Commit fb493724 authored by Jean-Paul Smets's avatar Jean-Paul Smets

Code review and refactoring based on Document API.

git-svn-id: https://svn.erp5.org/repos/public/erp5/trunk@13631 20353a03-c40f-0410-a6d1-a30d3c3de9de
parent d7067b8b
...@@ -42,6 +42,9 @@ from Products.ERP5Type.WebDAVSupport import TextContent ...@@ -42,6 +42,9 @@ from Products.ERP5Type.WebDAVSupport import TextContent
from Products.ERP5Type.Message import Message from Products.ERP5Type.Message import Message
from Products.ERP5Type.Utils import convertToUpperCase, convertToMixedCase from Products.ERP5Type.Utils import convertToUpperCase, convertToMixedCase
from Products.ERP5.Document.Url import UrlMixIn from Products.ERP5.Document.Url import UrlMixIn
from Products.ERP5.Tool.ContributionTool import MAX_REPEAT
from zLOG import LOG
_MARKER = [] _MARKER = []
VALID_ORDER_KEY_LIST = ('user_login', 'content', 'file_name', 'input') VALID_ORDER_KEY_LIST = ('user_login', 'content', 'file_name', 'input')
...@@ -51,6 +54,47 @@ def makeSortedTuple(kw): ...@@ -51,6 +54,47 @@ def makeSortedTuple(kw):
items.sort() items.sort()
return tuple(items) return tuple(items)
class SnapshotMixin:
"""
This class provides a generic API to store in the ZODB
PDF snapshots of objects and documents with the
goal to keep a facsimile copy of documents as they
were at a given date.
"""
# Declarative security
security = ClassSecurityInfo()
security.declareObjectProtected(Permissions.AccessContentsInformation)
security.declareProtected(Permissions.ModifyPortalContent, 'createSnapshot')
def createSnapshot(self):
"""
Create a snapshot (PDF). This is the normal way to modifiy
snapshot_data. Once a snapshot is taken, a new snapshot
can not be taken.
NOTE: use getSnapshotData and hasSnapshotData accessors
to access a snapshot.
NOTE2: implementation of createSnapshot should probably
be delegated to a types base method since this it
is configuration dependent.
"""
if self.hasSnapshotData():
raise ConversionError('This document already has a snapshot.')
self._setSnapshotData(self.convert(format='pdf'))
security.declareProtected(Permissions.ManagePortal, 'deleteSnapshot')
def deleteSnapshot(self):
"""
Deletes the snapshot - in theory this should never be done.
It is there for programmers and system administrators.
"""
try:
del(self.snapshot_data)
except AttributeError:
pass
class ConversionError(Exception):pass class ConversionError(Exception):pass
class ConversionCacheMixin: class ConversionCacheMixin:
...@@ -107,8 +151,7 @@ class ConversionCacheMixin: ...@@ -107,8 +151,7 @@ class ConversionCacheMixin:
security.declareProtected(Permissions.View, 'getCacheTime') security.declareProtected(Permissions.View, 'getCacheTime')
def getCacheTime(self, **format): def getCacheTime(self, **format):
""" """
Checks when Checks when if ever was the file produced
if ever was the file produced
""" """
self.updateConversionCache() self.updateConversionCache()
return self._cached_time.get(makeSortedTuple(format), 0) return self._cached_time.get(makeSortedTuple(format), 0)
...@@ -148,7 +191,7 @@ if ever was the file produced ...@@ -148,7 +191,7 @@ if ever was the file produced
tformat = makeSortedTuple(format) tformat = makeSortedTuple(format)
return self._cached_mime.get(tformat, ''), self._cached_data.get(tformat, '') return self._cached_mime.get(tformat, ''), self._cached_data.get(tformat, '')
security.declareProtected(Permissions.View, 'getConversionCacheInfo') security.declareProtected(Permissions.ViewManagementScreens, 'getConversionCacheInfo')
def getConversionCacheInfo(self): def getConversionCacheInfo(self):
""" """
Get cache details as string (for debugging) Get cache details as string (for debugging)
...@@ -173,7 +216,7 @@ if ever was the file produced ...@@ -173,7 +216,7 @@ if ever was the file produced
return s return s
class Document(XMLObject, UrlMixIn): class Document(XMLObject, UrlMixIn, ConversionCacheMixin, SnapshotMixin):
""" """
Document is an abstract class with all methods Document is an abstract class with all methods
related to document management in ERP5. This includes related to document management in ERP5. This includes
...@@ -335,6 +378,8 @@ class Document(XMLObject, UrlMixIn): ...@@ -335,6 +378,8 @@ class Document(XMLObject, UrlMixIn):
# Regular expressions # Regular expressions
href_parser = re.compile('<a[^>]*href=[\'"](.*?)[\'"]',re.IGNORECASE) href_parser = re.compile('<a[^>]*href=[\'"](.*?)[\'"]',re.IGNORECASE)
