erp5_web: add tools to export web page as single file (mhtml or embedded html)

bt5/erp5_web/ExtensionTemplateItem/portal_components/extension.erp5.WebUtility.py

+##############################################################################
+#
+# Copyright (c) 2016 Nexedi SA and Contributors. All Rights Reserved.
+#
+# WARNING: This program as such is intended to be used by professional
+# programmers who take the whole responsibility of assessing all potential
+# consequences resulting from its eventual inadequacies and bugs
+# End users who are looking for a ready-to-use solution with commercial
+# garantees and support are strongly advised to contract a Free Software
+# Service Company
+#
+# This program is Free Software; you can redistribute it and/or
+# modify it under the terms of the GNU General Public License
+# as published by the Free Software Foundation; either version 2
+# of the License, or (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software
+# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
+#
+##############################################################################
+
+from HTMLParser import HTMLParser
+class HtmlParseHelper(HTMLParser):
+  """
+  Listens to all the HTMLParser methods and push results in a list of tuple.
+  Tuple contains every method arguments, for instance the `handle_starttag`
+  method pushes `('starttag', tag, attrs)` to the tuple list.
+
+  See https://docs.python.org/2/library/htmlparser.html
+  """
+  def __init__(self, *args, **kw):
+    HTMLParser.__init__(self, *args, **kw)
+    self.result = []
+  def handle_starttag(self, tag, attrs):
+    self.result.append(("starttag", tag, attrs))
+  def handle_startendtag(self, tag, attrs):
+    self.result.append(("startendtag", tag, attrs))
+  def handle_endtag(self, tag):
+    self.result.append(("endtag", tag))
+  def handle_data(self, data):
+    self.result.append(("data", data))
+  def handle_entityref(self, name):
+    self.result.append(("entityref", name))
+  def handle_charref(self, name):
+    self.result.append(("charref", name))
+  def handle_comment(self, data):
+    self.result.append(("comment", data))
+  def handle_decl(self, decl):
+    self.result.append(("decl", decl))
+  def handle_pi(self, data):
+    self.result.append(("pi", data))
+  def unknown_decl(self, data):
+    self.result.append(("unknown_decl", data))
+
+def parseHtml(text):
+  """
+  Parses a string and returns html parts as tuple list.
+
+  Example:
+  input: 'Click <a href="destination">here</a> to see the documentation.'
+  return: [
+    ('data', 'Click '),
+    ('starttag', 'a', ('href', 'destination')),
+    ('data', 'here'),
+    ('endtag', 'a'),
+    ('data', ' to see the documentation'),
+  ]
+  """
+  hr = HtmlParseHelper()
+  hr.feed(text)
+  hr.close()
+  return hr.result
+
+import re
+def partition(text, separatorRegexp):
+  """
+  partition("abcba", re.compile("(b)")) -> [
+    ("a",),
+    ("b", "b"),
+    ("c",),
+    ("b", "b"),
+    ("a",),
+  ]
+  """
+  result = []
+  lastIndex = 0
+  for match in separatorRegexp.finditer(text):
+    result.append((text[lastIndex:match.start()],))
+    result.append((match.group(0),) + match.groups())
+    lastIndex = match.end()
+  result.append((text[lastIndex:],))
+  return result
+
+css_comment_filter_re = re.compile(r"/\*((?:[^\*]|\*[^/])*)\*/")
+#css_url_re = re.compile(r"""(:[ \t]*url\()((")([^"]*)"|(')([^']*)'|([^\)]*))\)""")
+css_url_re = re.compile(r"""(:[ \t]*url\()(\s*(")([^"]*)"\s*|\s*(')([^']*)'\s*|([^\)]*))\)""")
+def parseCssForUrl(text):
+  """
+  return tuple list like: [
+    ("data", ""),
+    ("comment", "/* set body background image */", " set body background image "),
+    ("data", "\nbody {\n  background-image: url("),
+    ("url", "  'http://ima.ge/bg.png' ", "http://ima.ge/bg.png", "'"),
+    ("data", ");\n}\n"),
+  ]
+  """
+  result = []
+  parts = partition(text, css_comment_filter_re)  # filter comments
+  i = 0
+  for part in parts:
+    i += 1
+    if i % 2 == 0:  # comment
+      result.append(("comment", part[0], part[1]))
+    else:  # non comment
+      parts = partition(part[0], css_url_re)
+      data = ""
+      j = 0
+      for part in parts:
+        j += 1
+        if j % 2 == 1:  # css data
+          data += part[0]
+        else:  # url
+          result.append(("data", data + part[1]))
+          result.append(("url", part[2], (part[4] or part[6] or part[7] or "").strip(), part[3] or part[5] or ""))
+          data = ")"
+      result.append(("data", data))
+  return result

bt5/erp5_web/ExtensionTemplateItem/portal_components/extension.erp5.WebUtility.xml

