Changing : %s"%fn+self.result + logging.debug("uploadatffinallythread changing:%s"%fn+self.result) founds[0].getObject().manage_addCDLIFileObject('',comment,SESSION['author'],file=os.path.join(SESSION['tmpdir'],fn),from_tmp=True) - if i==200: - i=0 - transaction.get().commit() - logging.info("changing: do commit") + if i%200==0: + transaction.get().commit() + logging.debug("uploadatffinallythread changing: do commit") transaction.get().commit() - logging.info("changing: last commit") + logging.debug("uploadatffinallythread changing: last commit") #now add the new files newPs=SESSION['newPs'] if len(newPs)>0: tmpDir=SESSION['tmpdir'] - logging.info("adding start") + logging.debug("uploadatffinallythread adding start") self.result="
Adding files
"+self.result #TODO: make this configurable, at the moment base folder for the files has to be cdli_main - ctx2.importFiles(comment=comment,author=str(username) ,folderName=tmpDir, files=newPs,ext=self) - logging.info("adding finished") - + logging.debug("uploadatffinallythread adding finished") #unlock locked files? if unlock: - logging.info("unlocking start") + logging.debug("uploadatffinallythread unlocking start") self.result="Unlock files
"+self.result unlockFns=[] for x in os.listdir(SESSION['tmpdir']): if not x in SESSION['errors']: unlockFns.append(x) - logging.info("unlocking have now what to unlock") + + logging.debug("unlocking have now what to unlock") for fn in unlockFns: #logging.info("will unlock: %s"%fn) @@ -317,18 +377,18 @@ class uploadATFfinallyThread(Thread): if len(founds)>0: #logging.info("unlock: %s"%founds[0].getObject().getId()) SESSION['author']=str(username) - founds[0].getObject().lockedBy="" - logging.info("unlocking done") + + logging.debug("uploadatffinallythread unlocking done") #if a basketname is given, add files to the basket if not (basketname ==''): - logging.info("add to basket %s"%basketname) + logging.debug("uploadatffinallythread add to basket %s"%basketname) self.result="Add to basket
"+self.result basketId=ctx2.basketContainer.getBasketIdfromName(basketname) if not basketId: # create new basket - logging.info("create basket %s"%basketname) + logging.debug("uploadatffinallythread create basket %s"%basketname) self.result="Create a new basket
"+self.result ob=ctx2.basketContainer.addBasket(basketname) basketId=ob.getId() @@ -336,11 +396,11 @@ class uploadATFfinallyThread(Thread): ids=os.listdir(SESSION['tmpdir']) basket.addObjects(ids,deleteOld=True,username=str(username)) + logging.debug("uploadatffinallythread uploadfinally done") + if RESPONSE is not None: RESPONSE.redirect(self.aq_parent.absolute_url()) - - logging.info("uploadfinally done") return True class tmpStore(SimpleItem): @@ -572,7 +632,7 @@ class CDLIBasketContainer(OrderedFolder) ret+=str(object[0].getData())+"\n" elif current=="yes": #search current object - logging.info("crrent: %s"%object[1].getId().split(".")[0]) + logging.debug("current: %s"%object[1].getId().split(".")[0]) founds=self.CDLICatalog.search({'title':object[1].getId().split(".")[0]}) if len(founds)>0: ret+=str(founds[0].getObject().getLastVersion().getData())+"\n" @@ -912,7 +972,7 @@ class CDLIBasket(Folder,CatalogAware): def searchInBasket(self,indexName,searchStr,regExp=False): """searchInBasket""" - lst=self.searchInLineIndexDocs(indexName,searchStr,uniq=True,regExp=regExp) + lst=self.searchInLineIndexDocs(indexName,searchStr,uniq=True,regExp=regExp) #TODO: fix this ret={} lv=self.getLastVersion() @@ -1450,10 +1510,14 @@ class CDLIFileObject(CatalogAware,extVer security=ClassSecurityInfo() - - security.declarePublic('makeThisVersionCurrent') - security.declareProtected('manage','index_html') + + security.declarePublic('view') + view = PageTemplateFile('zpt/viewCDLIFile.zpt', globals()) + + security.declarePublic('editATF') + editATF = PageTemplateFile('zpt/editATFFile.zpt', globals()) + def PrincipiaSearchSource(self): """Return cataloguable key for ourselves.""" return str(self) @@ -1463,32 +1527,26 @@ class CDLIFileObject(CatalogAware,extVer pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','makeThisVersionCurrent.zpt')).