body_parser = re.compile('<body[^>]*>(.*?)</body>', re.IGNORECASE + re.DOTALL)
title_parser = re.compile('<title[^>]*>(.*?)</title>', re.IGNORECASE + re.DOTALL)
# Declarative security # Declarative security
security = ClassSecurityInfo() security = ClassSecurityInfo()
...@@ -347,25 +392,25 @@ class Document(XMLObject, UrlMixIn): ...@@ -347,25 +392,25 @@ class Document(XMLObject, UrlMixIn):
, PropertySheet.DublinCore , PropertySheet.DublinCore
, PropertySheet.Version , PropertySheet.Version
, PropertySheet.Document , PropertySheet.Document
, PropertySheet.Snapshot
, PropertySheet.ExternalDocument , PropertySheet.ExternalDocument
, PropertySheet.Url , PropertySheet.Url
, PropertySheet.Periodicity , PropertySheet.Periodicity
, PropertySheet.Snapshot
) )
# Declarative interfaces # Declarative interfaces
__implements__ = () __implements__ = ()
searchable_property_list = ('title', 'description', 'id', 'reference', searchable_property_list = ('asText', 'title', 'description', 'id', 'reference',
'version', 'short_title', 'keyword', 'version', 'short_title',
'subject', 'source_reference', 'source_project_title') 'subject', 'source_reference', 'source_project_title',)
data = '' # some day this will be in property sheets data = '' # some day this will be in property sheets
base_format = 'base storage format' base_format = 'base storage format'
### Content processing methods ### Content processing methods
security.declareProtected(Permissions.View, 'index_html') security.declareProtected(Permissions.View, 'index_html')
def index_html(self, REQUEST, RESPONSE, format=None, force=0, **kw): def index_html(self, REQUEST, RESPONSE, format=None, **kw):
""" """
We follow here the standard Zope API for files and images We follow here the standard Zope API for files and images
and extend it to support format conversion. The idea and extend it to support format conversion. The idea
...@@ -382,16 +427,14 @@ class Document(XMLObject, UrlMixIn): ...@@ -382,16 +427,14 @@ class Document(XMLObject, UrlMixIn):
Should return appropriate format (calling convert Should return appropriate format (calling convert
if necessary) and set headers. if necessary) and set headers.
format - the format specied in the form of an extension format -- the format specied in the form of an extension
string (ex. jpeg, html, text, txt, etc.) string (ex. jpeg, html, text, txt, etc.)
force - convert doc even if it has a cached version which seems to be up2date
**kw can be various things - e.g. resolution **kw -- can be various things - e.g. resolution
TODO: TODO:
- implement guards API so that conversion to certain - implement guards API so that conversion to certain
formats require certain permission formats require certain permission
- force parameter should be somehow restricted
to prevent denial of service attack
""" """
pass pass
...@@ -421,6 +464,7 @@ class Document(XMLObject, UrlMixIn): ...@@ -421,6 +464,7 @@ class Document(XMLObject, UrlMixIn):
else: else:
val = list(val) val = list(val)
return val return val
searchable_text = reduce(add, map(lambda x: getPropertyListOrValue(x), searchable_text = reduce(add, map(lambda x: getPropertyListOrValue(x),
self.searchable_property_list)) self.searchable_property_list))
searchable_text = ' '.join(searchable_text) searchable_text = ' '.join(searchable_text)
...@@ -429,23 +473,30 @@ class Document(XMLObject, UrlMixIn): ...@@ -429,23 +473,30 @@ class Document(XMLObject, UrlMixIn):
# Compatibility with CMF Catalog # Compatibility with CMF Catalog
SearchableText = getSearchableText SearchableText = getSearchableText
security.declareProtected(Permissions.AccessContentsInformation, 'isExternalDocument')
def isExternalDocument(self):
"""
Return true if this document was obtained from an external source
"""
return bool(self.getUrlString())
### Relation getters ### Relation getters
security.declareProtected(Permissions.View, 'getSearchableReferenceList')
def getSearchableReferenceList(self): def getSearchableReferenceList(self):
""" """
Public Method
This method returns a list of dictionaries which can This method returns a list of dictionaries which can
be used to find objects by reference. It uses for be used to find objects by reference. It uses for
that a regular expression defined at system level that a regular expression defined at system level
preferences. preferences.
