web, PortalTransforms: py3

bt5/erp5_web/DocumentTemplateItem/portal_components/document.erp5.WebSite.py

@@ -174,11 +174,7 @@ class WebSite(WebSection):
           if sub_path in section_dict:
             del section_dict[sub_path]
 
-      section_list = section_dict.values()
-
       # Sort by Index
-      section_list.sort(key=lambda x: x.getIntIndex())
-
-      return section_list
+      return sorted(section_dict.values(), key=lambda x: x.getIntIndex())
     else:
       return []
\ No newline at end of file

bt5/erp5_web/SkinTemplateItem/portal_skins/erp5_web/Base_convertHtmlToSingleFile.py

@@ -14,7 +14,8 @@ TODO: export same components into one mhtml attachment if possible.
 """
 # ERP5 web uses format= argument, which is also a python builtin
 # pylint: disable=redefined-builtin
-
+import six
+from Products.PythonScripts.standard import html_quote
 from zExceptions import Unauthorized
 from base64 import b64encode, b64decode
 portal = context.getPortalObject()
@@ -27,10 +28,11 @@ mhtml_message = {
 }
 
 def main(data):
-  if isinstance(data, str):
+  if isinstance(data, bytes):
     data = data.decode("utf-8")
   data = u"".join([fn(p) for fn, p in handleHtmlPartList(parseHtml(data))])
-  data = data.encode("utf-8")
+  if six.PY2:
+    data = data.encode("utf-8")
   if format == "mhtml":
     mhtml_message["attachment_list"].insert(0, {
       "mime_type": "text/html",
@@ -75,7 +77,7 @@ def strHtmlPart(part):
   part_type = part[0]
   if part_type in ("starttag", "startendtag"):
     tag, attrs = handleHtmlTag(part[1], part[2])
-    attrs_str = " ".join(["%s=\"%s\"" % (escapeHtml(k), escapeHtml(v or "")) for k, v in attrs])
+    attrs_str = " ".join(["%s=\"%s\"" % (html_quote(k), html_quote(v or "")) for k, v in attrs])
     return "<%s%s%s>" % (tag, " " + attrs_str if attrs_str else "", " /" if part_type == "startendtag" else "")
   if part_type == "endtag":
     return "</%s>" % part[1]
@@ -191,7 +193,7 @@ def handleHrefObject(obj, src, default_mimetype="text/html", default_data="<p>Li
         data = str(obj.data or "")
       else:
         data = getattr(obj, "getData", lambda: str(obj))() or ""
-      if isinstance(data, unicode):
+      if six.PY2 and isinstance(data, unicode):
         data = data.encode("utf-8")
       return handleLinkedData(mime, data, src)
     return handleLinkedData(default_mimetype, default_data, src)
@@ -201,7 +203,7 @@ def handleHrefObject(obj, src, default_mimetype="text/html", default_data="<p>Li
   # use the same behavior as when we call a script from browser URL bar.
   if not hasattr(obj, "getPortalType") and callable(obj):
     mime, data = "text/html", obj()
-    if isinstance(data, unicode):
+    if six.PY2 and isinstance(data, unicode):
       data = data.encode("utf-8")
     return handleLinkedData(mime, data, src)
 
@@ -270,7 +272,7 @@ def handleLinkedData(mime, data, href):
     })
     return url
   else:
-    return "data:%s;base64,%s" % (mime, b64encode(data))
+    return "data:%s;base64,%s" % (mime, b64encode(data.encode()).decode())
 
 def makeHrefAbsolute(href):
   if isHrefAnAbsoluteUrl(href) or not isHrefAUrl(href):
@@ -325,7 +327,8 @@ def replaceFromDataUri(data_uri, replacer):
   if ";base64" in header:
     is_base64 = True
     data = b64decode(data)
-  data = replacer(data)
+  if not is_base64:
+    data = replacer(data)
   return "%s,%s" % (header, b64encode(data) if is_base64 else data)
 
 def extractUrlSearch(url):
@@ -346,9 +349,6 @@ def parseUrlSearch(search):
 def parseHtml(text):
   return context.Base_parseHtml(text)
 
