############################################################################## # # Copyright (c) 2007 Nexedi SARL and Contributors. All Rights Reserved. # Jean-Paul Smets <jp@nexedi.com> # # WARNING: This program as such is intended to be used by professional # programmers who take the whole responsability 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 adviced 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. # ############################################################################## import cStringIO import re import string import socket import md5 import urllib2, urllib from AccessControl import ClassSecurityInfo, getSecurityManager from Globals import InitializeClass, DTMLFile from Products.ERP5Type.Tool.BaseTool import BaseTool from Products.ERP5Type import Permissions from Products.ERP5 import _dtmldir from Products.ERP5.Document.Url import no_crawl_protocol_list, no_host_protocol_list from zLOG import LOG from DateTime import DateTime from Acquisition import aq_base # Install openers import ContributionOpener opener = urllib2.build_opener(ContributionOpener.DirectoryFileHandler) urllib2.install_opener(opener) # A temporary hack until urllib2 supports timeout setting - XXX import socket socket.setdefaulttimeout(600) # 1 minute timeout # Global parameters TEMP_NEW_OBJECT_KEY = '_v_new_object' MAX_REPEAT = 10 _marker = [] # Create a new marker object. class ContributionTool(BaseTool): """ ContributionTool provides an abstraction layer to unify the contribution of documents into an ERP5 Site. ContributionTool needs to be configured in portal_types (allowed contents) so that it can store Text, Spreadsheet, PDF, etc. The main method of ContributionTool is newContent. This method can be provided various parameters from which the portal type and document metadata can be derived. Configuration Scripts: - ContributionTool_getPropertyDictFromFileName: receives file name and a dict derived from filename by regular expression, and does any necesary operations (e.g. mapping document type id onto a real portal_type). Problems which are not solved - handling of relative links in HTML contents (or others...) some text rewriting is necessary. """ title = 'Contribution Tool' id = 'portal_contributions' meta_type = 'ERP5 Contribution Tool' portal_type = 'Contribution Tool' # Regular expressions simple_normaliser = re.compile('#.*') # Declarative Security security = ClassSecurityInfo() security.declareProtected(Permissions.ManagePortal, 'manage_overview' ) manage_overview = DTMLFile( 'explainContributionTool', _dtmldir ) security.declarePrivate('findTypeName') def findTypeName(self, file_name, document, container=None): """ Finds the appropriate portal type based on the file name or if necessary the content of the document. NOTE: XXX This implementation can be greatly accelerated by caching a dict resulting which combines getContentTypeRegistryTypeDict and valid_portal_type_list """ def getContentTypeRegistryTypeDict(): result = {} for id, pred in self.content_type_registry.listPredicates(): (p, type) = pred result[type] = None return result portal_type = None # We should only consider those portal_types which share the # same constructor with the current object and which are not # part of the definitions of content_type_registry. For # example if content type registry has a definition for # RSS feed, then there is no reason to consider this type # whenever receiving some text/html content although both # types share the same constructor. However, if Memo has # same constructor as Text and Memo is not in content_type_registry # then it should be considered. extra_valid_portal_type_list = [] content_registry_type_dict = getContentTypeRegistryTypeDict() portal_type_tool = self.portal_types for pt in portal_type_tool.objectValues(): if hasattr(pt, 'factory') and pt.factory == portal_type_tool[document.getPortalType()].factory: if not content_registry_type_dict.has_key(pt.id): extra_valid_portal_type_list.append(pt.id) if not extra_valid_portal_type_list: # There is really no ambiguity here # The portal_type set by PUT_factory is appropriate # This is the best case we can get # LOG('findTypeName no ambiguity', 0, document.portal_type) return document.portal_type valid_portal_type_list = [document.portal_type] + extra_valid_portal_type_list # LOG('valid_portal_type_list', 0, str(valid_portal_type_list)) # If a container is defined, filter valid portal types with allowedContentTypes if container is not None: allowed_type_list = map(lambda x: x.id, container.allowedContentTypes()) # LOG('allowed_type_list', 0, str(allowed_type_list)) valid_portal_type_list = filter(lambda x: x in allowed_type_list, valid_portal_type_list) # LOG('filtered valid_portal_type_list', 0, str(valid_portal_type_list)) # Check if there is any intersection with index portal types # If not, we do not need to even check if content is an index is_index_candidate = False for index_type in self.getPortalCrawlerIndexTypeList(): if index_type in valid_portal_type_list: is_index_candidate = True candidate_index_type = index_type if is_index_candidate and document.isIndexContent(container=container): # If this document has to be created inside an External Source (container) # we need to analyse