""" """
text = self.getSearchableText() text = self.getSearchableText() # XXX getSearchableText or asText ?
regexp = self.portal_preferences.getPreferredDocumentReferenceRegularExpression() regexp = self.portal_preferences.getPreferredDocumentReferenceRegularExpression()
try: try:
rx_search = re.compile(regexp) rx_search = re.compile(regexp)
except TypeError: # no regexp in preference except TypeError: # no regexp in preference
self.log('please set document reference regexp in preferences') LOG('ERP5/Document/Document.getSearchableReferenceList', 0,
return [] 'Document regular expression must be set in portal preferences')
return ()
res = rx_search.finditer(text) res = rx_search.finditer(text)
res = [(r.group(), r.groupdict()) for r in res] res = [(r.group(), r.groupdict()) for r in res]
return res return res
...@@ -785,8 +836,8 @@ class Document(XMLObject, UrlMixIn): ...@@ -785,8 +836,8 @@ class Document(XMLObject, UrlMixIn):
Based on the document content, find out as many properties as needed. Based on the document content, find out as many properties as needed.
returns properties which should be set on the document returns properties which should be set on the document
""" """
# XXX this method should first make sure we have text content if not self.hasBaseData():
# or do a conversion self.convertToBaseFormat()
method = self._getTypeBasedMethod('getPropertyDictFromContent', method = self._getTypeBasedMethod('getPropertyDictFromContent',
fallback_script_id='Document_getPropertyDictFromContent') fallback_script_id='Document_getPropertyDictFromContent')
return method() return method()
...@@ -821,7 +872,7 @@ class Document(XMLObject, UrlMixIn): ...@@ -821,7 +872,7 @@ class Document(XMLObject, UrlMixIn):
### Metadata disovery and ingestion methods ### Metadata disovery and ingestion methods
security.declareProtected(Permissions.ModifyPortalContent, 'discoverMetadata') security.declareProtected(Permissions.ModifyPortalContent, 'discoverMetadata')
def discoverMetadata(self, file_name, user_login=None): # XXX Was filename always there ? at least make it optional def discoverMetadata(self, file_name=None, user_login=None):
""" """
This is the main metadata discovery function - controls the process This is the main metadata discovery function - controls the process
of discovering data from various sources. The discovery itself is of discovering data from various sources. The discovery itself is
...@@ -843,88 +894,39 @@ class Document(XMLObject, UrlMixIn): ...@@ -843,88 +894,39 @@ class Document(XMLObject, UrlMixIn):
order_list.reverse() order_list.reverse()
# Start with everything until content - build a dictionary according to the order # Start with everything until content - build a dictionary according to the order
content_index = order_list.index('content')
kw = {} kw = {}
first_list = order_list[:content_index] for order_id in order_list:
for order_id in first_list:
if order_id not in VALID_ORDER_KEY_LIST: if order_id not in VALID_ORDER_KEY_LIST:
# Prevent security attack or bad preferences # Prevent security attack or bad preferences
raise AttributeError, "%s is not in valid order key list" % order_id raise AttributeError, "%s is not in valid order key list" % order_id
method_id = 'getPropertyDictFrom%s' % convertToUpperCase(order_id) method_id = 'getPropertyDictFrom%s' % convertToUpperCase(order_id)
method = getattr(self, method_id) method = getattr(self, method_id)
if order_id == 'file_name': if order_id == 'file_name':
result = method(file_name) if file_name is not None: result = method(file_name)
elif order_id == 'user_login': elif order_id == 'user_login':
result = method(user_login) if user_login is not None: result = method(user_login)
else: else:
result = method() result = method()
if result is not None: if result is not None:
# LOG('discoverMetadata %s' % order_id, 0, repr(result))
kw.update(result) kw.update(result)
# Edit content # Prepare the content edit parameters - portal_type should not be changed
try: try:
del(kw['portal_type']) del(kw['portal_type'])
except KeyError: except KeyError:
pass pass
self.edit(**kw) self._edit(**kw) # Try not to invoke an automatic transition here
self.finishIngestion() # Finish ingestion by calling method
# Finish in second stage self.reindexObject()
self.activate().finishMetadataDiscovery(user_login=user_login)
security.declareProtected(Permissions.ModifyPortalContent, 'finishMetadataDiscovery')
def finishMetadataDiscovery(self, user_login):
"""
This is called by portal_activities, to leave time-consuming procedures
for later. It converts what needs conversion to base, and
does things that can be done only after it is converted).