-def escapeHtml(text):
-  return text.replace("&", "&amp;").replace("<", "&lt;").replace(">", "&gt;").replace("\"", "&quot;")
-
 def anny(iterable, key=None):
   for i in iterable:
     if key:

bt5/erp5_web/TestTemplateItem/portal_components/test.erp5.testERP5Web.py

@@ -268,6 +268,7 @@ class TestERP5Web(ERP5TypeTestCase):
     page.edit(text_content='<p>Hé Hé Hé!</p>', content_type='text/html')
     self.tic()
     self.assertEqual('Hé Hé Hé!', page.asText().strip())
+    self.assertIn('Hé Hé Hé!', page.getSearchableText())
 
   def test_WebPageAsTextHTMLEntities(self):
     """Check if Web Page's asText() converts html entities properly
@@ -1032,12 +1033,10 @@ Hé Hé Hé!""", page.asText().strip())
     web_section_portal_type = 'Web Section'
     web_section = website.newContent(portal_type=web_section_portal_type)
 
-    content = '<p>initial text</p>'
-    new_content = '<p>modified text<p>'
     document = portal.web_page_module.newContent(portal_type='Web Page',
             id='document_cache',
             reference='NXD-Document.Cache',
-            text_content=content)
+            text_content='<p>initial text</p>')
     document.publish()
     self.tic()
     self.assertEqual(document.asText().strip(), 'initial text')
@@ -1051,15 +1050,15 @@ Hé Hé Hé!""", page.asText().strip())
     # Through the web_site.
     path = website.absolute_url_path() + '/NXD-Document.Cache'
     response = self.publish(path, self.credential)
-    self.assertNotEqual(response.getBody().find(content), -1)
+    self.assertIn(b'<p>initial text</p>', response.getBody())
 
     # Through a web_section.
     path = web_section.absolute_url_path() + '/NXD-Document.Cache'
     response = self.publish(path, self.credential)
-    self.assertNotEqual(response.getBody().find(content), -1)
+    self.assertIn(b'<p>initial text</p>', response.getBody())
 
     # modified the web_page content
-    document.edit(text_content=new_content)
+    document.edit(text_content='<p>modified text<p>')
     self.assertEqual(document.asText().strip(), 'modified text')
     self.tic()
 
@@ -1067,12 +1066,12 @@ Hé Hé Hé!""", page.asText().strip())
     # Through the web_site.
     path = website.absolute_url_path() + '/NXD-Document.Cache'
     response = self.publish(path, self.credential)
-    self.assertNotEqual(response.getBody().find(new_content), -1)
+    self.assertIn(b'<p>modified</p>', response.getBody())
 
     # Through a web_section.
     path = web_section.absolute_url_path() + '/NXD-Document.Cache'
     response = self.publish(path, self.credential)
-    self.assertNotEqual(response.getBody().find(new_content), -1)
+    self.assertIn(b'<p>modified</p>', response.getBody())
 
   def test_13a_DocumentMovedCache(self):
     """
@@ -1123,12 +1122,10 @@ Hé Hé Hé!""", page.asText().strip())
     web_section_portal_type = 'Web Section'
     web_section = website.newContent(portal_type=web_section_portal_type)
 
-    content = '<p>initial text</p>'
-    new_content = '<p>modified text</p>'
     document = portal.web_page_module.newContent(portal_type='Web Page',
             id='document_cache',
             reference='NXD-Document.Cache',
-            text_content=content)
+            text_content='<p>initial text</p>')
     document.publish()
     self.tic()
     self.assertEqual(document.asText().strip(), 'initial text')
@@ -1136,16 +1133,16 @@ Hé Hé Hé!""", page.asText().strip())
     # Through the web_site.
     path = website.absolute_url_path() + '/NXD-Document.Cache'
     response = self.publish(path, self.credential)
-    self.assertNotEqual(response.getBody().find(content), -1)
+    self.assertIn(b'<p>initial text</p>', response.getBody())
     # Through a web_section.
     path = web_section.absolute_url_path() + '/NXD-Document.Cache'
     response = self.publish(path, self.credential)
-    self.assertNotEqual(response.getBody().find(content), -1)
+    self.assertIn(b'<p>initial text</p>', response.getBody())
 
