--- cdli/cdli_files.py 2007/03/19 18:34:01 1.67 +++ cdli/cdli_files.py 2008/06/26 18:08:02 1.82 @@ -26,7 +26,124 @@ import logging import transaction import copy import codecs +import sys +import cdliSplitter + + +def unicodify(s): + """decode str (utf-8 or latin-1 representation) into unicode object""" + if not s: + return u"" + if isinstance(s, str): + try: + return s.decode('utf-8') + except: + return s.decode('latin-1') + else: + return s + +def utf8ify(s): + """encode unicode object or string into byte string in utf-8 representation. + assumes string objects to be utf-8""" + if not s: + return "" + if isinstance(s, str): + return s + else: + return s.encode('utf-8') + +def formatAtfHtml(l): + """escape special ATF characters for HTML""" + if not l: + return "" + + # replace & + l = l.replace('&','&') + # replace angular brackets + l = l.replace('<','<') + l = l.replace('>','>') + return l + +def formatAtfLineHtml(l, nolemma=True): + """format ATF line for HTML""" + if not l: + return "" + + if nolemma: + # ignore lemma lines + if l.lstrip().startswith('#lem:'): + return "" + + return formatAtfHtml(l) + + + +def formatAtfFullLineNum(txt, nolemma=True): + """format full line numbers in ATF text""" + # surface codes + surfaces = {'@obverse':'obv', + '@reverse':'rev', + '@surface':'surface', + '@edge':'edge', + '@left':'left', + '@right':'right', + '@top':'top', + '@bottom':'bottom', + '@face':'face', + '@seal':'seal'} + + if not txt: + return "" + + ret = [] + surf = "" + col = "" + for line in txt.splitlines(): + line = unicodify(line) + if line and line[0] == '@': + # surface or column + words = line.split(' ') + if words[0] in surfaces: + surf = line.replace(words[0],surfaces[words[0]]).strip() + + elif words[0] == '@column': + col = ' '.join(words[1:]) + + elif line and line[0] in '123456789': + # ordinary line -> add line number + line = "%s:%s:%s"%(surf,col,line) + + ret.append(line) + + return '\n'.join(ret) + + +def generateXMLReturn(hash): + """erzeugt das xml file als returnwert fuer uploadATFRPC""" + + ret="" + + ret+="" + for error in hash['errors']: + ret+="""%s"""%error + + ret+="" + + ret+="" + for changed in hash['changed']: + ret+="""%s"""%changed + ret+="" + + ret+="" + for new in hash['newPs']: + ret+=""""""%new + ret+="" + + ret+="" + return ret + + def unique(s): """Return a list of the elements in s, but without duplicates. @@ -99,7 +216,7 @@ def unique(s): class BasketContent(SimpleItem): """classe fuer den Inhalt eines Baskets""" - + def __init__(self,content=[]): """content""" self.contentList=content[0:] @@ -196,6 +313,7 @@ class uploadATFfinallyThread(Thread): self.result+="

Start processing

" #shall I only upload the changed files? + logging.debug("uploadATFfinally procedure: %s"%procedure) if procedure=="uploadchanged": changed=[x[0] for x in SESSION.get('changed',[])] uploadFns=changed+SESSION.get('newPs',[]) @@ -209,48 +327,48 @@ class uploadATFfinallyThread(Thread): #or maybe nothing elif procedure=="noupload": - return True + return True else: uploadFns=[] #do first the changed files i=0 for fn in uploadFns: + logging.debug("uploadATFfinally uploadFn=%s"%fn) i+=1 founds=ctx2.CDLICatalog.search({'title':fn}) if len(founds)>0: SESSION['author']=str(username) self.result="

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) @@ -259,18 +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() @@ -278,11 +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): @@ -341,6 +459,7 @@ class uploadATFThread(Thread): ctx = self.getContext(app,serverport=self.serverport) logging.info("run intern") try: + logging.info("created: %s"%idTmp) ctx.temp_folder._setObject(idTmp,tmpStore(idTmp)) except: logging.error("thread upload: %s %s"%sys.exc_info()[0:2]) @@ -355,6 +474,7 @@ class uploadATFThread(Thread): conn.close() + return getattr(ctx.temp_folder,idTmp) def getResult(self): """method for accessing result""" @@ -425,11 +545,13 @@ class uploadATFThread(Thread): #if not than