This example supposes your test is in the *tests* directory of PortalTransforms
+and your input and output files respectively in *tests/input* and
+*tests/output*.
\ No newline at end of file

product/PortalTransforms/docs/makefile

+REST2HTML=html.py --compact-lists --date --generator
+
+all: user_manual.html dev_manual.html
+
+user_manual.html: user_manual.rst
+	$(REST2HTML) user_manual.rst user_manual.html
+
+dev_manual.html: dev_manual.rst
+	$(REST2HTML) dev_manual.rst dev_manual.html
+
+clean:
+	rm *.html

product/PortalTransforms/docs/uno_setup.rst

+===================================
+How to setup PyUNO for zope
+===================================
+
+:Author: Junyong Pan <panjy at zopechina.com>, Anton Stonor <stonor@giraffen.dk>
+:Date: $Date: 2003-08-12 02:50:50 -0800 (Tue, 12 Aug 2003) $
+:Version: $Revision: 1.5 $
+
+(to be refined)
+
+Portal Transforms allows you to convert Word documents to HTML. A cool
+feature
+if you want to preview Word documents at your web site or use Word as a web
+authoring tool.
+
+To do the actual transform, Portal Transforms rely on a third party
+application
+to do the heavy lifting. If you have not installed such an application,
+Portal
+Transforms will not perfom Word to HTML transforms.
+
+One of the options is Open Office. It is not the easiest application to
+set up
+to work with Portal Transforms, but it works on both Windows and Unix
+and delivers
+fairly good HTML.
+
+Problems
+====================
+- PyUNO is cool, but PyUNO now ship with its own python intepreter, which is not compatible with zope's
+- PyUNO is not threadsafe now.
+
+SETTING UP OPEN OFFICE ON WINDOWS
+=======================================
+
+WARNING: You can setup pyuno, but you can't use it concurrently. see 'Install oood'
+
+1) Install Open Office 2.0
+
+   Just run the standard installer.
+
+   Pyuno in this version ship with python 2.3, which is compatible with Zope 2.7
+
+2) Set the environment PATH
+Add the Open Office program dir to the Windows PATH, e.g.
+C:\Program Files\OpenOffice.org 1.9.82\program
+
+See this article on how to set the Windows PATH:
+http://vlaurie.com/computers2/Articles/environment.htm
+
+You can also look at the python.bat (located in your Open Office program
+dir)
+for inspiration.
+
+3) Set the PYTHONPATH
+You need to add these directories to the PYTHONPATH:
+a) The Open Office program dir (e.g. C:\Program Files\OpenOffice.org
+1.9.82\program)
+b) The Open Office python lib dir (e.g. C:\Program Files\OpenOffice.org
+1.9.82\program\python-core-2.3.4\lib)
+
+From the Windows system shell, just run e.g.:
+set PYTHONPATH= C:\Program Files\OpenOffice.org 1.9.82\program
+set PYTHONPATH=  C:\Program Files\OpenOffice.org
+1.9.82\program\python-core-2.3.4\lib
+
+You can also look at the python.bat (located in your Open Office program
+dir) for inspiration.
+
+4) Start Open Office as UNO-server
+Run soffice "-accept=socket,host=localhost,port=2002;urp;"
+
+5) Now it should work
+
+For Debian Linux Users
+=========================
+
+see: http://bibus-biblio.sourceforge.net/html/en/LinuxInstall.html
+
+1. install version 1.1, which doesn't contain pyuno::
+
+    apt-get install openoffice
+
+2. install a version of pyuno which enable ucs4 unicode
+
+   - you can download at http://sourceforge.net/projects/bibus-biblio/
+
+   - copy to /usr/lib/openoffice/program
+
+3. set up environment variables
+
+    OPENOFFICE_PATH="/usr/lib/openoffice/program"
+    export PYTHONPATH="$OPENOFFICE_PATH"
+    export LD_LIBRARY_PATH="$OPENOFFICE_PATH"
+
+Install oood
+===================
+
+Note, this product is for linux only
+
+http://udk.openoffice.org/python/oood/
+
+UNDERSTANDING OPEN OFFICE AND UNO
+=============================================
+Open Office allows programmers to remotely control it. Portal Transforms
+takes
+advantage of this opportunity by scripting Open Office from Python. It
+is possible
+through PyUNO that exposes the Open Office API in Python.
+
+Now, you can't download and install PyUNO as a module for your the Python
+interpreter that is running your Zope server. PyUNO only comes bundled
+with Open
+Office and the Python that is distributed with Open Office. To make
+PyUNO work
+from within your standard Python you must expand the PYTHONPATH as done
+above so
+Python also will look inside Open Office for modules. If it works you
+should be
+able to start up a Python shell and do
+
+>>>import uno
+
+In some cases you can be unlucky and the Python used for Zope is not in
+sync with
+the Python that is distributed with Open Office. That is solved by
+rebuilding
+Python -- a task that is beyond the scope of this guide.
+

product/PortalTransforms/docs/user_manual.rst

+=============================
+Portal Transforms'User manual
+=============================
+
+:Author: Sylvain Thénault
+:Contact: syt@logilab.fr
+:Date: $Date: 2005-08-19 23:43:41 +0200 (Fre, 19 Aug 2005) $
+:Version: $Revision: 1.7 $
+:Web site: http://sourceforge.net/projects/archetypes
+
+.. contents::
+
+
+What does this package provide ?
+================================
+
+This package is both a python library for MIME type based content
+transformation, including a command line tool, (what i call the python package)
+and a Zope product providing two new tools for the CMF (what i call the Zope
+product). A python only distribution will be available, where all the Zope
+specific files won't be included.
+
+
+
+Python side
+===========
+
+The *transform* command line tool
+`````````````````````````````````
+
+    command line tool for MIME type based transformation
+
+    USAGE: transform [OPTIONS] input_file output_file
+
+    OPTIONS:
+     -h / --help
+       display this help message and exit.
+
+     -o / --output <output mime type>
+       output MIME type. (conflict with --transform)
+
+     -t / --transform <transform id>
+       id of the transform to apply. (conflict with --output)
+
+    EXAMPLE:
+     $ transform -o text/html dev_manual.rst dev_manual.html
+
+
+Customization hook
+``````````````````
+
+You can customize the transformation engine by providing a module
+"transform_customize" somewhere in your Python path. The module must provide a
+*initialize* method which will take the engine as only argument. This method
+will have the reponsability to initialize the engine with desired
+transformations. When it's not found, the *initialize* method from the
+*transforms* subpackage will be used.
+
+
+
+Zope side
+=========
+
+
+The MIME types registry
+```````````````````````
+
+This tool registered known MIME types. The information associated with a MIME
+type are :
+
+  * a title
+
+  * a list rfc-2046 types
+
+  * a list of files extensions
+
+  * a binary flag
+
+  * an icon path
+
+You can see regitered types by going to the *mimetypes_registry* object at the
+root of your CMF site, using the ZMI. There you can modify existent information
+or add / delete types. This product cames with a default set of MIME types icons
+located in portal_skins/mimetypes_icons.
+
+
+The tranformation tool
+``````````````````````
+
+It's a MIME type based transformation engine. It's has been designed to
+transform portal content from a given MIME type to another. You can add / delete
+transformations by going to the *portal_transforms* object at the root of your
+CMF site, using the ZMI. Some transformations are configurable, but not all. A
+transform is a Python object implementing a special interface. See the
+developper documentation if you're interested in writing a new
+transformation.
+
+
+Archetypes integration
+``````````````````````
+
+Archetypes will use this product for automatic transformation if you have
+configurated it to use the new base unit (set USE_NEW_BASEUNIT to 1 in the
+Archetypes config.py). If you're using the old base unit (still default in 1.0),
+the transform tool won't be used (at least by the Archetypes library).
+
+
+Default transformations
+=======================
+
+The default transformations are described here. They are separated in two groups,
+safe and unsafe. Safe transforms are located in the *transforms* directory of this
+product. Unsafe transforms are located in the *unsafe_transforms* directory and
+are not registered by default. Moreover, there is no __init__.py file in this
+directory so it requires a manual intervention to make them addable to the
+transforms tool. Usually unsafe transforms are so called since they allow
+configuration of a path to a binary executable on the server, which may be
+indesirable for Zope service providers. 
+
+
+Safe transforms
+```````````````
+
+*st* 
+  transforms Structured Text to HTML. Not configurable.
+
+*rest*
+  transforms Re Structured Text to HTML. You need docutils to use this
+  transformation. Not configurable.
+
+*word_to_html*
+  transforms M$ Word file to HTML, using either COM (on windows), wvWare or
+  PyUNO (from OpenOffice.org). Not configurable.
+
+*pdf_to_html*
+  transforms Adobe PDF to HTML. This transforms requires the "pdftohtml"
+  program. Not Configurable.
+
+*lynx_dump*
+  transforms HTML to plain text. This transforms requires the "lynx"
+  program. Not Configurable.
+
+*python*
+  transforms Python source code to colorized HTML. You can configure used
+  colors.
+
+*text_to_html*
+  transforms plain text file to HTML by replacing new lines with
+  <br/>. You can configure allowable inputs for this transform.
+
+*rest_to_text*
+  This is an example use of the *identity* transform, which does
+  basically nothing :). It's used here to transform ReST files
+  (text/x-rst) to text/plain. You can configure allowable inputs and
+  outuput on this transform.
+
+
+Unsafe transforms
+`````````````````
+
+*command*
+  this is a fully configurable transform based on external commands. For
+  instance, you can obtain the same transformation as the previous
+  *lynx_dump*:
+  
+  1. add a new transform named "lynx_dump" with
+     "Products.PortalTransforms.unsafe_transforms.command" as module
+     (this supposes that you've added a __init__.py file to the
+     unsafe_transforms directory to make them importable). 
+  2. go to the configure tab of this transform and set the following
+     parameters :
+
+     :binary_path:  '/usr/bin/lynx' 
+
+     :command_line: '-dump %s'
+
+     :input:        'text/html'
+
+     :output:       'text/plain'
+
+
+*xml*
+  this transform has been designed to handle XML file on a doctype / DTD
+  basis. All the real transformation work is done by a xslt processor. This
+  transform only associate XSLT to doctypes or DTD, and use give the correct
+  transformation to the processor when some content has to be
+  transformed.
+
+  FIXME: add an example on how to setup docbook transform.
+
+
+
+Advanced features
+=================
+
+Transformation chains
+`````````````````````
+
+A transformation chain is an ordered suite of transformations. A chain
+itselve is a transformation. You can build a transformations chain
+using the ZMI.
+
+
+Transformation policy
+`````````````````````
+
+You can set a simple transformation policies for the transforms
+tool. A policy say that when you try to convert content to a given
+MIME type, you have to include a given transformation. For instance,
+imagine you have a *html_tidy* tranformation which tidy HTML page, you
+can say that the transformation path to text/html should include the
+*html_tidy* transform.
+
+
+Caches
+``````
+
+For efficiency, transformation's result are cached. You can set the
+life time of a cached result using the ZMI. This is a time exprimed in
+seconds.
+

product/PortalTransforms/implements.zcml

+<configure
+    xmlns="http://namespaces.zope.org/five"
+    >
+
+  <implements
+      class=".chain.chain"
+      interface=".z3.interfaces.IChain"
+      />   
+
+  <implements
+      class=".chain.chain"
+      interface=".z3.interfaces.ITransform"
+      />   
+
+  <implements
+      class=".data.datastream"
+      interface=".z3.interfaces.IDataStream"
+      />   
+
+  <implements
+      class=".Transform.Transform"
+      interface=".z3.interfaces.ITransform"
+      />   
+
+  <implements
+      class=".TransformEngine.TransformTool"
+      interface=".z3.interfaces.IEngine"
+      />   
+
+  <implements
+      class=".libtransforms.commandtransform.commandtransform"
+      interface=".z3.interfaces.ITransform"
+      />   
+
+  <implements
+      class=".libtransforms.commandtransform.popentransform"
+      interface=".z3.interfaces.ITransform"
+      />  
+
+  <implements
+      class=".libtransforms.retransform.retransform"
+      interface=".z3.interfaces.ITransform"
+      />   
+
+  <!-- TODO: more -->
+
+</configure>

product/PortalTransforms/interfaces.py

+from Interface import Interface, Attribute
+
+class idatastream(Interface):
+    """data stream, is the result of a transform"""
+
+    def setData(value):
+        """set the main data produced by a transform, i.e. usually a string"""
+
+    def getData():
+        """provide access to the transformed data object, i.e. usually a string.
+        This data may references subobjects.
+        """
+
+    def setSubObjects(objects):
+        """set a dict-like object containing subobjects.
+        keys should be object's identifier (e.g. usually a filename) and
+        values object's content.
+        """
+
+    def getSubObjects():
+        """return a dict-like object with any optional subobjects associated
+        with the data"""
+
+    def getMetadata():
+        """return a dict-like object with any optional metadata from
+        the transform
+        You can modify the returned dictionnary to add/change metadata
+        """
+        
+    def isCacheable():
+        """Return a bool which indicates wether the result should be cached
+        
+        Default is true
+        """
+        
+    def setCachable(value):
+        """Set cacheable flag to yes or no
+        """
+
+class itransform(Interface):
+    """A transformation plugin -- tranform data somehow
+    must be threadsafe and stateless"""
+
+#     inputs = Attribute("""list of imimetypes (or registered rfc-2046
+#     names) this transform accepts as inputs.""")
+
+#     output = Attribute("""output imimetype as instance or rfc-2046
+#     name""")
+
+#     output_encoding = Attribute("""output encoding of this transform.
+#     If not specified, the transform should output the same encoding as received data
+#     """)
+
+    def name(self):
+        """return the name of the transform instance"""
+
+    def convert(data, idata, filename=None, **kwargs):
+        """convert the data, store the result in idata and return that
+
+        optional argument filename may give the original file name of received data
+
+        additional arguments given to engine's convert, convertTo or __call__ are
+        passed back to the transform
+        
+        The object on which the translation was invoked is available as context
+        (default: None)
+        """
+
+
+class ichain(itransform):
+
+    def registerTransform(transform, condition=None):
+        """Append a transform to the chain"""
+
+
+class iengine(Interface):
+
+    def registerTransform(transform):
+        """register a transform
+
+        transform must implements itransform
+        """
+
+    def unregisterTransform(name):
+        """ unregister a transform
+        name is the name of a registered transform
+        """
+
+    def convertTo(mimetype, orig, data=None, object=None, context=None, **kwargs):
+        """Convert orig to a given mimetype
+
+        * orig is an encoded string
+
+        * data an optional idatastream object. If None a new datastream will be
+        created and returned
+
+        * optional object argument is the object on which is bound the data.
+        If present that object will be used by the engine to bound cached data.
+        
+        * optional context argument is the object on which the transformation
+          was called.
+
+        * additional arguments (kwargs) will be passed to the transformations.
+
+        return an object implementing idatastream or None if no path has been
+        found.
+        """
+
+    def convert(name, orig, data=None, context=None, **kwargs):
+        """run a tranform of a given name on data
+
+        * name is the name of a registered transform
+
+        see convertTo docstring for more info
+        """
+
+    def __call__(name, orig, data=None, context=None, **kwargs):
+        """run a transform by its name, returning the raw data product
+
+        * name is the name of a registered transform.
+
+        return an encoded string.
+        see convert docstring for more info on additional arguments.
+        """
+

product/PortalTransforms/libtransforms/__init__.py

+""" package containing some utilities which aim to facilitae transformation writing
+"""

product/PortalTransforms/libtransforms/commandtransform.py

+import os
+import sys
+import tempfile
+import re
+import shutil
+from os.path import join, basename
+
+from Products.PortalTransforms.interfaces import itransform
+from Products.PortalTransforms.libtransforms.utils import bin_search, sansext, getShortPathName
+
+class commandtransform:
+    """abstract class for external command based transform
+    """
+    __implements__ = itransform
+
+    def __init__(self, name=None, binary=None, **kwargs):
+        if name is not None:
+            self.__name__ = name
+        if binary is not None:
+            self.binary = bin_search(binary)
+            self.binary = getShortPathName(self.binary)
+
+    def name(self):
+        return self.__name__
+
+    def initialize_tmpdir(self, data, **kwargs):
+        """create a temporary directory, copy input in a file there
+        return the path of the tmp dir and of the input file
+        """
+        tmpdir = tempfile.mktemp()
+        os.mkdir(tmpdir)
+        filename = kwargs.get("filename", '')
+        fullname = join(tmpdir, basename(filename))
+        filedest = open(fullname , "wb").write(data)
+        return tmpdir, fullname
+
+    def subObjects(self, tmpdir):
+        imgs = []
+        for f in os.listdir(tmpdir):
+            result = re.match("^.+\.(?P<ext>.+)$", f)
+            if result is not None:
+                ext = result.group('ext')
+                if ext in ('png', 'jpg', 'gif'):
+                    imgs.append(f)
+        path = join(tmpdir, '')
+        return path, imgs
+
+    def fixImages(self, path, images, objects):
+        for image in images:
+            objects[image] = open(join(path, image), 'rb').read()
+
+    def cleanDir(self, tmpdir):
+        shutil.rmtree(tmpdir)
+
+class popentransform:
+    """abstract class for external command based transform
+
+    Command must read from stdin and write to stdout
+    """
+    __implements__ = itransform
+
+    binaryName = ""
+    binaryArgs = ""
+    useStdin = True
+
+    def __init__(self, name=None, binary=None, binaryArgs=None, useStdin=None,
+                 **kwargs):
+        if name is not None:
+            self.__name__ = name
+        if binary is not None:
+            self.binary = bin_search(binary)
+        else:
+            self.binary = bin_search(self.binaryName)
+        if binaryArgs is not None:
+            self.binaryArgs = binaryArgs
+        if useStdin is not None:
+            self.useStdin = useStdin
+
+    def name(self):
+        return self.__name__
+
+    def getData(self, couterr):
+        return couterr.read()
+
+    def convert(self, data, cache, **kwargs):
+        command = "%s %s" % (self.binary, self.binaryArgs)
+        if not self.useStdin:
+            tmpfile, tmpname = tempfile.mkstemp(text=False) # create tmp
+            os.write(tmpfile, data) # write data to tmp using a file descriptor
+            os.close(tmpfile)       # close it so the other process can read it
+            command = command % { 'infile' : tmpname } # apply tmp name to command
+
+        cin, couterr = os.popen4(command, 'b')
+
+        if self.useStdin:
+            cin.write(str(data))
+
+        status = cin.close()
+
+        out = self.getData(couterr)
+        couterr.close()
+
+        if not self.useStdin:
+            # remove tmp file
+            os.unlink(tmpname)
+
+        cache.setData(out)
+        return cache

product/PortalTransforms/libtransforms/piltransform.py

+from Products.PortalTransforms.interfaces import itransform
+from StringIO import StringIO
+import PIL.Image
+
+class PILTransforms:
+    __implements__ = itransform
+    __name__  = "piltransforms"
+    def __init__(self, name=None):
+         if name is not None:
+            self.__name__ = name
+
+    def name(self):
+        return self.__name__
+
+    def convert(self, orig, data, **kwargs):
+        imgio = StringIO()
+        orig = StringIO(orig)
+        newwidth = kwargs.get('width',None)
+        newheight = kwargs.get('height',None)
+        pil_img = PIL.Image.open(orig)
+        if(self.format in ['jpeg','ppm']):
+            pil_img.draft("RGB", pil_img.size)
+            pil_img = pil_img.convert("RGB")
+        if(newwidth or newheight):
+            pil_img.thumbnail((newwidth,newheight),PIL.Image.ANTIALIAS)
+        pil_img.save(imgio,self.format)
+        data.setData(imgio.getvalue())
+        return data
+
+
+
+def register():
+    return PILTransforms()

product/PortalTransforms/libtransforms/retransform.py

+from Products.PortalTransforms.interfaces import itransform
+import re
+
+class retransform:
+    """abstract class for regex transforms (re.sub wrapper)"""
+    __implements__ = itransform
+
+    inputs  = ('text/',)
+
+    def __init__(self, name, *args):
+        self.__name__ = name
+        self.regexes = []
+        for pat, repl in args:
+            self.addRegex(pat, repl)
+
+    def name(self):
+        return self.__name__
+
+    def addRegex(self, pat, repl):
+        r = re.compile(pat)
+        self.regexes.append((r, repl))
+
+    def convert(self, orig, data, **kwargs):
+        for r, repl in self.regexes:
+            orig = r.sub(repl, orig)
+        data.setData(orig)
+        return data