"""
self.convertToBaseFormat()
# Get the order from preferences
# Preference is made of a sequence of 'user_login', 'content', 'file_name', 'input'
method = self._getTypeBasedMethod('getPreferredDocumentMetadataDiscoveryOrderList',
fallback_script_id = 'Document_getPreferredDocumentMetadataDiscoveryOrderList')
order_list = list(method())
order_list.reverse()
# do content and everything after content
content_index = order_list.index('content')
second_list = order_list[content_index:]
kw = {}
for order_id in second_list:
if order_id not in VALID_ORDER_KEY_LIST:
# Prevent security attack or bad preferences
raise AttributeError, "%s is not in valid order key list" % order_id
method_id = 'getPropertyDictFrom%s' % convertToUpperCase(order_id)
method = getattr(self, method_id)
if order_id == 'file_name':
result = method(self.getSourceReference())
elif order_id == 'user_login':
result = method(user_login)
else:
result = method()
if result is not None:
kw.update(result)
# Edit content
try:
del(kw['portal_type'])
except KeyError:
pass
self.edit(**kw)
# Erase backup attributes
if hasattr(self, '_backup_input'):
delattr(self, '_backup_input')
# Finish ingestion by calling method
self.finishIngestion()
security.declareProtected(Permissions.ModifyPortalContent, 'finishIngestion') security.declareProtected(Permissions.ModifyPortalContent, 'finishIngestion')
def finishIngestion(self): def finishIngestion(self):
""" """
Finish the ingestion process by calling the appropriate script Finish the ingestion process by calling the appropriate script. This
script can for example allocate a reference number automatically if
no reference was defined.
""" """
return self._getTypeBasedMethod('finishIngestion', return self._getTypeBasedMethod('finishIngestion',
fallback_script_id='Document_finishIngestion') fallback_script_id='Document_finishIngestion')
...@@ -939,39 +941,86 @@ class Document(XMLObject, UrlMixIn): ...@@ -939,39 +941,86 @@ class Document(XMLObject, UrlMixIn):
string (ex. jpeg, html, text, txt, etc.) string (ex. jpeg, html, text, txt, etc.)
**kw can be various things - e.g. resolution **kw can be various things - e.g. resolution
Default implementation returns an empty string (html, text)
or raises an error.
TODO: TODO:
- implement guards API so that conversion to certain - implement guards API so that conversion to certain
formats require certain permission formats require certain permission
""" """
pass if format == 'html':
return 'text/html', ''
if format in ('text', 'txt'):
return 'text/plain', ''
raise NotImplementedError
security.declareProtected(Permissions.View, 'asText') security.declareProtected(Permissions.View, 'asText')
def asText(self): def asText(self):
""" """
Converts the content of the document to a textual representation. Converts the content of the document to a textual representation.
""" """
return self.convert(format='txt') mime, data = self.convert(format='txt')
return data
security.declareProtected(Permissions.View, 'asHTML') security.declareProtected(Permissions.View, 'asHTML')
def asHTML(self): def asHTML(self):
""" """
Returns a complete HTML representation of the document Returns a complete HTML representation of the document
(with body tags, etc.). (with body tags, etc.). Adds if necessary a base
""" tag so that the document can be displayed in an iframe
return self.convert(format='html') or standalone.