add filename to the list of newfiles data=file(os.path.join(dir,fn)).read() - #status,msg=checkFile(fn,data,dir) - status=True - msg="" + status,msg=checkFile(fn,data,dir) + #status=True + + if not status: # error errors.append((fn,msg)) + else: if len(founds)==0: newPs.append(fn) @@ -453,7 +575,7 @@ class uploadATFThread(Thread): stObj.returnValue={} stObj.returnValue['errors']=errors - + stObj.returnValue['newPs']=newPs stObj.returnValue['tmpdir']=dir stObj.returnValue['basketLen']=basketLen @@ -477,6 +599,46 @@ class CDLIBasketContainer(OrderedFolder) security=ClassSecurityInfo() meta_type="CDLIBasketContainer" + def getPNumbersOfBasket(self,basketName): + """get all pnumbers of a basket as a list, returns an empty list if basket not found + @param basketName: name of the basket + """ + ret=[] + basketId=self.getBasketIdfromName(basketName) + if not basketId: + return [] + + ob=getattr(self,basketId).getContent() + + ret=[x[0].split(".")[0] for x in ob] + + return ret + + security.declareProtected('manage','getBasketAsOneFile') + def getBasketAsOneFile(self,basketName,current="no"): + """returns all files of the basket combined in one file + @param basketName: Name of the basket + @param current: (optional) if current is set to "yes" then the most current version of + all files are downloaded and not the versions of the files as stored in the basket + """ + ret="" + basketId=self.getBasketIdfromName(basketName) + if not basketId: + return "" + + ob=getattr(self,basketId).getLastVersion() + for object in ob.getContent(): + if current=="no": #version as they are in the basket + ret+=str(object[0].getData())+"\n" + elif current=="yes": + #search current object + 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" + return ret + + security.declareProtected('manage','upDateBaskets') def upDateBaskets(self): """update content in to objects""" @@ -504,12 +666,19 @@ class CDLIBasketContainer(OrderedFolder) trash.manage_pasteObjects(cut) security.declareProtected('manage','manageBaskets') - def manageBaskets(self,ids,submit,REQUEST=None,RESPONSE=None): + def manageBaskets(self,submit,ids=None,basket1="",basket2="",joinBasket="",subtractBasket="",REQUEST=None,RESPONSE=None): """manage baskets, delete or copy""" if submit=="delete": self.deleteBaskets(ids) - + elif submit=="join": + flag,msg=self.joinBasket(joinBasket, ids) + logging.info("joining %s %s"%(flag,msg)) + + elif submit=="subtract": + logging.info("BBBb %s %s"%(basket1,basket2)) + flag,msg=self.subtractBasket(subtractBasket, basket1,basket2) + logging.info("subtract %s %s"%(flag,msg)) if RESPONSE: RESPONSE.redirect(self.absolute_url()) @@ -540,7 +709,7 @@ class CDLIBasketContainer(OrderedFolder) return pt(basketId=basketId,basketName=basketName) - security.declareProtected('View','index_html') + security.declareProtected('manage','index_html') def index_html(self): """stanadard ansicht""" @@ -632,9 +801,83 @@ class CDLIBasketContainer(OrderedFolder) baskets.sort(sortComment) return baskets + + + def subtractBasket(self,newBasket,basket1,basket2): + """subtract basket2 from basket1 + (i.e. newbasket will contain alle elements of basket1 which are not in basket2), + if basket2 contains files which are not in basket1, then theses files fill be ignored + + @param newbasket: name of the new basket + @param basket1: basket where basket2 will be subtracted from + @param basket2: see above + + """ + logging.info("CCCCC %s %s"%(basket1,basket2)) + + try: + newB=self.addBasket(newBasket) + except: + return False, "cannot create the new basket" + + - - + + + bas2= getattr(self,basket2) + bas2content=bas2.getContent() + bas2ids=[x[0] for x in bas2content] + + + + bas1= getattr(self,basket1) + bas1content=bas1.getContent() + + + newBasketContent={} + + for id,version in bas1content: + if not (id in bas2ids): + newBasketContent[id]=version + + username=self.getActualUserName() + + logging.info("sbc %s"%newBasketContent) + newB.addObjectsWithVersion(newBasketContent,username=username,catalog=self.CDLICatalog) + + return True, "" + + + def joinBasket(self,newBasket,oldBaskets): + """join two baskets + @param newbasket: name of the new basket + @param oldbaskets: list of baskets to be joined + """ + try: + newB=self.addBasket(newBasket) + except: + return False, "cannot create the new basket" + + newBasketContent={} + for ob in oldBaskets: + x= getattr(self,ob,None) + if x is None: + return False, "cannot find basket: %s"%ob + + ids=x.getContent() # hole den Inhalt + + for id,version in ids: + if newBasketContent.has_key(id): # p number gibt's schon + newBasketContent[id]=max(newBasketContent[id],version) # speichere die groessere Versionsnumber + else: + newBasketContent[id]=version + username=self.getActualUserName() + + logging.info("nbc %s"%newBasketContent) + newB.addObjectsWithVersion(newBasketContent,username=username,catalog=self.CDLICatalog) + + return True, "" + def getNewId(self): """createIds""" last=getattr(self,'last',0) @@ -670,13 +913,13 @@ class CDLIBasketContainer(OrderedFolder) """get name of the actualuser""" return str(self.REQUEST['AUTHENTICATED_USER']) - + security.declareProtected('manage','addBasket') def addBasket(self,newBasketName): """add a new basket""" ob=manage_addCDLIBasket(self,newBasketName) return ob - + def storeInBasket(self,submit,ids=None,newBasketName=None,fromFileList=None,RESPONSE=None,REQUEST=None): """store it""" if not ids: @@ -726,7 +969,43 @@ class CDLIBasket(Folder,CatalogAware): meta_type="CDLIBasket" default_catalog="CDLIBasketCatalog" + def searchInBasket(self,indexName,searchStr,regExp=False): + """searchInBasket""" + + lst=self.searchInLineIndexDocs(indexName,searchStr,uniq=True,regExp=regExp) #TODO: fix this + ret={} + + lv=self.getLastVersion() + + for obj in lv.content.getContent(): + id=obj[1].getId().split(".")[0] + if id in lst: + + ret[id]=self.showWordInFile(id,searchStr,lineList=self.getLinesFromIndex(indexName,searchStr,id,regExp=regExp),regExp=regExp,indexName=indexName) + + + pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','searchResultsInBasket')).__of__(self) + return pt(result=ret,indexName=indexName,regExp=regExp,word=searchStr) + + + + + def searchInBasket_v1(self,searchStr): + """search occurences of searchStr in files im basket""" + ret=[] + lv=self.getLastVersion() + logging.info("searching") + for obj in lv.content.getContent(): + txt=obj[0].getData() + for x in txt.split("\n"): + logging.info("search %s"%x) + if re.match(searchStr,x): + ret.append(x) + + return "\n".join(ret) + + def getFile(self,obj): return obj[1] @@ -778,7 +1057,11 @@ class CDLIBasket(Folder,CatalogAware): self.shortDescription=shortDescription self.comment=comment - + def getActualUserName(self): + """get name of the actualuser""" + + return str(self.REQUEST['AUTHENTICATED_USER']) + def getLastVersion(self): """hole letzte version""" @@ -842,6 +1125,56 @@ class CDLIBasket(Folder,CatalogAware): return obj + def addObjectsWithVersion(self,ids,deleteOld=None,username=None,catalog=None): + """generate a new version of the basket with objects added, + hier wird jedoch nicht die letzte Version jedes Files hinzugefuegt, s + ondern ids is ein Tupel mit der Id (d.h. der p-number) und der Versionsnummer. + """ + logging.info("add to basket (%s)"%(self.getId())) + lastVersion=self.getLastVersion() + + if not catalog: + catalog=self.CDLICatalog + + if lastVersion is None: + oldContent=[] + else: + oldContent=lastVersion.content.getContent() + + if deleteOld: + oldContent=[] + + newContent=[] + added=0 + + for id,version in ids.iteritems(): + logging.info("adding %s %s"%(id,version)) + id=id.split(".")