__of__(self) return pt() + + security.declarePublic('makeThisVersionCurrent') def makeThisVersionCurrent(self,comment,author,RESPONSE=None): """copy this version to current""" parent=self.aq_parent - - - newversion=parent.manage_addCDLIFileObject('',comment,author) - newversion.manage_upload(self.getData()) + parent.manage_addVersionedFileObject(id=None,vC=comment,author=author,file=self.getData(),RESPONSE=RESPONSE) + #newversion=parent.manage_addCDLIFileObject('',comment,author) + #newversion.manage_upload(self.getData()) - if RESPONSE is not None: - RESPONSE.redirect(self.aq_parent.absolute_url()+'/history') - + #if RESPONSE is not None: + # RESPONSE.redirect(self.aq_parent.absolute_url()+'/history') return True - security.declarePublic('view') - def getFormattedData(self): """fromat text""" data=self.getData() # return re.sub("\s\#lem"," #lem",data) #remove return vor #lem return re.sub("#lem"," #lem",data) #remove return vor #lem - def view(self): - """view file""" - pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','viewCDLIFile.zpt')).__of__(self) - return pt() security.declarePublic('getPNumber') def getPNumber(self): @@ -1517,13 +1575,14 @@ class CDLIFileObject(CatalogAware,extVer return txt.group(2) except: return "ERROR" + manage_addCDLIFileObjectForm=DTMLFile('dtml/fileAdd', globals(),Kind='CDLIFileObject',kind='CDLIFileObject', version='1') -def manage_addCDLIFileObject(self,id,vC='',author='', file='',title='',precondition='', content_type='', +def manage_addCDLIFileObject(self,id,vC='',author='', file='',title='',versionNumber=0, + precondition='', content_type='', from_tmp=False,REQUEST=None): """Add a new File object. - Creates a new File object 'id' with the contents of 'file'""" id=str(id) @@ -1536,11 +1595,8 @@ def manage_addCDLIFileObject(self,id,vC= self=self.this() # First, we create the file without data: - self._setObject(id, CDLIFileObject(id,title,'',content_type, precondition)) + self._setObject(id, CDLIFileObject(id,title,versionNumber=versionNumber,versionComment=vC,time=time.localtime(),author=author)) fob = self._getOb(id) - fob.versionComment=str(vC) - fob.time=time.localtime() - setattr(fob,'author',author) # Now we "upload" the data. By doing this in two steps, we # can use a database trick to make the upload more efficient. @@ -1548,14 +1604,17 @@ def manage_addCDLIFileObject(self,id,vC= if file and not from_tmp: fob.manage_upload(file) elif file and from_tmp: - fob.manage_upload_from_tmp(file) + fob.manage_file_upload(file) # manage_upload_from_tmp doesn't exist in ExtFile2 + # fob.manage_upload_from_tmp(file) # manage_upload_from_tmp doesn't exist in ExtFile2 if content_type: fob.content_type=content_type - logging.debug("reindex1: %s"%repr(self)) + #logging.debug("manage_add: lastversion=%s"%self.getData()) + logging.debug("reindex1: %s in %s"%(repr(self),repr(self.default_catalog))) self.reindex_object() + #logging.debug("manage_add: fob_data=%s"%fob.getData()) logging.debug("reindex2: %s in %s"%(repr(fob), repr(fob.default_catalog))) - fob.reindex_object() + fob.index_object() if REQUEST is not None: REQUEST['RESPONSE'].redirect(self.absolute_url()+'/manage_main') @@ -1566,20 +1625,26 @@ class CDLIFile(extVersionedFile,CatalogA security=ClassSecurityInfo() meta_type="CDLI file" + content_meta_type = ["CDLI File Object"] + default_catalog='CDLICatalog' + security.declareProtected('manage','index_html') - #security.declarePublic('history') + def getLastVersionData(self): """get last version data""" - return self.getLastVersion().getData() + return self.getData() def getLastVersionFormattedData(self): """get last version data""" - return self.getLastVersion().getFormattedData() + return self.getContentObject().getFormattedData() + + def getTextId(self): + """returns P-number of text""" + # assuming that its the beginning of the title + return