"""
if self.hasConversion(format='base-html'):
mime, data = self.getConversion(format='base-html')
return data
mime, html = self.convert(format='html')
if self.getUrlString():
# If a URL is defined, add the base tag
# if base is defined yet.
html = str(html)
if not html.find('<base') >= 0:
base = '<base href="%s">' % self.getContentBaseURL()
html = html.replace('<head>', '<head>%s' % base)
# We do not implement cache yet since it increases ZODB
# for probably no reason. More research needed
# self.setConversion(html, mime='text/html', format='base-html')
return html
security.declareProtected(Permissions.View, 'asStrippedHTML') security.declareProtected(Permissions.View, 'asStrippedHTML')
def asStrippedHTML(self): def asStrippedHTML(self):
""" """
Returns a stipped HTML representation of the document Returns a stripped HTML representation of the document
(without body tags, etc.) which can be used to inline (without html and body tags, etc.) which can be used to inline
a preview of the document. a preview of the document.
""" """
return self.convert(format='html') if self.hasConversion(format='stripped-html'):
mime, data = self.getConversion(format='stripped-html')
return data
mime, html = self.convert(format='html')
body_list = re.findall(self.body_parser, str(html))
if len(body_list):
return body_list[0]
return html
security.declareProtected(Permissions.AccessContentsInformation, 'getContentInformation')
def getContentInformation(self):
"""
Returns the content information from the HTML conversion.
The default implementation tries to build a dictionnary
from the HTML conversion of the document and extract
the document title.
"""
result = {}
html = self.asHTML()
if not html: return result
title_list = re.findall(self.title_parser, str(html))
if title_list:
result['title'] = title_list[0]
return result
# Base format support # Base format support
security.declareProtected(Permissions.ModifyPortalContent, 'convertToBaseFormat') security.declareProtected(Permissions.ModifyPortalContent, 'convertToBaseFormat')
def convertToBaseFormat(self, REQUEST=None): def convertToBaseFormat(self):
""" """
Converts the content of the document to a base format Converts the content of the document to a base format
which is later used for all conversions. This method which is later used for all conversions. This method
...@@ -990,7 +1039,8 @@ class Document(XMLObject, UrlMixIn): ...@@ -990,7 +1039,8 @@ class Document(XMLObject, UrlMixIn):
Use accessors (getBaseData, setBaseData, hasBaseData, etc.) Use accessors (getBaseData, setBaseData, hasBaseData, etc.)
""" """
try: try:
msg = self._convertToBaseFormat() msg = self._convertToBaseFormat() # Call implemetation method
self.clearConversionCache() # Conversion cache is now invalid
if msg is None: if msg is None:
msg = 'Converted to %s.' % self.base_format msg = 'Converted to %s.' % self.base_format
self.convertFile(comment=msg) # Invoke workflow method self.convertFile(comment=msg) # Invoke workflow method
...@@ -1025,36 +1075,6 @@ class Document(XMLObject, UrlMixIn): ...@@ -1025,36 +1075,6 @@ class Document(XMLObject, UrlMixIn):
""" """
raise NotImplementedError raise NotImplementedError
# Snapshot methods - XXX since this can be useful beyond
# documents, it should be moved to MixIn class which may
# be used, for example, to take a snapshot of an invoice.
security.declareProtected(Permissions.ModifyPortalContent, 'createSnapshot')
def createSnapshot(self):
"""
Create a snapshot (PDF). This is the normal way to modifiy
snapshot_data. Once a snapshot is taken, a new snapshot
can not be taken.
NOTE: use getSnapshotData and hasSnapshotData accessors
to access a snapshot.
NOTE2: implementation of createSnapshot should probably
be delegated to a types base method since this it
is configuration dependent.
"""
pass
security.declareProtected(Permissions.ManagePortal, 'deleteSnapshot')
def deleteSnapshot(self):
"""
Deletes the snapshot - in theory this should never be done.
It is there for programmers and system administrators.
"""
try:
del(self.snapshot_data)
except AttributeError:
pass
# Transformation API # Transformation API
security.declareProtected(Permissions.ModifyPortalContent, 'populateContent') security.declareProtected(Permissions.ModifyPortalContent, 'populateContent')
def populateContent(self): def populateContent(self):
...@@ -1091,12 +1111,12 @@ class Document(XMLObject, UrlMixIn): ...@@ -1091,12 +1111,12 @@ class Document(XMLObject, UrlMixIn):
return re.findall(self.href_parser, str(html_content)) return re.findall(self.href_parser, str(html_content))
security.declareProtected(Permissions.ModifyPortalContent, 'updateContentFromURL') security.declareProtected(Permissions.ModifyPortalContent, 'updateContentFromURL')
def updateContentFromURL(self): def updateContentFromURL(self, repeat=MAX_REPEAT, crawling_depth=0):
""" """
Download and update content of this document from its source URL. Download and update content of this document from its source URL.