[0] # title nur die pnumber ohne atf + + try: + founds=catalog.search({'title':id}) + except: + founds=[] + logging.info(" found %s "%(founds)) + for found in founds: + if found.getObject() not in oldContent: + + #TODO: was passiert wenn, man eine Object dazufŸgt, das schon da ist aber eine neuere version + newContent.append((found.getObject().getVersions()[version-1][1],found.getObject())) + added+=1 + + content=oldContent+newContent + if not username: + logging.error("XXXXXXXXXXX %s"%repr(self)) + user=self.getActualUserName() + else: + user = username + + ob=manage_addCDLIBasketVersion(self,user,comment="",basketContent=content) + logging.info("add to basket (%s) done"%(self.getId())) + return added + + def addObjects(self,ids,deleteOld=None,username=None): """generate a new version of the basket with objects added""" logging.info("add to basket (%s)"%(self.getId())) @@ -881,7 +1214,17 @@ class CDLIBasket(Folder,CatalogAware): - + def getContent(self): + """print content""" + ret=[] + + lv=self.getLastVersion() + for obj in lv.content.getContent(): + logging.info("XXXXXXXXXX %s"%repr(obj)) + ret.append((obj[1].getId(),obj[0].versionNumber)) + + return ret + def getContentIds(self): """print basket content""" ret=[] @@ -968,7 +1311,22 @@ class CDLIBasketVersion(Implicit,Persist return True return False - security.declareProtected('View','downloadObjectsAsOneFile') + def downloadListOfPnumbers(self): + """download pnumbers of the basket as list""" + + basket_name=self.aq_parent.title + + ids=self.getContent() # get the list of objects + logging.error(ids) + ret="\n".join([x[1].getId().split(".")[0] for x in ids]) + + self.REQUEST.RESPONSE.setHeader("Content-Disposition","""attachement; filename="%s.txt" """%basket_name) + self.REQUEST.RESPONSE.setHeader("Content-Type","application/octet-stream") + length=len(ret) + self.REQUEST.RESPONSE.setHeader("Content-Length",length) + self.REQUEST.RESPONSE.write(ret) + + security.declareProtected('manage','downloadObjectsAsOneFile') def downloadObjectsAsOneFile(self,lock=None,procedure=None,REQUEST=None,check="yes",current="no"): """download all selected files in one file""" @@ -1040,7 +1398,7 @@ class CDLIBasketVersion(Implicit,Persist ret+=str(object[0].getData())+"\n" elif current=="yes": #search current object - founds=self.CDLICatalog.search({'title':object[0].getId()}) + founds=self.CDLICatalog.search({'title':object[1].getId().split(".")[0]}) if len(founds)>0: ret+=str(founds[0].getObject().getLastVersion().getData())+"\n" @@ -1086,7 +1444,7 @@ class CDLIBasketVersion(Implicit,Persist def __init__(self,id,user,comment="",basketContent=[]): """ init a basket version""" self.id=id - self.coment=comment + self.comment=comment self._setObject('content',BasketContent(basketContent)) #self.basketContent=basketContent[0:]a self.user=user @@ -1100,7 +1458,7 @@ class CDLIBasketVersion(Implicit,Persist """get Comment""" return self.comment - security.declareProtected('View','index_html') + security.declareProtected('manage','index_html') def index_html(self): """view the basket""" @@ -1152,9 +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) @@ -1164,31 +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("\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): @@ -1217,13 +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) @@ -1236,47 +1595,56 @@ 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._getOb(id).versionComment=str(vC) - self._getOb(id).time=time.localtime() - - setattr(self._getOb(id),'author',author) - + self._setObject(id, CDLIFileObject(id,title,versionNumber=versionNumber,versionComment=vC,time=time.localtime(),author=author)) + fob = self._getOb(id) # Now we "upload" the data. By doing this in two steps, we # can use a database trick to make the upload more efficient. if file and not from_tmp: - self._getOb(id).manage_upload(file) + fob.manage_upload(file) elif file and from_tmp: - self._getOb(id).