+<?xml version="1.0"?>
+<ZopeData>
+  <record id="1" aka="AAAAAAAAAAE=">
+    <pickle>
+      <global name="Extension Component" module="erp5.portal_type"/>
+    </pickle>
+    <pickle>
+      <dictionary>
+        <item>
+            <key> <string>default_reference</string> </key>
+            <value> <string>WebUtility</string> </value>
+        </item>
+        <item>
+            <key> <string>description</string> </key>
+            <value>
+              <none/>
+            </value>
+        </item>
+        <item>
+            <key> <string>id</string> </key>
+            <value> <string>extension.erp5.WebUtility</string> </value>
+        </item>
+        <item>
+            <key> <string>portal_type</string> </key>
+            <value> <string>Extension Component</string> </value>
+        </item>
+        <item>
+            <key> <string>sid</string> </key>
+            <value>
+              <none/>
+            </value>
+        </item>
+        <item>
+            <key> <string>text_content_error_message</string> </key>
+            <value>
+              <tuple/>
+            </value>
+        </item>
+        <item>
+            <key> <string>text_content_warning_message</string> </key>
+            <value>
+              <tuple/>
+            </value>
+        </item>
+        <item>
+            <key> <string>version</string> </key>
+            <value> <string>erp5</string> </value>
+        </item>
+        <item>
+            <key> <string>workflow_history</string> </key>
+            <value>
+              <persistent> <string encoding="base64">AAAAAAAAAAI=</string> </persistent>
+            </value>
+        </item>
+      </dictionary>
+    </pickle>
+  </record>
+  <record id="2" aka="AAAAAAAAAAI=">
+    <pickle>
+      <global name="PersistentMapping" module="Persistence.mapping"/>
+    </pickle>
+    <pickle>
+      <dictionary>
+        <item>
+            <key> <string>data</string> </key>
+            <value>
+              <dictionary>
+                <item>
+                    <key> <string>component_validation_workflow</string> </key>
+                    <value>
+                      <persistent> <string encoding="base64">AAAAAAAAAAM=</string> </persistent>
+                    </value>
+                </item>
+              </dictionary>
+            </value>
+        </item>
+      </dictionary>
+    </pickle>
+  </record>
+  <record id="3" aka="AAAAAAAAAAM=">
+    <pickle>
+      <global name="WorkflowHistoryList" module="Products.ERP5Type.patches.WorkflowTool"/>
+    </pickle>
+    <pickle>
+      <tuple>
+        <none/>
+        <list>
+          <dictionary>
+            <item>
+                <key> <string>action</string> </key>
+                <value> <string>validate</string> </value>
+            </item>
+            <item>
+                <key> <string>validation_state</string> </key>
+                <value> <string>validated</string> </value>
+            </item>
+          </dictionary>
+        </list>
+      </tuple>
+    </pickle>
+  </record>
+</ZopeData>

bt5/erp5_web/SkinTemplateItem/portal_skins/erp5_web/Base_formatAttachmentListToMIMEMultipartString.py