product/PortalTransforms/libtransforms/utils.py

+import re
+import os
+import sys
+from sgmllib import SGMLParser
+
+try:
+    import win32api
+    WIN32 = True
+except ImportError:
+    WIN32 = False
+
+class MissingBinary(Exception): pass
+
+envPath = os.environ['PATH']
+bin_search_path = [path for path in envPath.split(os.pathsep)
+                   if os.path.isdir(path)]
+
+cygwin = 'c:/cygwin'
+
+# cygwin support
+if sys.platform == 'win32' and os.path.isdir(cygwin):
+    for p in ['/bin', '/usr/bin', '/usr/local/bin' ]:
+        p = os.path.join(cygwin, p)
+        if os.path.isdir(p):
+            bin_search_path.append(p)
+
+if sys.platform == 'win32':
+    extensions = ('.exe', '.com', '.bat', )
+else:
+    extensions = ()
+
+def bin_search(binary):
+    """search the bin_search_path for a given binary returning its fullname or
+       raises MissingBinary"""
+    result = None
+    mode   = os.R_OK | os.X_OK
+    for path in bin_search_path:
+        for ext in ('', ) + extensions:
+            pathbin = os.path.join(path, binary) + ext
+            if os.access(pathbin, mode) == 1:
+                result = pathbin
+                break
+    if not result:
+        raise MissingBinary('Unable to find binary "%s" in %s' % 
+                             (binary, os.pathsep.join(bin_search_path)))
+    else:
+        return result
+
+def getShortPathName(binary):
+    if WIN32:
+        try:
+            binary = win32api.GetShortPathName(binary)
+        except win32api.error:
+            log("Failed to GetShortPathName for '%s'" % binary)
+    return binary
+
+def sansext(path):
+    return os.path.splitext(os.path.basename(path))[0]
+
+
+##########################################################################
+# The code below is taken from CMFDefault.utils to remove
+# dependencies for Python-only installations
+##########################################################################
+
+def bodyfinder(text):
+    """ Return body or unchanged text if no body tags found.
+
+    Always use html_headcheck() first.
+    """
+    lowertext = text.lower()
+    bodystart = lowertext.find('<body')
+    if bodystart == -1:
+        return text
+    bodystart = lowertext.find('>', bodystart) + 1
+    if bodystart == 0:
+        return text
+    bodyend = lowertext.rfind('</body>', bodystart)
+    if bodyend == -1:
+        return text
+    return text[bodystart:bodyend]
+
+
+#
+#   HTML cleaning code
+#
+
+# These are the HTML tags that we will leave intact
+VALID_TAGS = { 'a'          : 1
+             , 'b'          : 1
+             , 'base'       : 0
+             , 'blockquote' : 1
+             , 'body'       : 1
+             , 'br'         : 0
+             , 'caption'    : 1
+             , 'cite'       : 1
+             , 'code'       : 1
+             , 'div'        : 1
+             , 'dl'         : 1
+             , 'dt'         : 1
+             , 'dd'         : 1
+             , 'em'         : 1
+             , 'h1'         : 1
+             , 'h2'         : 1
+             , 'h3'         : 1
+             , 'h4'         : 1
+             , 'h5'         : 1
+             , 'h6'         : 1
+             , 'head'       : 1
+             , 'hr'         : 0
+             , 'html'       : 1
+             , 'i'          : 1
+             , 'img'        : 0
+             , 'kbd'        : 1
+             , 'li'         : 1
+             , 'meta'       : 0
+             , 'ol'         : 1
+             , 'p'          : 1
+             , 'pre'        : 1
+             , 'span'       : 1
+             , 'strong'     : 1
+             , 'table'      : 1
+             , 'tbody'      : 1
+             , 'td'         : 1
+             , 'th'         : 1
+             , 'title'      : 1
+             , 'tr'         : 1
+             , 'tt'         : 1
+             , 'u'          : 1
+             , 'ul'         : 1
+             }
+
+NASTY_TAGS = { 'script'     : 1
+             , 'object'     : 1
+             , 'embed'      : 1
+             , 'applet'     : 1
+             }
+
+class IllegalHTML( ValueError ):
+    pass
+
+class StrippingParser( SGMLParser ):
+    """ Pass only allowed tags;  raise exception for known-bad.  """
+
+    from htmlentitydefs import entitydefs # replace entitydefs from sgmllib
+
+    def __init__( self ):
+
+        SGMLParser.__init__( self )
+        self.result = ""
+
+    def handle_data( self, data ):
+
+        if data:
+            self.result = self.result + data
+
+    def handle_charref( self, name ):
+
+        self.result = "%s&#%s;" % ( self.result, name )
+
+    def handle_entityref(self, name):
+
+        if self.entitydefs.has_key(name):
+            x = ';'
+        else:
+            # this breaks unstandard entities that end with ';'
+            x = ''
+
+        self.result = "%s&%s%s" % (self.result, name, x)
+
+    def unknown_starttag(self, tag, attrs):
+
+        """ Delete all tags except for legal ones.
+        """
+        if VALID_TAGS.has_key(tag):
+
+            self.result = self.result + '<' + tag
+
+            for k, v in attrs:
+
+                if k.lower().startswith( 'on' ):
+                    raise IllegalHTML, 'Javascipt event "%s" not allowed.' % k
+
+                if v.lower().startswith( 'javascript:' ):
+                    raise IllegalHTML, 'Javascipt URI "%s" not allowed.' % v
+
+                self.result = '%s %s="%s"' % (self.result, k, v)
+
+            endTag = '</%s>' % tag
+            if VALID_TAGS.get(tag):
+                self.result = self.result + '>'
+            else:
+                self.result = self.result + ' />'
+
+        elif NASTY_TAGS.get( tag ):
+            raise IllegalHTML, 'Dynamic tag "%s" not allowed.' % tag
+
+        else:
+            pass    # omit tag
+
+    def unknown_endtag(self, tag):
+
+        if VALID_TAGS.get( tag ):
+
+            self.result = "%s</%s>" % (self.result, tag)
+            remTag = '</%s>' % tag
+
+
+def scrubHTML( html ):
+    """ Strip illegal HTML tags from string text.  """
+    parser = StrippingParser()
+    parser.feed( html )
+    parser.close()
+    return parser.result

product/PortalTransforms/libtransforms/zope27rest.py

+##############################################################################
+#
+# Copyright (c) 2001 Zope Corporation and Contributors. All Rights Reserved.
+#
+# This software is subject to the provisions of the Zope Public License,
+# Version 2.0 (ZPL).  A copy of the ZPL should accompany this distribution.
+# THIS SOFTWARE IS PROVIDED "AS IS" AND ANY AND ALL EXPRESS OR IMPLIED
+# WARRANTIES ARE DISCLAIMED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+# WARRANTIES OF TITLE, MERCHANTABILITY, AGAINST INFRINGEMENT, AND FITNESS
+# FOR A PARTICULAR PURPOSE
+#
+##############################################################################
+"""Wrapper to integrate reStructuredText into Zope
+
+This implementation requires docutils 0.3.4+ from http://docutils.sf.net/
+
+Based on the new implementation of Zope 2.7.1 altered for PortalTransforms
+
+"""
+try:
+    import docutils
+except ImportError:
+    raise ImportError, 'Please install docutils 0.3.3+ from http://docutils.sourceforge.net/#download.'
+
+version = docutils.__version__.split('.')
+if version < ['0', '3', '3']:
+    raise ImportError, """Old version of docutils found:
+Got: %(version)s, required: 0.3.3+
+Please remove docutils from %(path)s and replace it with a new version. You
+can download docutils at http://docutils.sourceforge.net/#download.
+""" % {'version' : docutils.__version__, 'path' : docutils.__path__[0] }
+
+
+import sys, os, locale
+##from App.config import getConfiguration
+from docutils.core import publish_parts
+
+# get encoding
+##default_enc = sys.getdefaultencoding()
+##default_output_encoding = getConfiguration().rest_output_encoding or default_enc
+##default_input_encoding = getConfiguration().rest_input_encoding or default_enc
+default_enc = 'utf-8'
+default_output_encoding = default_enc
+default_input_encoding = default_enc
+
+# starting level for <H> elements (default behaviour inside Zope is <H3>)
+default_level = 3
+##initial_header_level = getConfiguration().rest_header_level or default_level
+initial_header_level = default_level
+
+# default language
+##default_lang = getConfiguration().locale or locale.getdefaultlocale()[0]
+default_lang = locale.getdefaultlocale()[0]
+if default_lang and '_' in default_lang:
+    default_lang = default_lang[:default_lang.index('_')]
+
+
+class Warnings:
+
+    def __init__(self):
+        self.messages = []
+
+    def write(self, message):
+        self.messages.append(message)
+
+def render(src,
+           writer='html4css1',
+           report_level=1,
+           stylesheet='default.css',
+           input_encoding=default_input_encoding,
+           output_encoding=default_output_encoding,
+           language_code=default_lang,
+           initial_header_level = initial_header_level,
+           settings = {}):
+    """get the rendered parts of the document the and warning object
+    """
+    # Docutils settings:
+    settings = settings.copy()
+    settings['input_encoding'] = input_encoding
+    settings['output_encoding'] = output_encoding
+    settings['stylesheet'] = stylesheet
+    settings['language_code'] = language_code
+    # starting level for <H> elements:
+    settings['initial_header_level'] = initial_header_level + 1
+    # set the reporting level to something sane:
+    settings['report_level'] = report_level
+    # don't break if we get errors:
+    settings['halt_level'] = 6
+    # remember warnings:
+    settings['warning_stream'] = warning_stream = Warnings()
+
+    parts = publish_parts(source=src, writer_name=writer,
+                          settings_overrides=settings,
+                          config_section='zope application')
+
+    return parts, warning_stream
+
+def HTML(src,
+         writer='html4css1',
+         report_level=1,
+         stylesheet='default.css',
+         input_encoding=default_input_encoding,
+         output_encoding=default_output_encoding,
+         language_code=default_lang,
+         initial_header_level = initial_header_level,
+         warnings = None,
+         settings = {}):
+    """ render HTML from a reStructuredText string 
+
+        - 'src'  -- string containing a valid reST document
+
+        - 'writer' -- docutils writer 
+
+        - 'report_level' - verbosity of reST parser
+
+        - 'stylesheet' - Stylesheet to be used
+
+        - 'input_encoding' - encoding of the reST input string
+
+        - 'output_encoding' - encoding of the rendered HTML output
+        
+        - 'report_level' - verbosity of reST parser
+
+        - 'language_code' - docutils language
+        
+        - 'initial_header_level' - level of the first header tag
+        
+        - 'warnings' - will be overwritten with a string containing the warnings
+        
+        - 'settings' - dict of settings to pass in to Docutils, with priority
+
+    """
+    parts, warning_stream = render(src,
+                                   writer = writer,
+                                   report_level = report_level,
+                                   stylesheet = stylesheet,
+                                   input_encoding = input_encoding,
+                                   output_encoding = output_encoding,
+                                   language_code=language_code,
+                                   initial_header_level = initial_header_level,
+                                   settings = settings)
+
+    header = '<h%(level)s class="title">%(title)s</h%(level)s>\n' % {
+                  'level': initial_header_level,
+                  'title': parts['title'],
+             }
+
+    body = '%(docinfo)s%(body)s' % {
+                  'docinfo': parts['docinfo'],
+                  'body': parts['body'],
+             }
+
+    if parts['title']:
+        output = header + body
+    else:
+        output = body
+
+    warnings = ''.join(warning_stream.messages)
+
+    return output.encode(output_encoding)
+
+
+__all__ = ("HTML", 'render')

product/PortalTransforms/tests/__init__.py

+"""import this"""

product/PortalTransforms/tests/framework.py

+##############################################################################
+#
+# ZopeTestCase 
+#
+# COPY THIS FILE TO YOUR 'tests' DIRECTORY.
+#
+# This version of framework.py will use the SOFTWARE_HOME
+# environment variable to locate Zope and the Testing package.
+#
+# If the tests are run in an INSTANCE_HOME installation of Zope,
+# Products.__path__ and sys.path with be adjusted to include the
+# instance's Products and lib/python directories respectively.
+#
+# If you explicitly set INSTANCE_HOME prior to running the tests,
+# auto-detection is disabled and the specified path will be used 
+# instead.
+#
+# If the 'tests' directory contains a custom_zodb.py file, INSTANCE_HOME
+# will be adjusted to use it.
+#
+# If you set the ZEO_INSTANCE_HOME environment variable a ZEO setup 
+# is assumed, and you can attach to a running ZEO server (via the 
+# instance's custom_zodb.py).
+#
+##############################################################################
+#
+# The following code should be at the top of every test module:
+#
+# import os, sys
+# if __name__ == '__main__':
+#     execfile(os.path.join(sys.path[0], 'framework.py'))
+#
+# ...and the following at the bottom:
+#
+# if __name__ == '__main__':
+#     framework()
+#
+##############################################################################
+
+__version__ = '0.2.3'
+
+# Save start state
+#
+__SOFTWARE_HOME = os.environ.get('SOFTWARE_HOME', '')
+__INSTANCE_HOME = os.environ.get('INSTANCE_HOME', '')
+
+if __SOFTWARE_HOME.endswith(os.sep):
+    __SOFTWARE_HOME = os.path.dirname(__SOFTWARE_HOME)
+
+if __INSTANCE_HOME.endswith(os.sep):
+    __INSTANCE_HOME = os.path.dirname(__INSTANCE_HOME)
+
+# Find and import the Testing package
+#
+if not sys.modules.has_key('Testing'):
+    p0 = sys.path[0]
+    if p0 and __name__ == '__main__':
+        os.chdir(p0)
+        p0 = ''
+    s = __SOFTWARE_HOME
+    p = d = s and s or os.getcwd()
+    while d:
+        if os.path.isdir(os.path.join(p, 'Testing')):
+            zope_home = os.path.dirname(os.path.dirname(p))
+            sys.path[:1] = [p0, p, zope_home]
+            break
+        p, d = s and ('','') or os.path.split(p)
+    else:
+        print 'Unable to locate Testing package.',
+        print 'You might need to set SOFTWARE_HOME.'
+        sys.exit(1)
+
+import Testing, unittest
+execfile(os.path.join(os.path.dirname(Testing.__file__), 'common.py'))
+
+# Include ZopeTestCase support
+#
+if 1:   # Create a new scope
+
+    p = os.path.join(os.path.dirname(Testing.__file__), 'ZopeTestCase')
+
+    if not os.path.isdir(p):
+        print 'Unable to locate ZopeTestCase package.',
+        print 'You might need to install ZopeTestCase.'
+        sys.exit(1)
+
+    ztc_common = 'ztc_common.py'
+    ztc_common_global = os.path.join(p, ztc_common)
+
+    f = 0
+    if os.path.exists(ztc_common_global):
+        execfile(ztc_common_global)
+        f = 1
+    if os.path.exists(ztc_common):
+        execfile(ztc_common)
+        f = 1
+
+    if not f:
+        print 'Unable to locate %s.' % ztc_common
+        sys.exit(1)
+
+# Debug
+#
+print 'SOFTWARE_HOME: %s' % os.environ.get('SOFTWARE_HOME', 'Not set')
+print 'INSTANCE_HOME: %s' % os.environ.get('INSTANCE_HOME', 'Not set')
+sys.stdout.flush()
+

product/PortalTransforms/tests/input/org-news.xml

+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!DOCTYPE rss PUBLIC "-//Netscape Communications//DTD RSS 0.91//EN" "http://my.netscape.com/publish/formats/rss-0.91.dtd">
+<rss version="0.91"><channel><title>Logilab.org news</title><language>en</language><item><title>xmltools 1.3.7</title><descr>bugfix in namespace handling</descr></item><item><title>Python-logic</title><descr>Set up of the Python-Logic special interest group</descr></item><item><title>PyReverse 0.2.3</title><descr>New features and bug fixes</descr></item><item><title>xmltools 1.3.6</title><descr>Uses the new APIs in pyxml-0.7 and 4Suite-0.12.0</descr></item><item><title>hmm-0.2</title><descr>New learning algorithms available</descr></item><item><title>Version 1.2a1 is out</title><descr>Overall refactoring of the engine. Backward incompatible changes
+      in the syntax of recipes and in modules, in order to ease product development.</descr></item><item><title>XMLdiff v0.5.3 (bug fixes)</title><descr>Version 0.5.3 fixes packaging bugs.</descr></item><item><title>hmm-0.1</title><descr>hmm is a module for Hidden Markov Model manipulation.</descr></item><item><title>PyReverse 0.1 (new product)</title><descr>
+Beta release for this set of tools for reverse engineering python code
+    </descr></item><item><title>PyPaSax 0.3 (bug fixes)</title><descr>A few changes in the DTD, improved PyXML compatibility</descr></item><item><title>XMLdiff v0.5.2 (bug fixes)</title><descr>Version 0.5.2 fixes several bugs.</descr></item><item><title>Version 1.1 is out</title><descr>bugfixes over beta 3.</descr></item><item><title>Version 1.1b3 is out</title><descr>Great speed improvement for Horn. All-in-one windows installer.</descr></item><item><title>xmltools-1.3.5</title><descr>Version 1.3.5 code cleanup.</descr></item><item><title>xmltools-1.3.4</title><descr>Version 1.3.4 fixes a sever encoding bug that could cause crashes on windows machines.</descr></item><item><title>Version 1.1b1 is out</title><descr>Version 1.1b1 drops support for Python 1.5.2 in favor of Python 2.1, and features a new version of Horn, with localization support</descr></item><item><title>XMLdiff v0.5 (algorithm change, bug fixes)</title><descr>Version 0.5. The new algorithm makes it now usable either on big
+documents and really faster in any cases. Fixes Unicode problem.</descr></item><item><title>XMLtools v1.3.1 (bugfixes)</title><descr>Version 1.3.1. This release fixes some minor glitches that had slipped in 1.3.</descr></item><item><title>XMLdiff v0.2 (performance improvement)</title><descr>Version 0.2. Huge performance improvement, and output cleanup.</descr></item><item><title>XMLdiff v0.1.1 (beta release)</title><descr>Version 0.1.1. Fully functionnal. Beta release.</descr></item><item><title>XPathVis v1.0beta (beta release)</title><descr>Version 1.0beta. Works nicely.</descr></item><item><title>XMLtools v1.3 (new features)</title><descr>Version 1.3. This release is compatible with Python 2.x and Unicode. It is not guaranteed to work with Python 1.5.2.</descr></item><item><title>Narval on developerWorks</title><descr>An Introduction to Narval was published on developerWorks.</descr><link>http://www-106.ibm.com/developerworks/library/l-ai/</link></item><item><title>Version 1.0.1 is out</title><descr>Version 1.0.1 is a bugfix release.</descr></item><item><title>Narval reviewed on AI.About.com</title><descr>AI.About.com published a review of Narval.</descr><link>http://ai.about.com/compute/ai/library/weekly/aa060801a.htm</link></item><item><title>Narval at BotShow 2001</title><descr>Narval was presented at the first BotShow event. The slides will soon be available online.</descr><link>http://www.ptolemee.com/botshow/text/text_fr/edito/edito_set.html</link></item><item><title>Version 1.0 is out</title><descr>Version 1.0. Celebration time, come on!</descr></item><item><title>Network-boot-HOWTO v0.2.1</title><descr>Version 0.2.1 is out.</descr></item><item><title>GuessLang v0.1.0 (beta release)</title><descr>Version 0.1.0 is out.</descr></item><item><title>Network-boot-HOWTO v0.1.1</title><descr>Version 0.1.1 is out.</descr></item><item><title>PyPaSax v0.1</title><descr>Version 0.1 is out.</descr></item><item><title>RC2 is out</title><descr>Release Candidate 2 is out. French documentation will be updated within a few days. We also released several applications (or maybe extension sets?) that are in alpha/beta stage. Give them a try!</descr></item><item><title>VCalSax v0.1 (beta)</title><descr>Version 0.1 is out. Still beta, but fully functional.</descr></item><item><title>Talk at LinuxExpo in English</title><descr>A translation of the talk we gave at Linux Expo 2001 is available on-line.</descr><link>http://www.logilab.com/press/linux-expo2001/</link></item><item><title>RC1 is out</title><descr>Release Candidate 1 is out. Documentation will be updated within a few days. Please help us test this one so that we can release 1.0 quicker.</descr></item><item><title>XMLtools v1.2 (stable release)</title><descr>Version 1.2 is released. Bugfixes, mainly..</descr></item><item><title>Application section on web site</title><descr>We just added a new applications section on Logilab.org web site.</descr><link>http://www.logilab.org/narval/app.html</link></item><item><title>WMgMon v0.4.0</title><descr>version 0.4.0 is out. Bugfixes and new monitor functions. </descr></item><item><title>XmlTools v1.1</title><descr>version 1.1 is out. New features in XmlTree.</descr></item><item><title>Beta5 is out</title><descr>Beta 5 is out. Lots of bugfixes in Narval and Horn, client server communication between the kernel and the graphical interface using SOAP. Windows specific bugfixes.</descr></item><item><title>XmlTools v1.0</title><descr>Initial release.</descr></item><item><title>PyGantt v0.6.0</title><descr>Version 0.6.0 released. New features added.</descr></item><item><title>Beta4 is out</title><descr>Beta 4 is out. Improved Windows compatibility. New features and bugfixes in both Narval and Horn.</descr></item><item><title>Article on Narval in Linux Gazette</title><descr>We published an article in the #59 issue of the Linux Gazette. It describes Narval and its use to set up Gazo, the assistant-coordinator for the translation of the Linux Gazette.</descr><link>http://www.linuxgazette.com/issue59/chauvat.html</link></item><item><title>Beta3 is out</title><descr>Beta 3 is out. No more memory leaks (almost). Time conditions work correctly. A step can be an XSL transform. Changes in the Narval DTD.</descr></item><item><title>Logilab invited at Linux Expo</title><descr>We got invited to give a talk at Linux Expo in Paris, France. The talk will be geared toward business uses of Narval. The title will be Using XML and Intelligent Personnal Assistants to enhance groupware and workflow enterprise applications. Come and meet with us!</descr><link>http://www.linuxexpoparis.com/EN/conferences</link></item><item><title>Beta2 is out</title><descr>Beta 2 is out. Installation is much easier. Tutorial. GUI improvements. Bugfixes.</descr></item><item><title>Beta1 is out</title><descr>Beta 1 is out. Many bug fixes.</descr></item><item><title>Beta0 is out</title><descr>Beta 0 is out. Logilab.org is one-line.</descr><link>http://www.logilab.org</link></item></channel></rss>

product/PortalTransforms/tests/input/rest1.rst

+
+This is a test of the *reST* transform
+    o one
+    o two
+    o three
+

product/PortalTransforms/tests/input/rest2.rst

+Heading 1
+=========
+
+Some text.
+
+Heading 2
+---------
+
+Some text, bla ble bli blo blu. Yes, i know this is Stupid_.
+
+.. _Stupid: http://www.example.com
+

product/PortalTransforms/tests/input/rest3.rst

+=====
+Title
+=====
+
+--------
+Subtitle
+--------
+
+This is a test document to make sure subtitle gets the right heading.
+
+Now the real heading
+====================
+
+The brown fox jumped over the lazy dog.
+
+With a subheading
+------------------
+
+Some text, bla ble bli blo blu. Yes, i know this is Stupid_.
+
+.. _Stupid: http://www.example.com
+

product/PortalTransforms/tests/input/rss2html.xslt

+<?xml version="1.0" encoding="ISO-8859-1"?>
+
+<xsl:transform xmlns:xsl='http://www.w3.org/1999/XSL/Transform' version='1.0'>
+
+<xsl:strip-space elements='*'/>
+<xsl:output method='xml'/>
+
+<!-- Narval prototype ====================================================== -->
+
+<al:prototype xmlns:al="http://www.logilab.org/namespaces/Narval/1.2">
+<al:description lang="fr">Transforme du RSS en du HTML.</al:description>
+<al:description lang="en">Turns RSS into HTML.</al:description>
+
+<al:input id="input"><al:match>rss</al:match></al:input>
+<al:output id="output" list="yes"><al:match>html-body</al:match></al:output>
+</al:prototype>
+
+<!-- root ================================================================== -->
+ 
+<xsl:template match='rss/rss/channel'>
+  <html-body>
+    <h2>
+      <xsl:value-of select='title'/>
+    </h2>
+    <p>
+      <xsl:element name='a'>
+        <xsl:attribute name='href'><xsl:value-of select='link'/></xsl:attribute>
+        <xsl:value-of select='title'/>
+      </xsl:element>
+      <em><xsl:value-of select='description'/></em>
+    </p>
+    <table>
+      <xsl:apply-templates select='item'/>
+    </table>
+  </html-body>
+</xsl:template>
+
+<xsl:template match='item'>
+    <tr>
+      <td>
+        <xsl:element name='a'>
+          <xsl:attribute name='href'><xsl:value-of select='link'/></xsl:attribute>
+          <xsl:value-of select='title'/>
+        </xsl:element>
+        <xsl:apply-templates mode='multi' select='description'/>
+      </td>
+    </tr>
+</xsl:template>
+
+</xsl:transform>