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: - self._getOb(id).content_type=content_type + fob.content_type=content_type + #logging.debug("manage_add: lastversion=%s"%self.getData()) + logging.debug("reindex1: %s in %s"%(repr(self),repr(self.default_catalog))) self.reindex_object() - self._getOb(id).reindex_object() + #logging.debug("manage_add: fob_data=%s"%fob.getData()) + logging.debug("reindex2: %s in %s"%(repr(fob), repr(fob.default_catalog))) + fob.index_object() if REQUEST is not None: REQUEST['RESPONSE'].redirect(self.absolute_url()+'/manage_main') + class CDLIFile(extVersionedFile,CatalogAware): """CDLI file""" + security=ClassSecurityInfo() meta_type="CDLI file" + content_meta_type = ["CDLI File Object"] + default_catalog='CDLICatalog' - #security.declarePublic('history') + security.declareProtected('manage','index_html') + 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""" @@ -1321,6 +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""" @@ -1346,58 +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:] - - - if not newName=='': - self.title=newName[0:] - - + ob = self.addContentObject(id, vC, author, file, title, changeName=changeName, newName=newName, from_tmp=from_tmp, + precondition=precondition, content_type=content_type) - - 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): @@ -1411,8 +1754,6 @@ def manage_addCDLIFile(self,id,title,loc tryToggle=True tryCount=0 - - self._setObject(id,newObj) getattr(self,id).reindex_object() @@ -1421,6 +1762,14 @@ def manage_addCDLIFile(self,id,title,loc RESPONSE.redirect('manage_main') +def checkUTF8(data): + """check utf 8""" + try: + data.encode('utf-8') + return True + except: + return False + def checkFile(filename,data,folder): """check the files""" @@ -1431,17 +1780,11 @@ def checkFile(filename,data,folder): return False,"P missing in the filename" elif len(fn[0])!=7: return False,"P number has not the right length 6" + elif not checkUTF8(data): + return False,"not utf-8" else: - fn=os.path.join(folder,filename) - stin,out=os.popen4("/usr/bin/atfcheck.plx %s"%fn) - value=out.read() - ret= out.close() - - if value: - - return False,"atf checker error: %s"%value - else: - return True,"" + return True,"" + def splitatf(fh,dir=None,ext=None): """split it""" @@ -1449,7 +1792,13 @@ def splitatf(fh,dir=None,ext=None): nf=None i=0 - for lineTmp in fh.readlines(): + #ROC: why split \n first and then \r??? + if (type(fh) is StringType) or (type(fh) is UnicodeType): + iter=fh.split("\n") + else: + iter=fh.readlines() + + for lineTmp in iter: lineTmp=lineTmp.replace(codecs.BOM_UTF8,'') # make sure that all BOM are removed.. for line in lineTmp.split("\r"): #logging.log("Deal with: %s"%line) @@ -1486,7 +1835,9 @@ def splitatf(fh,dir=None,ext=None): nf.close() except: pass - fh.close() + + if not((type(fh) is StringType) or (type(fh) is UnicodeType)): + fh.close() return ret,len(os.listdir(dir)) @@ -1495,34 +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 hinzufŸgen 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""" @@ -1532,38 +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() - logging.error(obj.getData()) - return obj.getData()[0:] + obj = self[fn].getContentObject() + + 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) @@ -1644,7 +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 @@ -1671,7 +2026,8 @@ 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 transaction.get().commit() @@ -1690,7 +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 @@ -1698,16 +2054,16 @@ 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 - security.declareProtected('View','index_html') + security.declareProtected('manage','index_html') def index_html(self): """main""" ext=self.ZopeFind(self,obj_ids=["index.html"]) @@ -1754,172 +2110,199 @@ 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 - def findWordRegExp(self,searchTerm): - """find all words in index which match regexp in SearchTerm""" - ret=[] - for x in self.lineIndex.iterkeys(): - if re.match(searchTerm,x): - ret.append(x) - return ret + file_catalog = 'CDLICatalog' - def searchRegExpInLineIndexDocs(self,searchTerm): - """search in inLineIndex with regexp""" - if not searchTerm: - return [] - ret=[] - words=self.findWordRegExp(searchTerm) # suche nach allen Treffern - logging.info("wd:%s"%words) - for word in words: - ret+=self.searchInLineIndexDocs(word) - - return unique(ret) - - def showInLineIndex(self): - """get the index for debug purposes""" - print "show" - for x in self.lineIndex.