its content to determine whether it is or not # an index document. Index documents should not be searchable as documents # and should not be considered in the depth calculation of the crawling # process return candidate_index_type # We suppose that there is only one index type in allowed content types # Check if the filename tells which portal_type this is portal_type_list = self.getPropertyDictFromFileName(file_name).get('portal_type', []) if isinstance(portal_type_list, str): portal_type_list = [portal_type_list] portal_type_list = filter(lambda x: x in valid_portal_type_list, portal_type_list) if not portal_type_list: portal_type_list = valid_portal_type_list if len(portal_type_list) == 1: # if we have only one, then this is it # LOG('findTypeName single portal_type_list', 0, portal_type_list[0]) return portal_type_list[0] # If it is still None, we need to read the document # to check which of the candidates is suitable # Let us give a chance to getPropertyDictFromContent to # tell us what is the portal type of this document content_portal_type_list = document.getPropertyDictFromContent().get('portal_type', None) if content_portal_type_list: if isinstance(portal_type, str): content_portal_type_list = [content_portal_type_list] # Filter valid candidates content_portal_type_list = filter(lambda x: x in portal_type_list, content_portal_type_list) if content_portal_type_list: # if we have more than one, then return the first one # LOG('findTypeName from content', 0, content_portal_type_list[0]) return content_portal_type_list[0] # If portal_type_list is not empty, return the first one # LOG('findTypeName from first portal_type_list', 0, portal_type_list[0]) return portal_type_list[0] security.declareProtected(Permissions.AddPortalContent, 'newContent') def newContent(self, id=None, portal_type=None, url=None, container=None, container_path=None, discover_metadata=1, temp_object=0, user_login=None, data=None, file_name=None, **kw): """ The newContent method is overriden to implement smart content creation by detecting the portal type based on whatever information was provided and finding out the most appropriate module to store the content. user_login is the name under which the content will be created XXX - this is a security hole which needs to be fixed by making sure only Manager can use this parameter container -- if specified, it is possible to define where to contribute the content. Else, ContributionTool tries to guess. container_path -- if specified, defines the container path and has precedence over container url -- if specified, content is download from the URL. NOTE: We always generate ID. So, we must prevent using the one which we were provided. """ if file_name is not None: kw['file_name'] = file_name if data is not None: kw['data'] = data # This is only used to make sure # we can pass file as parameter to ZPublisher # whenever we ingest email # Temp objects use the standard newContent from Folder if temp_object: # For temp_object creation, use the standard method return BaseTool.newContent(self, id=id, portal_type=portal_type, temp_object=1, **kw) # Try to find the file_name file_name = None mime_type = None if not url: # check if file was provided file = kw.get('file', None) if file is not None: file_name = file.filename else: # some channels supply data and file-name separately # this is the case for example for email ingestion # in this case, we build a file wrapper for it data = kw.get('data', None) if data is not None: file_name = kw.get('file_name', None) if file_name is not None: file = cStringIO.StringIO() file.write(data) file.seek(0) kw['file'] = file del kw['data'] del kw['file_name'] else: # build a new file from the url url_file = urllib2.urlopen(url) data = url_file.read() # time out must be set or ... too long XXX file = cStringIO.StringIO() file.write(data) file.seek(0) # Create a file name based on the URL and quote it file_name = url.split('/')[-1] or url.split('/')[-2] file_name = urllib.quote(file_name, safe='') file_name = file_name.replace('%', '') # For URLs, we want an id by default equal to the encoded URL if id is None: id = self.encodeURL(url) if hasattr(url_file, 'headers'): headers = url_file.headers if hasattr(headers, 'type'): mime_type = headers.type kw['file'] = file # If the portal_type was provided, we can go faster if portal_type and container is None: # We know the portal_type, let us find the default module # and use it as container try: container = self.getDefaultModule(portal_type) except ValueError: container = None if portal_type and container is not None: # We could simplify things here and return a document immediately # NOTE: we use the module ID generator rather than the provided ID #document = module.newContent(portal_type=portal_type, **kw) #if discover_metadata: # document.activate().discoverMetadata(file_name=file_name, user_login=user_login) #return document pass # XXX - This needs to be implemented once the rest is stable # From here, there is no hope unless a file was provided if file is None: raise ValueError, "could not determine portal type" # So we will simulate WebDAV to get an empty object # with PUT_factory - we provide the mime_type as # parameter # LOG('new content', 0, "%s -- %s" % (file_name, mime_type)) ob = self.PUT_factory(file_name, mime_type, None) # Raise an error if we could not guess the portal type if ob is None: raise ValueError, "Could not determine the document type" # Prevent any reindexing operations ob.isIndexable = 0 # Then put the file inside ourselves for a short while BaseTool._setObject(self, file_name, ob) document = BaseTool._getOb(self, file_name) try: # Then edit the document contents (so that upload can happen) document._edit(set_filename__=0, **kw) if url: document.fromURL(url) finally: # Remove the object from ourselves BaseTool._delObject(self, file_name) # Move the document to where it belongs if container_path is not None: container = self.getPortalObject().restrictedTraverse(container_path) document = self._setObject(file_name, ob, user_login=user_login, id=id, container=container, discover_metadata=discover_metadata, ) document = self._getOb(file_name) # Call _getOb to purge cache # Notify workflows #document.notifyWorkflowCreated() # Allow reindexing, reindex it and return the document delattr(document, 'isIndexable') document.reindexObject() return document security.declareProtected( Permissions.AddPortalContent, 'newXML' ) def newXML(self, xml): """ Create a new content based on XML data. This is intended for contributing to ERP5 from another application. """ pass security.declareProtected(Permissions.ModifyPortalContent,'getPropertyDictFromFileName') def getPropertyDictFromFileName(self, file_name): """ Gets properties from filename. File name is parsed with a regular expression set in preferences. The regexp should contain named groups. """ if file_name is None: return {} property_dict = {} rx_src = self.portal_preferences.getPreferredDocumentFileNameRegularExpression() if rx_src in ('', None): # we must have file name regular expression defined in preferences raise ValueError, '[DMS] No file name regular expression defined in preferences.' if rx_src: rx_parse = re.compile(rx_src) if rx_parse is not None: try: property_dict = rx_parse.match(file_name).groupdict() except AttributeError: # no match pass method = self._getTypeBasedMethod('getPropertyDictFromFileName', fallback_script_id = 'ContributionTool_getPropertyDictFromFileName') property_dict = method(file_name, property_dict) if property_dict.get('portal_type', None) is not None: # we have to return portal_type as a tuple # because we should allow for having multiple candidate types property_dict['portal_type'] = (property_dict['portal_type'],) else: # we have to find candidates by file extenstion if file_name.rfind('.')!= -1: ext = file_name.split('.')[-1] property_dict['portal_type'] = self.ContributionTool_getCandidateTypeListByExtension(ext) return property_dict # WebDAV virtual folder support def _setObject(self, name, ob, user_login=None, container=None, id=None, discover_metadata=1): """ The strategy is to let NullResource.PUT do everything as usual and at the last minute put the object in a different location with a different portal type. This means that NullResource.PUT creates an empty document with PUT_factory then upload document data by invoking PUT on the empty document and finally sets the object. By overriding _setObject we get a chance to fix the portal_type of the document (as long as the one we find is compatible) and move the document to the appropriate module. content_type_registry must be set up so that an appropriate portal_type with appropriate meta_type is found for every kind of document. However, a different portal_type might be used in the end. The ContributionTool instance must be configured in such way that _verifyObjectPaste will return TRUE. Refer to: NullResource.PUT """ # _setObject is called by constructInstance at a time # when the object has no portal_type defined yet. It # will be removed later on. We can safely store the # document inside us at this stage. Else we # must find out where to store it. if not ob.__dict__.has_key('portal_type'): BaseTool._setObject(self, name, ob) document = self[name] else: # We give the system a last chance to analyse the # portal_type based on the document content # (ex. a Memo is a kind of Text which can be identified # by the fact it includes some specific content) portal_type = self.findTypeName(name, ob.__of__(self), container=container) if portal_type is None: portal_type = ob.portal_type ob._setPortalTypeName(portal_type) # This is redundant with finishConstruction # but necessary to move objects to appropriate # location based on their content. Since the # object is already constructed here, we # can safely change its portal_type # Now we know the portal_type, let us find the module # to which we should move the document to if container is None: module = self.getDefaultModule(ob.portal_type) else: module = container if id is None: new_id = module.generateNewId() else: new_id = id ob.id = new_id existing_document = module.get(new_id, None) if existing_document is None: # There is no preexisting document - we can therefore # set the new object module._setObject(new_id, ob) # We can now discover metadata document = module[new_id] if discover_metadata: # Metadata disovery is done as an activity by default # If we need to discoverMetadata synchronously, it