+"""
+Usage:
+
+formatAttachmentListToMIMEMultipartString(
+  subtype="related",
+  header_dict={
+    "From": "<Saved by ERP5>",
+    "Subject": "Document Title",
+  },
+  param_list=[("type", "text/html")],
+  attachment_list=[
+    {
+      "mime_type": "text/html",
+      "charset": "utf-8",
+      "encode": "quoted-printable",
+      "header_dict": {"Content-Location": "https://www.erp5.com/My.Web.Page"},  # only add headers
+      "data": "<!DOCTYPE ...>.....................</...>",
+    },
+    {
+      "mime_type": "image/png",
+      "add_header_list": [("Content-Location", "https://www.erp5.com/My.Image")],
+      "data": "\x00............\x01",
+    }
+  ]
+);
+
+Only attachtment_list property is mandatory.
+
+Note: text/* content will not be automatically encoded to quoted-printable
+because this encoding can lose some characters like "\r" and possibly others.
+Default text/* is encoded in 7or8bit.
+
+To send specific encoded data, please make your attachment dict look like:
+
+{
+  "mime_type": "text/html",
+  "encode": "noop",
+  "add_header_list": [("Content-Transfer-Encoding", "my-encoding")],
+  "data": encodestring(html_data),
+}
+"""
+
+from email.encoders import encode_noop, encode_7or8bit, \
+                           encode_base64 as original_encode_base64
+from email.mime.base import MIMEBase
+from email.mime.text import MIMEText
+from email.mime.image import MIMEImage
+from email.mime.audio import MIMEAudio
+from email.mime.application import MIMEApplication
+from email.mime.multipart import MIMEMultipart
+import quopri
+
+def formatMultipartMessageToRFC2822String(msg):
+  """
+  The `msg.as_string()` method does not exactly follow the RFC2822. The EOL are
+  not CRLF ("\r\n") by default, so we have to replace the actual newlines
+  (LF "\n") by CRLF if necessary.
+
+  Note: The first space in each line of a multiline header will be replaced by a
+  tabulation to make some mhtml viewers able to parse it, even if a simple space
+  follows the RFC2822.
+  """
+  as_string = msg.as_string()  # it also forces the boundary generation
+  if as_string.split("\n", 1)[0].endswith("\r"):
+    return as_string
+  boundary = msg.get_boundary()
+  parts = as_string.split("\n--" + boundary)
+  parts[0] = "\r\n".join(parts[0].split("\n")).replace("\r\n ", "\r\n\t")
+  i = 0
+  for part in parts[1:]:
+    i += 1
+    partsplit = part.split("\n\n", 1)
+    partsplit[0] = "\r\n".join(partsplit[0].split("\n")).replace("\r\n ", "\r\n\t")
+    parts[i] = "\r\n\r\n".join(partsplit)
+  return ("\r\n--" + boundary).join(parts)
+
+def encode_quopri(msg):
+  """Same as encoders.encode_quopri except that spaces are kept
+  when possible and end of lines are converted to CRLF ("\r\n")
+  when necessary.
+  """
+  orig = msg.get_payload()
+  encdata = quopri.encodestring(orig).replace("=\n", "=\r\n")
+  msg.set_payload(encdata)
+  msg.add_header("Content-Transfer-Encoding", "quoted-printable")
+
+def encode_base64(msg):
+  """Extend encoders.encode_base64 to return CRLF at end of lines"""
+  original_encode_base64(msg)
+  msg.set_payload(msg.get_payload().replace("\n", "\r\n"))
+
+outer = MIMEMultipart(subtype)
+for key, value in param_list:
+  outer.set_param(key, value)
+if boundary is not None:
+  outer.set_boundary(boundary)
+if replace_header_list is not None:
+  for key, value in replace_header_list:
+    outer.replace_header(key, value)
+if header_dict is not None:  # adds headers, does not replace or set
+  for key, value in header_dict.items():
+    outer.add_header(key, value)
+if add_header_list is not None:
+  for key, value in add_header_list:
+    outer.add_header(key, value)
+for attachment in attachment_list:
+  mime_type = attachment.get("mime_type", "application/octet-stream")
+  data = attachment.get("data", "")
+  encoding = attachment.get("encode")
+  if encoding not in ("base64", "quoted-printable", "7or8bit", "noop", None):
+    raise ValueError("unknown attachment encoding %r" % encoding)
+  main_type, sub_type = mime_type.split("/")
+  if encoding is None:
+    if main_type == "image":
+      if sub_type == "svg+xml":
+        part = MIMEImage(data, sub_type, encode_quopri)  # should we trust the mime_type ?
+      else:
+        part = MIMEImage(data, sub_type, encode_base64)
+    elif main_type == "text":
+      part = MIMEText(data, sub_type, attachment.get("charset", "us-ascii"))
+    elif main_type == "audio":
+      part = MIMEAudio(data, sub_type, encode_base64)
+    elif main_type == "application":
+      part = MIMEApplication(data, sub_type, encode_noop)
+      if sub_type == "javascript":
+        encode_quopri(part)
+      else:
+        encode_base64(part)
+    else:
+      part = MIMEBase(main_type, sub_type)
+      part.set_payload(data)
+      encode_base64(part)
+  else:
+    part = MIMEBase(main_type, sub_type)
+    part.set_payload(data)
+    if encoding == "base64":
+      encode_base64(part)
+    elif encoding == "quoted-printable":
+      encode_quopri(part)
+    elif encoding == "7or8bit":
+      encode_7or8bit(part)
+    else:  # elif encoding == "noop":
+      encode_noop(part)
+  for key, value in attachment.get("replace_header_list", []):
+    part.replace_header(key, value)
+  for key, value in attachment.get("header_dict", {}).items():  # adds headers, does not replace or set
+    part.add_header(key, value)
+  for key, value in attachment.get("add_header_list", []):
+    part.add_header(key, value)
+  if attachment.get("filename", None) is not None:
+    part.add_header("Content-Disposition", "attachment", attachment["filename"])
+  outer.attach(part)
+
+#return outer.as_string()
+return formatMultipartMessageToRFC2822String(outer)

bt5/erp5_web/SkinTemplateItem/portal_skins/erp5_web/Base_formatAttachmentListToMIMEMultipartString.xml