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.lineIndex - - def searchInLineIndexDocs(self,word,uniq=True,regExp=False): - """search occurences""" - - if regExp: - return self.searchRegExpInLineIndexDocs(word) - - try: - lst=list(self.lineIndex.get(word).keys()) - except: - lst=[] - if uniq: - return unique(lst) - else: - return lst - - def getLinesFromIndex(self,word,doc,regExp=False): - """get lines""" - if not regExp: - return self.lineIndex.get(word)[doc] - else: # wenn regexp, suche welches word - for w in self.findWordRegExp(word): - if self.lineIndex.get(w): # ein word in im dex gefunden - try: - dc=self.lineIndex.get(word)[doc] - return dc # und ein document dann gib es zurueck - except: - pass #andernfalls weiter - - def cleanInLineIndex(self): - """delete InlineIndex""" - for x in list(self.lineIndex.keys()): - del(self.lineIndex[x]) - print [x for x in self.lineIndex.keys()] - - return "ok" + # word splitter for search + splitter = {'words':cdliSplitter.wordSplitter(), + 'graphemes':cdliSplitter.graphemeSplitter()} - def storeInLineIndex(self,key,value): - """store in index""" - - if (not hasattr(self,'lineIndex')) or (type(self.lineIndex) is DictType): - self.lineIndex=OOBTree() - li=self.lineIndex - - 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.lineIndex=li - - transaction.get().commit() + + def deleteFiles(self,ids): + """delete files""" + for id in ids: + founds=self.CDLICatalog.search({'title':id.split(".")[0]}) + if founds: + logging.debug("deleting %s"%founds) + folder=founds[0].getObject().aq_parent #get the parent folder of the object + logging.debug("deleting from %s"%folder) + cut=folder.delete([founds[0].getId]) #cut it out + + + + 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): - """show a file""" - f=self.CDLICatalog({'title':fileId}) + def showFile(self,fileId,wholePage=False): + """show a file + @param fileId: P-Number of the document to be displayed + """ + f=getattr(self, self.file_catalog).search({'textid':fileId}) if not f: return "" - return f[0].getObject().getLastVersionFormattedData() + if wholePage: + logging.debug("show whole page") + return f[0].getObject().getContentObject().view() + else: + return f[0].getObject().getLastVersionFormattedData() - def showLineFromFile(self,fileId,lineNum,word): - """get line lineNum fromFileId""" - - file=self.showFile(fileId) - #str="^%s\.[^%s\.]*%s[^\n]*\n"%(lineNum,lineNum,word) - #str="^%s\..*?%s[^\n]*\n"%(lineNum,word) - - #print str - #m=re.search(str,file,flags=re.M|re.DOTALL) - #if m: - # return m.group() - #else: - # return "" - #ret=lineNum+"." - #splitted=file.split(lineNum+".") - #if len(splitted)>1: - #for part in splitted[1:]: - #if part.find(word)>-1: - # for x in part.split("\n"): - #ret+=x - #if x.find(word)>-1: - #break - #break; - #return ret - def showWordInFile(self,fileId,word,lineList=None): - """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) - + file = formatAtfFullLineNum(self.getFile(fileId)) ret=[] - for line in file.split("\n"): - 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) + + # 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: + #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): - """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)) + + 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.showFile(fileId) - tagStr="""%s""" + file=self.getFile(fileId) + tagStart=u'' + tagEnd=u'' + tagStr=tagStart + u'%%s' + tagEnd ret=[] - for line in file.split("\n"): - 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.replace(word,tagStr%word)) - else: # nimm alles ohne line check - ret.append(line.replace(word,tagStr%word)) - else: - ret.append(line) - return "
\n".join(ret) + + # 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.splitlines(): + line = unicodify(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'