self.title[:7] #security.declarePublic('history') - - def history(self): """history""" @@ -1624,6 +1689,12 @@ class CDLIFile(extVersionedFile,CatalogA #return [x.getObject() for x in context.CDLIBasketCatalog.search({'getFileNamesInLastVersion':self.getId()})] + def _newContentObject(self, id, title='', versionNumber=0, versionComment=None, time=None, author=None): + """factory for content objects. to be overridden in derived classes.""" + logging.debug("_newContentObject(CDLI)") + return CDLIFileObject(id,title,versionNumber=versionNumber,versionComment=versionComment,time=time,author=author) + + def addCDLIFileObjectForm(self): """add a new version""" @@ -1649,58 +1720,27 @@ class CDLIFile(extVersionedFile,CatalogA except: pass - - if changeName=="yes": - filename=file.filename - self.title=filename[max(filename.rfind('/'), - filename.rfind('\\'), - filename.rfind(':'), - )+1:] - + ob = self.addContentObject(id, vC, author, file, title, changeName=changeName, newName=newName, from_tmp=from_tmp, + precondition=precondition, content_type=content_type) - if not newName=='': - self.title=newName[0:] - - - - - positionVersionNum=getattr(self,'positionVersionNum','front') - - if positionVersionNum=='front': - id="V%i"%self.getVersion()+"_"+self.title - else: - tmp=os.path.splitext(self.title) - if len(tmp)>1: - id=tmp[0]+"_V%i"%self.getVersion()+tmp[1] - else: - id=tmp[0]+"_V%i"%self.getVersion() - - - manage_addCDLIFileObject(self,id,vC,author,file,id,precondition, content_type,from_tmp=from_tmp) - #objs=self.ZopeFind(self,obj_ids=[id])[0][1].setVersionNumber(int(self.getVersion())) - objs=getattr(self,id).setVersionNumber(int(self.getVersion())) try: - #FIXME: wozu ist das gut? - self.REQUEST.SESSION['objID_parent']=self.getId() + #FIXME: wozu ist das gut? + self.REQUEST.SESSION['objID_parent']=self.getId() except: - pass + pass if RESPONSE: - - obj=self.ZopeFind(self,obj_ids=[id])[0][1] - if obj.getSize()==0: - self.REQUEST.SESSION['objID']=obj.getId() + if ob.getSize()==0: + self.REQUEST.SESSION['objID']=ob.getId() pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','errorUploadFile')).__of__(self) return pt() - else: if come_from and (come_from!=""): - RESPONSE.redirect(come_from+"?change="+self.getId()) + RESPONSE.redirect(come_from+"?change="+self.getId()) else: RESPONSE.redirect(self.REQUEST['URL2']+'?uploaded=%s'%self.title) - else: - return self.ZopeFind(self,obj_ids=[id])[0][1] + return ob def manage_addCDLIFileForm(self): @@ -1714,8 +1754,6 @@ def manage_addCDLIFile(self,id,title,loc tryToggle=True tryCount=0 - - self._setObject(id,newObj) getattr(self,id).reindex_object() @@ -1723,6 +1761,7 @@ def manage_addCDLIFile(self,id,title,loc if RESPONSE is not None: RESPONSE.redirect('manage_main') + def checkUTF8(data): """check utf 8""" try: @@ -1753,6 +1792,7 @@ def splitatf(fh,dir=None,ext=None): nf=None i=0 + #ROC: why split \n first and then \r??? if (type(fh) is StringType) or (type(fh) is UnicodeType): iter=fh.split("\n") else: @@ -1806,34 +1846,37 @@ class CDLIFileFolder(extVersionedFileFol security=ClassSecurityInfo() meta_type="CDLI Folder" - filesMetaType=['CDLI file'] - folderMetaType=['CDLI Folder'] - default_catalog='CDLICatalog' - defaultFileCatalog=default_catalog #wenn dieses definiert ist, wird beim hinzufgen einer neuen version eines files dieser catalog neuiniziert + file_meta_type=['CDLI file'] + folder_meta_type=['CDLI Folder'] + + file_catalog='CDLICatalog' + #downloadCounter=0 # counts how many download for all files currently run, be mehr als 5 wird verweigert. tmpStore2={} + + def _newVersionedFile(self, id, title='', lockedBy=None, author=None): + """factory for versioned files. to be overridden in derived classes.""" + logging.debug("_newVersionedFile(CDLI)") + return CDLIFile(id, title, lockedBy=lockedBy, author=author) + def setTemp(self,name,value): """set tmp""" setattr(self,name,value) + deleteFileForm = PageTemplateFile("zpt/doDeleteFile", globals()) - def delete(self,ids): - """delete