product/PortalTransforms/tests/input/test_lynx.html

+<?xml version="1.0" encoding="utf-8" ?>
+<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
+<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
+<head>
+<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
+<meta name="generator" content="Docutils 0.2.8: http://docutils.sourceforge.net/" />
+<title>Copying Docutils</title>
+<meta name="author" content="David Goodger" />
+<meta name="date" content="2002-10-03" />
+<link rel="stylesheet" href="tools/stylesheets/default.css" type="text/css" />
+</head>
+<body>
+<div class="document" id="copying-docutils">
+<h1 class="title">Copying Docutils</h1>
+<table class="docinfo" frame="void" rules="none">
+<col class="docinfo-name" />
+<col class="docinfo-content" />
+<tbody valign="top">
+<tr><th class="docinfo-name">Author:</th>
+<td>David Goodger</td></tr>
+<tr><th class="docinfo-name">Contact:</th>
+<td><a class="first last reference" href="mailto:goodger&#64;users.sourceforge.net">goodger&#64;users.sourceforge.net</a></td></tr>
+<tr><th class="docinfo-name">Date:</th>
+<td>2002-10-03</td></tr>
+<tr class="field"><th class="docinfo-name">Web site:</th><td class="field-body"><a class="reference" href="http://docutils.sourceforge.net/">http://docutils.sourceforge.net/</a></td>
+</tr>
+</tbody>
+</table>
+<p>Most of the files included in this project are in the public domain,
+and therefore have no license requirement and no restrictions on
+copying or usage.  The exceptions are:</p>
+<ul class="simple">
+<li>docutils/optik.py, copyright Gregory P. Ward, released under a
+BSD-style license (which can be found in the module's source code).</li>
+<li>docutils/roman.py, copyright by Mark Pilgrim, released under the
+<a class="reference" href="http://www.python.org/2.1.1/license.html">Python 2.1.1 license</a>.</li>
+<li>test/difflib.py, copyright by the Python Software Foundation,
+released under the <a class="reference" href="http://www.python.org/2.2/license.html">Python 2.2 license</a>.  This file is included for
+compatibility with Python versions less than 2.2; if you have Python
+2.2 or higher, difflib.py is not needed and may be removed.  (It's
+only used to report test failures anyhow; it isn't installed
+anywhere.  The included file is a pre-generator version of the
+difflib.py module included in Python 2.2.)</li>
+</ul>
+<p>(Disclaimer: I am not a lawyer.)  Both the BSD license and the Python
+license are <a class="reference" href="http://opensource.org/licenses/">OSI-approved</a> and <a class="reference" href="http://www.gnu.org/philosophy/license-list.html">GPL-compatible</a>.  Although complicated
+by multiple owners and lots of legalese, the Python license basically
+lets you copy, use, modify, and redistribute files as long as you keep
+the copyright attribution intact, note any changes you make, and don't
+use the owner's name in vain.  The BSD license is similar.</p>
+</div>
+<hr class="footer"/>
+<div class="footer">
+Generated on: 2003-04-19 15:32 UTC.
+Generated by <a class="reference" href="http://docutils.sourceforge.net/">Docutils</a> from <a class="reference" href="http://docutils.sourceforge.net/rst.html">reStructuredText</a> source.
+</div>
+</body>
+</html>

product/PortalTransforms/tests/input/test_python.py

+""" nice docstring """
+
+class A : pass
+
+# comment
+
+def inc(i):
+    return i+1
+    
+def greater(a, b):
+    """foo <html />"""
+    return a > b

product/PortalTransforms/tests/input/test_safehtml.html

+<?xml version="1.0" encoding="utf-8" ?>
+<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
+<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
+<head>
+<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
+<title>Test page for save html rendering</title>
+<meta name="date" content="2005-07-22" />
+</head>
+<body>
+<h1>Test page</h1>
+<table>
+<tr>
+<th>Test1</th>
+<td>test2</td>
+</tr>
+</table>
+<p>This is a text used as a blind text.</p>
+<ul>
+<li>A sample list item1</li>
+<li>A sample list item2</li>
+</ul>
+<p>This is again a blind text with a<br>line break.</p>
+<div>
+Can we <q>quote</q> or write something we <del>didn't</del> mean to write? Or how is <ins>this</ins> instead?
+</div>
+<hr>
+<div>
+<a href="http://www.plone.org"><img src="http://www.plone.org/logo.jpg"/></a> is just great.
+</div>
+</body>
+</html>

product/PortalTransforms/tests/output/demo1.html

+<A name=1></a>Chapter 44<br>
+Writing Basic Unit Tests<br>
+Difficulty<br>
+Newcomer<br>
+Skills<br>
+• All you need to know is some Python.<br>
+Problem/Task<br>
+As you know by now, Zope 3 gains its incredible stability from testing any code in great detail. The<br>currently most common method is to write unit tests. This chapter introduces unit tests – which<br>are Zope 3 independent – and introduces some of the subtleties.<br>
+Solution<br>
+44.1<br>
+Implementing the Sample Class<br>
+Before we can write tests, we have to write some code that we can test. Here, we will implement<br>a simple class called Sample with a public attribute title and description that is accessed<br>via getDescription() and mutated using setDescription(). Further, the description must be<br>either a regular or unicode string.<br>
+Since this code will not depend on Zope, open a file named test sample.py anywhere and add<br>
+the following class:<br>
+1 Sample(object):<br>
+2<br>
+&quot;&quot;&quot;A trivial Sample object.&quot;&quot;&quot;<br>
+3<br>
+4<br>
+title = None<br>
+5<br>
+6<br>
+def __init__(self):<br>
+7<br>
+&quot;&quot;&quot;Initialize object.&quot;&quot;&quot;<br>
+8<br>
+self._description = ’’<br>
+9<br>
+1<br>
+<hr>
+<A name=2></a>2<br>
+CHAPTER 44. WRITING BASIC UNIT TESTS<br>
+10<br>
+def setDescription(self, value):<br>
+11<br>
+&quot;&quot;&quot;Change the value of the description.&quot;&quot;&quot;<br>
+12<br>
+assert isinstance(value, (str, unicode))<br>
+13<br>
+self._description = value<br>
+14<br>
+15<br>
+def getDescription(self):<br>
+16<br>
+&quot;&quot;&quot;Change the value of the description.&quot;&quot;&quot;<br>
+17<br>
+return self._description<br>
+Line 4: The title is just publicly declared and a value of None is given. Therefore this is just<br>a regular attribute.<br>
+Line 8: The actual description string will be stored in description.<br>
+Line 12: Make sure that the description is only a regular or unicode string, like it was stated in<br>the requirements.<br>
+If you wish you can now manually test the class with the interactive Python shell. Just start<br>
+Python by entering python in your shell prompt. Note that you should be in the directory in<br>which test sample.py is located when starting Python (an alternative is of course to specify the<br>directory in your PYTHONPATH.)<br>
+1 &gt;&gt;&gt; from test_sample import Sample<br>2 &gt;&gt;&gt; sample = Sample()<br>
+3 &gt;&gt;&gt; print sample.title<br>4 None<br>
+5 &gt;&gt;&gt; sample.title = ’Title’<br>
+6 &gt;&gt;&gt; print sample.title<br>7 Title<br>
+8 &gt;&gt;&gt; print sample.getDescription()<br>9<br>
+10 &gt;&gt;&gt; sample.setDescription(’Hello World’)<br>
+11 &gt;&gt;&gt; print sample.getDescription()<br>12 Hello World<br>
+13 &gt;&gt;&gt; sample.setDescription(None)<br>
+14 Traceback (most recent call last):<br>
+15<br>
+File &quot;&lt;stdin&gt;&quot;, line 1, in ?<br>
+16<br>
+File &quot;test_sample.py&quot;, line 31, in setDescription<br>
+17<br>
+assert isinstance(value, (str, unicode))<br>
+18 AssertionError<br>
+As you can see in the last test, non-string object types are not allowed as descriptions and an<br>
+AssertionError is raised.<br>
+44.2<br>
+Writing the Unit Tests<br>
+The goal of writing the unit tests is to convert this informal, manual, and interactive testing session<br>into a formal test class. Python provides already a module called unittest for this purpose, which<br>is a port of the Java-based unit testing product, JUnit, by Kent Beck and Erich Gamma. There are<br>three levels to the testing framework (this list deviates a bit from the original definitions as found<br>in the Python library documentation. 1).<br>
+1 http://www.python.org/doc/current/lib/module-unittest.html<br>
+<hr>
+<A name=3></a>44.2. WRITING THE UNIT TESTS<br>
+3<br>
+The smallest unit is obviously the “test”, which is a single method in a TestCase class that<br>
+tests the behavior of a small piece of code or a particular aspect of an implementation. The “test<br>case” is then a collection tests that share the same setup/inputs. On top of all of this sits the “test<br>suite” which is a collection of test cases and/or other test suites. Test suites combine tests that<br>should be executed together. With the correct setup (as shown in the example below), you can<br>then execute test suites. For large projects like Zope 3, it is useful to know that there is also the<br>concept of a test runner, which manages the test run of all or a set of tests. The runner provides<br>useful feedback to the application, so that various user interaces can be developed on top of it.<br>
+But enough about the theory. In the following example, which you can simply put into the same<br>
+file as your code above, you will see a test in common Zope 3 style.<br>
+1 import unittest<br>2<br>
+3 class SampleTest(unittest.TestCase):<br>4<br>
+&quot;&quot;&quot;Test the Sample class&quot;&quot;&quot;<br>
+5<br>
+6<br>
+def test_title(self):<br>
+7<br>
+sample = Sample()<br>
+8<br>
+self.assertEqual(sample.title, None)<br>
+9<br>
+sample.title = ’Sample Title’<br>
+10<br>
+self.assertEqual(sample.title, ’Sample Title’)<br>
+11<br>
+12<br>
+def test_getDescription(self):<br>
+13<br>
+sample = Sample()<br>
+14<br>
+self.assertEqual(sample.getDescription(), ’’)<br>
+15<br>
+sample._description = &quot;Description&quot;<br>
+16<br>
+self.assertEqual(sample.getDescription(), ’Description’)<br>
+17<br>
+18<br>
+def test_setDescription(self):<br>
+19<br>
+sample = Sample()<br>
+20<br>
+self.assertEqual(sample._description, ’’)<br>
+21<br>
+sample.setDescription(’Description’)<br>
+22<br>
+self.assertEqual(sample._description, ’Description’)<br>
+23<br>
+sample.setDescription(u’Description2’)<br>
+24<br>
+self.assertEqual(sample._description, u’Description2’)<br>
+25<br>
+self.assertRaises(AssertionError, sample.setDescription, None)<br>
+26<br>
+27<br>
+28 def test_suite():<br>29<br>
+return unittest.TestSuite((<br>
+30<br>
+unittest.makeSuite(SampleTest),<br>
+31<br>
+))<br>
+32<br>
+33 if __name__ == ’__main__’:<br>34<br>
+unittest.main(defaultTest=’test_suite’)<br>
+Line 3–4: We usually develop test classes which must inherit from TestCase. While often not<br>done, it is a good idea to give the class a meaningful docstring that describes the purpose of the<br>tests it includes.<br>
+Line 6, 12 &amp; 18: When a test case is run, a method called runTests() is executed. While it<br>is possible to overrride this method to run tests differently, the default option will look for any<br>method whose name starts with test and execute it as a single test. This way we can create<br>a “test method” for each aspect, method, function or property of the code to be tested. This<br>default is very sensible and is used everywhere in Zope 3.<br>
+<hr>
+<A name=4></a>4<br>
+CHAPTER 44. WRITING BASIC UNIT TESTS<br>
+Note that there is no docstring for test methods. This is intentional. If a docstring is specified,<br>it is used instead of the method name to identify the test. When specifying a docstring, we have<br>noticed that it is very difficult to identify the test later; therefore the method name is a much<br>better choice.<br>
+Line 8, 10, 14, . . . : The TestCase class implements a handful of methods that aid you with the<br>testing. Here are some of the most frequently used ones. For a complete list see the standard<br>Python documentation referenced above.<br>
+• assertEqual(first,second[,msg])<br>
+Checks whether the first and second value are equal. If the test fails, the msg or None<br>is returned.<br>
+• assertNotEqual(first,second[,msg])<br>
+This is simply the opposite to assertEqual() by checking for non-equality.<br>
+• assertRaises(exception,callable,...)<br>
+You expect the callable to raise exception when executed. After the callable you can<br>specify any amount of positional and keyword arguments for the callable. If you expect<br>a group of exceptions from the execution, you can make exception a tuple of possible<br>exceptions.<br>
+• assert (expr[,msg])<br>
+Assert checks whether the specified expression executes correctly. If not, the test fails and<br>msg or None is returned.<br>
+• failUnlessEqual()<br>
+This testing method is equivalent to assertEqual().<br>
+• failUnless(expr[,msg])<br>
+This method is equivalent to assert (expr[,msg]).<br>
+• failif()<br>
+This is the opposite to failUnless().<br>
+• fail([msg])<br>
+Fails the running test without any evaluation. This is commonly used when testing various<br>possible execution paths at once and you would like to signify a failure if an improper path<br>was taken.<br>
+Line 6–10: This method tests the title attribute of the Sample class. The first test should<br>be of course that the attribute exists and has the expected initial value (line 8). Then the title<br>attribute is changed and we check whether the value was really stored. This might seem like<br>overkill, but later you might change the title in a way that it uses properties instead. Then it<br>becomes very important to check whether this test still passes.<br>
+Line 12–16: First we simply check that getDescription() returns the correct default value.<br>Since we do not want to use other API calls like setDescription() we set a new value of the<br>description via the implementation-internal description attribute (line 15). This is okay! Unit<br>tests can make use of implementation-specific attributes and methods. Finally we just check that<br>the correct value is returned.<br>
+<hr>
+<A name=5></a>44.3. RUNNING THE TESTS<br>
+5<br>
+Line 18–25: On line 21–24 it is checked that both regular and unicode strings are set correctly.<br>In the last line of the test we make sure that no other type of objects can be set as a description<br>and that an error is raised.<br>
+28–31: This method returns a test suite that includes all test cases created in this module. It is<br>used by the Zope 3 test runner when it picks up all available tests. You would basically add the<br>line unittest.makeSuite(TestCaseClass) for each additional test case.<br>
+33–34: In order to make the test module runnable by itself, you can execute unittest.main()<br>when the module is run.<br>
+44.3<br>
+Running the Tests<br>
+You can run the test by simply calling pythontest sample.py from the directory you saved the<br>file in. Here is the result you should see:<br>
+.<br>--------------------------------------------------------------------<br>n 3 tests in 0.001s<br>
+The three dots represent the three tests that were run. If a test had failed, it would have been<br>
+reported pointing out the failing test and providing a small traceback.<br>
+When using the default Zope 3 test runner, tests will be picked up as long as they follow some<br>
+conventions.<br>
+• The tests must either be in a package or be a module called tests.<br>
+• If tests is a package, then all test modules inside must also have a name starting with test,<br>
+as it is the case with our name test sample.py.<br>
+• The test module must be somewhere in the Zope 3 source tree, since the test runner looks<br>
+only for files there.<br>
+In our case, you could simply create a tests package in ZOPE3/src (do not forget the<br>
+init .<br>
+py file). Then place the test sample.py file into this directory.<br>
+You you can use the test runner to run only the sample tests as follows from the Zope 3 root<br>
+directory:<br>
+python test.py -vp tests.test_sample<br>
+The -v option stands for verbose mode, so that detailed information about a test failure is<br>
+provided. The -p option enables a progress bar that tells you how many tests out of all have been<br>completed. There are many more options that can be specified. You can get a full list of them with<br>the option -h: pythontest.py-h.<br>
+The output of the call above is as follows:<br>
+nfiguration file found.<br>nning UNIT tests at level 1<br>nning UNIT tests from /opt/zope/Zope3<br>
+3/3 (100.0%): test_title (tests.test_sample.SampleTest)<br>
+--------------------------------------------------------------------<br>n 3 tests in 0.002s<br>
+<hr>
+<A name=6></a>6<br>
+CHAPTER 44. WRITING BASIC UNIT TESTS<br>
+nning FUNCTIONAL tests at level 1<br>nning FUNCTIONAL tests from /opt/zope/Zope3<br>
+--------------------------------------------------------------------<br>n 0 tests in 0.000s<br>
+Line 1: The test runner uses a configuration file for some setup. This allows developers to use<br>the test runner for other projects as well. This message simply tells us that the configuration file<br>was found.<br>
+Line 2–8: The unit tests are run. On line 4 you can see the progress bar.<br>
+Line 9–15: The functional tests are run, since the default test runner runs both types of tests.<br>Since we do not have any functional tests in the specified module, there are no tests to run. To<br>just run the unit tests, use option -u and -f for just running the functional tests. See “Writing<br>Functional Tests” for more detials on functional tests.<br>
+<hr>
+<A name=7></a>44.3. RUNNING THE TESTS<br>
+7<br>
+Exercises<br>
+1. It is not very common to do the setup – in our case sample=Sample() – in every test<br>
+method. Instead there exists a method called setUp() and its counterpart tearDown that<br>are run before and after each test, respectively. Change the test code above, so that it uses<br>the setUp() method. In later chapters and the rest of the book we will frequently use this<br>method of setting up tests.<br>
+2. Currently the test setDescription() test only verifies that None is not allowed as input<br>
+value.<br>
+(a) Improve the test, so that all other builtin types are tested as well.<br>
+(b) Also, make sure that any objects inheriting from str or unicode pass as valid values.<br>
+<hr>
\ No newline at end of file

product/PortalTransforms/tests/output/demo1.html.nofilename

+<A name=1></a>Chapter 44<br>
+Writing Basic Unit Tests<br>
+Difficulty<br>
+Newcomer<br>
+Skills<br>
+• All you need to know is some Python.<br>
+Problem/Task<br>
+As you know by now, Zope 3 gains its incredible stability from testing any code in great detail. The<br>currently most common method is to write unit tests. This chapter introduces unit tests – which<br>are Zope 3 independent – and introduces some of the subtleties.<br>
+Solution<br>
+44.1<br>
+Implementing the Sample Class<br>
+Before we can write tests, we have to write some code that we can test. Here, we will implement<br>a simple class called Sample with a public attribute title and description that is accessed<br>via getDescription() and mutated using setDescription(). Further, the description must be<br>either a regular or unicode string.<br>
+Since this code will not depend on Zope, open a file named test sample.py anywhere and add<br>
+the following class:<br>
+1 Sample(object):<br>
+2<br>
+&quot;&quot;&quot;A trivial Sample object.&quot;&quot;&quot;<br>
+3<br>
+4<br>
+title = None<br>
+5<br>
+6<br>
+def __init__(self):<br>
+7<br>
+&quot;&quot;&quot;Initialize object.&quot;&quot;&quot;<br>
+8<br>
+self._description = ’’<br>
+9<br>
+1<br>
+<hr>
+<A name=2></a>2<br>
+CHAPTER 44. WRITING BASIC UNIT TESTS<br>
+10<br>
+def setDescription(self, value):<br>
+11<br>
+&quot;&quot;&quot;Change the value of the description.&quot;&quot;&quot;<br>
+12<br>
+assert isinstance(value, (str, unicode))<br>
+13<br>
+self._description = value<br>
+14<br>
+15<br>
+def getDescription(self):<br>
+16<br>
+&quot;&quot;&quot;Change the value of the description.&quot;&quot;&quot;<br>
+17<br>
+return self._description<br>
+Line 4: The title is just publicly declared and a value of None is given. Therefore this is just<br>a regular attribute.<br>
+Line 8: The actual description string will be stored in description.<br>
+Line 12: Make sure that the description is only a regular or unicode string, like it was stated in<br>the requirements.<br>
+If you wish you can now manually test the class with the interactive Python shell. Just start<br>
+Python by entering python in your shell prompt. Note that you should be in the directory in<br>which test sample.py is located when starting Python (an alternative is of course to specify the<br>directory in your PYTHONPATH.)<br>
+1 &gt;&gt;&gt; from test_sample import Sample<br>2 &gt;&gt;&gt; sample = Sample()<br>
+3 &gt;&gt;&gt; print sample.title<br>4 None<br>
+5 &gt;&gt;&gt; sample.title = ’Title’<br>
+6 &gt;&gt;&gt; print sample.title<br>7 Title<br>
+8 &gt;&gt;&gt; print sample.getDescription()<br>9<br>
+10 &gt;&gt;&gt; sample.setDescription(’Hello World’)<br>
+11 &gt;&gt;&gt; print sample.getDescription()<br>12 Hello World<br>
+13 &gt;&gt;&gt; sample.setDescription(None)<br>
+14 Traceback (most recent call last):<br>
+15<br>
+File &quot;&lt;stdin&gt;&quot;, line 1, in ?<br>
+16<br>
+File &quot;test_sample.py&quot;, line 31, in setDescription<br>
+17<br>
+assert isinstance(value, (str, unicode))<br>
+18 AssertionError<br>
+As you can see in the last test, non-string object types are not allowed as descriptions and an<br>
+AssertionError is raised.<br>
+44.2<br>
+Writing the Unit Tests<br>
+The goal of writing the unit tests is to convert this informal, manual, and interactive testing session<br>into a formal test class. Python provides already a module called unittest for this purpose, which<br>is a port of the Java-based unit testing product, JUnit, by Kent Beck and Erich Gamma. There are<br>three levels to the testing framework (this list deviates a bit from the original definitions as found<br>in the Python library documentation. 1).<br>
+1 http://www.python.org/doc/current/lib/module-unittest.html<br>
+<hr>
+<A name=3></a>44.2. WRITING THE UNIT TESTS<br>
+3<br>
+The smallest unit is obviously the “test”, which is a single method in a TestCase class that<br>
+tests the behavior of a small piece of code or a particular aspect of an implementation. The “test<br>case” is then a collection tests that share the same setup/inputs. On top of all of this sits the “test<br>suite” which is a collection of test cases and/or other test suites. Test suites combine tests that<br>should be executed together. With the correct setup (as shown in the example below), you can<br>then execute test suites. For large projects like Zope 3, it is useful to know that there is also the<br>concept of a test runner, which manages the test run of all or a set of tests. The runner provides<br>useful feedback to the application, so that various user interaces can be developed on top of it.<br>
+But enough about the theory. In the following example, which you can simply put into the same<br>
+file as your code above, you will see a test in common Zope 3 style.<br>
+1 import unittest<br>2<br>
+3 class SampleTest(unittest.TestCase):<br>4<br>
+&quot;&quot;&quot;Test the Sample class&quot;&quot;&quot;<br>
+5<br>
+6<br>
+def test_title(self):<br>
+7<br>
+sample = Sample()<br>
+8<br>
+self.assertEqual(sample.title, None)<br>
+9<br>
+sample.title = ’Sample Title’<br>
+10<br>
+self.assertEqual(sample.title, ’Sample Title’)<br>
+11<br>
+12<br>
+def test_getDescription(self):<br>
+13<br>
+sample = Sample()<br>
+14<br>
+self.assertEqual(sample.getDescription(), ’’)<br>
+15<br>
+sample._description = &quot;Description&quot;<br>
+16<br>
+self.assertEqual(sample.getDescription(), ’Description’)<br>
+17<br>
+18<br>
+def test_setDescription(self):<br>
+19<br>
+sample = Sample()<br>
+20<br>
+self.assertEqual(sample._description, ’’)<br>
+21<br>
+sample.setDescription(’Description’)<br>
+22<br>
+self.assertEqual(sample._description, ’Description’)<br>
+23<br>
+sample.setDescription(u’Description2’)<br>
+24<br>
+self.assertEqual(sample._description, u’Description2’)<br>
+25<br>
+self.assertRaises(AssertionError, sample.setDescription, None)<br>
+26<br>
+27<br>
+28 def test_suite():<br>29<br>
+return unittest.TestSuite((<br>
+30<br>
+unittest.makeSuite(SampleTest),<br>
+31<br>
+))<br>
+32<br>
+33 if __name__ == ’__main__’:<br>34<br>
+unittest.main(defaultTest=’test_suite’)<br>
+Line 3–4: We usually develop test classes which must inherit from TestCase. While often not<br>done, it is a good idea to give the class a meaningful docstring that describes the purpose of the<br>tests it includes.<br>
+Line 6, 12 &amp; 18: When a test case is run, a method called runTests() is executed. While it<br>is possible to overrride this method to run tests differently, the default option will look for any<br>method whose name starts with test and execute it as a single test. This way we can create<br>a “test method” for each aspect, method, function or property of the code to be tested. This<br>default is very sensible and is used everywhere in Zope 3.<br>
+<hr>
+<A name=4></a>4<br>
+CHAPTER 44. WRITING BASIC UNIT TESTS<br>
+Note that there is no docstring for test methods. This is intentional. If a docstring is specified,<br>it is used instead of the method name to identify the test. When specifying a docstring, we have<br>noticed that it is very difficult to identify the test later; therefore the method name is a much<br>better choice.<br>
+Line 8, 10, 14, . . . : The TestCase class implements a handful of methods that aid you with the<br>testing. Here are some of the most frequently used ones. For a complete list see the standard<br>Python documentation referenced above.<br>
+• assertEqual(first,second[,msg])<br>
+Checks whether the first and second value are equal. If the test fails, the msg or None<br>is returned.<br>
+• assertNotEqual(first,second[,msg])<br>
+This is simply the opposite to assertEqual() by checking for non-equality.<br>
+• assertRaises(exception,callable,...)<br>
+You expect the callable to raise exception when executed. After the callable you can<br>specify any amount of positional and keyword arguments for the callable. If you expect<br>a group of exceptions from the execution, you can make exception a tuple of possible<br>exceptions.<br>
+• assert (expr[,msg])<br>
+Assert checks whether the specified expression executes correctly. If not, the test fails and<br>msg or None is returned.<br>
+• failUnlessEqual()<br>
+This testing method is equivalent to assertEqual().<br>
+• failUnless(expr[,msg])<br>
+This method is equivalent to assert (expr[,msg]).<br>
+• failif()<br>
+This is the opposite to failUnless().<br>
+• fail([msg])<br>
+Fails the running test without any evaluation. This is commonly used when testing various<br>possible execution paths at once and you would like to signify a failure if an improper path<br>was taken.<br>
+Line 6–10: This method tests the title attribute of the Sample class. The first test should<br>be of course that the attribute exists and has the expected initial value (line 8). Then the title<br>attribute is changed and we check whether the value was really stored. This might seem like<br>overkill, but later you might change the title in a way that it uses properties instead. Then it<br>becomes very important to check whether this test still passes.<br>
+Line 12–16: First we simply check that getDescription() returns the correct default value.<br>Since we do not want to use other API calls like setDescription() we set a new value of the<br>description via the implementation-internal description attribute (line 15). This is okay! Unit<br>tests can make use of implementation-specific attributes and methods. Finally we just check that<br>the correct value is returned.<br>
+<hr>
+<A name=5></a>44.3. RUNNING THE TESTS<br>
+5<br>
+Line 18–25: On line 21–24 it is checked that both regular and unicode strings are set correctly.<br>In the last line of the test we make sure that no other type of objects can be set as a description<br>and that an error is raised.<br>
+28–31: This method returns a test suite that includes all test cases created in this module. It is<br>used by the Zope 3 test runner when it picks up all available tests. You would basically add the<br>line unittest.makeSuite(TestCaseClass) for each additional test case.<br>
+33–34: In order to make the test module runnable by itself, you can execute unittest.main()<br>when the module is run.<br>
+44.3<br>
+Running the Tests<br>
+You can run the test by simply calling pythontest sample.py from the directory you saved the<br>file in. Here is the result you should see:<br>
+.<br>--------------------------------------------------------------------<br>n 3 tests in 0.001s<br>
+The three dots represent the three tests that were run. If a test had failed, it would have been<br>
+reported pointing out the failing test and providing a small traceback.<br>
+When using the default Zope 3 test runner, tests will be picked up as long as they follow some<br>
+conventions.<br>
+• The tests must either be in a package or be a module called tests.<br>
+• If tests is a package, then all test modules inside must also have a name starting with test,<br>
+as it is the case with our name test sample.py.<br>
+• The test module must be somewhere in the Zope 3 source tree, since the test runner looks<br>
+only for files there.<br>
+In our case, you could simply create a tests package in ZOPE3/src (do not forget the<br>
+init .<br>
+py file). Then place the test sample.py file into this directory.<br>
+You you can use the test runner to run only the sample tests as follows from the Zope 3 root<br>
+directory:<br>
+python test.py -vp tests.test_sample<br>
+The -v option stands for verbose mode, so that detailed information about a test failure is<br>
+provided. The -p option enables a progress bar that tells you how many tests out of all have been<br>completed. There are many more options that can be specified. You can get a full list of them with<br>the option -h: pythontest.py-h.<br>
+The output of the call above is as follows:<br>
+nfiguration file found.<br>nning UNIT tests at level 1<br>nning UNIT tests from /opt/zope/Zope3<br>
+3/3 (100.0%): test_title (tests.test_sample.SampleTest)<br>
+--------------------------------------------------------------------<br>n 3 tests in 0.002s<br>
+<hr>
+<A name=6></a>6<br>
+CHAPTER 44. WRITING BASIC UNIT TESTS<br>
+nning FUNCTIONAL tests at level 1<br>nning FUNCTIONAL tests from /opt/zope/Zope3<br>
+--------------------------------------------------------------------<br>n 0 tests in 0.000s<br>
+Line 1: The test runner uses a configuration file for some setup. This allows developers to use<br>the test runner for other projects as well. This message simply tells us that the configuration file<br>was found.<br>
+Line 2–8: The unit tests are run. On line 4 you can see the progress bar.<br>
+Line 9–15: The functional tests are run, since the default test runner runs both types of tests.<br>Since we do not have any functional tests in the specified module, there are no tests to run. To<br>just run the unit tests, use option -u and -f for just running the functional tests. See “Writing<br>Functional Tests” for more detials on functional tests.<br>
+<hr>
+<A name=7></a>44.3. RUNNING THE TESTS<br>
+7<br>
+Exercises<br>
+1. It is not very common to do the setup – in our case sample=Sample() – in every test<br>
+method. Instead there exists a method called setUp() and its counterpart tearDown that<br>are run before and after each test, respectively. Change the test code above, so that it uses<br>the setUp() method. In later chapters and the rest of the book we will frequently use this<br>method of setting up tests.<br>
+2. Currently the test setDescription() test only verifies that None is not allowed as input<br>
+value.<br>
+(a) Improve the test, so that all other builtin types are tested as well.<br>
+(b) Also, make sure that any objects inheriting from str or unicode pass as valid values.<br>
+<hr>
\ No newline at end of file

product/PortalTransforms/tests/output/logo.ppm

+P6
+24 23
+255
+̙̙
\ No newline at end of file

product/PortalTransforms/tests/output/org-news.html

+<?xml version="1.0"?>
+Logilab.org newsen<tr><td><a href="">xmltools 1.3.7</a></td></tr><tr><td><a href="">Python-logic</a></td></tr><tr><td><a href="">PyReverse 0.2.3</a></td></tr><tr><td><a href="">xmltools 1.3.6</a></td></tr><tr><td><a href="">hmm-0.2</a></td></tr><tr><td><a href="">Version 1.2a1 is out</a></td></tr><tr><td><a href="">XMLdiff v0.5.3 (bug fixes)</a></td></tr><tr><td><a href="">hmm-0.1</a></td></tr><tr><td><a href="">PyReverse 0.1 (new product)</a></td></tr><tr><td><a href="">PyPaSax 0.3 (bug fixes)</a></td></tr><tr><td><a href="">XMLdiff v0.5.2 (bug fixes)</a></td></tr><tr><td><a href="">Version 1.1 is out</a></td></tr><tr><td><a href="">Version 1.1b3 is out</a></td></tr><tr><td><a href="">xmltools-1.3.5</a></td></tr><tr><td><a href="">xmltools-1.3.4</a></td></tr><tr><td><a href="">Version 1.1b1 is out</a></td></tr><tr><td><a href="">XMLdiff v0.5 (algorithm change, bug fixes)</a></td></tr><tr><td><a href="">XMLtools v1.3.1 (bugfixes)</a></td></tr><tr><td><a href="">XMLdiff v0.2 (performance improvement)</a></td></tr><tr><td><a href="">XMLdiff v0.1.1 (beta release)</a></td></tr><tr><td><a href="">XPathVis v1.0beta (beta release)</a></td></tr><tr><td><a href="">XMLtools v1.3 (new features)</a></td></tr><tr><td><a href="http://www-106.ibm.com/developerworks/library/l-ai/">Narval on developerWorks</a></td></tr><tr><td><a href="">Version 1.0.1 is out</a></td></tr><tr><td><a href="http://ai.about.com/compute/ai/library/weekly/aa060801a.htm">Narval reviewed on AI.About.com</a></td></tr><tr><td><a href="http://www.ptolemee.com/botshow/text/text_fr/edito/edito_set.html">Narval at BotShow 2001</a></td></tr><tr><td><a href="">Version 1.0 is out</a></td></tr><tr><td><a href="">Network-boot-HOWTO v0.2.1</a></td></tr><tr><td><a href="">GuessLang v0.1.0 (beta release)</a></td></tr><tr><td><a href="">Network-boot-HOWTO v0.1.1</a></td></tr><tr><td><a href="">PyPaSax v0.1</a></td></tr><tr><td><a href="">RC2 is out</a></td></tr><tr><td><a href="">VCalSax v0.1 (beta)</a></td></tr><tr><td><a href="http://www.logilab.com/press/linux-expo2001/">Talk at LinuxExpo in English</a></td></tr><tr><td><a href="">RC1 is out</a></td></tr><tr><td><a href="">XMLtools v1.2 (stable release)</a></td></tr><tr><td><a href="http://www.logilab.org/narval/app.html">Application section on web site</a></td></tr><tr><td><a href="">WMgMon v0.4.0</a></td></tr><tr><td><a href="">XmlTools v1.1</a></td></tr><tr><td><a href="">Beta5 is out</a></td></tr><tr><td><a href="">XmlTools v1.0</a></td></tr><tr><td><a href="">PyGantt v0.6.0</a></td></tr><tr><td><a href="">Beta4 is out</a></td></tr><tr><td><a href="http://www.linuxgazette.com/issue59/chauvat.html">Article on Narval in Linux Gazette</a></td></tr><tr><td><a href="">Beta3 is out</a></td></tr><tr><td><a href="http://www.linuxexpoparis.com/EN/conferences">Logilab invited at Linux Expo</a></td></tr><tr><td><a href="">Beta2 is out</a></td></tr><tr><td><a href="">Beta1 is out</a></td></tr><tr><td><a href="http://www.logilab.org">Beta0 is out</a></td></tr>

product/PortalTransforms/tests/output/rest1.html

+<p>This is a test of the *reST* transform<br />    o one<br />    o two<br />    o three</p>
\ No newline at end of file

product/PortalTransforms/tests/output/rest1.out

+<dl class="docutils">
+<dt>This is a test of the <em>reST</em> transform</dt>
+<dd>o one
+o two
+o three</dd>
+</dl>

product/PortalTransforms/tests/output/rest1.rst

+
+This is a test of the *reST* transform
+    o one
+    o two
+    o three
+

product/PortalTransforms/tests/output/rest2.out

+<h2 class="title">Heading 1</h2>
+<p>Some text.</p>
+<div class="section" id="heading-2">
+<h3><a name="heading-2">Heading 2</a></h3>
+<p>Some text, bla ble bli blo blu. Yes, i know this is <a class="reference" href="http://www.example.com">Stupid</a>.</p>
+</div>

product/PortalTransforms/tests/output/rest3.out

+<h2 class="title">Title</h2>
+<h3 class="subtitle">Subtitle</h3>
+<p>This is a test document to make sure subtitle gets the right heading.</p>
+<div class="section" id="now-the-real-heading">
+<h3><a name="now-the-real-heading">Now the real heading</a></h3>
+<p>The brown fox jumped over the lazy dog.</p>
+<div class="section" id="with-a-subheading">
+<h4><a name="with-a-subheading">With a subheading</a></h4>
+<p>Some text, bla ble bli blo blu. Yes, i know this is <a class="reference" href="http://www.example.com">Stupid</a>.</p>
+</div>
+</div>

product/PortalTransforms/tests/output/test_html_to_text.txt

+
+
+
+
+
+
+Copying Docutils
+
+
+
+
+
+ 
+ Copying Docutils
+ 
+
+
+
+Author:
+David Goodger
+Contact:
+goodger@users.sourceforge.net
+Date:
+2002-10-03
+Web site:http://docutils.sourceforge.net/
+
+
+
+ Most of the files included in this project are in the public domain,
+and therefore have no license requirement and no restrictions on
+copying or usage.  The exceptions are:
+ 
+docutils/optik.py, copyright Gregory P. Ward, released under a
+BSD-style license (which can be found in the module's source code).
+docutils/roman.py, copyright by Mark Pilgrim, released under the
+Python 2.1.1 license.
+test/difflib.py, copyright by the Python Software Foundation,
+released under the Python 2.2 license.  This file is included for
+compatibility with Python versions less than 2.2; if you have Python
+2.2 or higher, difflib.py is not needed and may be removed.  (It's
+only used to report test failures anyhow; it isn't installed
+anywhere.  The included file is a pre-generator version of the
+difflib.py module included in Python 2.2.)
+
+ (Disclaimer: I am not a lawyer.)  Both the BSD license and the Python
+license are OSI-approved and GPL-compatible.  Although complicated
+by multiple owners and lots of legalese, the Python license basically
+lets you copy, use, modify, and redistribute files as long as you keep
+the copyright attribution intact, note any changes you make, and don't
+use the owner's name in vain.  The BSD license is similar.
+
+ 
+ 
+Generated on: 2003-04-19 15:32 UTC.
+Generated by Docutils from reStructuredText source.
+
+
+

product/PortalTransforms/tests/output/test_lynx.txt

+Copying Docutils
+
+    Author:  David Goodger
+   Contact:  goodger@users.sourceforge.net
+     Date:   2002-10-03
+   Web site: http://docutils.sourceforge.net/
+
+   Most of the files included in this project are in the public domain,
+   and therefore have no license requirement and no restrictions on
+   copying or usage. The exceptions are:
+     * docutils/optik.py, copyright Gregory P. Ward, released under a
+       BSD-style license (which can be found in the module's source
+       code).
+     * docutils/roman.py, copyright by Mark Pilgrim, released under the
+       Python 2.1.1 license.
+     * test/difflib.py, copyright by the Python Software Foundation,
+       released under the Python 2.2 license. This file is included for
+       compatibility with Python versions less than 2.2; if you have
+       Python 2.2 or higher, difflib.py is not needed and may be removed.
+       (It's only used to report test failures anyhow; it isn't installed
+       anywhere. The included file is a pre-generator version of the
+       difflib.py module included in Python 2.2.)
+
+   (Disclaimer: I am not a lawyer.) Both the BSD license and the Python
+   license are OSI-approved and GPL-compatible. Although complicated by
+   multiple owners and lots of legalese, the Python license basically
+   lets you copy, use, modify, and redistribute files as long as you keep
+   the copyright attribution intact, note any changes you make, and don't
+   use the owner's name in vain. The BSD license is similar.
+     _________________________________________________________________
+
+   Generated on: 2003-04-19 15:32 UTC. Generated by Docutils from
+   reStructuredText source.

product/PortalTransforms/tests/output/test_python.out

+<pre class="python">
+<span style="color: #004080;">&quot;&quot;&quot; nice docstring &quot;&quot;&quot;</span>
+
+<span style="color: #C00000;">class</span> <span style="color: #000000;">A</span> <span style="color: #0000C0;">:</span> <span style="color: #C00000;">pass</span>
+
+<span style="color: #008000;"># comment
+</span>
+<span style="color: #C00000;">def</span> <span style="color: #000000;">inc</span><span style="color: #0000C0;">(</span><span style="color: #000000;">i</span><span style="color: #0000C0;">)</span><span style="color: #0000C0;">:</span>
+    <span style="color: #C00000;">return</span> <span style="color: #000000;">i</span><span style="color: #0000C0;">+</span><span style="color: #0080C0;">1</span>
+
+<span style="color: #C00000;">def</span> <span style="color: #000000;">greater</span><span style="color: #0000C0;">(</span><span style="color: #000000;">a</span><span style="color: #0000C0;">,</span> <span style="color: #000000;">b</span><span style="color: #0000C0;">)</span><span style="color: #0000C0;">:</span>
+    <span style="color: #004080;">&quot;&quot;&quot;foo &lt;html /&gt;&quot;&quot;&quot;</span>
+    <span style="color: #C00000;">return</span> <span style="color: #000000;">a</span> <span style="color: #0000C0;">&gt;</span> <span style="color: #000000;">b</span>
+</pre>

product/PortalTransforms/tests/output/test_safe.html

+<h1>Test page</h1>
+<table>
+<tr>
+<th>Test1</th>
+<td>test2</td>
+</tr>
+</table>
+<p>This is a text used as a blind text.</p>
+<ul>
+<li>A sample list item1</li>
+<li>A sample list item2</li>
+</ul>
+<p>This is again a blind text with a<br />line break.</p>
+<div>
+Can we <q>quote</q> or write something we <del>didn't</del> mean to write? Or how is <ins>this</ins> instead?
+</div>
+<hr />
+<div>
+<a href="http://www.plone.org"><img src="http://www.plone.org/logo.jpg" /></a> is just great.
+</div>
\ No newline at end of file

product/PortalTransforms/tests/output/test_word.html

+<br />
+
+
+
+<p><div name="Default" align="left" style="  padding: 0.00mm 0.00mm 0.00mm 0.00mm; "> 
+
+<p style="text-indent: 0.00mm; text-align: left; line-height: 4.166667mm; color: Black; background-color: White; ">
+how odd: blank named file in directory
+</p></div>
+

product/PortalTransforms/tests/runalltests.py

+#
+# Runs all tests in the current directory
+#
+# Execute like:
+#   python runalltests.py
+#
+# Alternatively use the testrunner: 
+#   python /path/to/Zope/utilities/testrunner.py -qa
+#
+
+import os, sys
+if __name__ == '__main__':
+    execfile(os.path.join(sys.path[0], 'framework.py')) 
+
+import unittest
+TestRunner = unittest.TextTestRunner
+suite = unittest.TestSuite()
+
+tests = os.listdir(os.curdir)
+tests = [n[:-3] for n in tests if n.startswith('test') and n.endswith('.py')]
+
+for test in tests:
+    m = __import__(test)
+    if hasattr(m, 'test_suite'):
+        suite.addTest(m.test_suite())
+
+if __name__ == '__main__':
+    TestRunner().run(suite)
+

product/PortalTransforms/tests/runtests.sh.example

+#!/bin/bash
+#
+# example test runner shell script
+#
+
+# full path to the python interpretor
+export PYTHON="/usr/local/bin/python2.3"
+
+# path to ZOPE_HOME/lib/python
+export SOFTWARE_HOME="/opt/zope/releases/Zope-2_7-branch/lib/python"
+
+
+# path to your instance. Don't set it if you aren't having  instance
+export INSTANCE_HOME="/opt/zope/instances/plone21/"
+
+
+${PYTHON} runalltests.py

product/PortalTransforms/tests/test_engine.py

+import os, sys
+if __name__ == '__main__':
+    execfile(os.path.join(sys.path[0], 'framework.py'))
+
+from Testing import ZopeTestCase
+from Products.Archetypes.tests.atsitetestcase import ATSiteTestCase
+
+from Products.PortalTransforms.utils import TransformException
+from Products.PortalTransforms.interfaces import *
+from Products.PortalTransforms.chain import chain
+
+import urllib
+import time
+import re
+
+class BaseTransform:
+    def name(self):
+        return getattr(self, '__name__', self.__class__.__name__)
+
+
+class HtmlToText(BaseTransform):
+    __implements__ = itransform
+    inputs = ('text/html',)
+    output = 'text/plain'
+
+    def __call__(self, orig, **kwargs):
+        orig = re.sub('<[^>]*>(?i)(?m)', '', orig)
+        return urllib.unquote(re.sub('\n+', '\n', orig)).strip()
+
+    def convert(self, orig, data, **kwargs):
+        orig = self.__call__(orig)
+        data.setData(orig)
+        return data
+
+class HtmlToTextWithEncoding(HtmlToText):
+    output_encoding = 'ascii'
+
+class FooToBar(BaseTransform):
+    __implements__ = itransform
+    inputs = ('text/*',)
+    output = 'text/plain'
+
+    def __call__(self, orig, **kwargs):
+        orig = re.sub('foo', 'bar', orig)
+        return urllib.unquote(re.sub('\n+', '\n', orig)).strip()
+
+    def convert(self, orig, data, **kwargs):
+        orig = self.__call__(orig)
+        data.setData(orig)
+        return data
+
+
+class TransformNoIO(BaseTransform):
+    __implements__ = itransform
+
+class BadTransformMissingImplements(BaseTransform):
+    __implements__ = None
+    inputs = ('text/*',)
+    output = 'text/plain'
+
+class BadTransformBadMIMEType1(BaseTransform):
+    __implements__ = itransform
+    inputs = ('truc/muche',)
+    output = 'text/plain'
+
+class BadTransformBadMIMEType2(BaseTransform):
+    __implements__ = itransform
+    inputs = ('text/plain',)
+    output = 'truc/muche'
+
+class BadTransformNoInput(BaseTransform):
+    __implements__ = itransform
+    inputs = ()
+    output = 'text/plain'
+
+class BadTransformWildcardOutput(BaseTransform):
+    __implements__ = itransform
+    inputs = ('text/plain',)
+    output = 'text/*'
+
+
+
+class TestEngine(ATSiteTestCase):
+
+    def afterSetUp(self):
+        ATSiteTestCase.afterSetUp(self)
+        self.engine = self.portal.portal_transforms
+        self.data = '<b>foo</b>'
+
+    def register(self):
+        #A default set of transforms to prove the interfaces work
+        self.engine.registerTransform(HtmlToText())
+        self.engine.registerTransform(FooToBar())
+
+    def testRegister(self):
+        self.register()
+
+    def testFailRegister(self):
+        register = self.engine.registerTransform
+        self.assertRaises(TransformException, register, TransformNoIO())
+        self.assertRaises(TransformException, register, BadTransformMissingImplements())
+        self.assertRaises(TransformException, register, BadTransformBadMIMEType1())
+        self.assertRaises(TransformException, register, BadTransformBadMIMEType2())
+        self.assertRaises(TransformException, register, BadTransformNoInput())
+        self.assertRaises(TransformException, register, BadTransformWildcardOutput())
+
+    def testCall(self):
+        self.register()
+        data = self.engine('HtmlToText', self.data)
+        self.failUnlessEqual(data, "foo")
+
+        data = self.engine('FooToBar', self.data)
+        self.failUnlessEqual(data, "<b>bar</b>")
+
+    def testConvert(self):
+        self.register()
+
+        data = self.engine.convert('HtmlToText', self.data)
+        self.failUnlessEqual(data.getData(), "foo")
+        self.failUnlessEqual(data.getMetadata()['mimetype'], 'text/plain')
+        self.failUnlessEqual(data.getMetadata().get('encoding'), None)
+        self.failUnlessEqual(data.name(), "HtmlToText")
+
+        self.engine.registerTransform(HtmlToTextWithEncoding())
+        data = self.engine.convert('HtmlToTextWithEncoding', self.data)
+        self.failUnlessEqual(data.getMetadata()['mimetype'], 'text/plain')
+        self.failUnlessEqual(data.getMetadata()['encoding'], 'ascii')
+        self.failUnlessEqual(data.name(), "HtmlToTextWithEncoding")
+
+    def testConvertTo(self):
+        self.register()
+
+        data = self.engine.convertTo('text/plain', self.data, mimetype="text/html")
+        self.failUnlessEqual(data.getData(), "foo")
+        self.failUnlessEqual(data.getMetadata()['mimetype'], 'text/plain')
+        self.failUnlessEqual(data.getMetadata().get('encoding'), None)
+        self.failUnlessEqual(data.name(), "text/plain")
+
+        self.engine.unregisterTransform('HtmlToText')
+        self.engine.unregisterTransform('FooToBar')
+        self.engine.registerTransform(HtmlToTextWithEncoding())
+        data = self.engine.convertTo('text/plain', self.data, mimetype="text/html")
+        self.failUnlessEqual(data.getMetadata()['mimetype'], 'text/plain')
+        # HtmlToTextWithEncoding. Now None is the right 
+        #self.failUnlessEqual(data.getMetadata()['encoding'], 'ascii')
+        # XXX the new algorithm is choosing html_to_text instead of 
+        self.failUnlessEqual(data.getMetadata()['encoding'], None)
+        self.failUnlessEqual(data.name(), "text/plain")
+
+    def testChain(self):
+        self.register()
+        hb = chain('hbar')
+        hb.registerTransform(HtmlToText())
+        hb.registerTransform(FooToBar())
+
+        self.engine.registerTransform(hb)
+        cache = self.engine.convert('hbar', self.data)
+        self.failUnlessEqual(cache.getData(), "bar")
+        self.failUnlessEqual(cache.name(), "hbar")
+
+    def testSame(self):
+        data = "This is a test"
+        mt = "text/plain"
+        out = self.engine.convertTo('text/plain', data, mimetype=mt)
+        self.failUnlessEqual(out.getData(), data)
+        self.failUnlessEqual(out.getMetadata()['mimetype'], 'text/plain')
+
+    def testCache(self):
+        data = "This is a test"
+        other_data = 'some different data'
+        mt = "text/plain"
+        self.engine.max_sec_in_cache = 20
+        out = self.engine.convertTo(mt, data, mimetype=mt, object=self)
+        self.failUnlessEqual(out.getData(), data, out.getData())
+        out = self.engine.convertTo(mt, other_data, mimetype=mt, object=self)
+        self.failUnlessEqual(out.getData(), data, out.getData())
+        self.engine.max_sec_in_cache = -1
+        out = self.engine.convertTo(mt, data, mimetype=mt, object=self)
+        self.failUnlessEqual(out.getData(), data, out.getData())
+        out = self.engine.convertTo(mt, other_data, mimetype=mt, object=self)
+        self.failUnlessEqual(out.getData(), other_data, out.getData())
+
+
+def test_suite():
+    from unittest import TestSuite, makeSuite
+    suite = TestSuite()
+    suite.addTest(makeSuite(TestEngine))
+    return suite
+
+if __name__ == '__main__':
+    framework()

product/PortalTransforms/tests/test_graph.py

+import os, sys
+if __name__ == '__main__':
+    execfile(os.path.join(sys.path[0], 'framework.py'))
+
+from Testing import ZopeTestCase
+from Products.Archetypes.tests.atsitetestcase import ATSiteTestCase
+
+from utils import input_file_path
+FILE_PATH = input_file_path("demo1.pdf")
+
+class TestGraph(ATSiteTestCase):
+
+    def afterSetUp(self):
+        ATSiteTestCase.afterSetUp(self)
+        self.engine = self.portal.portal_transforms
+
+    def testGraph(self):
+        ### XXX Local file and expected output
+        data = open(FILE_PATH, 'r').read()
+        out = self.engine.convertTo('text/plain', data, filename=FILE_PATH)
+        assert(out.getData())
+
+
+def test_suite():
+    from unittest import TestSuite, makeSuite
+    suite = TestSuite()
+    suite.addTest(makeSuite(TestGraph))
+    return suite
+
+if __name__ == '__main__':
+    framework()

product/PortalTransforms/tests/test_transforms.py

+from __future__ import nested_scopes
+
+import os, sys
+if __name__ == '__main__':
+    execfile(os.path.join(sys.path[0], 'framework.py'))
+
+from Testing import ZopeTestCase
+from Products.Archetypes.tests.atsitetestcase import ATSiteTestCase
+
+from utils import input_file_path, output_file_path, normalize_html,\
+     load, matching_inputs
+from Products.PortalTransforms.data import datastream
+from Products.PortalTransforms.interfaces import idatastream
+from Products.MimetypesRegistry.MimeTypesTool import MimeTypesTool
+from Products.PortalTransforms.TransformEngine import TransformTool
+
+from Products.PortalTransforms.libtransforms.utils import MissingBinary
+from Products.PortalTransforms.transforms.image_to_gif import image_to_gif
+from Products.PortalTransforms.transforms.image_to_png import image_to_png
+from Products.PortalTransforms.transforms.image_to_jpeg import image_to_jpeg
+from Products.PortalTransforms.transforms.image_to_bmp import image_to_bmp
+from Products.PortalTransforms.transforms.image_to_tiff import image_to_tiff
+from Products.PortalTransforms.transforms.image_to_ppm  import image_to_ppm
+from Products.PortalTransforms.transforms.image_to_pcx  import image_to_pcx
+
+from os.path import exists
+import sys
+# we have to set locale because lynx output is locale sensitive !
+os.environ['LC_ALL'] = 'C'
+
+class TransformTest(ATSiteTestCase):
+
+    def do_convert(self, filename=None):
+        if filename is None and exists(self.output + '.nofilename'):
+            output = self.output + '.nofilename'
+        else:
+            output = self.output
+        input = open(self.input)
+        orig = input.read()
+        input.close()
+        data = datastream(self.transform.name())
+        res_data = self.transform.convert(orig, data, filename=filename)
+        self.assert_(idatastream.isImplementedBy(res_data))
+        got = res_data.getData()
+        try:
+            output = open(output)
+        except IOError:
+            import sys
+            print >>sys.stderr, 'No output file found.'
+            print >>sys.stderr, 'File %s created, check it !' % self.output
+            output = open(output, 'w')
+            output.write(got)
+            output.close()
+            self.assert_(0)
+        expected = output.read()
+        if self.normalize is not None:
+            expected = self.normalize(expected)
+            got = self.normalize(got)
+        output.close()
+
+        self.assertEquals(got, expected,
+                          '[%s]\n\n!=\n\n[%s]\n\nIN %s(%s)' % (
+            got, expected, self.transform.name(), self.input))
+        self.assertEquals(self.subobjects, len(res_data.getSubObjects()),
+                          '%s\n\n!=\n\n%s\n\nIN %s(%s)' % (
+            self.subobjects, len(res_data.getSubObjects()), self.transform.name(), self.input))
+
+    def testSame(self):
+        self.do_convert(filename=self.input)
+
+    def testSameNoFilename(self):
+        self.do_convert()
+
+    def __repr__(self):
+        return self.transform.name()
+
+class PILTransformsTest(ATSiteTestCase):
+    def afterSetUp(self):
+        ATSiteTestCase.afterSetUp(self)
+        self.pt = self.portal.portal_transforms
+
+    def test_image_to_bmp(self):
+        self.pt.registerTransform(image_to_bmp())
+        imgFile = open(input_file_path('logo.jpg'), 'rb')
+        data = imgFile.read()
+        self.failUnlessEqual(self.portal.mimetypes_registry.classify(data),'image/jpeg')
+        data = self.pt.convertTo(target_mimetype='image/x-ms-bmp',orig=data)
+        self.failUnlessEqual(data.getMetadata()['mimetype'], 'image/x-ms-bmp')
+
+    def test_image_to_gif(self):
+        self.pt.registerTransform(image_to_gif())
+        imgFile = open(input_file_path('logo.png'), 'rb')
+        data = imgFile.read()
+        self.failUnlessEqual(self.portal.mimetypes_registry.classify(data),'image/png')
+        data = self.pt.convertTo(target_mimetype='image/gif',orig=data)
+        self.failUnlessEqual(data.getMetadata()['mimetype'], 'image/gif')
+
+    def test_image_to_jpeg(self):
+        self.pt.registerTransform(image_to_jpeg())
+        imgFile = open(input_file_path('logo.gif'), 'rb')
+        data = imgFile.read()
+        self.failUnlessEqual(self.portal.mimetypes_registry.classify(data),'image/gif')
+        data = self.pt.convertTo(target_mimetype='image/jpeg',orig=data)
+        self.failUnlessEqual(data.getMetadata()['mimetype'], 'image/jpeg')
+
+    def test_image_to_png(self):
+        self.pt.registerTransform(image_to_png())
+        imgFile = open(input_file_path('logo.jpg'), 'rb')
+        data = imgFile.read()
+        self.failUnlessEqual(self.portal.mimetypes_registry.classify(data),'image/jpeg')
+        data = self.pt.convertTo(target_mimetype='image/png',orig=data)
+        self.failUnlessEqual(data.getMetadata()['mimetype'], 'image/png')
+
+    def test_image_to_pcx(self):
+        self.pt.registerTransform(image_to_pcx())
+        imgFile = open(input_file_path('logo.gif'), 'rb')
+        data = imgFile.read()
+        self.failUnlessEqual(self.portal.mimetypes_registry.classify(data),'image/gif')
+        data = self.pt.convertTo(target_mimetype='image/pcx',orig=data)
+        self.failUnlessEqual(data.getMetadata()['mimetype'], 'image/pcx')
+
+    def test_image_to_ppm(self):
+        self.pt.registerTransform(image_to_ppm())
+        imgFile = open(input_file_path('logo.png'), 'rb')
+        data = imgFile.read()
+        self.failUnlessEqual(self.portal.mimetypes_registry.classify(data),'image/png')
+        data = self.pt.convertTo(target_mimetype='image/x-portable-pixmap',orig=data)
+        self.failUnlessEqual(data.getMetadata()['mimetype'], 'image/x-portable-pixmap')
+
+    def test_image_to_tiff(self):
+        self.pt.registerTransform(image_to_tiff())
+        imgFile = open(input_file_path('logo.jpg'), 'rb')
+        data = imgFile.read()
+        self.failUnlessEqual(self.portal.mimetypes_registry.classify(data),'image/jpeg')
+        data = self.pt.convertTo(target_mimetype='image/tiff',orig=data)
+        self.failUnlessEqual(data.getMetadata()['mimetype'], 'image/tiff')
+
+
+TRANSFORMS_TESTINFO = (
+    ('Products.PortalTransforms.transforms.pdf_to_html',
+     "demo1.pdf", "demo1.html", None, 0
+     ),
+    ('Products.PortalTransforms.transforms.word_to_html',
+     "test.doc", "test_word.html", normalize_html, 0
+     ),
+    ('Products.PortalTransforms.transforms.lynx_dump',
+     "test_lynx.html", "test_lynx.txt", None, 0
+     ),
+    ('Products.PortalTransforms.transforms.html_to_text',
+     "test_lynx.html", "test_html_to_text.txt", None, 0
+     ),
+    ('Products.PortalTransforms.transforms.identity',
+     "rest1.rst", "rest1.rst", None, 0
+     ),
+    ('Products.PortalTransforms.transforms.text_to_html',
+     "rest1.rst", "rest1.html", None, 0
+     ),
+    ('Products.PortalTransforms.transforms.safe_html',
+     "test_safehtml.html", "test_safe.html", None, 0
+     ),
+    ('Products.PortalTransforms.transforms.image_to_bmp',
+     "logo.jpg", "logo.bmp", None, 0
+     ),
+    ('Products.PortalTransforms.transforms.image_to_gif',
+     "logo.bmp", "logo.gif", None, 0
+     ),
+    ('Products.PortalTransforms.transforms.image_to_jpeg',
+     "logo.gif", "logo.jpg", None, 0
+     ),
+    ('Products.PortalTransforms.transforms.image_to_png',
+     "logo.bmp", "logo.png", None, 0
+     ),
+    ('Products.PortalTransforms.transforms.image_to_ppm',
+     "logo.gif", "logo.ppm", None, 0
+     ),
+    ('Products.PortalTransforms.transforms.image_to_tiff',
+     "logo.png", "logo.tiff", None, 0
+     ),
+    ('Products.PortalTransforms.transforms.image_to_pcx',
+     "logo.png", "logo.pcx", None, 0
+     ),
+    )
+
+def initialise(transform, normalize, pattern):
+    global TRANSFORMS_TESTINFO
+    for fname in matching_inputs(pattern):
+        outname = '%s.out' % fname.split('.')[0]
+        #print transform, fname, outname
+        TRANSFORMS_TESTINFO += ((transform, fname, outname, normalize, 0),)
+
+
+# ReST test cases
+initialise('Products.PortalTransforms.transforms.rest', normalize_html, "rest*.rst")
+# Python test cases
+initialise('Products.PortalTransforms.transforms.python', normalize_html, "*.py")
+
+# FIXME missing tests for image_to_html, st
+
+TR_NAMES = None
+
+def make_tests(test_descr=TRANSFORMS_TESTINFO):
+    """generate tests classes from test info
+
+    return the list of generated test classes
+    """
+    tests = []
+    for _transform, tr_input, tr_output, _normalize, _subobjects in test_descr:
+        # load transform if necessary
+        if type(_transform) is type(''):
+            try:
+                _transform = load(_transform).register()
+            except MissingBinary:
+                # we are not interessted in tests with missing binaries
+                continue
+            except:
+                import traceback
+                traceback.print_exc()
+                continue
+
+        if TR_NAMES is not None and not _transform.name() in TR_NAMES:
+            print 'skip test for', _transform.name()
+            continue
+
+        class TransformTestSubclass(TransformTest):
+            input = input_file_path(tr_input)
+            output = output_file_path(tr_output)
+            transform = _transform
+            normalize = lambda x, y: _normalize(y)
+            subobjects = _subobjects
+
+        tests.append(TransformTestSubclass)
+
+    tests.append(PILTransformsTest)
+    return tests
+
+
+def test_suite():
+    from unittest import TestSuite, makeSuite
+    suite = TestSuite()
+    for test in make_tests():
+        suite.addTest(makeSuite(test))
+    return suite
+
+if __name__ == '__main__':
+    framework()

product/PortalTransforms/tests/utils.py

+import re
+import glob
+from unittest import TestSuite
+from sys import modules
+from os.path import join, abspath, dirname, basename
+
+def normalize_html(s):
+    s = re.sub(r"\s+", " ", s)
+    s = re.sub(r"(?s)\s+<", "<", s)
+    s = re.sub(r"(?s)>\s+", ">", s)
+    s = re.sub(r"\r", "", s)
+    return s
+
+def build_test_suite(package_name,module_names,required=1):
+    """
+    Utlitity for building a test suite from a package name
+    and a list of modules.
+
+    If required is false, then ImportErrors will simply result
+    in that module's tests not being added to the returned
+    suite.
+    """
+
+    suite = TestSuite()
+    try:
+        for name in module_names:
+            the_name = package_name+'.'+name
+            __import__(the_name,globals(),locals())
+            suite.addTest(modules[the_name].test_suite())
+    except ImportError:
+        if required:
+            raise
+    return suite
+
+PREFIX = abspath(dirname(__file__))
+
+def input_file_path(file):
+    return join(PREFIX, 'input', file)
+
+def output_file_path(file):
+    return join(PREFIX, 'output', file)
+
+def matching_inputs(pattern):
+    return [basename(path) for path in glob.glob(join(PREFIX, "input", pattern))]
+
+def load(dotted_name, globals=None):
+    """ load a python module from it's name """
+    mod = __import__(dotted_name, globals)
+    components = dotted_name.split('.')
+    for comp in components[1:]:
+        mod = getattr(mod, comp)
+    return mod

product/PortalTransforms/tests/word.py

+from rigging import transformer
+
+import os
+from stat import ST_MTIME
+
+
+## BIG BAD FUNCTIONAL TEST OF OOo Word Conversion
+## The interfaces work, but are not quite what we need
+## I might have to back fill a chain from source/dest graphing
+
+file = "/tmp/word.doc"
+
+class curry:
+    def __init__(self, func, *fixed_args):
+        self.func = func
+        self.fixed_args = fixed_args
+
+    def __call__(self, *variable_args):
+        return apply(self.func, self.fixed_args +
+          variable_args)
+
+data = open("/tmp/word.doc", "r").read()
+
+data = transformer.convert("WordToHtml", data, filename="word.doc")
+print data.getData()

product/PortalTransforms/transforms/__init__.py

+### Register Transforms
+### This is interesting because we don't expect all transforms to be
+### available on all platforms. To do this we allow things to fail at
+### two levels
+### 1) Imports
+###    If the import fails the module is removed from the list and
+###    will not be processed/registered
+### 2) Registration
+###    A second phase happens when the loaded modules register method
+###    is called and this produces an instance that will used to
+###    implement the transform, if register needs to fail for now it
+###    should raise an ImportError as well (dumb, I know)
+
+from logging import DEBUG, ERROR
+from Products.PortalTransforms.utils import log
+from Products.PortalTransforms.libtransforms.utils import MissingBinary
+modules = [
+    'st',             # zopish
+    'rest',           # docutils
+    'word_to_html',   # uno, com, wvware
+    'safe_html',      # extract <body> and remove potentially harmful tags
+    'html_body',      # extract only the contents of the <body> tag
+    'html_to_text',   # re based transform
+    'text_to_html',   # wrap text in a verbatim env
+    'text_pre_to_html', # wrap text into a pre
+    'pdf_to_html',    # sf.net/projects/pdftohtml
+    'pdf_to_text',    # www.foolabs.com/xpdf
+    'rtf_to_html',    # sf.net/projects/rtf-converter
+    'rtf_to_xml',     # sf.net/projects/rtf2xml
+    'image_to_png',   # transforms any image to a PNG image
+    'image_to_gif',   # transforms any image to a GIF image
+    'image_to_jpeg',  # transforms any image to a JPEG image
+    'image_to_pcx',   # transforms any image to a PCX image
+    'image_to_ppm',   # transforms any image to a PPM image
+    'image_to_tiff',  # transforms any image to a TIFF image
+    'image_to_bmp',   # transforms any image to a BMP image
+    'lynx_dump',      # lynx -dump
+    'python',         # python source files, no dependancies
+    'identity',       # identity transform, no dependancies
+    ]
+
+g = globals()
+transforms = []
+for m in modules:
+    try:
+        ns = __import__(m, g, g, None)
+        transforms.append(ns.register())
+    except ImportError, e:
+        msg = "Problem importing module %s : %s" % (m, e)
+        log(msg, severity=ERROR)
+    except MissingBinary, e:
+        log(str(e), severity=DEBUG)
+    except Exception, e:
+        import traceback
+        traceback.print_exc()
+        log("Raised error %s for %s" % (e, m), severity=ERROR)
+
+def initialize(engine):
+    for transform in transforms:
+        engine.registerTransform(transform)

product/PortalTransforms/transforms/broken.py

+from Products.PortalTransforms.interfaces import itransform
+from Products.PortalTransforms.utils import log
+WARNING=100
+
+class BrokenTransform:
+    __implements__ = itransform
+
+    __name__ = "broken transform"
+    inputs  = ("BROKEN",)
+    output = "BROKEN"
+    
+    def __init__(self, id, module, error):
+        self.id = id
+        self.module = module
+        self.error = error
+
+    def name(self):
+        return self.__name__
+
+    def convert(self, orig, data, **kwargs):
+        # do the format
+        msg = "Calling convert on BROKEN transform %s (%s). Error: %s" % \
+              (self.id, self.module, self.error)
+        log(msg, severity=WARNING)
+        print msg
+        data.setData('')
+        return data
+
+def register():
+    return broken()

product/PortalTransforms/transforms/html_body.py

+from Products.PortalTransforms.interfaces import itransform
+from Products.CMFDefault.utils import bodyfinder
+
+class HTMLBody:
+    """Simple transform which extracts the content of the body tag"""
+
+    __implements__ = itransform
+
+    __name__ = "html_body"
+    inputs   = ('text/html',)
+    output = "text/html"
+
+    def __init__(self, name=None):
+        self.config_metadata = {
+            'inputs' : ('list', 'Inputs', 'Input(s) MIME type. Change with care.'),
+            }
+        if name:
+            self.__name__ = name
+
+    def name(self):
+        return self.__name__
+
+    def __getattr__(self, attr):
+        if attr == 'inputs':
+            return self.config['inputs']
+        if attr == 'output':
+            return self.config['output']
+        raise AttributeError(attr)
+
+    def convert(self, orig, data, **kwargs):
+        body = bodyfinder(orig)
+        data.setData(body)
+        return data
+
+def register():
+    return HTMLBody()

product/PortalTransforms/transforms/html_to_text.py

+from Products.PortalTransforms.libtransforms.retransform import retransform
+
+class html_to_text(retransform):
+    inputs  = ('text/html',)
+    output = 'text/plain'
+
+def register():
+    # XXX convert entites with htmlentitydefs.name2codepoint ?
+    return html_to_text("html_to_text",
+                       ('<script [^>]>.*</script>(?im)', ''),
+                       ('<style [^>]>.*</style>(?im)', ''),
+                       ('<head [^>]>.*</head>(?im)', ''),
+                       ('(?im)<(h[1-6r]|address|p|ul|ol|dl|pre|div|center|blockquote|form|isindex|table)(?=\W)[^>]*>', ' '),
+                       ('<[^>]*>(?i)(?m)', ''),
+                       )

product/PortalTransforms/transforms/identity.py

+"""
+A simple identity transform
+"""
+
+__revision__ = '$Id: identity.py 4787 2005-08-19 21:43:41Z dreamcatcher $'
+
+from Products.PortalTransforms.interfaces import itransform
+
+class IdentityTransform:
+    """ Identity transform
+
+    return content unchanged.
+    """
+    __implements__ = (itransform,)
+
+    __name__ = "rest_to_text"
+
+    def __init__(self, name=None, **kwargs):
+        self.config = {
+            'inputs'       : ('text/x-rst',),
+            'output'      : 'text/plain',
+            }
+        self.config_metadata = {
+            'inputs'       : ('list', 'Inputs', 'Input(s) MIME type. Change with care.'),
+            'output'      : ('string', 'Output', 'Output MIME type. Change with care.'),
+            }
+        self.config.update(kwargs)
+
+    def __getattr__(self, attr):
+        if attr == 'inputs':
+            return self.config['inputs']
+        if attr == 'output':
+            return self.config['output']
+        raise AttributeError(attr)
+
+    def name(self):
+        return self.__name__
+
+    def convert(self, data, cache, **kwargs):
+        cache.setData(data)
+        return cache
+
+def register():
+    return IdentityTransform()

product/PortalTransforms/transforms/image_to_bmp.py

+from Products.PortalTransforms.libtransforms.piltransform import PILTransforms
+
+class image_to_bmp(PILTransforms):
+    __name__  = "image_to_bmp"
+    inputs    = ('image/*', )
+    output   = 'image/x-ms-bmp'
+    format  = 'bmp'
+
+
+def register():
+    return image_to_bmp()

product/PortalTransforms/transforms/image_to_gif.py

+from Products.PortalTransforms.libtransforms.piltransform import PILTransforms
+
+class image_to_gif(PILTransforms):
+    __name__  = "image_to_gif"
+    inputs    = ('image/*', )
+    output   = 'image/gif'
+    format  = 'gif'
+
+
+def register():
+    return image_to_gif()

product/PortalTransforms/transforms/image_to_html.py

+from Products.PortalTransforms.interfaces import itransform
+
+class image_to_html:
+    __implements__ = itransform
+
+    __name__  = "image_to_html"
+    inputs    = ('image/*', )
+    output   = 'text/html'
+
+    def name(self):
+        return self.__name__
+
+    def convert(self, data, cache, **kwargs):
+        imageName = kwargs.get("image")
+        cache.setData('<img src="%s"/>' %imageName)
+        return cache

product/PortalTransforms/transforms/image_to_jpeg.py

+from Products.PortalTransforms.libtransforms.piltransform import PILTransforms
+
+class image_to_jpeg(PILTransforms):
+    __name__  = "image_to_jpeg"
+    inputs    = ('image/*', )
+    output   = 'image/jpeg'
+    format  = 'jpeg'
+
+
+def register():
+    return image_to_jpeg()

product/PortalTransforms/transforms/image_to_pcx.py

+from Products.PortalTransforms.libtransforms.piltransform import PILTransforms
+
+class image_to_pcx(PILTransforms):
+    __name__  = "image_to_pcx"
+    inputs    = ('image/*', )
+    output   = 'image/pcx'
+    format  = 'pcx'
+
+
+def register():
+    return image_to_pcx()

product/PortalTransforms/transforms/image_to_png.py

+from Products.PortalTransforms.libtransforms.piltransform import PILTransforms
+
+class image_to_png(PILTransforms):
+    __name__  = "image_to_png"
+    inputs    = ('image/*', )
+    output   = 'image/png'
+    format  = 'png'
+
+def register():
+    return image_to_png()

product/PortalTransforms/transforms/image_to_ppm.py

+from Products.PortalTransforms.libtransforms.piltransform import PILTransforms
+
+class image_to_ppm(PILTransforms):
+    __name__  = "image_to_ppm"
+    inputs    = ('image/*', )
+    output   = 'image/x-portable-pixmap'
+    format  = 'ppm'
+
+
+def register():
+    return image_to_ppm()

product/PortalTransforms/transforms/image_to_tiff.py

+from Products.PortalTransforms.libtransforms.piltransform import PILTransforms
+
+class image_to_tiff(PILTransforms):
+    __name__  = "image_to_tiff"
+    inputs    = ('image/*', )
+    output   = 'image/tiff'
+    format  = 'tiff'
+
+
+def register():
+    return image_to_tiff()

product/PortalTransforms/transforms/lynx_dump.py

+"""
+Uses lynx -dump
+"""
+from Products.PortalTransforms.interfaces import itransform
+from Products.PortalTransforms.libtransforms.commandtransform import commandtransform
+from Products.PortalTransforms.libtransforms.commandtransform import popentransform
+import os
+
+class lynx_dump(popentransform):
+    __implements__ = itransform
+
+    __name__ = "lynx_dump"
+    inputs   = ('text/html',)
+    output  = 'text/plain'
+    
+    __version__ = '2004-07-02.1'
+
+    binaryName = "lynx"
+    # XXX does -stdin work on windows?
+    binaryArgs = "-dump -crawl -stdin"
+    useStdin = True
+    
+    def getData(self, couterr):
+        lines = [ line for line in couterr.readlines() ]
+        return ''.join(lines[3:])
+
+class old_lynx_dump(commandtransform):
+    __implements__ = itransform
+
+    __name__ = "lynx_dump"
+    inputs   = ('text/html',)
+    output  = 'text/plain'
+
+    binaryName = "lynx"
+    binaryArgs = "-dump"
+
+    def __init__(self):
+        commandtransform.__init__(self, binary=self.binaryName)
+
+    def convert(self, data, cache, **kwargs):
+        kwargs['filename'] = 'unknown.html'
+        tmpdir, fullname = self.initialize_tmpdir(data, **kwargs)
+        outname = "%s/%s.txt" % (tmpdir, orig_name)
+        self.invokeCommand(tmpdir, fullname, outname)
+        text = self.astext(outname)
+        self.cleanDir(tmpdir)
+        cache.setData(text)
+        return cache
+
+    def invokeCommand(self, tmpdir, inputname, outname):
+        os.system('cd "%s" && %s %s "%s" 1>"%s" 2>/dev/null' % \
+               (tmpdir, self.binary, self.binaryArgs, inputname, outname))
+
+    def astext(self, outname):
+        txtfile = open("%s" % (outname), 'r')
+        txt = txtfile.read()
+        txtfile.close()
+        return txt
+
+def register():
+    return lynx_dump()

product/PortalTransforms/transforms/office_com.py

+import win32com, sys, string, win32api, traceback, re, tempfile, os
+import win32com.client
+# from win32com.test.util import CheckClean
+import pythoncom
+from win32com.client import gencache
+from win32com.client import constants, Dispatch
+from pywintypes import Unicode
+import os.path
+
+from Products.PortalTransforms.libtransforms.commandtransform import commandtransform
+from Products.PortalTransforms.libtransforms.utils import bodyfinder, scrubHTML
+
+class document(commandtransform):
+
+    def __init__(self, name, data):
+        """Initialization: create tmp work
+        directory and copy the document into a file"""
+        commandtransform.__init__(self, name)
+        name = self.name()
+        if not name.endswith('.doc'):
+            name = name + ".doc"
+        self.tmpdir, self.fullname = self.initialize_tmpdir(data, filename=name)
+
+    def convert(self):
+        try:
+            # initialize COM for multi-threading, ignoring any errors
+            # when someone else has already initialized differently.
+            pythoncom.CoInitializeEx(pythoncom.COINIT_MULTITHREADED)
+        except pythoncom.com_error:
+            pass
+
+        word = Dispatch("Word.Application")
+        word.Visible = 0
+        word.DisplayAlerts = 0
+        doc = word.Documents.Open(self.fullname)
+        # Let's set up some html saving options for this document
+        doc.WebOptions.RelyOnCSS = 1
+        doc.WebOptions.OptimizeForBrowser = 1
+        doc.WebOptions.BrowserLevel = 0 # constants.wdBrowserLevelV4
+        doc.WebOptions.OrganizeInFolder = 0
+        doc.WebOptions.UseLongFileNames = 1
+        doc.WebOptions.RelyOnVML = 0
+        doc.WebOptions.AllowPNG = 1
+        # And then save the document into HTML
+        doc.SaveAs(FileName = "%s.htm" % (self.fullname),
+                   FileFormat = 8) # constants.wdFormatHTML)
+
+        # TODO -- Extract Metadata (author, title, keywords) so we
+        # can populate the dublin core
+        # Converter will need to be extended to return a dict of
+        # possible MD fields
+
+        doc.Close()
+        # word.Quit()
+
+    def html(self):
+        htmlfile = open(self.fullname + '.htm', 'r')
+        html = htmlfile.read()
+        htmlfile.close()
+        html = scrubHTML(html)
+        body = bodyfinder(html)
+        return body
+
+## This function has to be done. It's more difficult to delete the temp
+## directory under Windows, because there is sometimes a directory in it.
+##    def cleanDir(self, tmpdir):

product/PortalTransforms/transforms/office_uno.py

+import re, os, tempfile
+import uno
+import unohelper
+from com.sun.star.beans import PropertyValue
+
+from Products.PortalTransforms.libtransforms.commandtransform import commandtransform
+from Products.PortalTransforms.libtransforms.utils import bodyfinder, scrubHTML
+
+class document(commandtransform):
+
+    def __init__(self, name, data):
+        """ Initialization: create tmp work directory and copy the
+        document into a file"""
+ 
+        commandtransform.__init__(self, name)
+        name = self.name()
+        self.tmpdir, self.fullname = self.initialize_tmpdir(data, filename=name)
+        self.outputfile = self.fullname + '.html'
+
+    def convert(self):
+        """Convert the document"""
+
+        localContext = uno.getComponentContext()
+        resolver = localContext.ServiceManager.createInstanceWithContext(
+                       'com.sun.star.bridge.UnoUrlResolver', localContext )
+        ctx = resolver.resolve(
+                       'uno:socket,host=localhost,port=2002;'
+                       'urp;StarOffice.ComponentContext')
+        smgr = ctx.ServiceManager
+        desktop = smgr.createInstanceWithContext(
+                       'com.sun.star.frame.Desktop', ctx)
+
+        # load the document
+        url = unohelper.systemPathToFileUrl(self.fullname)
+        doc = desktop.loadComponentFromURL(url, '_blank', 0, ())
+
+        filterName = 'swriter: HTML (StarWriter)'
+        storeProps = (
+                       PropertyValue('FilterName', 0, filterName, 0),
+                      )
+
+        # pre-create a empty file for security reason
+        url = unohelper.systemPathToFileUrl(self.outputfile)
+        doc.storeToURL(url, storeProps)
+
+        try:
+            doc.close(True)
+        except com.sun.star.util.CloseVetoException:
+            pass
+
+        # maigic to release some resource
+        ctx.ServiceManager
+
+    def html(self):
+        htmlfile = open(self.outputfile, 'r')
+        html = htmlfile.read()
+        htmlfile.close()
+        html = scrubHTML(html)
+        body = bodyfinder(html)
+        return body

product/PortalTransforms/transforms/office_wvware.py

+import re, tempfile
+import os, os.path
+from Products.PortalTransforms.libtransforms.utils import bin_search, \
+     sansext, bodyfinder, scrubHTML
+from Products.PortalTransforms.libtransforms.commandtransform import commandtransform
+
+class document(commandtransform):
+
+    def __init__(self, name, data):
+        """ Initialization: create tmp work directory and copy the
+        document into a file"""
+        commandtransform.__init__(self, name, binary="wvHtml")
+        name = self.name()
+        if not name.endswith('.doc'):
+            name = name + ".doc"
+        self.tmpdir, self.fullname = self.initialize_tmpdir(data, filename=name)
+
+    def convert(self):
+        "Convert the document"
+        tmpdir = self.tmpdir
+
+        # for windows, install wvware from GnuWin32 at C:\Program Files\GnuWin32\bin
+        # you can use:
+        # wvware.exe -c ..\share\wv\wvHtml.xml --charset=utf-8 -d d:\temp d:\temp\test.doc > test.html
+
+        if os.name == 'posix':
+            os.system('cd "%s" && %s --charset=utf-8 "%s" "%s.html"' % (tmpdir, self.binary,
+                                                             self.fullname,
+                                                             self.__name__))
+
+    def html(self):
+        htmlfile = open("%s/%s.html" % (self.tmpdir, self.__name__), 'r')
+        html = htmlfile.read()
+        htmlfile.close()
+        html = scrubHTML(html)
+        body = bodyfinder(html)
+        return body

product/PortalTransforms/transforms/pdf_to_html.py

+"""
+Uses the http://sf.net/projects/pdftohtml bin to do its handy work
+
+"""
+from Products.PortalTransforms.interfaces import itransform
+from Products.PortalTransforms.libtransforms.utils import bin_search, sansext
+from Products.PortalTransforms.libtransforms.commandtransform import commandtransform
+from Products.PortalTransforms.libtransforms.commandtransform import popentransform
+from Products.CMFDefault.utils import bodyfinder
+import os
+
+class popen_pdf_to_html(popentransform):
+    __implements__ = itransform
+    
+    __version__ = '2004-07-02.01'
+
+    __name__ = "pdf_to_html"
+    inputs   = ('application/pdf',)
+    output  = 'text/html'
+    output_encoding = 'utf-8'
+
+    binaryName = "pdftohtml"
+    binaryArgs = "%(infile)s -noframes -stdout -enc UTF-8"
+    useStdin = False
+
+    def getData(self, couterr):
+        return bodyfinder(couterr.read())
+
+class pdf_to_html(commandtransform):
+    __implements__ = itransform
+
+    __name__ = "pdf_to_html"
+    inputs   = ('application/pdf',)
+    output  = 'text/html'
+    output_encoding = 'utf-8'
+
+    binaryName = "pdftohtml"
+    binaryArgs = "-noframes -enc UTF-8"
+
+    def __init__(self):
+        commandtransform.__init__(self, binary=self.binaryName)
+
+    def convert(self, data, cache, **kwargs):
+        kwargs['filename'] = 'unknown.pdf'
+
+        tmpdir, fullname = self.initialize_tmpdir(data, **kwargs)
+        html = self.invokeCommand(tmpdir, fullname)
+        path, images = self.subObjects(tmpdir)
+        objects = {}
+        if images:
+            self.fixImages(path, images, objects)
+        self.cleanDir(tmpdir)
+        cache.setData(bodyfinder(html))
+        cache.setSubObjects(objects)
+        return cache
+
+    def invokeCommand(self, tmpdir, fullname):
+        if os.name=='posix':
+            cmd = 'cd "%s" && %s %s "%s" 2>error_log 1>/dev/null' % (
+                   tmpdir, self.binary, self.binaryArgs, fullname)
+        else:
+            cmd = 'cd "%s" && %s %s "%s"' % (
+                  tmpdir, self.binary, self.binaryArgs, fullname)
+        os.system(cmd)
+        try:
+            htmlfilename = os.path.join(tmpdir, sansext(fullname) + '.html')
+            htmlfile = open(htmlfilename, 'r')
+            html = htmlfile.read()
+            htmlfile.close()
+        except:
+            try:
+                return open("%s/error_log" % tmpdir, 'r').read()
+            except:
+                return "transform failed while running %s (maybe this pdf file doesn't support transform)" % cmd
+        return html
+
+def register():
+    return pdf_to_html()

product/PortalTransforms/transforms/pdf_to_text.py

+"""
+Uses the xpdf (www.foolabs.com/xpdf)
+"""
+
+from Products.PortalTransforms.interfaces import itransform
+from Products.PortalTransforms.libtransforms.utils import bin_search, sansext
+from Products.PortalTransforms.libtransforms.commandtransform import commandtransform
+from Products.PortalTransforms.libtransforms.commandtransform import popentransform
+import os
+
+class pdf_to_text(popentransform):
+    __implements__ = itransform
+
+    __name__ = "pdf_to_text"
+    inputs   = ('application/pdf',)
+    output  = 'text/plain'
+    output_encoding = 'utf-8'
+    
+    __version__ = '2004-07-02.01'
+
+    binaryName = "pdftotext"
+    binaryArgs = "%(infile)s -enc UTF-8 -"
+    useStdin = False
+
+class old_pdf_to_text(commandtransform):
+    __implements__ = itransform
+
+    __name__ = "pdf_to_text"
+    inputs   = ('application/pdf',)
+    output  = 'text/plain'
+    output_encoding = 'utf-8'
+
+    binaryName = "pdftotext"
+
+    def __init__(self):
+        commandtransform.__init__(self, binary=self.binaryName)
+
+    def convert(self, data, cache, **kwargs):
+        kwargs['filename'] = 'unkown.pdf'
+
+        tmpdir, fullname = self.initialize_tmpdir(data, **kwargs)
+        text = self.invokeCommand(tmpdir, fullname)
+        path, images = self.subObjects(tmpdir)
+        objects = {}
+        if images:
+            self.fixImages(path, images, objects)
+        self.cleanDir(tmpdir)
+        cache.setData(text)
+        cache.setSubObjects(objects)
+        return cache
+
+    def invokeCommand(self, tmpdir, fullname):
+        # FIXME: windows users...
+        textfile = "%s/%s.txt" % (tmpdir, sansext(fullname))
+        cmd = 'cd "%s" && %s -enc UTF-8 "%s" "%s" 2>error_log 1>/dev/null' % (
+            tmpdir, self.binary, fullname, textfile)
+        os.system(cmd)
+        try:
+            text = open(textfile).read()
+        except:
+            try:
+                return open("%s/error_log" % tmpdir, 'r').read()
+            except:
+                return ''
+        return text
+
+def register():
+    return pdf_to_text()

product/PortalTransforms/transforms/python.py

+"""
+Original code from active state recipe
+        'Colorize Python source using the built-in tokenizer'
+
+----------------------------------------------------------------------------
+     MoinMoin - Python Source Parser
+
+ This code is part of MoinMoin (http://moin.sourceforge.net/) and converts
+ Python source code to HTML markup, rendering comments, keywords, operators,
+ numeric and string literals in different colors.
+
+ It shows how to use the built-in keyword, token and tokenize modules
+ to scan Python source code and re-emit it with no changes to its
+ original formatting (which is the hard part).
+"""
+__revision__ = '$Id: python.py 3661 2005-02-23 17:05:31Z tiran $'
+
+import string
+import keyword, token, tokenize
+from cStringIO import StringIO
+
+from Products.PortalTransforms.interfaces import itransform
+from DocumentTemplate.DT_Util import html_quote
+
+## Python Source Parser #####################################################
+
+_KEYWORD = token.NT_OFFSET + 1
+_TEXT    = token.NT_OFFSET + 2
+
+class Parser:
+    """ Send colored python source.
+    """
+
+    def __init__(self, raw, tags, out):
+        """ Store the source text.
+        """
+        self.raw = string.strip(string.expandtabs(raw))
+        self.out = out
+        self.tags = tags
+
+    def format(self):
+        """ Parse and send the colored source.
+        """
+        # store line offsets in self.lines
+        self.lines = [0, 0]
+        pos = 0
+        while 1:
+            pos = string.find(self.raw, '\n', pos) + 1
+            if not pos: break
+            self.lines.append(pos)
+        self.lines.append(len(self.raw))
+
+        # parse the source and write it
+        self.pos = 0
+        text = StringIO(self.raw)
+        self.out.write('<pre class="python">\n')
+        try:
+            tokenize.tokenize(text.readline, self)
+        except tokenize.TokenError, ex:
+            msg = ex[0]
+            line = ex[1][0]
+            self.out.write("<h5 class='error>'ERROR: %s%s</h5>" % (
+                msg, self.raw[self.lines[line]:]))
+        self.out.write('\n</pre>\n')
+
+    def __call__(self, toktype, toktext, (srow,scol), (erow,ecol), line):
+        """ Token handler.
+        """
+        #print "type", toktype, token.tok_name[toktype], "text", toktext,
+        #print "start", srow,scol, "end", erow,ecol, "<br>"
+
+        ## calculate new positions
+        oldpos = self.pos
+        newpos = self.lines[srow] + scol
+        self.pos = newpos + len(toktext)
+
+        ## handle newlines
+        if toktype in [token.NEWLINE, tokenize.NL]:
+            self.out.write('\n')
+            return
+
+        ## send the original whitespace, if needed
+        if newpos > oldpos:
+            self.out.write(self.raw[oldpos:newpos])
+
+        ## skip indenting tokens
+        if toktype in [token.INDENT, token.DEDENT]:
+            self.pos = newpos
+            return
+
+        ## map token type to a group
+        if token.LPAR <= toktype and toktype <= token.OP:
+            toktype = 'OP'
+        elif toktype == token.NAME and keyword.iskeyword(toktext):
+            toktype = 'KEYWORD'
+        else:
+            toktype = tokenize.tok_name[toktype]
+
+        open_tag = self.tags.get('OPEN_'+toktype, self.tags['OPEN_TEXT'])
+        close_tag = self.tags.get('CLOSE_'+toktype, self.tags['CLOSE_TEXT'])
+
+        ## send text
+        self.out.write(open_tag)
+        self.out.write(html_quote(toktext))
+        self.out.write(close_tag)
+
+
+
+class PythonTransform:
+    """Colorize Python source files
+    """
+    __implements__ = itransform
+
+    __name__ = "python_to_html"
+    inputs  = ("text/x-python",)
+    output = "text/html"
+
+    config = {
+        'OPEN_NUMBER':       '<span style="color: #0080C0;">',
+        'CLOSE_NUMBER':      '</span>',
+        'OPEN_OP':           '<span style="color: #0000C0;">',
+        'CLOSE_OP':          '</span>',
+        'OPEN_STRING':       '<span style="color: #004080;">',
+        'CLOSE_STRING':      '</span>',
+        'OPEN_COMMENT':      '<span style="color: #008000;">',
+        'CLOSE_COMMENT':      '</span>',
+        'OPEN_NAME':         '<span style="color: #000000;">',
+        'CLOSE_NAME':        '</span>',
+        'OPEN_ERRORTOKEN':   '<span style="color: #FF8080;">',
+        'CLOSE_ERRORTOKEN':  '</span>',
+        'OPEN_KEYWORD':      '<span style="color: #C00000;">',
+        'CLOSE_KEYWORD':     '</span>',
+        'OPEN_TEXT':         '',
+        'CLOSE_TEXT':        '',
+        }
+
+    def name(self):
+        return self.__name__
+
+    def convert(self, orig, data, **kwargs):
+        dest = StringIO()
+        Parser(orig, self.config, dest).format()
+        data.setData(dest.getvalue())
+        return data
+
+
+def register():
+    return PythonTransform()

product/PortalTransforms/transforms/rest.py

+from Products.PortalTransforms.interfaces import itransform
+from reStructuredText import HTML
+import sys
+
+class rest:
+    __implements__ = itransform
+
+    __name__ = "rest_to_html"
+    inputs  = ("text/x-rst", "text/restructured",)
+    output = "text/html"
+
+    def name(self):
+        return self.__name__
+
+    def convert(self, orig, data, **kwargs):
+        # do the format
+        encoding        = kwargs.get('encoding', 'utf-8')
+        input_encoding  = kwargs.get('input_encoding', encoding)
+        output_encoding = kwargs.get('output_encoding', encoding)
+        language        = kwargs.get('language', 'en') 
+        warnings        = kwargs.get('warnings', None) 
+        settings = {'documentclass': '',
+                    'traceback': 1,
+               }
+        html = HTML(orig, 
+                    input_encoding=input_encoding, 
+                    output_encoding=output_encoding, 
+                    language_code=language, 
+                    warnings=warnings, 
+                    settings=settings)
+        html = html.replace(' class="document"', '', 1)
+        data.setData(html)
+        return data
+
+def register():
+    return rest()

product/PortalTransforms/transforms/rtf_to_html.py

+"""
+Uses the http://freshmeat.net/projects/rtfconverter/ bin to do its handy work
+"""
+
+from Products.PortalTransforms.interfaces import itransform
+from Products.PortalTransforms.libtransforms.utils import bin_search, sansext
+from Products.PortalTransforms.libtransforms.commandtransform import commandtransform
+from Products.CMFDefault.utils import bodyfinder
+import os
+
+class rtf_to_html(commandtransform):
+    __implements__ = itransform
+
+    __name__ = "rtf_to_html"
+    inputs   = ('application/rtf',)
+    output  = 'text/html'
+
+    binaryName = "rtf-converter"
+
+    def __init__(self):
+        commandtransform.__init__(self, binary=self.binaryName)
+
+    def convert(self, data, cache, **kwargs):
+        kwargs['filename'] = 'unknow.rtf'
+
+        tmpdir, fullname = self.initialize_tmpdir(data, **kwargs)
+        html = self.invokeCommand(tmpdir, fullname)
+        path, images = self.subObjects(tmpdir)
+        objects = {}
+        if images:
+            self.fixImages(path, images, objects)
+        self.cleanDir(tmpdir)
+        cache.setData(bodyfinder(html))
+        cache.setSubObjects(objects)
+        return cache
+
+    def invokeCommand(self, tmpdir, fullname):
+        # FIXME: windows users...
+        htmlfile = "%s/%s.html" % (tmpdir, sansext(fullname))
+        cmd = 'cd "%s" && %s -o %s "%s" 2>error_log 1>/dev/null' % (
+            tmpdir, self.binary, htmlfile, fullname)
+        os.system(cmd)
+        try:
+            html = open(htmlfile).read()
+        except:
+            try:
+                return open("%s/error_log" % tmpdir, 'r').read()
+            except:
+                return ''
+        return html
+
+def register():
+    return rtf_to_html()

product/PortalTransforms/transforms/rtf_to_xml.py

+"""
+Uses the http://sf.net/projects/rtf2xml bin to do its handy work
+
+"""
+from Products.PortalTransforms.interfaces import itransform
+from Products.PortalTransforms.libtransforms.utils import bin_search, sansext
+from Products.PortalTransforms.libtransforms.commandtransform import commandtransform
+import os
+
+class rtf_to_xml(commandtransform):
+    __implements__ = itransform
+
+    __name__ = "rtf_to_xml"
+    inputs   = ('application/rtf',)
+    output  = 'text/xml'
+
+    binaryName = "rtf2xml"
+
+    def __init__(self):
+        commandtransform.__init__(self, binary=self.binaryName)
+
+    def convert(self, data, cache, **kwargs):
+        kwargs['filename'] = 'unknown.rtf'
+
+        tmpdir, fullname = self.initialize_tmpdir(data, **kwargs)
+        xml = self.invokeCommand(tmpdir, fullname)
+        path, images = self.subObjects(tmpdir)
+        objects = {}
+        if images:
+            self.fixImages(path, images, objects)
+        self.cleanDir(tmpdir)
+        cache.setData(xml)
+        cache.setSubObjects(objects)
+        return cache
+
+    def invokeCommand(self, tmpdir, fullname):
+        # FIXME: windows users...
+        xmlfile = "%s/%s.xml" % (tmpdir, sansext(fullname))
+        cmd = 'cd "%s" && %s -o %s "%s" 2>error_log 1>/dev/null' % (
+            tmpdir, self.binary, xmlfile, fullname)
+        os.system(cmd)
+        try:
+            xml = open(xmlfile).read()
+        except:
+            try:
+                return open("%s/error_log" % tmpdir, 'r').read()
+            except:
+                return ''
+        return xml
+
+def register():
+    return rtf_to_xml()

product/PortalTransforms/transforms/safe_html.py

+import logging
+from sgmllib import SGMLParser
+
+from Products.PortalTransforms.interfaces import itransform
+from Products.PortalTransforms.utils import log
+from Products.CMFDefault.utils import bodyfinder
+from Products.CMFDefault.utils import IllegalHTML
+from Products.CMFDefault.utils import SimpleHTMLParser
+from Products.CMFDefault.utils import VALID_TAGS
+from Products.CMFDefault.utils import NASTY_TAGS
+
+# tag mapping: tag -> short or long tag
+VALID_TAGS = VALID_TAGS.copy()
+NASTY_TAGS = NASTY_TAGS.copy()
+
+# add some tags to allowed types. This should be fixed in CMFDefault
+VALID_TAGS['ins'] = 1
+VALID_TAGS['del'] = 1
+VALID_TAGS['q'] = 1
+VALID_TAGS['map']=1
+VALID_TAGS['area']=1
+
+msg_pat = """
+<div class="system-message">
+<p class="system-message-title">System message: %s</p>
+%s</d>
+"""
+
+class StrippingParser(SGMLParser):
+    """Pass only allowed tags;  raise exception for known-bad.
+
+    Copied from Products.CMFDefault.utils
+    Copyright (c) 2001 Zope Corporation and Contributors. All Rights Reserved.
+    """
+
+    from htmlentitydefs import entitydefs # replace entitydefs from sgmllib
+
+    def __init__(self, valid, nasty, remove_javascript, raise_error):
+        SGMLParser.__init__( self )
+        self.result = []
+        self.valid = valid
+        self.nasty = nasty
+        self.remove_javascript = remove_javascript
+        self.raise_error = raise_error
+        self.suppress = False
+
+    def handle_data(self, data):
+        if self.suppress: return
+        if data:
+            self.result.append(data)
+
+    def handle_charref(self, name):
+        if self.suppress: return
+        self.result.append('&#%s;' % name)
+
+    def handle_comment(self, comment):
+        pass
+
+    def handle_decl(self, data):
+        pass
+
+    def handle_entityref(self, name):
+        if self.suppress: return
+        if self.entitydefs.has_key(name):
+            x = ';'
+        else:
+            # this breaks unstandard entities that end with ';'
+            x = ''
+
+        self.result.append('&%s%s' % (name, x))
+
+    def unknown_starttag(self, tag, attrs):
+        """ Delete all tags except for legal ones.
+        """
+
+        if self.suppress: return
+        #if self.remove_javascript and tag == script and :
+
+        if self.valid.has_key(tag):
+            self.result.append('<' + tag)
+
+            for k, v in attrs:
+                if self.remove_javascript and k.strip().lower().startswith('on'):
+                    if not self.raise_error: continue
+                    else: raise IllegalHTML, 'Javascript event "%s" not allowed.' % k
+                elif self.remove_javascript and v.strip().lower().startswith('javascript:' ):
+                    if not self.raise_error: continue
+                    else: raise IllegalHTML, 'Javascript URI "%s" not allowed.' % v
+                else:
+                    self.result.append(' %s="%s"' % (k, v))
+
+            #UNUSED endTag = '</%s>' % tag
+            if self.valid.get(tag):
+                self.result.append('>')
+            else:
+                self.result.append(' />')
+        elif self.nasty.has_key(tag):
+            self.suppress = True
+            if self.raise_error:
+                raise IllegalHTML, 'Dynamic tag "%s" not allowed.' % tag
+        else:
+            # omit tag
+            pass
+
+    def unknown_endtag(self, tag):
+        if self.nasty.has_key(tag):
+            self.suppress = False
+        if self.suppress: return
+        if self.valid.get(tag):
+            self.result.append('</%s>' % tag)
+            #remTag = '</%s>' % tag
+
+    def getResult(self):
+        return ''.join(self.result)
+
+def scrubHTML(html, valid=VALID_TAGS, nasty=NASTY_TAGS,
+              remove_javascript=True, raise_error=True):
+
+    """ Strip illegal HTML tags from string text.
+    """
+    parser = StrippingParser(valid=valid, nasty=nasty,
+                             remove_javascript=remove_javascript,
+                             raise_error=raise_error)
+    parser.feed(html)
+    parser.close()
+    return parser.getResult()
+
+class SafeHTML:
+    """Simple transform which uses CMFDefault functions to
+    clean potentially bad tags.
+    Tags must explicit be allowed in valid_tags to pass. Only the tags
+    themself are removed, not their contents. If tags are
+    removed and in nasty_tags, they are removed with
+    all of their contents.
+    Be aware that you may need to clean the cache to let a Object
+    call the transform again.
+    """
+
+    __implements__ = itransform
+
+    __name__ = "safe_html"
+    inputs   = ('text/html',)
+    output = "text/x-html-safe"
+
+    def __init__(self, name=None, **kwargs):
+
+
+        self.config = {
+            'inputs': self.inputs,
+            'output': self.output,
+            'valid_tags': VALID_TAGS,
+            'nasty_tags': NASTY_TAGS,
+            'remove_javascript': 1,
+            'disable_transform': 0,
+            }
+
+        self.config_metadata = {
+            'inputs' : ('list', 'Inputs', 'Input(s) MIME type. Change with care.'),
+            'valid_tags' : ('dict',
+                            'valid_tags',
+                            'List of valid html-tags, value is 1 if they ' +
+                            'have a closing part (e.g. <p>...</p>) and 0 for empty ' +
+                            'tags (like <br />). Be carefull!',
+                            ('tag', 'value')),
+            'nasty_tags' : ('dict',
+                            'nasty_tags',
+                            'Dynamic Tags that are striped with ' +
+                            'everything they contain (like applet, object). ' +
+                            'They are only deleted if they are not marked as valid_tags.',
+                            ('tag', 'value')),
+            'remove_javascript' : ("int",
+                                   'remove_javascript',
+                                   '1 to remove javascript attributes that begin with on (e.g. onClick) ' +
+                                   'and attributes where the value starts with "javascript:" ' +
+                                   '(e.g. <a href="javascript:function()". ' +
+                                   'This does not effect <script> tags. 0 to leave the attributes.'),
+            'disable_transform' : ("int",
+                                   'disable_transform',
+                                   'If 1, nothing is done.')
+            }
+
+        self.config.update(kwargs)
+
+        if name:
+            self.__name__ = name
+
+    def name(self):
+        return self.__name__
+
+    def __getattr__(self, attr):
+        if attr == 'inputs':
+            return self.config['inputs']
+        if attr == 'output':
+            return self.config['output']
+        raise AttributeError(attr)
+
+    def convert(self, orig, data, **kwargs):
+        # note if we need an upgrade.
+        if not self.config.has_key('disable_transform'):
+            log(logging.ERROR, 'PortalTransforms safe_html transform needs to be '
+                'updated. Please re-install the PortalTransforms product to fix.')
+
+        # if we have a config that we don't want to delete
+        # we need a disable option
+        if self.config.get('disable_transform'):
+            data.setData(orig)
+            return data
+
+        try:
+            safe = scrubHTML(
+                bodyfinder(orig),
+                valid=self.config.get('valid_tags', {}),
+                nasty=self.config.get('nasty_tags', {}),
+                remove_javascript=self.config.get('remove_javascript', True),
+                raise_error=False)
+        except IllegalHTML, inst:
+            data.setData(msg_pat % ("Error", str(inst)))
+        else:
+            data.setData(safe)
+        return data
+
+def register():
+    return SafeHTML()

product/PortalTransforms/transforms/st.py

+from StructuredText.StructuredText import HTML
+from Products.PortalTransforms.interfaces import itransform
+
+DEFAULT_STX_LEVEL = 2
+STX_LEVEL = DEFAULT_STX_LEVEL
+
+class st:
+    __implements__ = itransform
+
+    __name__ = "st_to_html"
+    inputs   = ("text/structured",)
+    output   = "text/html"
+
+    def name(self):
+        return self.__name__
+
+    def convert(self, orig, data, level=None, **kwargs):
+        if level is None:
+            level = STX_LEVEL
+        data.setData(HTML(orig, level=level, header=0))
+        return data
+
+def register():
+    return st()

product/PortalTransforms/transforms/text_pre_to_html.py

+from Products.PortalTransforms.interfaces import itransform
+from DocumentTemplate.DT_Util import html_quote
+
+__revision__ = '$Id: text_pre_to_html.py 3658 2005-02-23 16:29:54Z tiran $'
+
+class TextPreToHTML:
+    """simple transform which wraps raw text into a <pre> tag"""
+
+    __implements__ = itransform
+
+    __name__ = "text-pre_to_html"
+    inputs   = ('text/plain-pre',)
+    output = "text/html"
+
+    def __init__(self, name=None):
+        self.config_metadata = {
+            'inputs' : ('list', 'Inputs', 'Input(s) MIME type. Change with care.'),
+            }
+        if name:
+            self.__name__ = name
+
+    def name(self):
+        return self.__name__
+
+    def __getattr__(self, attr):
+        if attr == 'inputs':
+            return self.config['inputs']
+        if attr == 'output':
+            return self.config['output']
+        raise AttributeError(attr)
+
+    def convert(self, orig, data, **kwargs):
+        data.setData('<pre class="data">%s</pre>' % html_quote(orig))
+        return data
+
+def register():
+    return TextPreToHTML()

product/PortalTransforms/transforms/text_to_html.py

+from Products.PortalTransforms.interfaces import itransform
+from DocumentTemplate.DT_Util import html_quote
+
+__revision__ = '$Id: text_to_html.py 4787 2005-08-19 21:43:41Z dreamcatcher $'
+
+class TextToHTML:
+    """simple transform which wrap raw text in a verbatim environment"""
+
+    __implements__ = itransform
+
+    __name__ = "text_to_html"
+    output = "text/html"
+
+    def __init__(self, name=None, inputs=('text/plain',)):
+        self.config = { 'inputs' : inputs, }
+        self.config_metadata = {
+            'inputs' : ('list', 'Inputs', 'Input(s) MIME type. Change with care.'),
+            }
+        if name:
+            self.__name__ = name
+
+    def name(self):
+        return self.__name__
+
+    def __getattr__(self, attr):
+        if attr == 'inputs':
+            return self.config['inputs']
+        if attr == 'output':
+            return self.config['output']
+        raise AttributeError(attr)
+
+    def convert(self, orig, data, **kwargs):
+        # Replaces all line breaks with a br tag, and wraps it in a p tag.
+        data.setData('<p>%s</p>' % html_quote(orig.strip()).replace('\n', '<br />'))
+        return data
+
+def register():
+    return TextToHTML()

product/PortalTransforms/transforms/uno.ini.in

+UNO_TYPES=file:///usr/lib/openoffice/program/applicat.rdb
+UNO_SERVICES=file:///usr/lib/openoffice/program/applicat.rdb
+

product/PortalTransforms/transforms/uno.py

+import os, os.path, sys
+
+def package_home(globals_dict):
+    __name__=globals_dict['__name__']
+    m=sys.modules[__name__]
+    if hasattr(m,'__path__'):
+        r=m.__path__[0]
+    elif "." in __name__:
+        r=sys.modules[__name__[:__name__.rfind('.')]].__path__[0]
+    else:
+        r=__name__
+    return os.path.abspath(os.path.join(os.getcwd(), r))
+
+
+setup_ini = 'file://%s/uno.ini' % package_home(globals())
+
+import PyUNO
+
+
+class uno:
+    def __init__ ( self, connection='socket,host=localhost,port=2002;urp', setup=setup_ini ):
+        """ do the bootstrap
+
+        connection can be one or more of the following:
+
+        socket,
+        host = localhost | <hostname> | <ip-addr>,
+        port = <port>,
+        service = soffice,
+        user = <username>,
+        password = <password>
+        ;urp
+
+        """
+        self.XComponentContext = PyUNO.bootstrap ( setup )
+        self.XUnoUrlResolver, o = \
+                              self.XComponentContext.ServiceManager.createInstanceWithContext ( 'com.sun.star.bridge.UnoUrlResolver', self.XComponentContext )
+        self.XNamingService, o = self.XUnoUrlResolver.resolve ( 'uno:%s;StarOffice.NamingService' % connection )
+        self.XMultiServiceFactory, o = self.XNamingService.getRegisteredObject ('StarOffice.ServiceManager')
+        self.XComponentLoader, o = \
+                               self.XMultiServiceFactory.createInstance ( 'com.sun.star.frame.Desktop' )
+
+
+    def new ( self, what, where='_blank', no=0, propertyValues=() ):
+        return self.XComponentLoader.loadComponentFromURL (
+                what, where, no, propertyValues )
+
+
+    def newIdlStruct ( self, type ):
+        return PyUNO.createIdlStruct ( self.XMultiServiceFactory, type )
+
+
+    def newCalc (self):
+        return self.new ('private:factory/scalc')
+
+
+    def newWriter (self):
+        return self.new ('private:factory/swriter')
+
+
+    def newImpress (self):
+        return self.new ('private:factory/simpress')
+
+
+    def newDraw (self):
+        return self.new ('private:factory/sdraw')
+
+
+    def newPropertyValue (self, propertyValue={} ):
+        property = self.newIdlStruct  ( 'com.sun.star.beans.PropertyValue' )
+
+        if propertyValue.has_key('Name'):
+            property.Name = propertyValue['Name']
+
+        if propertyValue.has_key('Value'):
+            property.Value = propertyValue['Value']
+
+        if propertyValue.has_key('State'):
+            property.State = propertyValue['State']
+
+        if propertyValue.has_key('Handle'):
+            property.Handle = propertyValue['Handle']
+
+        return property
+
+
+    def newPropertyValues ( self, propertyValues=[] ):
+        list = ()
+
+        l = len(propertyValues)
+        for p in range (l):
+            list = list + ( self.newPropertyValue ( propertyValues.pop(0) ), )
+
+        return list
+
+
+    def newBoolean ( self, bool=0 ):
+        if bool:
+            return PyUNO.true()
+        else:
+            return PyUNO.false()

product/PortalTransforms/transforms/word_to_html.py

+from Products.PortalTransforms.interfaces import itransform
+
+EXTRACT_BODY  = 1
+EXTRACT_STYLE = 0
+
+FIX_IMAGES    = 1
+IMAGE_PREFIX  = "img_"
+
+# disable office_uno because it doesn't support multithread yet
+ENABLE_UNO = False
+
+import os
+if os.name == 'posix':
+    try:
+        if ENABLE_UNO:
+            from office_uno import document
+        else:
+            raise
+    except:
+        from office_wvware import document
+else:
+    try:
+        if ENABLE_UNO:
+            from office_uno import document
+        else:
+            raise
+    except:
+        from office_com import document
+
+import os.path
+
+class word_to_html:
+    __implements__ = itransform
+
+    __name__ = "word_to_html"
+    inputs   = ('application/msword',)
+    output  = 'text/html'
+    output_encoding = 'utf-8'
+
+    tranform_engine = document.__module__
+
+    def name(self):
+        return self.__name__
+
+    def convert(self, data, cache, **kwargs):
+        orig_file = 'unknown.doc'
+
+        doc = document(orig_file, data)
+        doc.convert()
+        html = doc.html()
+
+        path, images = doc.subObjects(doc.tmpdir)
+        objects = {}
+        if images:
+            doc.fixImages(path, images, objects)
+        doc.cleanDir(doc.tmpdir)
+
+        cache.setData(html)
+        cache.setSubObjects(objects)
+        return cache
+
+def register():
+    return word_to_html()

product/PortalTransforms/unsafe_transforms/build_transforms.py

+"""try to build some usefull transformations with the command and xml
+transforms and the available binaries
+"""
+from Products.PortalTransforms.libtransforms.utils import bin_search, MissingBinary
+
+COMMAND_CONFIGS = (
+    ('lynx_dump', '.html',
+     {'binary_path'  : 'lynx',
+      'command_line' : '-dump %s',
+      'inputs'       : ('text/html',),
+      'output'       : 'text/plain',
+      }),
+    ('tidy_html', '.html',
+     {'binary_path'  : 'tidy',
+      'command_line' : '%s',
+      'inputs'       : ('text/html',),
+      'output'       : 'text/html',
+      }),
+    ('rtf_to_html', None,
+     {'binary_path'  : 'unrtf',
+      'command_line' : '%s',
+      'inputs'       : ('application/rtf',),
+      'output'       : 'text/html',
+      }),
+    ('ppt_to_html', None,
+     {'binary_path'  : 'ppthtml',
+      'command_line' : '%s',
+      'inputs'       : ('application/vnd.ms-powerpoint',),
+      'output'       : 'text/html',
+      }),
+    ('excel_to_html', None,
+     {'binary_path'  : 'xlhtml',
+      'command_line' : '-nh -a %s',
+      'inputs'       : ('application/vnd.ms-excel',),
+      'output'       : 'text/html',
+      }),
+    ('ps_to_text', None,
+     {'binary_path'  : 'ps2ascii',
+      'command_line' : '%s',
+      'inputs'       : ('application/postscript',),
+      'output'       : 'text/plain',
+      }),
+    )
+
+TRANSFORMS = {}
+
+from command import ExternalCommandTransform
+for tr_name, extension, config in COMMAND_CONFIGS:
+    try:
+        bin = bin_search(config['binary_path'])
+    except MissingBinary:
+        print 'no such binary', config['binary_path']
+    else:
+        tr = ExternalCommandTransform(tr_name, extension)
+        tr.config['binary_path'] = bin
+        tr.__name__ = tr_name
+        tr.config = config
+        TRANSFORMS[tr_name] = tr
+
+XMLPROCS_CONF = {
+    'xsltproc' : '--catalogs --xinclude -o %(output)s %(transform)s %(input)s',
+    '4xslt' : ' -o %(output)s %(input)s %(transform)s'
+    }
+
+bin = None
+for proc in XMLPROCS_CONF.keys():
+    try:
+        bin = bin_search(proc)
+        break
+    except MissingBinary:
+        print 'no such binary', proc
+
+if bin is not None:
+    print 'Using %s as xslt processor' % bin
+    from xml import XsltTransform
+    for output in ('html', 'plain'):
+        name = "xml_to_" + output
+        command_line = XMLPROCS_CONF[proc]
+        tr = XsltTransform(name=name, inputs=('text/xml',), output='text/'+output,
+                           binary_path=bin, command_line=command_line)
+        TRANSFORMS[name] = tr
+
+def initialize(engine):
+    for transform in TRANSFORMS.values():
+        engine.registerTransform(transform)