+<?xml version="1.0"?>
+<ZopeData>
+  <record id="1" aka="AAAAAAAAAAE=">
+    <pickle>
+      <global name="PythonScript" module="Products.PythonScripts.PythonScript"/>
+    </pickle>
+    <pickle>
+      <dictionary>
+        <item>
+            <key> <string>Script_magic</string> </key>
+            <value> <int>3</int> </value>
+        </item>
+        <item>
+            <key> <string>_bind_names</string> </key>
+            <value>
+              <object>
+                <klass>
+                  <global name="NameAssignments" module="Shared.DC.Scripts.Bindings"/>
+                </klass>
+                <tuple/>
+                <state>
+                  <dictionary>
+                    <item>
+                        <key> <string>_asgns</string> </key>
+                        <value>
+                          <dictionary>
+                            <item>
+                                <key> <string>name_container</string> </key>
+                                <value> <string>container</string> </value>
+                            </item>
+                            <item>
+                                <key> <string>name_context</string> </key>
+                                <value> <string>context</string> </value>
+                            </item>
+                            <item>
+                                <key> <string>name_m_self</string> </key>
+                                <value> <string>script</string> </value>
+                            </item>
+                            <item>
+                                <key> <string>name_subpath</string> </key>
+                                <value> <string>traverse_subpath</string> </value>
+                            </item>
+                          </dictionary>
+                        </value>
+                    </item>
+                  </dictionary>
+                </state>
+              </object>
+            </value>
+        </item>
+        <item>
+            <key> <string>_params</string> </key>
+            <value> <string>attachment_list, subtype="mixed", header_dict=None, param_list=(), replace_header_list=None, add_header_list=None, boundary=None</string> </value>
+        </item>
+        <item>
+            <key> <string>id</string> </key>
+            <value> <string>Base_formatAttachmentListToMIMEMultipartString</string> </value>
+        </item>
+      </dictionary>
+    </pickle>
+  </record>
+</ZopeData>

bt5/erp5_web/SkinTemplateItem/portal_skins/erp5_web/Base_parseCssForUrl.xml

+<?xml version="1.0"?>
+<ZopeData>
+  <record id="1" aka="AAAAAAAAAAE=">
+    <pickle>
+      <global name="ExternalMethod" module="Products.ExternalMethod.ExternalMethod"/>
+    </pickle>
+    <pickle>
+      <dictionary>
+        <item>
+            <key> <string>_function</string> </key>
+            <value> <string>parseCssForUrl</string> </value>
+        </item>
+        <item>
+            <key> <string>_module</string> </key>
+            <value> <string>WebUtility</string> </value>
+        </item>
+        <item>
+            <key> <string>id</string> </key>
+            <value> <string>Base_parseCssForUrl</string> </value>
+        </item>
+        <item>
+            <key> <string>title</string> </key>
+            <value> <string></string> </value>
+        </item>
+      </dictionary>
+    </pickle>
+  </record>
+</ZopeData>

bt5/erp5_web/SkinTemplateItem/portal_skins/erp5_web/Base_parseHtml.xml

+<?xml version="1.0"?>
+<ZopeData>
+  <record id="1" aka="AAAAAAAAAAE=">
+    <pickle>
+      <global name="ExternalMethod" module="Products.ExternalMethod.ExternalMethod"/>
+    </pickle>
+    <pickle>
+      <dictionary>
+        <item>
+            <key> <string>_function</string> </key>
+            <value> <string>parseHtml</string> </value>
+        </item>
+        <item>
+            <key> <string>_module</string> </key>
+            <value> <string>WebUtility</string> </value>
+        </item>
+        <item>
+            <key> <string>id</string> </key>
+            <value> <string>Base_parseHtml</string> </value>
+        </item>
+        <item>
+            <key> <string>title</string> </key>
+            <value> <string></string> </value>
+        </item>
+      </dictionary>
+    </pickle>
+  </record>
+</ZopeData>

bt5/erp5_web/SkinTemplateItem/portal_skins/erp5_web/ERP5Site_getWebSiteDomainDict.py