this file, i.e. move into a trash folder""" - - found=self.ZopeFind(self,obj_ids=['.trash']) - - if len(found)<1: - manage_addCDLIFileFolder(self, '.trash',title="Trash") - trash=self._getOb('.trash') - else: - trash=found[0][1] - + def delete(self,ids,REQUEST=None): + """delete these files""" if type(ids) is not ListType: ids=[ids] - cut=self.manage_cutObjects(ids) - trash.manage_pasteObjects(cut) + + self.manage_delObjects(ids) + if REQUEST is not None: + return self.index_html() + + def getVersionNumbersFromIds(self,ids): """get the numbers of the current versions of documents described by their ids""" @@ -1843,38 +1886,39 @@ class CDLIFileFolder(extVersionedFileFol founds=self.CDLICatalog.search({'title':searchStr}) for found in founds: - lastVersion=found.getObject().getLastVersion() + lastVersion=found.getObject().getContentObject() ret.append((found.getId,lastVersion)) return ret def getFile(self,fn): """get the content of the file fn""" - founds=self.CDLICatalog.search({'title':fn}) - if not founds: - return "" + logging.debug("getFile: %s"%repr(fn)) + if not self.hasObject(fn): + # search deeper + founds=getattr(self, self.file_catalog).search({'textid':fn}) + if founds: + obj=founds[0].getObject().getContentObject() + else: + return "" else: - obj=founds[0].getObject().getLastVersion() + obj = self[fn].getContentObject() - return obj.getData()[0:] + return obj.getData()[0:] + def checkCatalog(self,fn): """check if fn is in the catalog""" #TODO add checkCatalog - def findObjectsFromListWithVersion(self,list,author=None): """find objects from a list with versions @param list: list of tuples (cdliFile,version) """ - - - #self.REQUEST.SESSION['fileIds']=list#store fieldIds in session for further usage #self.REQUEST.SESSION['searchList']=self.REQUEST.SESSION['fileIds'] - pt=getattr(self,'filelistVersioned.html') return pt(search=list,author=author) @@ -1955,7 +1999,7 @@ class CDLIFileFolder(extVersionedFileFol def sortF(x,y): return cmp(x[0],y[0]) - catalog=getattr(self,self.default_catalog) + catalog=getattr(self,self.file_catalog) #tf,tfilename=mkstemp() if not hasattr(self.temp_folder,'downloadCounter'): self.temp_folder.downloadCounter=0 @@ -1982,7 +2026,7 @@ class CDLIFileFolder(extVersionedFileFol #os.write(tf,obj.getLastVersion().data) if RESPONSE: - RESPONSE.write(obj.getLastVersion().getData()[0:]) + RESPONSE.write(obj.getData()[0:]) RESPONSE.write("\n") self.temp_folder.downloadCounter-=1 self._p_changed=1 @@ -2002,7 +2046,7 @@ class CDLIFileFolder(extVersionedFileFol def hasParent(self): """returns true falls subfolder""" - if self.aq_parent.meta_type in self.folderMetaType: + if self.aq_parent.meta_type in self.folder_meta_type: return True else: return False @@ -2010,11 +2054,11 @@ class CDLIFileFolder(extVersionedFileFol def getFolders(self): """get all subfolders""" ret=[] - folders=self.ZopeFind(self,obj_metatypes=self.folderMetaType) + folders=self.ZopeFind(self,obj_metatypes=self.folder_meta_type) for folder in folders: ret.append((folder[1], - len(self.ZopeFind(folder[1],obj_metatypes=self.folderMetaType)), - len(self.ZopeFind(folder[1],obj_metatypes=self.filesMetaType)) + len(self.ZopeFind(folder[1],obj_metatypes=self.folder_meta_type)), + len(self.ZopeFind(folder[1],obj_metatypes=self.file_meta_type)) )) return ret @@ -2066,243 +2110,192 @@ class CDLIRoot(Folder): """main folder for cdli""" meta_type="CDLIRoot" - downloadCounterBaskets=0# counts the current basket downloads if counter > 10 no downloads are possible + downloadCounterBaskets=0 # counts the current basket downloads if counter > 10 no downloads are possible + + file_catalog = 'CDLICatalog' + + # word splitter for search + splitter = {'words':cdliSplitter.wordSplitter(), + 'graphemes':cdliSplitter.graphemeSplitter()} + def deleteFiles(self,ids): - """delete files (resp. move into .trash folder)""" - # find or generete trash folder - - found=self.ZopeFind(self,obj_ids=['.trash']) - - if len(found)<1: - manage_addCDLIFileFolder(self, '.trash',title="Trash") - trash=self._getOb('.trash') - else: - logging.info(found) - trash=found[0][1] - - + """delete files""" for id in ids: founds=self.CDLICatalog.search({'title':id.split(".")