\n'.join(ret) + + + + def tagWordInFiles(self,fileIds,word,indexName='graphemes',regExp=False): + """ + get texts with highlighted word from all ids in list FileIds. + returns dict with id:text pairs. + """ + logging.debug("tagwordinfiles word='%s' index=%s file=%s"%(word,indexName,fileIds)) + return dict([(id,self.tagWordInFile(id, word, indexName, regExp)) for id in fileIds]) + + + def getFileVersionList(self, pnum): + """get the version history as a list for the translit file with the given pnum""" + f = getattr(self, self.file_catalog).search({'textid':pnum}) + if not f: + return [] + + return f[0].getObject().getVersionList() + def URLquote(self,str): """quote url""" @@ -1948,19 +2331,8 @@ class CDLIRoot(Folder): ret.append((f[0],un)) return ret - - def forceDahl(self): - "break all locks" - ret=[] - for f in self.ZopeFind(self,obj_metatypes="CDLI file",search_sub=1): - if str(f[1].lockedBy)=="dahl": - un=f[1].forceunlock() - - if un and un !="": - ret.append((f[0],un)) + - return ret - def getChangesByAuthor(self,author,n=100): """getChangesByAuthor""" zcat=self.CDLIObjectsCatalog @@ -2022,10 +2394,40 @@ class CDLIRoot(Folder): return ret - + def uploadATFRPC(self,data,username): + """upload an atffile via xml-rpc""" + uploader=uploadATFThread() + + #generate an random id for the upload object + from random import randint + if (not self.REQUEST.SESSION.get('idTmp',None)): + + idTmp=str(randint(0,1000000000)) + self.REQUEST.SESSION['idTmp']=idTmp + else: + idTmp=self.REQUEST.SESSION.get('idTmp',None) + + + uploader.set(data,0,username,idTmp) + + stObj=uploader.run() + + processor=uploadATFfinallyThread() + + basketname=stObj.returnValue['basketNameFromFile'] + + processor.set("uploadchanged",basketname=basketname,SESSION=stObj.returnValue,username=username,serverport=self.REQUEST['SERVER_PORT']) + + processor.run() + + + return generateXMLReturn(stObj.returnValue) + def uploadATF(self,repeat=None,upload=None,basketId=0,RESPONSE=None): - """standard ausgabe""" + """upload an atf file / basket file""" #self._v_uploadATF.returnValue=None + + #generate an random id for the upload thread from random import randint if (not self.REQUEST.SESSION.get('idTmp',None)): @@ -2034,8 +2436,10 @@ class CDLIRoot(Folder): else: idTmp=self.REQUEST.SESSION.get('idTmp',None) + threadName=repeat if not threadName or threadName=="": + #new thread not called from the waiting page tmpVar=False thread=uploadATFThread() @@ -2083,26 +2487,8 @@ class CDLIRoot(Folder): return pt(txt='/uploadATF',threadName=threadName) else: -# tmp={} -# for key in self._v_uploadATF[threadName].returnValue.keys(): -# t=self._v_uploadATF[threadName].returnValue[key] -# if type(t) is ListType: -# tmp[key]=self._v_uploadATF[threadName].returnValue[key][0:] -# else: -# tmp[key]=self._v_uploadATF[threadName].returnValue[key] -# repr(tmp[key]),repr(key) -# -# # - #tmp=self.cdli_main.tmpStore2[threadName] - tmp=getattr(self.temp_folder,idTmp).returnValue - - - - - #del(self.cdli_main.tmpStore2[threadName]) - - + pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','uploadCheck.zpt')).__of__(self) return pt(changed=tmp['changed'],lockerrors=tmp['lockerrors'],errors=tmp['errors'],dir=tmp['dir'],newPs=tmp['newPs'],basketLen=tmp['basketLen'],numberOfFiles=tmp['numberOfFiles'], @@ -2172,6 +2558,7 @@ class CDLIRoot(Folder): def importFiles(self,comment="",author="" ,folderName="/Users/dwinter/atf", files=None,ext=None): """import files""" + logging.debug("importFiles folderName=%s files=%s ext=%s"%(folderName,files,ext)) root=self.cdli_main count=0 if not files: @@ -2181,17 +2568,20 @@ class CDLIRoot(Folder): folder=f[0:3] f2=f[0:5] obj=self.ZopeFind(root,obj_ids=[folder]) + logging.debug("importFiles: folder=%s f2=%s obj=%s"%(folder,f2,obj)) if ext: - ext.result="

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) @@ -2202,22 +2592,15 @@ 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"