+# TODO: domain names should be exported to a web site property.
+#   domain_dict = {}
+#   for web_site in portal_catalog(portal_type="Web Site", validation_state="published"):
+#     domain = web_site.getDomainName("")
+#     if domain != "":
+#       domain_dict[domain] = web_site
+#   return domain_dict
+return {}

bt5/erp5_web/SkinTemplateItem/portal_skins/erp5_web/ERP5Site_getWebSiteDomainDict.xml

+<?xml version="1.0"?>
+<ZopeData>
+  <record id="1" aka="AAAAAAAAAAE=">
+    <pickle>
+      <global name="PythonScript" module="Products.PythonScripts.PythonScript"/>
+    </pickle>
+    <pickle>
+      <dictionary>
+        <item>
+            <key> <string>Script_magic</string> </key>
+            <value> <int>3</int> </value>
+        </item>
+        <item>
+            <key> <string>_bind_names</string> </key>
+            <value>
+              <object>
+                <klass>
+                  <global name="NameAssignments" module="Shared.DC.Scripts.Bindings"/>
+                </klass>
+                <tuple/>
+                <state>
+                  <dictionary>
+                    <item>
+                        <key> <string>_asgns</string> </key>
+                        <value>
+                          <dictionary>
+                            <item>
+                                <key> <string>name_container</string> </key>
+                                <value> <string>container</string> </value>
+                            </item>
+                            <item>
+                                <key> <string>name_context</string> </key>
+                                <value> <string>context</string> </value>
+                            </item>
+                            <item>
+                                <key> <string>name_m_self</string> </key>
+                                <value> <string>script</string> </value>
+                            </item>
+                            <item>
+                                <key> <string>name_subpath</string> </key>
+                                <value> <string>traverse_subpath</string> </value>
+                            </item>
+                          </dictionary>
+                        </value>
+                    </item>
+                  </dictionary>
+                </state>
+              </object>
+            </value>
+        </item>
+        <item>
+            <key> <string>_params</string> </key>
+            <value> <string></string> </value>
+        </item>
+        <item>
+            <key> <string>id</string> </key>
+            <value> <string>ERP5Site_getWebSiteDomainDict</string> </value>
+        </item>
+      </dictionary>
+    </pickle>
+  </record>
+</ZopeData>

bt5/erp5_web/SkinTemplateItem/portal_skins/erp5_web/WebPage_exportAsSingleFile.py

+"""
+Export the web page and its components to a single (m)html file.
+
+`format` parameter could also be "mhtml".
+
+TODO: export same components into one mhtml attachment if possible.
+"""
+
+from zExceptions import Unauthorized
+from base64 import b64encode, b64decode
+portal = context.getPortalObject()
+
+mhtml_message = {
+  "subtype": "related",
+  "param_list": [("type", "text/html")],
+  "header_dict": {"From": "<Saved by ERP5>", "Subject": "Untitled"},
+  "attachment_list": [],
+}
+
+def main():
+  data = context.getTextContent("").decode("utf-8")
+  data = "".join([fn(p) for fn, p in handleHtmlPartList(parseHtml(data))])
+  if format == "mhtml":
+    mhtml_message["header_dict"]["Subject"] = context.getTitle() or "Untitled"
+    mhtml_message["attachment_list"].insert(0, {
+      "mime_type": "text/html",
+      "encode": "quoted-printable",
+      "add_header_list": [("Content-Location", context.absolute_url())],
+      "data": str(data.encode("utf-8")),
+    })
+    res = context.Base_formatAttachmentListToMIMEMultipartString(**mhtml_message)
+    if REQUEST is not None:
+      REQUEST.RESPONSE.setHeader("Content-Type", "multipart/related")
+      REQUEST.RESPONSE.setHeader("Content-Disposition", 'attachment;filename="%s-%s-%s.mhtml"' % (
+        context.getReference("untitled").replace('"', '\\"'),
+        context.getVersion("001").replace('"', '\\"'),
+        context.getLanguage("en").replace('"', '\\"'),
+      ))
+    return res
+  if REQUEST is not None:
+    REQUEST.RESPONSE.setHeader("Content-Type", "text/html")
+    REQUEST.RESPONSE.setHeader("Content-Disposition", 'attachment;filename="%s-%s-%s.html"' % (
+      context.getReference("untitled").replace('"', '\\"'),
+      context.getVersion("001").replace('"', '\\"'),
+      context.getLanguage("en").replace('"', '\\"'),
+    ))
+  return data
+
+def handleHtmlTag(tag, attrs):
+  #if tag == "base": and "href" in attrs:  # should not exist in safe-html
+  #  NotImplemented
+  if tag == "object":
+    for i in range(len(attrs)):
+      if attrs[i][0] == "data":
+        attrs[i] = attrs[i][0], handleImageSource(attrs[i][1])
+  elif tag == "img":
+    for i in range(len(attrs)):
+      if attrs[i][0] == "src":
+        attrs[i] = attrs[i][0], handleImageSource(attrs[i][1])
+  elif tag == "link" and anny(attrs, key=lambda a: a[0] == "rel" and a[1] == "stylesheet"):
+    for i in range(len(attrs)):
+      if attrs[i][0] == "href":
+        attrs[i] = attrs[i][0], replaceFromDataUri(handleCssHref(attrs[i][1]), replaceCssUrl)
+  elif tag == "script":
+    for i in range(len(attrs)):
+      if attrs[i][0] == "src":
+        attrs[i] = attrs[i][0], handleJsSource(attrs[i][1])
+  else:
+    for i in range(len(attrs)):
+      if attrs[i][0] == "href" or attrs[i][0] == "src":
+        attrs[i] = attrs[i][0], makeHrefAbsolute(attrs[i][1])
+  return tag, attrs
+
+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])
+    return "<%s%s%s>" % (tag, " " + attrs_str if attrs_str else "", " /" if part_type == "startendtag" else "")
+  if part_type == "endtag":
+    return "</%s>" % part[1]
+  if part_type == "data":
+    return part[1]
+  if part_type == "entityref":
+    return "&%s;" % part[1]
+  if part_type == "charref":
+    return "&#%s;" % part[1]
+  if part_type == "comment":
+    return "<!--%s-->" % part[1]
+  if part_type in ("decl", "unknown_decl"):
+    return "<!%s>" % part[1]
+  if part_type == "pi":
+    return "<?%s>" % part[1]
+
+disallow_script = not allow_script
+def handleHtmlPartList(part_list):
+  res = []
+  style_data = ""
+  on_script = False
+  on_style = False
+  for part in part_list:
+    if on_script:
+      if part[0] == "endtag" and part[1] == "script":
+        on_script = False
+      # can only be data until </script> endtag
+    elif on_style:
+      if part[0] == "endtag" and part[1] == "style":
+        res.append((replaceCssUrl, style_data))
+        res.append((strHtmlPart, part))
+        style_data = ""
+        on_style = False
+      else:
+        # can only be data until </style> endtag
+        style_data += strHtmlPart(part)
+    else:
+      if part[0] == "starttag":
+        # when you save a page from a browser, every script tag are removed
+        if part[1] == "script" and disallow_script:
+          on_script = True
+          continue
+        elif part[1] == "style":
+          on_style = True
+      res.append((strHtmlPart, part))
+  return res
+
+def handleCssHref(href):
+  return handleHref(href)
+
+def handleJsSource(href):
+  return handleHref(href)
+
+def handleHref(href):
+  if not isHrefAUrl(href):
+    return href
+  try:
+    o = traverseHref(href)
+  except (KeyError, Unauthorized):
+    return makeHrefAbsolute(href)
+  return handleHrefObject(o, href)
+
+def handleImageSource(src):
+  if not isHrefAUrl(src):
+    return src
+  try:
+    o = traverseHref(src)
+  except (KeyError, Unauthorized):
+    return makeHrefAbsolute(src)
+  return handleImageSourceObject(o, src)
+
+def replaceCssUrl(data):
+  parts = context.Base_parseCssForUrl(data)
+  data = ""
+  for part in parts:
+    if part[0] == "url":
+      url = part[2]
+      if isHrefAUrl(url):
+        data += handleImageSource(url)
+      else:
+        data += part[1]
+    else:
+      data += part[1]
+  return data
+
+def handleImageSourceObject(o, src):
+  if hasattr(o, "convert"):
+    search = parseUrlSearch(extractUrlSearch(src))
+    format_kw = {}
+    for k, x in search:
+      if k == "format" and x is not None:
+        format_kw["format"] = x
+      elif k == "display" and x is not None:
+        format_kw["display"] = x
+    if format_kw:
+      mime, data = o.convert(**format_kw)
+      return handleLinkedData(mime, data, src)
+
+  return handleHrefObject(o, src, default_mimetype=bad_image_mime_type, default_data=bad_image_data)
+
+def handleHrefObject(o, src, default_mimetype="text/html", default_data="<p>Linked page not found</p>"):
+  # handle File portal_skins/folder/file.png
+  # XXX handle "?portal_skin=" parameter ?
+  if hasattr(o, "getContentType"):
+    mime = o.getContentType("")
+    if mime:
+      data = getattr(o, "getData", lambda: str(o))() or ""
+      if isinstance(data, unicode):
+        data = data.encode("utf-8")
+      return handleLinkedData(mime, data, src)
+    return handleLinkedData(default_mimetype, default_data, src)
+
+  # handle Object.view
+  # XXX handle url query parameters ? Not so easy because we need to
+  # use the same behavior as when we call a script from browser URL bar.
+  if not hasattr(o, "getPortalType") and callable(o):
+    mime, data = "text/html", o()
+    if isinstance(data, unicode):
+      data = data.encode("utf-8")
+    return handleLinkedData(mime, data, src)
+
+  return handleLinkedData(default_mimetype, default_data, src)
+
+bad_image_data_url = (
+  "data:image/png;base64," +  # little image showing cannot load image
+  "iVBORw0KGgoAAAANSUhEUgAAABEAAAATCAIAAAD5x3GmAAACWklEQVQokZWSX0hTcRTHz/" +
+  "3TunMmMyxrQUzEPQSCFEI0fCi0HmSKdsUGg3q2h5I99dj7bE8Nw6cwLDb3kO7JP5m6h0TE" +
+  "CmYQjJYgpaPc7q67+93de04P0zvnQ+CP78Pvdzgfzjnf3+GICE55+NMCACACACKOj49rmv" +
+  "afvNHRUZ4/KkBEjLFQKJRTjXyRTqigUSwWI6JQKGSaJhEREQ8ApmkCgFrif+8bJ7RfMAGA" +
+  "MRYMBsPhMCLWzFPUUdVI1cjjEj0usXLXdLJ6sTCx2jIBAd1otVVe11vPbKT1iqeJRMLKKp" +
+  "fLVYaoChxGEAwDbt0ZsNs4ABAEbiLyoqYOEax/ZyfsYmX4q5iCAABQd1aoen3UGmDt/zod" +
+  "/EWnuJczcgcIABzHu91um81W9YCI8Jga6rirqUV41O9pQqeDR6J6iRvs7VUeDFQZJCKEih" +
+  "DxfINemIioq4ms7GtrwkaH4KovZ2WfujLL1/SGiIgZZSmavj2Veto0GYXO7vzawo7saztX" +
+  "3JF9+bUF6Oyu8YAAtnLvNrJBAOPb7lbkizQyPZuWfX8+LeTaG00NHDe7r8Rmju0oQaawVA" +
+  "Eqga+/Xkc+B1vexDSJzx+AJvEtk1FDEHjLAEXfXdt7ZgEA0H754UjH2GZgWFGR2UVFxc3A" +
+  "sIh4yDDGFjPPdfxhAdea/Y87xpJy//bqnN3b05XK2r0928n55P2+w3kMw9CXmy/AE4u5Fw" +
+  "h89A/tLM9d6urxTr9/G4/74zMfBvt+rsxzRKTruqIojNUsgSRJB+vrqVcv705Fc8ViqVSS" +
+  "JMnpcMz5h/4B1Qxz9NOjZCgAAAAASUVORK5CYII="
+)
+bad_image_data = b64decode(bad_image_data_url.split(",", 1)[1])
+bad_image_mime_type = "image/png"
+
+request_protocol = context.REQUEST.SERVER_URL.split(":", 1)[0] + ":"
+site_object_dict = context.ERP5Site_getWebSiteDomainDict()
+base_url_root_object = portal
+base_url_object = context
+
+def handleLinkedData(mime, data, href):
+  if format == "mhtml":
+    url = makeHrefAbsolute(href)
+    mhtml_message["attachment_list"].append({
+      "mime_type": mime,
+      "encode": "quoted-printable" if mime.startswith("text/") else None,
+      "add_header_list": [("Content-Location", url)],
+      "data": str(data),
+    })
+    return url
+  else:
+    return "data:%s;base64,%s" % (mime, b64encode(data))
+
+def makeHrefAbsolute(href):
+  if isHrefAnAbsoluteUrl(href) or not isHrefAUrl(href):
+    return href
+  if href.startswith("//"):
+    return request_protocol + href
+  if href.startswith("/"):
+    return base_url_root_object.absolute_url() + href
+  return base_url_object.absolute_url() + "/" + href
+
+def isHrefAnAbsoluteUrl(href):
+  return href.startswith("https://") or href.startswith("http://")
+
+def isHrefAUrl(href):
+  return href.startswith("https://") or href.startswith("http://") or not href.split(":", 1)[0].isalpha()
+
+def traverseHref(url, allow_hash=False):
+  url = url.split("?")[0]
+  if not allow_hash:
+    url = url.split("#")[0]
+  if url.startswith("https://") or url.startswith("http://") or url.startswith("//"):  # absolute url possibly on other sites
+    site_url = "/".join(url.split("/", 3)[:3])
+    domain = url.split("/", 3)[2]
+    relative_path = url[len(site_url):]
+    relative_path = (relative_path[1:] if relative_path[:1] == "/" else relative_path)
+    site_object = site_object_dict.get(domain)
+    if site_object is None:
+      raise KeyError(relative_path.split("/")[0])
+    return site_object.restrictedTraverse(str(relative_path))
+  if url.startswith("/"):  # absolute path, relative url
+    return base_url_root_object.restrictedTraverse(str(url[1:]))
+  # relative url (just use a base url)
+  return base_url_object.restrictedTraverse(str(url))
+
+def replaceFromDataUri(data_uri, replacer):
+  header, data = data_uri.split(",")
+  if "text/css" not in header:
+    return data_uri
+  is_base64 = False
+  if ";base64" in header:
+    is_base64 = True
+    data = b64decode(data)
+  data = replacer(data)
+  return "%s,%s" % (header, b64encode(data) if is_base64 else data)
+
+def extractUrlSearch(url):
+  url = url.split("#", 1)[0].split("?", 1)
+  url[0] = ""
+  return "?".join(url)
+
+def parseUrlSearch(search):
+  if search[:1] == "?":
+    search = search[1:]
+  result = []
+  for part in search.split("&"):
+    k = part.split("=")
+    v = "=".join(k[1:]) if len(k) else None
+    result.append((k[0], v))
+  return result
+
+def parseHtml(text):
+  return context.Base_parseHtml(text)
+
+def escapeHtml(s):
+  return s.replace("&", "&amp;").replace("<", "&lt;").replace(">", "&gt;").replace("\"", "&quot;")
+
+def anny(iterable, key=None):
+  for i in iterable:
+    if key:
+      i = key(i)
+    if i:
+      return True
+  return False
+
+return main()