[0]}) if founds: - logging.info(founds) + logging.debug("deleting %s"%founds) folder=founds[0].getObject().aq_parent #get the parent folder of the object - logging.info(folder) - cut=folder.manage_cutObjects([founds[0].getId]) #cut it out - trash.manage_pasteObjects(cut) #paste it in the trash + logging.debug("deleting from %s"%folder) + cut=folder.delete([founds[0].getId]) #cut it out - def findWordRegExp(self,indexName,searchTerm): - """find all words in index which match regexp in SearchTerm - @param indexName: name of the index to be searched in - @param searchTerm: word to be searched""" - - ret=[] - for x in self.lineIndexes[indexName].iterkeys(): - if re.match(searchTerm,x): - ret.append(x) - return ret - - def searchRegExpInLineIndexDocs(self,indexName,searchTerm): - """search in inLineIndex with regexp - @param indexName: name of the index to be searched in - @param searchTerm: term to be searched - """ - if not searchTerm: - return [] - ret=[] - words=self.findWordRegExp(indexName,searchTerm) # suche nach allen Treffern - logging.info("wd:%s"%words) - for word in words: - - ret+=self.searchInLineIndexDocs(indexName,word) - - - x= unique(ret) - logging.info("words_done") - return x - - def showInLineIndex(self): - """get the index for debug purposes""" - print "show" - for key in self.lineIndexes.keys(): - logging.info("index:%s"%key) - for x in self.lineIndexes[key].iterkeys(): - logging.info("word:%s"%repr(x)) - #for y in self.lineIndex[x].iterkeys(): - # print "doc",repr(y),repr(self.lineIndex[x][y]) - - return self.lineIndexes - - def searchInLineIndexDocs(self,indexName,word,uniq=True,regExp=False): - """search occurences in an index - @param indexName: name of the index to be searched in - @param word: word to be searched - @param unique: (optional) unify the list of results - @param regExp: (optional) use regular expressions - """ - if regExp: - return self.searchRegExpInLineIndexDocs(indexName,word) - - try: - - lst=list(self.lineIndexes[indexName].get(word).keys()) - except: - logging.error("error: searchInLineIndexDocs (%s %s)"%(sys.exc_info()[0:2])) - lst=[] - if uniq: - return unique(lst) - else: - return lst - - def getLinesFromIndex(self,indexName,word,doc,regExp=False): - """return all lines from a document where word is found - @param indexName: Name of the index - @param word: word to be searched - @param doc: name of the document (usuallay the p-number) - @param regExp: (optional) use regExp - """ - - if not regExp: - return self.lineIndexes[indexName].get(word)[doc] - else: # wenn regexp, suche welches word - for w in self.findWordRegExp(indexName,word): - if self.lineIndexes[indexName].get(w): # ein word in im dex gefunden - try: - dc=self.lineIndex[indexName].get(word)[doc] - return dc # und ein document dann gib es zurueck - except: - pass #andernfalls weiter - - def cleanInLineIndex(self,indexName): - """empty an InlineIndex - @param indexName: name of the index - """ - for x in list(self.lineIndexes[indexName].keys()): - del(self.lineIndexes[indexName][x]) - print [x for x in self.lineIndexes[indexName].keys()] - - return "ok" - - def storeInLineIndex(self,indexName,key,value): - """store in index, key is normally a word or grapheme - and value is a tuple (documentname, line) where the word can be found - @param indexName: name of the index - @param key: key in index - @param value: value in index, value is a tuple (document name, line) - """ - logging.error("indexing: %s %s"%(indexName,key)) - if (not hasattr(self,'lineIndexes')): - - self.lineIndexes={} - - if self.lineIndexes.get(indexName,None) is None: - #index exisitiert noch nicht dann anlegen - - self.lineIndexes[indexName]=OOBTree() - lis=self.lineIndexes - li=lis[indexName] - - if li.has_key(key): - -# if li[key].has_key(value[0]) and (not (value[1] in li[key][value[0]])): - if li[key].has_key(value[0]): - tmp=li[key][value[0]] - tmp.append(value[1]) # add it if now in the array - li[key][value[0]]=tmp[0:] - else: - li[key][value[0]]=[value[1]] # new array for lines - - else: - - li[key]=OOBTree()# new btree for lines - li[key][value[0]]=[value[1]] - - - self.lineIndexes=lis - - transaction.get().commit() + def searchText(self, query, index='graphemes'): + """searches query in the fulltext index and returns a list of file ids/P-numbers""" + # see also: http://www.plope.com/Books/2_7Edition/SearchingZCatalog.stx#2-13 + logging.debug("searchtext for '%s' in index %s"%(query,index)) + #import Products.ZCTextIndex.QueryParser + #qp = QueryParser.QueryParser() + #logging.debug() + idxQuery = {index:{'query':query}} + idx = getattr(self, self.file_catalog) + # do search + resultset = idx.search(query_request=idxQuery,sort_index='textid') + # put only the P-Number in the result + results = [res.getId[:7] for res in resultset] + logging.debug("searchtext: found %d texts"%len(results)) + return results + + + def getFile(self, pnum): + """get the translit file with the given pnum""" + f = getattr(self, self.file_catalog).search({'textid':pnum}) + if not f: + return "" + return f[0].getObject().getData() + + def showFile(self,fileId,wholePage=False): """show a file @param fileId: P-Number of the document to be displayed """ - f=self.CDLICatalog({'title':fileId}) + f=getattr(self, self.file_catalog).search({'textid':fileId}) if not f: return "" if wholePage: - logging.info("whole") - return f[0].getObject().getLastVersion().view() + logging.debug("show whole page") + return f[0].getObject().getContentObject().view() else: return f[0].getObject().getLastVersionFormattedData() - def showWordInFile(self,fileId,word,lineList=None,regExp=True,indexName=""): - """get lines with word fromFileId""" + def showWordInFile(self,fileId,word,indexName='graphemes',regExp=False,): + """get lines with word from FileId""" + logging.debug("showwordinfile word='%s' index=%s file=%s"%(word,indexName,fileId)) - file=self.showFile(fileId) - logging.info("regEXP %s"%regExp) + file = formatAtfFullLineNum(self.getFile(fileId)) ret=[] - if regExp: # wenn regexp dann generiere alle worte aus der list die der regexp entsprechen - wordlist=self.findWordRegExp(indexName,word) - else: - wordlist=[word] - for line in file.split("\n"): - found=False + # add whitespace before and whitespace and line-end to splitter bounds expressions + bounds = self.splitter[indexName].bounds + splitexp = "(%s|\s)(%%s)(%s|\s|\Z)"%(bounds,bounds) + # clean word expression + # TODO: this should use QueryParser itself + # take out double quotes + word = word.replace('"','') + # take out ignorable signs + ignorable = self.splitter[indexName].ignorex + word = ignorable.sub('', word) + # compile into regexp objects and escape parens + wordlist = [re.compile(splitexp%re.escape(w)) for w in word.split(' ')] + + for line in file.splitlines(): for word in wordlist: - try: # just a hack because of possible unicode errors in line - if line.find(word)>-1: - if lineList: #liste of moeglichen Zeilennummern - num=line.split(".")[0] #Zeilenummer ist alles vor dem . in der Zeile - - if num in lineList: - - ret.append(line) - else: # nimm alles ohne line check - ret.append(line) - - break; - except: - pass + #logging.debug("showwordinfile: searching for %s in %s"%(word.pattern,ignoreable.sub('',line))) + if word.search(ignorable.sub('',line)): + line = formatAtfLineHtml(line) + ret.append(line) + break + return ret - def tagWordInFile(self,fileId,word,lineList=None,regExp=True,indexName=""): - """get lines with word fromFileId""" + + def showWordInFiles(self,fileIds,word,indexName='graphemes',regExp=False): + """ + get lines with word from all ids in list FileIds. + returns dict with id:lines pairs. + """ + logging.debug("showwordinfiles word='%s' index=%s file=%s"%(word,indexName,fileIds)) - file=self.showFile(fileId) - tagStr=u'%s' + return