     # Modify the web_page content
     # Use unrestrictedTraverse (XXX-JPS reason unknown)
     web_document = website.unrestrictedTraverse('web_page_module/%s' % document.getId())
-    web_document.edit(text_content=new_content)
+    web_document.edit(text_content='<p>modified text</p>')
     # Make sure cached is emptied
     self.assertFalse(web_document.hasConversion(format='txt'))
     self.assertFalse(document.hasConversion(format='txt'))
@@ -1170,14 +1167,14 @@ Hé Hé Hé!""", page.asText().strip())
     self.assertEqual(web_document.asText().strip(), 'modified text')
     path = web_section.absolute_url_path() + '/NXD-Document.Cache'
     response = self.publish(path, self.credential)
-    self.assertNotEqual(response.getBody().find(new_content), -1)
+    self.assertIn(b'<p>modified text</p>', response.getBody())
 
     # Through a web_site.
     web_document = website.restrictedTraverse('NXD-Document.Cache')
     self.assertEqual(web_document.asText().strip(), 'modified text')
     path = website.absolute_url_path() + '/NXD-Document.Cache'
     response = self.publish(path, self.credential)
-    self.assertNotEqual(response.getBody().find(new_content), -1)
+    self.assertIn(b'<p>modified text</p>', response.getBody())
 
   def test_14_AccessWebSiteForWithDifferentUserPreferences(self):
     """Check that Ram Cache Manager do not mix websection
@@ -1239,18 +1236,18 @@ Hé Hé Hé!""", page.asText().strip())
 
     # connect as administrator and check that only developper_mode is enable
     response = self.publish(websection_url, 'administrator:administrator')
-    self.assertIn('manage_main', response.getBody())
-    self.assertNotIn('manage_messages', response.getBody())
+    self.assertIn(b'manage_main', response.getBody())
+    self.assertNotIn(b'manage_messages', response.getBody())
 
     # connect as webeditor and check that only translator_mode is enable
     response = self.publish(websection_url, 'webeditor:webeditor')
-    self.assertNotIn('manage_main', response.getBody())
-    self.assertIn('manage_messages', response.getBody())
+    self.assertNotIn(b'manage_main', response.getBody())
+    self.assertIn(b'manage_messages', response.getBody())
 
     # anonymous user doesn't exists, check anonymous access without preferences
     response = self.publish(websection_url, 'anonymous:anonymous')
-    self.assertNotIn('manage_main', response.getBody())
-    self.assertNotIn('manage_messages', response.getBody())
+    self.assertNotIn(b'manage_main', response.getBody())
+    self.assertNotIn(b'manage_messages', response.getBody())
 
   def test_15_Check_LastModified_Header(self):
     """Checks that Last-Modified header set by caching policy manager
@@ -1416,7 +1413,7 @@ Hé Hé Hé!""", page.asText().strip())
     self.assertEqual(HTTP_OK, response.getStatus())
     self.assertEqual('text/html; charset=utf-8',
                       response.getHeader('content-type'))
-    self.assertIn("Data updated.", response.getBody())
+    self.assertIn(b"Data updated.", response.getBody())
 
     self.tic()
 
@@ -1472,7 +1469,7 @@ Hé Hé Hé!""", page.asText().strip())
     self.assertEqual(HTTP_OK, response.getStatus())
     self.assertEqual('text/html; charset=utf-8',
                       response.getHeader('content-type'))
-    self.assertIn("Data updated.", response.getBody())
+    self.assertIn(b"Data updated.", response.getBody())
 
     self.tic()
 

product/ERP5/bootstrap/erp5_core/DocumentTemplateItem/portal_components/document.erp5.TextDocument.py

@@ -163,7 +163,11 @@ class TextDocument(CachedConvertableMixin, BaseConvertableFileMixin, TextContent
         if mime_type == 'text/html':
           mime_type = 'text/x-html-safe'
         if src_mimetype != "image/svg+xml":
-          result = portal_transforms.convertToData(mime_type, text_content,
+          if six.PY2:
+            data = text_content
+          else:
+            data = text_content.encode()
+          result = portal_transforms.convertToData(mime_type, data,
                                                    object=self, context=self,
                                                    filename=filename,
                                                    mimetype=src_mimetype,
@@ -373,6 +377,7 @@ class TextDocument(CachedConvertableMixin, BaseConvertableFileMixin, TextContent
                                                   text_content, content_type)
       else:
         message = 'Conversion to base format succeeds'
+    # TODO(zope4py3): rethink this, shouldn't we store bytes in base data ?
     self._setBaseData(text_content)
     self._setBaseContentType(content_type)
     return message
@@ -385,15 +390,17 @@ class TextDocument(CachedConvertableMixin, BaseConvertableFileMixin, TextContent
     self._checkConversionFormatPermission(None)
     if default is _MARKER:
       text_content = self._baseGetTextContent()
-    text_content = self._baseGetTextContent(default)
+    else:
+      text_content = self._baseGetTextContent(default)
     if isinstance(text_content, bytes):
-      # XXX Zope4py3: should this return str ??
+      # TODO(Zope4py3): should this return str ??
       # We probably have "legacy" documents where `text_content` is a python2
       # str encoded as something else than utf-8.
       # Maybe we should introduce a new text_content_encoding property and
       # expose API to getRawTextContent (as bytes) and getTextContent would return
       # the decoded string.
       # XXX what about _convertToBaseFormat/guessCharsetAndConvert ???
+      LOG('TextDocument', WARNING, "getTextContent with bytes %s" % text_content)
       try:
         text_content = text_content.decode('utf-8')
       except UnicodeDecodeError:

product/ERP5/bootstrap/erp5_core/MixinTemplateItem/portal_components/mixin.erp5.TextConvertableMixin.py

@@ -32,6 +32,7 @@ from Products.ERP5Type.Globals import InitializeClass
 from Products.ERP5Type import Permissions
 from warnings import warn
 
+
 class TextConvertableMixin:
   """
   This class provides a generic implementation of ITextConvertable.
@@ -46,9 +47,9 @@ class TextConvertableMixin:
     """
     Converts the current document to plain text
     """
-    kw.pop('format', None)
-    _, data = self.convert(format='txt', **kw)
-    return str(data)
+    kw['format'] = 'txt'
+    _, data = self.convert(**kw)
+    return data
 
   security.declareProtected(Permissions.AccessContentsInformation,
                             'asRawText')
@@ -56,9 +57,9 @@ class TextConvertableMixin:
     """
     Converts the current document to plain text without substitution
     """
-    kw.pop('format', None)
-    _, data = self.convert(format='txt', substitute=False, **kw)
-    return str(data)
+    kw['format'] = 'txt'
+    kw['substitute'] = False
+    return self.asText(**kw)
 
   security.declareProtected(Permissions.AccessContentsInformation,
                             'asTextContent')

product/PortalTransforms/Transform.py

@@ -20,8 +20,15 @@ from Products.PortalTransforms.transforms.broken import BrokenTransform
 
 
 def import_from_name(module_name):
-    """ import and return a module by its name """
-    return __import__(module_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 as e:
+        raise ImportError(str(e))
+    return m
 
 def make_config_persistent(kwargs):
     """ iterates on the given dictionnary and replace list by persistent list,

product/PortalTransforms/libtransforms/commandtransform.py

@@ -154,7 +154,7 @@ class subprocesstransform:
         try:
             if not self.useStdin:
               stdin_file = tempfile.NamedTemporaryFile()
-              stdin_file.write( data)
+              stdin_file.write(data)
               stdin_file.seek(0)
               command = command % {'infile': stdin_file.name} # apply tmp name to command
               data = None

product/PortalTransforms/libtransforms/utils.py

@@ -230,9 +230,4 @@ class IllegalHTML( ValueError ):
 #                 j = i + len(toHandle)
 #         return j
 
-# def scrubHTML( html ):
-#     """ Strip illegal HTML tags from string text.  """
-#     parser = StrippingParser()
-#     parser.feed( html )
-#     parser.close()
-#     return parser.result
+from Products.PortalTransforms.transforms.safe_html import scrubHTML

product/PortalTransforms/transforms/html_to_text.py

@@ -21,14 +21,14 @@ def register():
             return unichr(result).encode('utf-8')
 
     return html_to_text("html_to_text",
-                       ('<script [^>]>.*</script>(?im)', ' '),
-                       ('<style [^>]>.*</style>(?im)', ' '),
-                       ('<head [^>]>.*</head>(?im)', ' '),
+                       ('(?im)<script [^>]>.*</script>', ' '),
+                       ('(?im)<style [^>]>.*</style>', ' '),
+                       ('(?im)<head [^>]>.*</head>', ' '),
 
                        # added for ERP5, we want to transform <br/> in newlines
-                       ('<br\s*/?>(?im)', '\n'),
+                       ('(?im)<br\s*/?>', '\n'),
 
                        ('(?im)</?(font|em|i|strong|b)(?=\W)[^>]*>', ''),
-                       ('<[^>]*>(?i)(?m)', ' '),
+                       ('(?i)(?m)<[^>]*>', ' '),
                        (r'&([a-zA-Z0-9#]*?);', sub_func),
                        )

product/PortalTransforms/transforms/safe_html.py

@@ -226,7 +226,7 @@ class StrippingParser(HTMLParser):
     def handle_data(self, data):
         if self.suppress: return
         data = html_quote(data)
-        if self.original_charset and isinstance(data, str):
+        if self.original_charset and isinstance(data, bytes):
             data = data.decode(self.original_charset)
         self.result.append(data)
 
@@ -332,7 +332,7 @@ class StrippingParser(HTMLParser):
                  k = len(self.rawdata)
              data = self.rawdata[i+9:k]
              j = k+3
-             if self.original_charset and isinstance(data, str):
+             if self.original_charset and isinstance(data, bytes):
                  data = data.decode(self.original_charset)
              self.result.append("<![CDATA[%s]]>" % data)
         else:
@@ -378,7 +378,7 @@ def scrubHTML(html, valid=VALID_TAGS, nasty=NASTY_TAGS,
     parser.feed(html)
     parser.close()
     result = parser.getResult()
-    if parser.original_charset and isinstance(result, str):
+    if parser.original_charset and isinstance(result, bytes):
         result = result.decode(parser.original_charset).encode(default_encoding)
     return result
 

product/PortalTransforms/transforms/text_pre_to_html.py

+import six
 from Products.PortalTransforms.interfaces import ITransform
 from zope.interface import implementer
 from DocumentTemplate.html_quote import html_quote
@@ -30,6 +31,7 @@ class TextPreToHTML:
         raise AttributeError(attr)
 
     def convert(self, orig, data, **kwargs):
+        orig = six.ensure_text(orig, errors='replace')
         data.setData('<pre class="data">%s</pre>' % html_quote(orig))
         return data
 

product/PortalTransforms/transforms/text_to_html.py

+import six
 from Products.PortalTransforms.interfaces import ITransform
 from zope.interface import implementer
 from DocumentTemplate.html_quote import html_quote
@@ -30,6 +31,7 @@ class TextToHTML:
         raise AttributeError(attr)
 
     def convert(self, orig, data, **kwargs):
+        orig = six.ensure_text(orig, errors='replace')
         # 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

product/PortalTransforms/utils.py

@@ -2,12 +2,13 @@
 """some common utilities
 """
 
+import io
 import six
 if six.PY2:
-    from email import message_from_file as message_from_bytes
+    from email import message_from_string as message_from_bytes
 else:
     from email import message_from_bytes
-from six.moves import cStringIO as StringIO
+
 
 class TransformException(Exception):
     pass
@@ -35,4 +36,6 @@ def safeToInt(value):
 
 def parseContentType(content_type):
   """Parses `text/plain;charset="utf-8"` to a email.Message object"""
-  return message_from_bytes(StringIO("Content-Type:" + content_type.replace("\r\n", "\r\n\t")))
+  return message_from_bytes(
+    b"Content-Type:"
+    + content_type.replace("\r\n", "\r\n\t").encode('utf-8'))