Implementation is handled by ContributionTool. Implementation is handled by ContributionTool.
""" """
self.portal_contributions.updateContentFromURL(self) self.portal_contributions.updateContentFromURL(self, repeat=repeat, crawling_depth=crawling_depth)
security.declareProtected(Permissions.ModifyPortalContent, 'crawlContent') security.declareProtected(Permissions.ModifyPortalContent, 'crawlContent')
def crawlContent(self): def crawlContent(self):
...@@ -1111,7 +1131,7 @@ class Document(XMLObject, UrlMixIn): ...@@ -1111,7 +1131,7 @@ class Document(XMLObject, UrlMixIn):
Returns the content base URL based on the actual content or Returns the content base URL based on the actual content or
on its URL. on its URL.
""" """
# XXX TODO - try to retrive base URL from content # XXX TODO - try to retrieve base URL from content
# If no base_url defined, define the base URL from our URL # If no base_url defined, define the base URL from our URL
base_url = self.asURL() base_url = self.asURL()
base_url_list = base_url.split('/') base_url_list = base_url.split('/')
...@@ -1123,7 +1143,7 @@ class Document(XMLObject, UrlMixIn): ...@@ -1123,7 +1143,7 @@ class Document(XMLObject, UrlMixIn):
def getNextAlarmDate(self): def getNextAlarmDate(self):
""" """
This method is only there to have something to test. This method is only there to have something to test.
Serious refactory of Alarm, Periodicity and CalendarPeriod Serious refactoring of Alarm, Periodicity and CalendarPeriod
classes is needed. classes is needed.
""" """
return DateTime() + .01 return DateTime() + .1
...@@ -42,18 +42,6 @@ from DateTime import DateTime ...@@ -42,18 +42,6 @@ from DateTime import DateTime
mimetypes.init() mimetypes.init()
rs=[]
rs.append(re.compile('<HEAD>.*</HEAD>',re.DOTALL|re.MULTILINE|re.IGNORECASE))
rs.append(re.compile('<!DOCTYPE[^>]*>'))
rs.append(re.compile('<.?(HTML|BODY)[^>]*>',re.DOTALL|re.MULTILINE|re.IGNORECASE))
def stripHtml(txt): # XXX-JPS to be moved to TextDocument
for r in rs:
txt=r.sub('',txt)
return txt
class File(Document, CMFFile, ConversionCacheMixin): class File(Document, CMFFile, ConversionCacheMixin):
""" """
A File can contain raw data which can be uploaded and downloaded. A File can contain raw data which can be uploaded and downloaded.
...@@ -86,15 +74,18 @@ class File(Document, CMFFile, ConversionCacheMixin): ...@@ -86,15 +74,18 @@ class File(Document, CMFFile, ConversionCacheMixin):
# Default Properties # Default Properties
property_sheets = ( PropertySheet.Base property_sheets = ( PropertySheet.Base
, PropertySheet.XMLObject
, PropertySheet.CategoryCore , PropertySheet.CategoryCore
, PropertySheet.DublinCore , PropertySheet.DublinCore
, PropertySheet.Version , PropertySheet.Version
, PropertySheet.Reference , PropertySheet.Reference
, PropertySheet.Document , PropertySheet.Document
, PropertySheet.Data , PropertySheet.Data
, PropertySheet.ExternalDocument
, PropertySheet.Url
, PropertySheet.Periodicity
) )
# Declarative interfaces # Declarative interfaces
#__implements__ = ( , ) #__implements__ = ( , )
...@@ -152,7 +143,7 @@ class File(Document, CMFFile, ConversionCacheMixin): ...@@ -152,7 +143,7 @@ class File(Document, CMFFile, ConversionCacheMixin):
self.reindexObject() self.reindexObject()
security.declarePrivate('_unpackData') security.declarePrivate('_unpackData')
def _unpackData(self,data): def _unpackData(self, data):
""" """
Unpack Pdata into string Unpack Pdata into string
""" """
...@@ -178,11 +169,9 @@ class File(Document, CMFFile, ConversionCacheMixin): ...@@ -178,11 +169,9 @@ class File(Document, CMFFile, ConversionCacheMixin):
return content_type return content_type