bt5/erp5_web/SkinTemplateItem/portal_skins/erp5_web/WebPage_exportAsSingleFile.xml

+<?xml version="1.0"?>
+<ZopeData>
+  <record id="1" aka="AAAAAAAAAAE=">
+    <pickle>
+      <global name="PythonScript" module="Products.PythonScripts.PythonScript"/>
+    </pickle>
+    <pickle>
+      <dictionary>
+        <item>
+            <key> <string>Script_magic</string> </key>
+            <value> <int>3</int> </value>
+        </item>
+        <item>
+            <key> <string>_bind_names</string> </key>
+            <value>
+              <object>
+                <klass>
+                  <global name="NameAssignments" module="Shared.DC.Scripts.Bindings"/>
+                </klass>
+                <tuple/>
+                <state>
+                  <dictionary>
+                    <item>
+                        <key> <string>_asgns</string> </key>
+                        <value>
+                          <dictionary>
+                            <item>
+                                <key> <string>name_container</string> </key>
+                                <value> <string>container</string> </value>
+                            </item>
+                            <item>
+                                <key> <string>name_context</string> </key>
+                                <value> <string>context</string> </value>
+                            </item>
+                            <item>
+                                <key> <string>name_m_self</string> </key>
+                                <value> <string>script</string> </value>
+                            </item>
+                            <item>
+                                <key> <string>name_subpath</string> </key>
+                                <value> <string>traverse_subpath</string> </value>
+                            </item>
+                          </dictionary>
+                        </value>
+                    </item>
+                  </dictionary>
+                </state>
+              </object>
+            </value>
+        </item>
+        <item>
+            <key> <string>_params</string> </key>
+            <value> <string>REQUEST=None, allow_script=False, format="embedded_html"</string> </value>
+        </item>
+        <item>
+            <key> <string>id</string> </key>
+            <value> <string>WebPage_exportAsSingleFile</string> </value>
+        </item>
+      </dictionary>
+    </pickle>
+  </record>
+</ZopeData>