product/PortalTransforms/unsafe_transforms/command.py

+"""
+A custom transform using external command
+"""
+
+__revision__ = '$Id: command.py 4439 2005-06-15 16:32:36Z panjunyong $'
+
+import os.path
+from os import popen3
+from Products.PortalTransforms.interfaces import itransform
+from Products.PortalTransforms.libtransforms.utils import bin_search, sansext
+from Products.PortalTransforms.libtransforms.commandtransform import commandtransform
+from Products.PortalTransforms.utils import log
+
+class ExternalCommandTransform(commandtransform):
+    """ Custom external command
+
+    transform content by launching an external command
+
+    the command should take the content in an input file (designed by '%s' in
+    the command line parameters) and return output on stdout.
+    Input and output mime types must be set correctly !
+    """
+    __implements__ = (itransform,)
+
+    __name__ = "command_transform"
+
+    def __init__(self, name=None, input_extension=None, **kwargs):
+        self.config = {
+            'binary_path'  : '',
+            'command_line' : '',
+            'inputs'       : ('text/plain',),
+            'output'       : 'text/plain',
+            }
+        self.config_metadata = {
+            'binary_path'  : ('string', 'Binary path',
+                              'Path of the executable on the server.'),
+            'command_line' : ('string', 'Command line',
+                              '''Additional command line option.
+There should be at least the input file (designed by "%(input)s").
+The transformation\'s result must be printed on stdout.
+'''),
+            'inputs'       : ('list', 'Inputs', 'Input(s) MIME type. Change with care.'),
+            'output'       : ('string', 'Output', 'Output MIME type. Change with care.'),
+            }
+        self.config.update(kwargs)
+        commandtransform.__init__(self, name=name, binary=self.config['binary_path'],  **kwargs)
+
+        # use the full binary path
+        self.config.update({'binary_path':self.binary})
+        self.input_extension = input_extension
+
+    def __getattr__(self, attr):
+        if attr == 'inputs':
+            return self.config['inputs']
+        if attr == 'output':
+            return self.config['output']
+        raise AttributeError(attr)
+
+
+    def convert(self, data, cache, **kwargs):
+        filename = kwargs.get('filename') or 'unknown'
+        if self.input_extension is not None:
+            kwargs['filename'] = 'unknown' + self.input_extension
+        else:
+            kwargs['filename'] = 'unknown' + os.path.splitext(filename)[-1]
+        tmpdir, fullname = self.initialize_tmpdir(data, **kwargs)
+
+        data = self.invokeCommand(fullname)
+        cache.setData(data)
+        path, images = self.subObjects(tmpdir)
+        objects = {}
+        if images:
+            self.fixImages(path, images, objects)
+            cache.setSubObjects(objects)
+        self.cleanDir(tmpdir)
+        return cache
+
+
+    def invokeCommand(self, input_name):
+        command = '%(binary_path)s %(command_line)s' % self.config
+        input, output, error = popen3(command % input_name)
+        input.close()
+        # first read stderr, else we may hang on stout
+        # but, still hang my windows, so commented it :-(
+        # error_data = error.read()
+        error_data = 'error while running "%s"' % (command % input_name)
+        error.close()
+        data = output.read()
+        output.close()
+        if error_data and not data:
+            data = error_data
+        else:
+            log('Error while running "%s":\n %s' % (command % input_name,
+                                                    error_data))
+        return data
+
+def register():
+    return ExternalCommandTransform()

product/PortalTransforms/unsafe_transforms/xml.py

+"""
+A custom transform using external command
+"""
+
+__revision__ = '$Id: xml.py 4787 2005-08-19 21:43:41Z dreamcatcher $'
+
+from os.path import join, dirname, exists
+import re
+from os import popen3, popen4, system
+from cStringIO import StringIO
+
+from Products.PortalTransforms.interfaces import itransform
+from Products.PortalTransforms.libtransforms.utils import bin_search, sansext
+from Products.PortalTransforms.libtransforms.commandtransform import commandtransform
+from Products.PortalTransforms.utils import log
+
+class XsltTransform(commandtransform):
+    """ Custom external command
+
+    transform xml content by launching an external XSLT processor
+
+    Input and output mime types must be set correctly !
+
+    You can associate different document type to different transformations.
+    """
+    __implements__ = (itransform,)
+
+    __name__ = "xml_to_html"
+
+    def __init__(self, name=None, **kwargs):
+        self.config = {
+            # sample configuration
+            'binary_path'  : bin_search('xsltproc'),
+            'command_line' : '%(transform)s %(input)s',
+            'inputs'       : ('text/xml',),
+            'output'       : 'text/html',
+            'output_encoding' : 'UTF-8',
+            'dtds'         : {
+            '-//OASIS//DTD DocBook V4.1//EN' : '/usr/share/sgml/docbook/xsl-stylesheets-1.29/html/docbook.xsl'
+            },
+            'default_transform': ''
+            }
+        self.config_metadata = {
+            'binary_path'  : ('string', 'Binary path',
+                              'Path of the executable on the server.'),
+            'command_line' : ('string', 'Command line',
+                              '''Additional command line option.
+There should be at least the input file (designed by "%(input)s") and the xsl
+file (designed by "%(transform)s").The transformation\'s result must be printed on stdout.
+'''),
+            'inputs'       : ('list', 'Inputs', 'Input(s) MIME type. Change with care.'),
+            'output'       : ('string', 'Output', 'Output MIME type. Change with care.'),
+            'output_encoding': ('string', 'Output encoding', 'Output encoding.'),
+            'dtds'         : ('dict', 'DTDs',
+                              'Association of public ids or dtds to XSL transformations.',
+                              ('Public id', 'XSLT path')),
+            'default_transform' : ('string', 'Default xslt',
+                                   'Default xslt, used when no specific transformation is found.'),
+            }
+        self.config.update(kwargs)
+        if name:
+            self.__name__ = name
+
+    def __getattr__(self, attr):
+        if attr == 'inputs':
+            return self.config['inputs']
+        if attr == 'output':
+            return self.config['output']
+        if attr == 'output_encoding':
+            return self.config['output_encoding']
+        raise AttributeError(attr)
+
+
+    def convert(self, data, cache, **kwargs):
+        base_name = sansext(kwargs.get("filename") or 'unknown.xml')
+        dtds = self.config['dtds']
+        tmpdir, fullname = self.initialize_tmpdir(data, filename=base_name)
+        try:
+            try:
+                doctype = get_doctype(data)
+            except DTException:
+                try:
+                    doctype = get_dtd(data)
+                except DTException:
+                    log('Unable to get doctype nor dtd in %s' % data)
+                    doctype = None
+            if doctype and dtds.has_key(doctype):
+                data = self.invokeCommand(fullname, dtds[doctype])
+            elif self.config['default_transform']:
+                data = self.invokeCommand(fullname, self.config['default_transform'])
+            cache.setData(data)
+            path, images = self.subObjects(tmpdir)
+            objects = {}
+            if images:
+                self.fixImages(path, images, objects)
+                cache.setSubObjects(objects)
+            return cache
+        finally:
+            self.cleanDir(tmpdir)
+
+
+    def invokeCommand(self, input_name, xsl):
+        dest_dir = dirname(input_name)
+        output_file = join(dirname(input_name), 'tr_output')
+        command = '%(binary_path)s %(command_line)s' % self.config
+        data = {'input': input_name, 'output': output_file, 'transform': xsl}
+        system(command % data)
+
+        if exists(output_file):
+            data = open(output_file).read()
+        else:
+            data = 'error occurs during transform. See error log'
+        return data
+
+
+
+def register():
+    return XsltTransform()
+
+DT_RGX = re.compile('<!DOCTYPE \w* PUBLIC \"([^"]*)\" \"([^"]*)\"')
+DT_RGX2 = re.compile('<!DOCTYPE \w* SYSTEM \"([^"]*)\"')
+
+class DTException(Exception): pass
+
+def get_doctype(data):
+    """ return the public id for the doctype given some raw xml data
+    """
+    if not hasattr(data, 'readlines'):
+        data = StringIO(data)
+    for line in data.readlines():
+        line = line.strip()
+        if not line:
+            continue
+        if line.startswith('<?xml') or line.startswith('<!-- '):
+            continue
+        m = DT_RGX.match(line)
+        if m is not None:
+            return m.group(1)
+        else:
+            raise DTException('Unable to match doctype in "%s"' % line)
+
+def get_dtd(data):
+    """ return the public id for the doctype given some raw xml data
+    """
+    if not hasattr(data, 'readlines'):
+        data = StringIO(data)
+    for line in data.readlines():
+        line = line.strip()
+        if not line:
+            continue
+        if line.startswith('<?xml') or line.startswith('<!-- '):
+            continue
+        m = DT_RGX.match(line)
+        if m is not None:
+            return m.group(2)
+        m = DT_RGX2.match(line)
+        if m is not None:
+            return m.group(1)
+        else:
+            raise DTException('Unable to match doctype in "%s"' % line)
+
+
+if __name__ == '__main__':
+    print get_doctype('''<?xml version="1.0" encoding="iso-8859-1"?>
+<!DOCTYPE article PUBLIC "-//LOGILAB/DTD DocBook V4.1.2-Based Extension V0.1//EN" "dcbk-logilab.dtd" []>
+
+<book id="devtools_user_manual" lang="fr">
+''')

product/PortalTransforms/utils.py

+"""some common utilities
+"""
+import logging
+from time import time
+from types import UnicodeType, StringType
+
+STRING_TYPES = (UnicodeType, StringType)
+
+class TransformException(Exception):
+    pass
+
+FB_REGISTRY = None
+
+# logging function
+logger = logging.getLogger('PortalTransforms')
+
+def log(message, severity=logging.INFO):
+    logger.log(severity, message)
+
+# directory where template for the ZMI are located
+import os.path
+_www = os.path.join(os.path.dirname(__file__), 'www')
+skins_dir = os.path.join(os.path.dirname(__file__), 'skins')
+
+from zExceptions import BadRequest
+
+# directory where template for the ZMI are located
+import os.path
+_www = os.path.join(os.path.dirname(__file__), 'www')
+skins_dir = None

product/PortalTransforms/version.txt

+1.4.0-final
\ No newline at end of file

product/PortalTransforms/www/addTransform.zpt

+<h1 tal:replace="structure here/manage_page_header|nothing">Header</h1>
+<h2 tal:define="manage_tabs_message options/manage_tabs_message | nothing"
+    tal:replace="structure here/manage_tabs">Tabs</h2>
+  
+
+  <form method="POST" action="manage_addTransform"
+        tal:attributes="action string:${here/absolute_url}/manage_addTransform;">
+    <div class="form-title">
+      Add a new transform
+    </div>
+    
+    <div tal:define="status python:request.get('portal_status', '')"
+      tal:condition="status"
+	     class="error"	
+      tal:content="status"
+	/>
+      
+      <table width="50%">
+        <tr>
+	  <td> ID</td>
+	  <td> 
+	    <input name="id" tal:attributes="value python:request.get('id', '');"/>
+	  </td>
+        </tr><tr>
+	  <td>	
+	    Module
+	  </td>
+	  <td>
+	    <input name="module" tal:attributes="value python:request.get('module', '');"/>
+	  </td>	
+	</tr>
+      </table>
+
+      <input type="submit"/>
+      </form>
+    
+<tal:footer tal:replace="structure here/manage_page_footer|nothing">footer</tal:footer>

product/PortalTransforms/www/addTransformsChain.zpt

+<h1 tal:replace="structure here/manage_page_header|nothing">Header</h1>
+<h2 tal:define="manage_tabs_message options/manage_tabs_message | nothing"
+    tal:replace="structure here/manage_tabs">Tabs</h2>
+  
+
+  <form method="POST" action="manage_addTransform"
+        tal:attributes="action string:${here/absolute_url}/manage_addTransformsChain;">
+    <div class="form-title">
+      Add a new transforms chain
+    </div>
+    
+    <div tal:define="status python:request.get('portal_status', '')"
+      tal:condition="status"
+	     class="error"	
+      tal:content="status"
+	/>
+      
+      <table width="50%">
+        <tr>
+	  <td> ID</td>
+	  <td> 
+	    <input name="id" tal:attributes="value python:request.get('id', '');"/>
+	  </td>
+        </tr><tr>
+	  <td>	
+	    Description
+	  </td>
+	  <td>
+	    <textarea name="description" tal:content="request/description | nothing"/>
+	  </td>	
+	</tr>
+      </table>
+
+      <input type="submit"/>
+      </form>
+    
+<tal:footer tal:replace="structure here/manage_page_footer|nothing">footer</tal:footer>

product/PortalTransforms/www/configureTransform.zpt

+<h1 tal:replace="structure here/manage_page_header|nothing">Header</h1>
+<h2 tal:define="manage_tabs_message options/manage_tabs_message | nothing"
+    tal:replace="structure here/manage_tabs">Tabs</h2>
+  
+
+  <form method="POST" 
+        tal:attributes="action string:${here/absolute_url}/set_parameters"
+        tal:define="params here/get_parameters">
+
+    <div class="form-title">
+      Configure transform
+    </div>
+    
+    <p>Transform inputs : <b tal:content="python:', '.join(here.inputs)"/>
+    <p>Transform output : <b tal:content="here/output"/>
+    
+    <p tal:condition="here/get_documentation" tal:content="here/get_documentation"/>
+
+    <div tal:define="status python:request.get('portal_status', '')"
+         tal:condition="status"
+         tal:content="status"
+	 class="error" />
+
+    <tal:block tal:condition="params">
+      <table width="80%">
+	<tr tal:repeat="param params">
+           <tal:block tal:define="meta python:here.get_parameter_infos(param);
+                                  type python:meta[0];
+                                  widget string:here/tr_widgets/macros/${type}_widget;
+                                  label python:meta[1] or param;">
+	    <td tal:content="label">Parameter's label</td>
+	    <td>             
+               <metal:block metal:use-macro="python:path(widget)" /> 
+	    </td>
+            <td tal:content="python:meta[2]">
+              field description
+            </td>
+          </tal:block>
+	</tr>
+      </table>
+
+      <input type="submit"/>
+    </tal:block>
+
+    <p tal:condition="not:params">
+      This transform has no configurable parameters.
+    </p>
+
+  </form>
+    
+<tal:footer tal:replace="structure here/manage_page_footer|nothing">footer</tal:footer>
+

product/PortalTransforms/www/editTransformationPolicy.zpt

+<h1 tal:replace="structure here/manage_page_header|nothing">Header</h1>
+<h2 tal:define="manage_tabs_message options/manage_tabs_message | nothing"
+    tal:replace="structure here/manage_tabs">Tabs</h2>
+
+
+   <div class="form-title">
+     Transformation policy
+   </div>
+
+   <div>
+     This page allow you to configure which transforms do you want to be applied
+     for a give output MIME type.
+   </div>
+ 
+   <hr/>
+
+   <form method="POST" action="manage_addPolicy">
+
+     <table>
+       <tr>
+	 <th align="left">output type</th>
+	 <th align="left">use transforms</th>
+       </tr>
+       <tr>
+         <td>
+	   <select name="output_mimetype" 
+                   tal:define="mimetypes here/mimetypes_registry/list_mimetypes;
+                               dummy mimetypes/sort">
+	     <option tal:repeat="mimetype mimetypes"
+		     tal:attributes="value mimetype;"
+		     tal:content="mimetype"/>
+	   </select>
+	 </td>
+         <td>
+	   <select name="required_transforms:list" multiple="multiple">
+	     <option tal:repeat="id here/objectIds"
+		     tal:attributes="value id;"
+		     tal:content="id"/>
+	   </select>
+         </td>
+       </tr>
+     </table>
+
+     <input type="submit" value="add"/>
+
+   </form>
+
+   <hr/>
+
+   <form method="POST" action="manage_delPolicies"
+         tal:define="policies here/listPolicies" tal:condition="policies">
+
+     <table>
+       <tr>
+	   <th colspan="2" align="left">output type</th>
+	   <th align="left">use transforms</th>
+       </tr>
+       <tr tal:repeat="policy policies">
+	 <td>
+	   <input type="checkbox" name="outputs:list"
+     	          tal:attributes="value python:policy[0]"/>
+         </td>
+	 <td tal:content="python:policy[0]">
+	 </td>
+	 <td tal:content="python:', '.join(policy[1])">
+	 </td>
+       </tr>
+     </table>
+
+     <input type="submit" value="delete selected"/>
+
+   </form>
+
+<tal:footer tal:replace="structure here/manage_page_footer|nothing">footer</tal:footer>

product/PortalTransforms/www/editTransformsChain.zpt

+<h1 tal:replace="structure here/manage_page_header|nothing">Header</h1>
+<h2 tal:define="manage_tabs_message options/manage_tabs_message | nothing"
+    tal:replace="structure here/manage_tabs">Tabs</h2>
+
+
+   <div class="form-title">
+     Chain <span tal:replace="here/title_or_id"/>.
+   </div>
+
+   <div align="right">
+     <form method="POST" action="manage_addObject">
+       <select name="id">
+         <option tal:repeat="id here/listAddableObjectIds"
+                 tal:attributes="value id;"
+                 tal:content="id"/>
+       </select>
+       <input type="submit" value="Add"/>
+     </form>
+   </div>
+
+   <p>Transform inputs : <b tal:content="python:', '.join(here.inputs)"/>
+   <p>Transform output : <b tal:content="here/output"/>
+
+
+   <div tal:content="here/description"/>
+
+   <div tal:define="status request/portal_status' | nothing"
+        tal:condition="status" class="error"
+        tal:content="status" />
+
+   <form method="POST" action="manage_delObjects"
+         tal:define="transforms here/objectValues" tal:condition="transforms">
+
+     <table width="60%" >
+       <tr>
+	 <th colspan="2">transform</th>
+	 <th>input</th>
+	 <th>output</th>
+	 <th colspan="2">&nbsp;</th>
+       </tr>
+
+       <tr tal:repeat="tr transforms">
+	   <td>
+	     <input type="checkbox" name="ids:list"
+     	            tal:attributes="value tr/getId"/>
+	   </td>
+	   <td tal:content="tr/title_or_id">
+	   </td>
+	   <td tal:content="python:tr.inputs[0]">
+	   </td>
+	   <td tal:content="tr/output">
+	   </td>
+	   <td>
+             <a tal:attributes="href string:${here/absolute_url}/move_object_down?id=${tr/getId}"
+                tal:condition="not:repeat/tr/end">
+  	       <img tal:attributes="src string:${here/portal_url}/down.png"/>
+             </a>
+	   </td>
+	   <td>
+             <a tal:attributes="href string:${here/absolute_url}/move_object_up?id=${tr/getId}"
+                tal:condition="not:repeat/tr/start">
+  	       <img tal:attributes="src string:${here/portal_url}/up.png"/>
+             </a>
+	   </td>
+       </tr>
+     </table>
+
+     <input type="submit" value="delete selected"/>
+
+   </form>
+
+
+<tal:footer tal:replace="structure here/manage_page_footer|nothing">footer</tal:footer>

product/PortalTransforms/www/reloadAllTransforms.zpt

+<h1 tal:replace="structure here/manage_page_header|nothing">Header</h1>
+<h2 tal:define="manage_tabs_message options/manage_tabs_message | nothing"
+    tal:replace="structure here/manage_tabs">Tabs</h2>
+  
+
+  <p>
+    Those Transformations have been reloaded :
+  </p>
+
+  <table border="1" tal:define="reloaded here/reloadTransforms">
+    <tr><th>name</th><th>module</th></tr>
+    <tr tal:repeat="tuple reloaded">
+      <td tal:content="python:tuple[0]"/>
+      <td tal:content="python:tuple[1]"/>
+    </tr>
+  </table>
+    
+<tal:footer tal:replace="structure here/manage_page_footer|nothing">footer</tal:footer>

product/PortalTransforms/www/reloadTransform.zpt

+<h1 tal:replace="structure here/manage_page_header|nothing">Header</h1>
+<h2 tal:define="manage_tabs_message options/manage_tabs_message | nothing"
+    tal:replace="structure here/manage_tabs">Tabs</h2>
+  
+
+  <p tal:define="dummy here/reload">Transformation reloaded <span
+  tal:condition="exists:here/module" tal:omit-tag="">(module <b tal:content="here/module"/>)</span>
+  </p>
+    
+<tal:footer tal:replace="structure here/manage_page_footer|nothing">footer</tal:footer>

product/PortalTransforms/www/setCacheTime.zpt

+<h1 tal:replace="structure here/manage_page_header|nothing">Header</h1>
+<h2 tal:define="manage_tabs_message options/manage_tabs_message | nothing"
+    tal:replace="structure here/manage_tabs">Tabs</h2>
+  
+
+  <form method="POST" action="manage_setCacheValidityTime"
+        tal:attributes="action
+        string:${here/absolute_url}/manage_setCacheValidityTime;">
+
+    <div class="form-title">
+      Manage transformation caches
+    </div>
+    
+    <div tal:define="status python:request.get('portal_status', '')"
+         class="error"	
+         tal:condition="status"
+         tal:content="status" />
+    
+    <table width="50%">
+      <tr>
+	<td>Lifetime of objects in cache (in seconds). O means infinity</td>
+	<td> 
+	  <input name="seconds" tal:attributes="value request/seconds | here/max_sec_in_cache | string:0;"/>
+	</td>
+      </tr>
+    </table>
+
+    <input type="submit"/>
+  </form>
+    
+<tal:footer tal:replace="structure here/manage_page_footer|nothing">footer</tal:footer>

product/PortalTransforms/www/tr_widgets.zpt

+<div metal:define-macro="int_widget">
+  <input tal:attributes="name param;
+			 value python:here.get_parameter_value(param);"/>
+</div>
+
+<div metal:define-macro="string_widget">
+  <input size="80"
+         tal:attributes="name param;
+			 value python:here.get_parameter_value(param);"/>
+</div>
+
+<div metal:define-macro="list_widget">
+  <textarea tal:attributes="name string:${param}:lines;"
+            tal:content="python: '\n'.join(here.get_parameter_value(param))"/>
+</div>
+
+<div metal:define-macro="dict_widget">
+              <table tal:define="titles python:meta[3];">
+                <tr> 
+                  <th tal:content="python:titles[0]"/>
+                  <th tal:content="python:titles[1]"/>
+                </tr><tr tal:define="values python:here.get_parameter_value(param).items();
+                                     dummy python:values.sort()"
+                         tal:repeat="key_val values">
+               
+                   <td><input type='text' size="30"
+                        tal:attributes="name python:param + '_key';
+                                        value python:key_val[0]"/></td>
+                   <td><input type='text' size="50"
+                        tal:attributes="name python:param + '_value';
+                                        value python:key_val[1]"/></td>
+                </tr><tr>
+                   <td><input type='text' size="30"
+                        tal:attributes="name python:param + '_key'"/></td>
+                   <td><input type='text' size="50" 
+                        tal:attributes="name python:param + '_value'"/></td>
+                </tr><tr>
+                   <td><input type='text' size="30"
+                        tal:attributes="name python:param + '_key'"/></td>
+                   <td><input type='text' size="50"
+                        tal:attributes="name python:param + '_value'"/></td>
+                </tr>
+
+              </table>
+</div>

product/PortalTransforms/z3/__init__.py

+# Five/Zope3 stuff
+

product/PortalTransforms/z3/interfaces.py

+# Five/Zope3 stuff
+