security.declareProtected(Permissions.ModifyPortalContent,'PUT') security.declareProtected(Permissions.ModifyPortalContent,'PUT')
def PUT(self,REQUEST,RESPONSE): def PUT(self, REQUEST, RESPONSE):
self.clearConversionCache() self.clearConversionCache()
CMFFile.PUT(self,REQUEST,RESPONSE) CMFFile.PUT(self, REQUEST, RESPONSE)
self.discoverMetadata(fname=self.getId())
# DAV Support # DAV Support
index_html = CMFFile.index_html index_html = CMFFile.index_html
...@@ -191,5 +180,3 @@ class File(Document, CMFFile, ConversionCacheMixin): ...@@ -191,5 +180,3 @@ class File(Document, CMFFile, ConversionCacheMixin):
manage_FTPget = CMFFile.manage_FTPget manage_FTPget = CMFFile.manage_FTPget
manage_FTPlist = CMFFile.manage_FTPlist manage_FTPlist = CMFFile.manage_FTPlist
manage_FTPstat = CMFFile.manage_FTPstat manage_FTPstat = CMFFile.manage_FTPstat
# vim: syntax=python shiftwidth=2
...@@ -94,12 +94,16 @@ class Image(File, OFSImage): ...@@ -94,12 +94,16 @@ class Image(File, OFSImage):
# Default Properties # Default Properties
property_sheets = ( PropertySheet.Base property_sheets = ( PropertySheet.Base
, PropertySheet.XMLObject
, PropertySheet.CategoryCore , PropertySheet.CategoryCore
, PropertySheet.DublinCore , PropertySheet.DublinCore
, PropertySheet.Version , PropertySheet.Version
, PropertySheet.Reference , PropertySheet.Reference
, PropertySheet.Document , PropertySheet.Document
, PropertySheet.Data , PropertySheet.Data
, PropertySheet.ExternalDocument
, PropertySheet.Url
, PropertySheet.Periodicity
) )
# #
...@@ -188,7 +192,7 @@ class Image(File, OFSImage): ...@@ -188,7 +192,7 @@ class Image(File, OFSImage):
if not self.hasConversion(display=display, format=format, if not self.hasConversion(display=display, format=format,
quality=quality, resolution=resolution): quality=quality, resolution=resolution):
# Generate photo on-the-fly # Generate photo on-the-fly
self._makeDisplayPhoto(display, 1, format=format, quality=quality, resolution=resolution) self._makeDisplayPhoto(display, format=format, quality=quality, resolution=resolution)
mime, image = self.getConversion(display=display, format=format, mime, image = self.getConversion(display=display, format=format,
quality=quality ,resolution=resolution) quality=quality ,resolution=resolution)
width, height = (image.width, image.height) width, height = (image.width, image.height)
...@@ -290,10 +294,11 @@ class Image(File, OFSImage): ...@@ -290,10 +294,11 @@ class Image(File, OFSImage):
if not self.hasConversion(display=display, format=format, if not self.hasConversion(display=display, format=format,
quality=quality,resolution=resolution): quality=quality,resolution=resolution):
# Generate photo on-the-fly # Generate photo on-the-fly
self._makeDisplayPhoto(display, 1, format=format, quality=quality,resolution=resolution) self._makeDisplayPhoto(display, format=format, quality=quality,resolution=resolution)
# Return resized image # Return resized image
mime, image = self.getConversion(display=display, format=format, mime, image = self.getConversion(display=display, format=format,
quality=quality ,resolution=resolution) quality=quality ,resolution=resolution)
RESPONSE.setHeader('Content-Type', mime)
return image.index_html(REQUEST, RESPONSE) return image.index_html(REQUEST, RESPONSE)
# Return original image # Return original image
...@@ -307,7 +312,6 @@ class Image(File, OFSImage): ...@@ -307,7 +312,6 @@ class Image(File, OFSImage):
def _resize(self, display, width, height, quality=75, format='', resolution=None): def _resize(self, display, width, height, quality=75, format='', resolution=None):
"""Resize and resample photo.""" """Resize and resample photo."""