must # be for user interface and should thus be handled by # ZODB scripts document.activate().discoverMetadata(file_name=name, user_login=user_login) else: if document.isExternalDocument(): document = existing_document # If this is an external document, update its content # document.activate().updateContentFromURL() # XXX I think this is no longer useful with alarms # XXX - Make sure this does not increase ZODB # XXX - what to do also with parameters (put again edit_kw) ? # Providing some information to the use about the fact # this was an existing document would also be great else: # We may have to implement additional revision support # to support in place contribution (ie. for a given ID) # but is this really useful ? raise NotImplementedError # Keep the document close to us - this is only useful for # file upload from webdav if not hasattr(self, '_v_document_cache'): self._v_document_cache = {} self._v_document_cache[name] = document.getRelativeUrl() # Return document to newContent method return document def _getOb(self, id, default=_marker): """ Check for volatile temp object info first and try to find it """ # Use the document cache if possible and return result immediately # this is only useful for webdav if hasattr(self, '_v_document_cache'): document_url = self._v_document_cache.get(id, None) if document_url is not None: del self._v_document_cache[id] return self.getPortalObject().unrestrictedTraverse(document_url) # Try first to return the real object inside # This is much safer than trying to access objects displayed by listDAVObjects # because the behaviour of catalog is unpredicatble if a string is passed # for a UID. For example # select path from catalog where uid = "001193.html"; # will return the same as # select path from catalog where uid = 1193; # This was the source of an error in which the contribution tool # was creating a web page and was returning a Base Category # when # o = folder._getOb(id) # was called in DocumentConstructor result = BaseTool._getOb(self, id, default=default) if result is not _marker: return result # Return an object listed by listDAVObjects uid = str(id).split('-')[-1] object = self.getPortalObject().portal_catalog.unrestrictedGetResultValue(uid=uid) if object is not None: return object.getObject() # Make sure this does not break security. XXX # Raise an AttributeError the same way as in OFS.ObjectManager._getOb raise AttributeError, id def listDAVObjects(self): """ Get all contents contributed by the current user. This is delegated to a script in order to help customisation. """ method = getattr(self, 'ContributionTool_getMyContentList', None) if method is not None: object_list = method() else: sm = getSecurityManager() user = sm.getUser() object_list = self.portal_catalog(portal_type=self.getPortalMyDocumentTypeList(), owner=str(user)) def wrapper(o_list): for o in o_list: o = o.getObject() reference = o.getReference() if reference: id = '%s-%s' % (reference, o.getUid()) else: id = '%s' % o.getUid() yield o.getObject().asContext(id=id) return wrapper(object_list) # Crawling methods security.declareProtected(Permissions.View, 'normaliseURL') def normaliseURL(self, url, base_url=None): """ Returns a normalised version of the url so that we do not download twice the same content. URL normalisation is an important part in crawlers. The current implementation is obviously simplistic. Refer to http://en.wikipedia.org/wiki/Web_crawler and study Harvestman for more ideas. """ url = self.simple_normaliser.sub('', url) url_split = url.split(':') url_protocol = url_split[0] if url_protocol in no_host_protocol_list: return url if base_url and len(url_split) == 1: # Make relative URL absolute url = '%s/%s' % (base_url, url) return url security.declareProtected(Permissions.View, 'encodeURL') def encodeURL(self, url): """ Returns the URL as an ID. ID should be chosen in such way that it is optimal with HBTreeFolder (ie. so that distribution of access time on a cluster is possible) NOTE: alternate approach is based on a url table and catalog lookup. It is faster ? Not sure. Since we must anyway insert objects in btrees and this is simimar in cost to accessing them. """ # Produce an MD5 from the URL hex_md5 = md5.md5(url).hexdigest() # Take the first part in the URL which is not empty # LOG("encodeURL", 0, url) url_segment = url.split(':')[1] url_segment_list = url_segment.split('/') url_domain = None for url_part in url_segment_list: if url_part: url_domain = url_part break # Return encoded url if url_domain: url_domain = urllib.quote(url_domain, safe='') url_domain = url_domain.replace('%', '') return "%s-%s" % (url_domain, hex_md5) return hex_md5 url = urllib.quote(url, safe='') url = url.replace('_', '__') url = url.replace('%', '_') return url security.declareProtected(Permissions.AddPortalContent, 'crawlContent') def crawlContent(self, content, container=None): """ Analyses content and download linked pages XXX: missing is the conversion of content local href to something valid. """ depth = content.getCrawlingDepth() if depth < 0: # Do nothing if crawling depth is reached # (this is not a duplicate code but a way to prevent # calling isIndexContent unnecessarily) return if not content.isIndexContent(): # Decrement depth only if it is a content document depth = depth - 1 if depth < 0: # Do nothing if crawling depth is reached return base_url = content.getContentBaseURL() url_list = map(lambda url: self.normaliseURL(url, base_url), set(content.getContentURLList())) for url in set(url_list): # LOG('trying to crawl', 0, url) # Some url protocols should not be crawled if url.split(':')[0] in no_crawl_protocol_list: continue if container is None: #if content.getParentValue() # in place of not ? container = content.getParentValue() # Calculate the id under which content will be stored id = self.encodeURL(url) # Try to access the document if it already exists document = container.get(id, None) if document is None: # XXX - This call is not working due to missing group_method_id # therefore, multiple call happen in parallel and eventually fail # (the same URL is created multiple times) # LOG('activate newContentFromURL', 0, url) self.activate(activity="SQLQueue").newContentFromURL(container_path=container.getRelativeUrl(), id=id, url=url, crawling_depth=depth) elif depth and document.getCrawlingDepth() < depth: # Update the crawling depth if necessary document._setCrawlingDepth(depth) document.activate().crawlContent() security.declareProtected(Permissions.AddPortalContent, 'updateContentFromURL') def updateContentFromURL(self, content, repeat=MAX_REPEAT, crawling_depth=0): """ Updates an existing content. """ # First, test if the document is updatable according to # its workflow states (if it has a workflow associated with) if content.isUpdatable(): # Step 0: update crawling_depth if required if crawling_depth > content.getCrawlingDepth(): content._setCrawlingDepth(crawling_depth) # Step 1: download new content try: url = content.asURL() data = urllib2.urlopen(url).read() file = cStringIO.StringIO() file.write(data) file.seek(0) except urllib2.HTTPError, error: if repeat == 0: # XXX - Call the extendBadURLList method,--NOT Implemented-- # IDEA : ajouter l'url en question dans une list "bad_url_list" puis lors du crawling au lieu que de boucler sur # la liste des url extraites de la page web on fait un test supplementaire qui verifie que l'url n'est pas # dans la liste bad_url_lis raise content.activate(at_date=DateTime() + 1).updateContentFromURL(repeat=repeat - 1) return except urllib2.URLError, error: if repeat == 0: # XXX - Call the extendBadURLList method,--NOT Implemented-- raise content.activate(at_date=DateTime() + 1).updateContentFromURL(repeat=repeat - 1) return # Step 2: compare and update if necessary (md5) # md5 stuff to compare contents new_content_md5 = md5.md5(data).hexdigest() content_md5 = content.getContentMd5() if content_md5 == new_content_md5: return content._edit(file=file)# Please make sure that if content is the same # we do not update it # This feature must be implemented by Base or File # not here (look at _edit in Base) # Step 3: convert to base format content.convertToBaseFormat() # Step 4: activate populate (unless interaction workflow does it) content.activate().populateContent() # Step 5: activate crawlContent depth = content.getCrawlingDepth() if depth > 0: content.activate().crawlContent() content.setContentMd5(new_content_md5) security.declareProtected(Permissions.AddPortalContent, 'newContentFromURL') def newContentFromURL(self, container_path=None, id=None, repeat=MAX_REPEAT, **kw): """ A wrapper method for newContent which provides extra safety in case or errors (ie. download, access, conflict, etc.). The method is able to handle a certain number of exceptions and can postpone itself through an activity based on the type of exception (ex. for a 404, postpone 1 day), using the at_date parameter and some standard values. NOTE: implementation needs to be done. """ # First of all, make sure do not try to create an existing document if container_path is not None and id is not None: container = self.restrictedTraverse(container_path) document = container.get(id, None) if document is not None: # Document aleardy exists: no need to keep on crawling return try: document = self.newContent(container_path=container_path, id=id, **kw) if document.isIndexContent() and document.getCrawlingDepth() >= 0: # If this is an index document, keep on crawling even if crawling_depth is 0 document.activate().crawlContent() elif document.getCrawlingDepth() > 0: # If this is an index document, stop crawling if crawling_depth is 0 document.activate().crawlContent() except urllib2.HTTPError, error: if repeat == 0: # here we must call the extendBadURLList method,--NOT Implemented-- # which had to add this url to bad URL list, so next time we avoid # crawling bad URL raise # Catch any HTTP error self.activate(at_date=DateTime() + 1).newContentFromURL( container_path=container_path, id=id, repeat=repeat - 1, **kw) except urllib2.URLError, error: if repeat == 0: # XXX - Call the extendBadURLList method, --NOT Implemented-- raise print error.reason #if getattr(error.reason,'args',None): #if error.reason.args[0] == socket.EAI_AGAIN: ## Temporary failure in name resolution - try again in 1 day self.activate(at_date=DateTime() + 1, activity="SQLQueue").newContentFromURL( container_path=container_path, id=id, repeat=repeat - 1, **kw) InitializeClass(ContributionTool)