dict([(id,self.showWordInFile(id, word, indexName, regExp)) for id in fileIds]) + + + def tagWordInFile(self,fileId,word,indexName='graphemes',regExp=False): + """get text with word highlighted from FileId""" + logging.debug("tagwordinfile word='%s' index=%s file=%s"%(word,indexName,fileId)) + + file=self.getFile(fileId) + tagStart=u'' + tagEnd=u'' + tagStr=tagStart + u'%%s' + tagEnd ret=[] - if regExp: # wenn regexp dann generiere alle worte aus der list die der regexp entsprechen - wordlist=self.findWordRegExp(indexName,word) - else: - wordlist=[word] + # add whitespace to splitter bounds expressions and compile into regexp object + bounds = self.splitter[indexName].bounds + wordsplit = re.compile("(%s|\s)"%bounds) + # clean word expression + # TODO: this should use QueryParser itself + word = word.replace('"','') # take out double quotes + # take out ignoreable signs + ignorable = self.splitter[indexName].ignorex + word = ignorable.sub('', word) + # split search terms by blanks + words = word.split(' ') + # split search terms again (for grapheme search with words) + splitwords = dict(((w,self.splitter[indexName].process([w])) for w in words)) - for line in file.split("\n"): + for line in file.splitlines(): line = unicodify(line) - found=False - for word in wordlist: - if line.find(word)>-1: #word ist gefunden dann makiere und breche die Schleife ab - if lineList: #liste of moeglichen Zeilennummern - num=line.split(".")[0] #Zeilenummer ist alles vor dem . in der Zeile - - if num in lineList: - - ret.append(line.replace(word,tagStr%word)) - - else: # nimm alles ohne line check - ret.append(line.replace(word,tagStr%word)) - found=True - break - if not found: #word wurde nicht gefunden keine makierung - ret.append(line) + # ignore lemma and other lines + if line.lstrip().startswith('#lem:'): + continue + # ignore p-num line + if line.startswith('&P'): + continue + # ignore version lines + if line.startswith('#version'): + continue + # ignore atf type lines + if line.startswith('#atf:'): + continue + + # first scan + hitwords = [] + for w in words: + if ignorable.sub('',line).find(w) > -1: + # word is in line + # append split word for grapheme search with words + hitwords.extend(splitwords[w]) + #hitwords.extend(wordsplit.split(w)) + + # examine hits closer + if hitwords: + # split line into words + parts = wordsplit.split(line) + line = "" + for p in parts: + #logging.debug("tagwordinfile: searching for %s in %s"%(p,hitwords)) + # reassemble line + if ignorable.sub('', p) in hitwords: + #logging.debug("tagwordinfile: found %s in %s"%(p,hitwords)) + # this part was found + line += tagStart + formatAtfHtml(p) + tagEnd + else: + line += formatAtfHtml(p) + + else: + # no hits + line = formatAtfHtml(line) + + ret.append(line) return u'adding: %s
"%f+ext.result + if not obj: manage_addCDLIFileFolder(root,folder,folder) fobj=getattr(root,folder) #transaction.get().commit() + else: fobj=obj[0][1] obj2=fobj.ZopeFind(fobj,obj_ids=[f2]) + logging.debug("importFiles: fobj=%s obj2=%s"%(fobj,obj2)) if not obj2: manage_addCDLIFileFolder(fobj,f2,f2) @@ -2587,22 +2584,15 @@ class CDLIRoot(Folder): file2=os.path.join(folderName,f) id=f - manage_addCDLIFile(fobj2,f,'','') - id=f - ob=fobj2._getOb(f) - ob.title=id - - manage_addCDLIFileObject(ob,id,comment,author,file2,content_type='',from_tmp=True) - self.CDLICatalog.catalog_object(ob) - #self.CDLICatalog.manage_catalogFoundItems(obj_ids=[id],search_sub=1) - #self.CDLICatalog.manage_catalogObject(self.REQUEST, self.REQUEST.RESPONSE, 'CDLICatalog', urlparse.urlparse(ob.absolute_url())[1]) + logging.debug("importFiles: addCDLIFile fobj2=%s, f=%s file2=%s"%(fobj2,repr(f),repr(file2))) + fobj2.addFile(vC='',file=file(file2),author=author,newName=f) count+=1 - if count > 1000: - print "committing" + if count%100==0: + logging.debug("importfiles: committing") transaction.get().commit() - count=0 - transaction.get().commit() + + transaction.get().commit() return "ok"