newimg = StringIO() newimg = StringIO()
os.putenv('TMPDIR', '/tmp') # because if we run zope as root, we have /root/tmp here and convert goes crazy
if sys.platform == 'win32': if sys.platform == 'win32':
from win32pipe import popen2 from win32pipe import popen2
...@@ -324,8 +328,9 @@ class Image(File, OFSImage): ...@@ -324,8 +328,9 @@ class Image(File, OFSImage):
imgout, imgin = popen2('convert -quality %s -geometry %sx%s - -' imgout, imgin = popen2('convert -quality %s -geometry %sx%s - -'
% (quality, width, height)) % (quality, width, height))
else: else:
LOG('Resolution',0,str(resolution)) # LOG('Resolution',0,str(resolution))
cmd = 'convert -density %sx%s -quality %s -geometry %sx%s - -' % (resolution, resolution, quality, width, height) cmd = 'convert -density %sx%s -quality %s -geometry %sx%s - -' % (resolution,
resolution, quality, width, height)
imgout, imgin = popen2(cmd) imgout, imgin = popen2(cmd)
imgin.write(str(self.getData())) imgin.write(str(self.getData()))
...@@ -357,9 +362,9 @@ class Image(File, OFSImage): ...@@ -357,9 +362,9 @@ class Image(File, OFSImage):
quality=quality,resolution=resolution)) quality=quality,resolution=resolution))
return image return image
def _makeDisplayPhoto(self, display, force=0, format='', quality=75, resolution=None): def _makeDisplayPhoto(self, display, format='', quality=75, resolution=None):
"""Create given display.""" """Create given display."""
if not self.hasConversion(display=display, format=format, quality=quality,resolution=resolution) or force: if not self.hasConversion(display=display, format=format, quality=quality,resolution=resolution):
image = self._getDisplayPhoto(display, format=format, quality=quality, resolution=resolution) image = self._getDisplayPhoto(display, format=format, quality=quality, resolution=resolution)
self.setConversion(image, mime=image.content_type, self.setConversion(image, mime=image.content_type,
display=display, format=format, display=display, format=format,
......
...@@ -125,7 +125,11 @@ class TextDocument(Document, TextContent): ...@@ -125,7 +125,11 @@ class TextDocument(Document, TextContent):
if format is None: if format is None:
# The default is to use ERP5 Forms to render the page # The default is to use ERP5 Forms to render the page
return self.view() return self.view()
return self.convert(format=format) mime, data = self.convert(format=format)
RESPONSE.setHeader('Content-Length', len(data))
RESPONSE.setHeader('Content-Type', mime)
RESPONSE.setHeader('Accept-Ranges', 'bytes')
return data
security.declareProtected(Permissions.View, 'convert') security.declareProtected(Permissions.View, 'convert')
def convert(self, format, **kw): def convert(self, format, **kw):
...@@ -136,30 +140,14 @@ class TextDocument(Document, TextContent): ...@@ -136,30 +140,14 @@ class TextDocument(Document, TextContent):
_setCacheHeaders(self, {'format' : format}) _setCacheHeaders(self, {'format' : format})
# Return the raw content # Return the raw content
if format == 'raw': if format == 'raw':
return self.getTextContent() return 'text/plain', self.getTextContent()
mime_type = getToolByName(self, 'mimetypes_registry').lookupExtension('name.%s' % format) mime_type = getToolByName(self, 'mimetypes_registry').lookupExtension('name.%s' % format)
src_mimetype = self.getTextFormat() src_mimetype = self.getTextFormat()
if not src_mimetype.startswith('text/'): if not src_mimetype.startswith('text/'):
src_mimetype = 'text/%s' % src_mimetype src_mimetype = 'text/%s' % src_mimetype
return getToolByName(self, 'portal_transforms').convertTo(mime_type, return mime_type, getToolByName(self, 'portal_transforms').convertTo(mime_type,
self.getTextContent(), object=self, mimetype=src_mimetype) self.getTextContent(), object=self, mimetype=src_mimetype)
def __call__(self): def __call__(self):
_setCacheHeaders(self, {}) _setCacheHeaders(self, {})
return Document.__call__(self) return Document.__call__(self)
### Content indexing methods
security.declareProtected(Permissions.View, 'getSearchableText')
def getSearchableText(self, md=None):
"""\
Used by the catalog for basic full text indexing
We should try to do some kind of file conversion here so that getTextContent
returns something more readable.
"""
searchable_text = "%s %s %s %s %s" % (self.getTitle(), self.getShortTitle(),
self.getDescription(),
self.getId(), self.getTextContent())
return searchable_text
# Compatibility with CMF Catalog / CPS sites
SearchableText = getSearchableText # XXX-JPS - Here wa have a security issue - ask seb what to do
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment