--- cdli/cdli_files.py 2008/01/18 20:53:49 1.80.2.18 +++ cdli/cdli_files.py 2011/07/20 09:33:35 1.119 @@ -1,10 +1,8 @@ """CDLI extensions of the filearchive""" from Products.versionedFile.extVersionedFile import * from Products.ZCatalog.CatalogPathAwareness import CatalogAware -from tempfile import mkstemp,mkdtemp import os.path import os -from types import * import urlparse import urllib import cgi @@ -21,1487 +19,23 @@ from ZPublisher.HTTPRequest import HTTPR from ZPublisher.HTTPResponse import HTTPResponse from ZPublisher.BaseRequest import RequestContainer import threading -from BTrees.OOBTree import OOBTree import logging import transaction import copy import codecs import sys - +from BTrees.IOBTree import IOBTree +from BTrees.OOBTree import OOBTree import cdliSplitter +from sets import Set +import md5 +from DownloadBasket import DownloadBasketFinallyThread +from types import * +import pickle +import tempfile + +from cdli_helpers import * - -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. - - For example, unique([1,2,3,1,2,3]) is some permutation of [1,2,3], - unique("abcabc") some permutation of ["a", "b", "c"], and - unique(([1, 2], [2, 3], [1, 2])) some permutation of - [[2, 3], [1, 2]]. - - For best speed, all sequence elements should be hashable. Then - unique() will usually work in linear time. - - If not possible, the sequence elements should enjoy a total - ordering, and if list(s).sort() doesn't raise TypeError it's - assumed that they do enjoy a total ordering. Then unique() will - usually work in O(N*log2(N)) time. - - If that's not possible either, the sequence elements must support - equality-testing. Then unique() will usually work in quadratic - time. - (from the python cookbook) - """ - - n = len(s) - if n == 0: - return [] - - # Try using a dict first, as that's the fastest and will usually - # work. If it doesn't work, it will usually fail quickly, so it - # usually doesn't cost much to *try* it. It requires that all the - # sequence elements be hashable, and support equality comparison. - u = {} - try: - for x in s: - u[x] = 1 - except TypeError: - del u # move on to the next method - else: - return u.keys() - - # We can't hash all the elements. Second fastest is to sort, - # which brings the equal elements together; then duplicates are - # easy to weed out in a single pass. - # NOTE: Python's list.sort() was designed to be efficient in the - # presence of many duplicate elements. This isn't true of all - # sort functions in all languages or libraries, so this approach - # is more effective in Python than it may be elsewhere. - try: - t = list(s) - t.sort() - except TypeError: - del t # move on to the next method - else: - assert n > 0 - last = t[0] - lasti = i = 1 - while i < n: - if t[i] != last: - t[lasti] = last = t[i] - lasti += 1 - i += 1 - return t[:lasti] - - # Brute force is all that's left. - u = [] - for x in s: - if x not in u: - u.append(x) - return u - - -class BasketContent(SimpleItem): - """classe fuer den Inhalt eines Baskets""" - - def __init__(self,content=[]): - """content""" - self.contentList=content[0:] - - def getContent(self): - """get content""" - - return self.contentList - - def setContent(self,content): - self.contentList=content[0:] - - def numberOfItems(self): - """number""" - - return len(self.getContent()) - - -class uploadATFfinallyThread(Thread): - """class for adding uploaded filed (temporarily stored in the staging area at /tmp""" - - def __init__(self): - """init for uploadATFfinallyThread""" - self.continueVar=True - self.returnValue=None - self.end=False - Thread.__init__(self) - - def set(self,procedure,comment="",basketname='',unlock=None,SESSION=None,username=None,serverport="8080"): - """set start values for the thread""" - self.procedure=procedure - self.comment=comment - self.basketname=basketname - self.unlock=unlock - self.SESSION=SESSION - self.username=username - self.serverport=serverport - - - def __call__(self): - """call of the thread (equals run)""" - self.run() - return True - - def getContext(self, app,serverport="8080"): - """get the context within the ZODB""" - - resp = HTTPResponse(stdout=None) - env = { - 'SERVER_NAME':'localhost', - 'SERVER_PORT':serverport, - 'REQUEST_METHOD':'GET' - } - req = HTTPRequest(None, env, resp) - return app.__of__(RequestContainer(REQUEST = req)) - - - def run(self): - """run""" - - self.result="" - #find context within ZODB - from Zope import DB - conn = DB.open() - root = conn.root() - app = root['Application'] - ctx = self.getContext(app,serverport=self.serverport) - - #add the files - self.uploadATFfinallyThread(ctx,self.procedure,comment=self.comment,basketname=self.basketname,unlock=self.unlock,SESSION=self.SESSION,username=self.username) - #commit the transactions - transaction.get().commit() - conn.close() - #set flag for end of this method - self.end=True - logging.info("ended") - return True - - def __del__(self): - """delete""" - - - - def getResult(self): - """method for accessing result""" - - return self.result - - def uploadATFfinallyThread(self,ctx,procedure,comment="",basketname='',unlock=None,RESPONSE=None,SESSION=None,username=None): - """upload the files""" - #TODO: make this configurable, at the moment, rootFolder for cdli has to be cdliRoot - ctx2=ctx.cdliRoot - - 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',[]) - - #or all - elif procedure=="uploadAll": - uploadFns=[] - for x in os.listdir(SESSION['tmpdir']): - if not x in SESSION['lockerrors']: - uploadFns.append(x) - - #or maybe nothing - elif procedure=="noupload": - 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==0: - transaction.get().commit() - logging.debug("uploadatffinallythread changing: do commit") - - transaction.get().commit() - logging.debug("uploadatffinallythread changing: last commit") - - #now add the new files - newPs=SESSION['newPs'] - if len(newPs)>0: - tmpDir=SESSION['tmpdir'] - 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.debug("uploadatffinallythread adding finished") - - #unlock locked files? - if unlock: - 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.debug("unlocking have now what to unlock") - - for fn in unlockFns: - #logging.info("will unlock: %s"%fn) - founds=ctx2.CDLICatalog.search({'title':fn}) - #logging.info("found it: %s"%repr(founds)) - if len(founds)>0: - #logging.info("unlock: %s"%founds[0].getObject().getId()) - SESSION['author']=str(username) - founds[0].getObject().lockedBy="" - - logging.debug("uploadatffinallythread unlocking done") - - #if a basketname is given, add files to the basket - if not (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.debug("uploadatffinallythread create basket %s"%basketname) - self.result="

Create a new basket

"+self.result - ob=ctx2.basketContainer.addBasket(basketname) - basketId=ob.getId() - basket=getattr(ctx2.basketContainer,str(basketId)) - 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()) - - return True - -class tmpStore(SimpleItem): - """simple item""" - meta_type="cdli_upload" - - def __init__(self,id): - """init tmp""" - self.id=id - -class uploadATFThread(Thread): - """class for checking the files befor uploading""" - - def __init__(self): - """initialise""" - - self.continueVar=True - self.returnValue=None - - Thread.__init__(self) - - - def set(self,upload,basketId,username,idTmp,serverport="8080"): - """set start values for the thread""" - self.result="" - self.upload=upload - self.basketId=basketId - self.username=username - self.serverport=serverport - self.idTmp=idTmp - - def __call__(self): - """call method """ - self.run() - return True - - def getContext(self, app,serverport="8080"): - """get the context within the ZODB""" - resp = HTTPResponse(stdout=None) - env = { - 'SERVER_NAME':'localhost', - 'SERVER_PORT':serverport, - 'REQUEST_METHOD':'GET' - } - req = HTTPRequest(None, env, resp) - return app.__of__(RequestContainer(REQUEST = req)) - - def run(self): - idTmp=self.idTmp - self.result="" - #find context within ZODB - from Zope import DB - conn = DB.open() - root = conn.root() - app = root['Application'] - 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]) - - logging.info("call thread intern") - self.uploadATFThread(ctx,self.upload,idTmp,self.basketId) - - #ctx.cdliRoot.cdli_main.tmpStore2[self.getName()[0:]]=self.returnValue - - - transaction.get().commit() - - conn.close() - - return getattr(ctx.temp_folder,idTmp) - - def getResult(self): - """method for accessing result""" - return self.result - - def uploadATFThread(self,ctx,upload,idTmp,basketId=0): - """upload an atf file""" - #TODO: add comments - #TODO: finish uploadATF - - stObj=getattr(ctx.temp_folder,idTmp) - logging.info("start, upload thread") - self.result="

I got your file, start now to split it into single atf-files!

" - - #make sure that id is a string and not an integer - basketId=str(basketId) - - #TODO: make this configurable, at the moment, rootFolder for cdli has to be cdliRoot - ctx2=ctx.cdliRoot - - #get temporary file for staging the downloaded and splitted files - dir=mkdtemp() - - - changed=[] # changed files - errors=[] # files with errors - lockerrors=[] # files with errors - - newPs=[] # new p filed - psNotInCatalog=[] # files not in the catalog - - #split the uploadedd atf file - basketNameFromFile, numberOfFiles=splitatf(upload,dir,ext=self) - - #find basketId if not set - - #get active abaket - if basketId == '0': - basketObj=ctx2.basketContainer.getActiveBasket() - if basketObj: - basketId=basketObj.getId() - - #if there is no active basket and no basketid given, id is empty, else get besketname and length - if basketId == '0': - basketNameFromId="" - basketLen=0 - else: - basketNameFromId=getattr(ctx2.basketContainer,basketId).title - basketLen=getattr(ctx2.basketContainer,basketId).getLastVersion().numberOfItems() - - logging.info("got the file, upload thread") - self.result+="""

I got the files

< - p>I am computing the differences to the exisiting files

""" - - #start to check the files - for fn in os.listdir(dir): - - self.result="

process:%s

"%fn+self.result - - # check if file is in the catalog - #TODO: checkCatalog is not implemented yet - if ctx2.cdli_main.checkCatalog(fn): - psNotInCatalog.append(fn) - - #check if p-file already at the server - founds=ctx2.CDLICatalog.search({'title':fn}) - - #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 - - - if not status: # error - errors.append((fn,msg)) - - else: - if len(founds)==0: - newPs.append(fn) - - #if p file alread at the server - for found in founds: - #analyse the differences to the actual file - obj=found.getObject() - - if (not (str(obj.lockedBy))=='') and (not (str(obj.lockedBy)==str(self.username))): - lockerrors.append((fn,str(obj.lockedBy))) - else: - - diffs=obj.diff(data) - if diffs[0]>0: - changed.append((obj,diffs)) #hochladen - - #ready, set the returnValues - self.result+="

Done

" - - stObj.returnValue={} - - stObj.returnValue['errors']=errors - - stObj.returnValue['newPs']=newPs - stObj.returnValue['tmpdir']=dir - stObj.returnValue['basketLen']=basketLen - stObj.returnValue['numberOfFiles']=numberOfFiles - stObj.returnValue['basketNameFromId']=basketNameFromId - stObj.returnValue['basketNameFromFile']=basketNameFromFile - stObj.returnValue['basketId']=basketId - stObj.returnValue['dir']=dir - #stObj.returnValue['changed']=copy.copy(changed) - stObj.returnValue['changed']=[(x[0].getId(),x[1][0]) for x in changed] - #stObj.returnValue['lockerrors']=[x[0].getId() for x in lockerrors] - stObj.returnValue['lockerrors']=[x for x in lockerrors] - self.returnValue=True - #ctx2.cdli_main.setTemp('v_uploadATF_returnValue',True) - - -class CDLIBasketContainer(OrderedFolder): - """contains the baskets""" - - - 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""" - - founds=self.ZopeFind(self,obj_metatypes=['CDLIBasketVersion'],search_sub=1) - - for found in founds: - found[1].updateBasket() - - security.declareProtected('manage','deleteBaskets') - def deleteBaskets(self,ids=None): - """delete baskets, i.e. move them into trash folder""" - - - found=self.ZopeFind(self,obj_ids=['trash']) - - if len(found)<1: - manage_addFolder(self, 'trash') - trash=self._getOb('trash') - else: - trash=found[0][1] - - if type(ids) is not ListType: - ids=[ids] - cut=self.manage_cutObjects(ids) - trash.manage_pasteObjects(cut) - - security.declareProtected('manage','manageBaskets') - 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()) - - security.declareProtected('View','getBasketIdfromName') - def getBasketIdfromName(self,basketname): - """get id from name""" - - for basket in self.ZopeFind(self,obj_metatypes=["CDLIBasket"]): - if basket[1].title==basketname: - return basket[0] - else: - None - - security.declareProtected('manage','uploadBasket_html') - - def uploadBasket_html(self,basketId='0'): - """upload an atf file, html form""" - - - basketId=str(basketId) - if not basketId=='0': - basketName=getattr(self.basketContainer,basketId).title - else: - basketName="" - - pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','uploadBasket_html.zpt')).__of__(self) - return pt(basketId=basketId,basketName=basketName) - - - security.declareProtected('manage','index_html') - def index_html(self): - """stanadard ansicht""" - - - - ext=self.ZopeFind(self,obj_ids=["index.html"]) - if ext: - return ext[0][1]() - - pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','BasketContainerMain')).__of__(self) - return pt() - - def getStorageFolderRoot(self): - """root des storage folders""" - return self.cdli_main - - def __init__(self,id,title): - """ init basket container""" - self.id=id - self.title=title - - - def getBasketsId(self): - """get all baskets als klartext""" - - ret="" - baskets=self.ZopeFind(self,obj_metatypes=['CDLIBasket']) - for basket in baskets: - com,user,time,values = basket[1].getContentIds() - ret+= "BASKET:"+com+"\t"+user+"\t"+time+"\n" - for x in values: - ret+= x[0]+"\t"+x[1]+"\n" - return ret - - def getBaskets(self,sortField='title'): - """get all baskets files""" - - def sortName(x,y): - return cmp(x[1].title.lower(),y[1].title.lower()) - - def sortDate(x,y): - return cmp(y[1].getLastVersion().getTime(),x[1].getLastVersion().getTime()) - - - def sortComment(x,y): - - - - try: - xc=getattr(x[1],'comment','ZZZZZZZZZZZZZ').lower() - except: - xc='ZZZZZZZZZZZZZ'.lower() - try: - yc=getattr(y[1],'comment','ZZZZZZZZZZZZZ').lower() - except: - yc='ZZZZZZZZZZZZZ'.lower() - - - if (xc=='') or (xc=='ZZZZZZZZZZZZZ'.lower()): - - try: - xc=x[1].getLastVersion().getComment().lower() - except: - xc='ZZZZZZZZZZZZZ'.lower() - - if (yc=='') or (yc=='ZZZZZZZZZZZZZ'.lower()): - try: - yc=y[1].getLastVersion().getComment().lower() - except: - yc='ZZZZZZZZZZZZZ'.lower() - - - return cmp(xc,yc) - - def sortAuthor(x,y): - - return cmp(x[1].getLastVersion().getUser().lower(),y[1].getLastVersion().getUser().lower()) - - baskets=self.ZopeFind(self,obj_metatypes=['CDLIBasket']) - - - if sortField=='title': - baskets.sort(sortName) - elif sortField=='date': - baskets.sort(sortDate) - elif sortField=='author': - baskets.sort(sortAuthor) - elif sortField=='comment': - 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) - last +=1 - while len(self.ZopeFind(self,obj_ids=[str(last)]))>0: - last+=1 - - self.last=last - return last - - def setActiveBasket(self,basketId,REQUEST=None): - """store active basketId in a cookie""" - self.REQUEST.RESPONSE.setCookie("CDLIActiveBasket",basketId,path="/") - try: - qs=cgi.parse_qs(REQUEST['QUERY_STRING']) - del(qs['basketId']) - except: - qs={} - if REQUEST: - REQUEST.RESPONSE.redirect(REQUEST['URL1']+'?'+urllib.urlencode(qs)) - - def getActiveBasket(self): - """get active basket from cookie""" - - id= self.REQUEST.cookies.get('CDLIActiveBasket',None) - if id: - obj=getattr(self,str(id),None) - else: - obj=None - return obj - - def getActualUserName(self): - """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: - ids=self.REQUEST.SESSION['fileIds'] - - if type(ids) is not ListType: - ids=[ids] - - if (submit.lower()=="store in new basket") or (submit.lower()=="new basket"): - basketRet=self.addBasket(newBasketName) - self.setActiveBasket(basketRet.getId()) - basket=getattr(self,basketRet.getId()) - elif (submit.lower()=="store in active basket") or (submit.lower()=="active basket"): - basket=self.getActiveBasket() - - added=basket.addObjects(ids) - back=self.REQUEST['HTTP_REFERER'].split("?")[0]+"?basketName="+basket.title+"&numberOfObjects="+str(added) - - - if fromFileList: - - return self.cdli_main.findObjectsFromList(list=self.REQUEST.SESSION['fileIds'],basketName=basket.title,numberOfObjects=added) - - if RESPONSE: - - RESPONSE.redirect(back) - - return True - -def manage_addCDLIBasketContainerForm(self): - """add the CDLIBasketContainer form""" - pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','addCDLIBasketContainer.zpt')).__of__(self) - return pt() - -def manage_addCDLIBasketContainer(self,id,title,RESPONSE=None): - """add the basket""" - ob=CDLIBasketContainer(id,title) - - self._setObject(id, ob) - - if RESPONSE is not None: - RESPONSE.redirect('manage_main') - -class CDLIBasket(Folder,CatalogAware): - """basket""" - - 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] - - def getFileLastVersion(self,obj): - return obj[0] - - def getFileNamesInLastVersion(self): - """get content of the last version as list""" - - return [x[1].getId() for x in self.getLastVersion().getContent()] - - - def isActual(self,obj): - """teste ob im basket die aktuelle version ist""" - actualNo=obj[1].getLastVersion().getVersionNumber() - storedNo=obj[0].getVersionNumber() - - founds=self.CDLICatalog.search({'title':obj[0].getId()}) - if len(founds)>0: - actualNo=founds[0].getObject().getLastVersion().getVersionNumber() - - if len(founds)>0 and founds[0].getObject().aq_parent.getId()==".trash": - return False, -1 - - if actualNo==storedNo: - return True , 0 - else: - return False, actualNo - - def history(self): - """history""" - - ext=self.ZopeFind(self.aq_parent,obj_ids=["history_template.html"]) - if ext: - return getattr(self,ext[0][1].getId())() - - pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','BasketHistory')).__of__(self) - return pt() - - def getStorageFolderRoot(self): - """root des storage folders""" - return self.aq_parent.cdli_main - - def __init__(self,id,title,shortDescription="",comment=""): - """init a CDLIBasket""" - - self.id=id - self.title=title - 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""" - - ids=[] - idsTmp= self.objectIds() - for x in idsTmp: - try: - ids.append(int(x)) - except: - pass - ids.sort() - - if len(ids)==0: - return None - else: - ob=getattr(self,str(ids[-1])) - - - return ob - - def getVersions(self): - """get versions""" - versions=self.ZopeFind(self,obj_metatypes=["CDLIBasketVersion"]) - return versions - - - def updateObjects(self,ids,RESPONSE=None,REQUEST=None): - """update ids, ids not in the basket the add""" - if type(ids) is not ListType: - ids=[ids] - - lastVersion=self.getLastVersion() - oldContent=lastVersion.content.getContent() - newContent=[] - - #first copy the old - for obj in oldContent: - if obj[1].getId() not in ids: - newContent.append(obj) - #now add the new - - for id in ids: - founds=self.CDLICatalog.search({'title':id}) - - 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().getLastVersion(),found.getObject())) - - - content=newContent - user=self.getActualUserName() - - ob=manage_addCDLIBasketVersion(self,user,comment="",basketContent=newContent) - - obj=self._getOb(ob.getId()) - if RESPONSE: - - RESPONSE.redirect(obj.absolute_url()) - - 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())) - lastVersion=self.getLastVersion() - - if lastVersion is None: - oldContent=[] - else: - oldContent=lastVersion.content.getContent() - - if deleteOld: - oldContent=[] - - newContent=[] - added=0 - for id in ids: - try: - founds=self.CDLICatalog.search({'title':id}) - except: - 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().getLastVersion(),found.getObject())) - added+=1 - - content=oldContent+newContent - if not username: - 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 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=[] - lv=self.getLastVersion() - for obj in lv.content.getContent(): - ret.append((obj[0].getId(),obj[1].getId())) - - - return lv.getComment(),lv.getUser(),lv.getTime(),ret - - def changeBasket(self,ids,submit,RESPONSE=None,REQUEST=None): - """change a basket""" - if submit=="update": - return self.updateObjects(ids,RESPONSE=RESPONSE,REQUEST=REQUEST) - elif submit=="delete": - return self.deleteObjects(ids,RESPONSE=RESPONSE,REQUEST=REQUEST) - - def deleteObjects(self,ids,RESPONSE=None,REQUEST=None): - """delete objects""" - - if type(ids) is not ListType: - ids=[ids] - - lastVersion=self.getLastVersion() - oldContent=lastVersion.content.getContent() - newContent=[] - for obj in oldContent: - if obj[1].getId() not in ids: - newContent.append(obj) - - - user=self.getActualUserName() - - ob=manage_addCDLIBasketVersion(self,user,comment="",basketContent=newContent) - - if RESPONSE: - obj=self._getOb(ob.getId()) - RESPONSE.redirect(obj.absolute_url()) - -def manage_addCDLIBasketForm(self): - """add the CDLIBasketContainer form""" - pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','addCDLIBasket.zpt')).__of__(self) - return pt() - -def manage_addCDLIBasket(self,title,shortDescription="",comment="",RESPONSE=None): - """add the basket""" - - id=str(self.getNewId()) - - ob=CDLIBasket(id,title,shortDescription,comment) - - self._setObject(id, ob) - - if RESPONSE is not None: - RESPONSE.redirect('manage_main') - else: - return ob - -class CDLIBasketVersion(Implicit,Persistent,Folder): - """version of a basket""" - - meta_type="CDLIBasketVersion" - security=ClassSecurityInfo() - - def updateBasket(self): - """update""" - try: - self._setObject('content',BasketContent(self.basketContent)) - except: - try: - if len(self.basketContent)>0: - self.content.setContent(self.basketContent) - except: - print "error",self.getId(),self.aq_parent.getId() - self.basketContent=[] - - - def containsNonActualFiles(self): - """returns True if basket contains one or more non current files""" - - objs=self.getContent() - for obj in objs: - if not self.isActual(obj)[0]: - return True - return False - - 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""" - - if self.temp_folder.downloadCounterBaskets > 10000: - return """I am sorry, currently the server has to many requests for downloads, please come back later!""" - - - if (check=="yes") and self.containsNonActualFiles(): - pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','downloadObjectAsOneFile_check.zpt')).__of__(self) - - return pt(lock=lock) - - else: - - return self.downloadObjectsAsOneFileFinally(lock=lock,procedure=procedure,REQUEST=REQUEST,current="no") - - def downloadObjectsAsOneFileFinally(self,lock=None,procedure=None,REQUEST=None,current="no"): - """print do the download""" - - ret="" - lockedObjects={} - - self.temp_folder.downloadCounterBaskets+=1 - self._p_changed=1 - transaction.get().commit() - - if lock: - - if str(self.REQUEST['AUTHENTICATED_USER'])=='Anonymous User': - self.temp_folder.downloadCounterBaskets-=1 - self._p_changed=1 - transaction.get().commit() - self.temp_folder.downloadCounterBaskets-=1 - self._p_changed=1 - transaction.get().commit() - return "please login first" - - #check if a locked object exist in the basket. - lockedObjects={} - for object in self.content.getContent(): - - if (not str(object[1].lockedBy)=="") and (not (str(object[1].lockedBy)==str(self.REQUEST['AUTHENTICATED_USER']))): - lockedObjects[object[1].title]=repr(object[1].lockedBy) - - - keys=lockedObjects.keys() - - - if len(keys)>0 and (not procedure): - self.REQUEST.SESSION['lockedObjects']=lockedObjects - pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','lockedObjects.zpt')).__of__(self) - - self.temp_folder.downloadCounterBaskets-=1 - self._p_changed=1 - transaction.get().commit() - - return pt() - - elif not procedure: #keine fails gesperrt dann alle donwloaden - procedure="downloadAll" - - - - - for object in self.content.getContent(): - - if (procedure=="downloadAll") or (object[1].lockedBy=='') or (object[1].lockedBy==self.REQUEST['AUTHENTICATED_USER']): - if current=="no": #version as they are in the basket - ret+=str(object[0].getData())+"\n" - elif current=="yes": - #search current object - founds=self.CDLICatalog.search({'title':object[1].getId().split(".")[0]}) - if len(founds)>0: - ret+=str(founds[0].getObject().getLastVersion().getData())+"\n" - - if lock and object[1].lockedBy=='': - object[1].lockedBy=self.REQUEST['AUTHENTICATED_USER'] - basket_name=self.aq_parent.title+"_V"+self.getId() - - #write basketname to header of atf file - ret="#basket: %s\n"%basket_name+ret - - self.temp_folder.downloadCounterBaskets-=1 - self._p_changed=1 - transaction.get().commit() - - self.REQUEST.RESPONSE.setHeader("Content-Disposition","""attachement; filename="%s.atf" """%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) - return True - - def numberOfItems(self): - """return anzahl der elemente im basket""" - return self.content.numberOfItems() - - def getTime(self): - """getTime""" - #return self.bobobase_modification_time().ISO() - - if hasattr(self,'time'): - return time.strftime("%Y-%m-%d %H:%M:%S",self.time) - elif hasattr(self,'timefixed'): - return self.timefixed - else: - setattr(self,'timefixed',self.bobobase_modification_time().ISO()) - return self.bobobase_modification_time().ISO() - - def getContent(self): - """get Basket Content""" - return self.content.getContent() - - - def __init__(self,id,user,comment="",basketContent=[]): - """ init a basket version""" - self.id=id - self.comment=comment - self._setObject('content',BasketContent(basketContent)) - #self.basketContent=basketContent[0:]a - self.user=user - self.time=time.localtime() - - def getUser(self): - """get user""" - return self.user - - def getComment(self): - """get Comment""" - return self.comment - - security.declareProtected('manage','index_html') - def index_html(self): - """view the basket""" - - if self.REQUEST.get('change',False): - ob=self.aq_parent.updateObjects(self.REQUEST['change']) - - self.REQUEST.RESPONSE.redirect(ob.absolute_url())#go to new basket, because changing generates a new basket - - pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','BasketVersionMain.zpt')).__of__(self) - return pt() - - def getObjUrl(self,result): - """getUrl of the version of the object""" - objId=result[1].getTitle() - founds=self.CDLICatalog.search({'title':objId}) - if len(founds)>0: - return founds[0].getObject().getLastVersion().absolute_url() - - else: #assume version number - splitted=objId.split("_") - founds=self.CDLICatalog.search({'title':splitted[1]}) - return founds[0].getObject().getLastVersion().absolute_url()+'/'+objId - -def manage_addCDLIBasketVersion(self,user,comment="",basketContent=[],RESPONSE=None): - """add a version""" - - #check for already existing versions - - lastVersion=self.getLastVersion() - if lastVersion is None: - newId=str(1) - else: - newId=str(int(lastVersion.getId())+1) - - ob=CDLIBasketVersion(newId,user,comment,basketContent) - - self._setObject(newId, ob) - - if RESPONSE is not None: - RESPONSE.redirect('manage_main') - else: - return ob - class CDLIFileObject(CatalogAware,extVersionedFileObject): """CDLI file object""" @@ -1522,8 +56,12 @@ class CDLIFileObject(CatalogAware,extVer """Return cataloguable key for ourselves.""" return str(self) + def setAuthor(self, author): + """change the author""" + self.author = author + def makeThisVersionCurrent_html(self): - """form for making this version current""" + """form for mthis version current""" pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','makeThisVersionCurrent.zpt')).__of__(self) return pt() @@ -1552,7 +90,7 @@ class CDLIFileObject(CatalogAware,extVer def getPNumber(self): """get the pnumber""" try: - txt=re.match("&[Pp](\d*)\s*=([^\r\n]*)",self.getData()[0:]) + txt=re.match("&[PpSs](\d*)\s*=([^\r\n]*)",self.getData()[0:]) except: txt=self.getData()[0:] @@ -1566,7 +104,7 @@ class CDLIFileObject(CatalogAware,extVer def getDesignation(self): """get the designation out of the file""" try: - txt=re.match("&[Pp](\d*)\s*=([^\r\n]*)",self.getData()[0:]) + txt=re.match("&[PpSs](\d*)\s*=([^\r\n]*)",self.getData()[0:]) except: txt=self.getData()[0:] @@ -1616,6 +154,7 @@ def manage_addCDLIFileObject(self,id,vC= logging.debug("reindex2: %s in %s"%(repr(fob), repr(fob.default_catalog))) fob.index_object() + self.CDLIRoot.updateOrAddToFileBTree(ob) if REQUEST is not None: REQUEST['RESPONSE'].redirect(self.absolute_url()+'/manage_main') @@ -1729,6 +268,9 @@ class CDLIFile(extVersionedFile,CatalogA except: pass + #self.cdliRoot.updateOrAddToFileBTree(self)# now update the object in the cache + + if RESPONSE: if ob.getSize()==0: self.REQUEST.SESSION['objID']=ob.getId() @@ -1764,10 +306,15 @@ def manage_addCDLIFile(self,id,title,loc def checkUTF8(data): """check utf 8""" + if not isinstance(data, str): + logging.error("checkUTF8 data is not string! (%s)"%repr(data)) + try: - data.encode('utf-8') + data.decode('utf-8') + logging.debug("checkUTF8: ok!") return True except: + logging.debug("checkUTF8: false!") return False @@ -1776,8 +323,8 @@ def checkFile(filename,data,folder): # first check the file name fn=filename.split(".") # no extension - if not fn[0][0]=="P": - return False,"P missing in the filename" + if not (fn[0][0]=="P" or fn[0][0]=="S"): + return False,"P/S missing in the filename" elif len(fn[0])!=7: return False,"P number has not the right length 6" elif not checkUTF8(data): @@ -1793,7 +340,7 @@ def splitatf(fh,dir=None,ext=None): i=0 #ROC: why split \n first and then \r??? - if (type(fh) is StringType) or (type(fh) is UnicodeType): + if isinstance(fh, basestring): iter=fh.split("\n") else: iter=fh.readlines() @@ -1801,7 +348,7 @@ def splitatf(fh,dir=None,ext=None): 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) + #logging.info("Deal with: %s"%line) if ext: i+=1 if (i%100)==0: @@ -1827,7 +374,7 @@ def splitatf(fh,dir=None,ext=None): if dir: filename=os.path.join(dir,filename) nf=file(filename,"w") - logging.info("open %s"%filename) + logging.debug("open %s"%filename) if nf: nf.write(line.replace("\n","")+"\n") @@ -1836,8 +383,9 @@ def splitatf(fh,dir=None,ext=None): except: pass - if not((type(fh) is StringType) or (type(fh) is UnicodeType)): + if not isinstance(fh, basestring): fh.close() + return ret,len(os.listdir(dir)) @@ -1931,10 +479,20 @@ class CDLIFileFolder(extVersionedFileFol return ret - def findObjectsFromList(self,enterList=None,display=False,start=None,upload=None,list=None,basketName=None,numberOfObjects=None,RESPONSE=None): + def expandFile(self,fileId,fileTree): + """wildcard in fileID suche alle Treffer""" + founds=self.CDLICatalog({'title':fileId}) + for found in founds: + fileTree.add(found.getId) + logging.debug("ADDD:"+found.getId) + + def findObjectsFromList(self,enterList=None,display=False,start=None,upload=None,list=None,basketName=None,numberOfObjects=None,RESPONSE=None,REQUEST=None,returnHash=False,hash=None): """findObjectsFromList (, TAB oder LINE separated)""" + logging.debug("start: findObjectsFromList") + #logging.debug("start: findObjectsFromList"+repr(list)) + if upload: # list from file upload txt=upload.read() @@ -1965,32 +523,83 @@ class CDLIFileFolder(extVersionedFileFol pt=getattr(self,'filelist.html') return pt(basketName=basketName,numberOfObjects=numberOfObjects) + + result =self.CDLICache.retrieve(hash) + if result: + logging.debug("give result from storage2") + return hash,result + if list is not None: # got already a list + + logging.debug(" ----List version") ret=[] + fileTree=Set() + for fileId in list: - if fileId.find("*"): #check for wildcards - fileId=fileId + + if fileId.find("*")>-1: #check for wildcards + self.expandFile(fileId,fileTree) + elif len(fileId.split("."))==1: fileId=fileId+".atf" + fileTree.add(fileId) + #logging.debug(" -----:"+fileId) + #ret+=self.CDLICatalog({'title':fileId}) + #x =self.getFileObject(fileId) + #if x is not None: + # ret.append(x) - ret+=self.CDLICatalog({'title':fileId}) + + + ids = fileTree & self.v_file_ids + #self.REQUEST.SESSION['fileIds']=ids#store fieldIds in session for further usage + l=makelist(fileTree)[0:] + #logging.debug("l-list:"+repr(l)) + self.REQUEST.SESSION['fileIds']=l#store fieldIds in session for further usage + self.REQUEST.SESSION['searchList']=l + #self.REQUEST.SESSION['searchList']=['P000001.atf'] + + + hash = md5.new(repr(makelist(fileTree))).hexdigest() # erzeuge hash als identification + self.REQUEST.SESSION['hash']=hash + #TODO: do I need garbage collection for v_tmpStore ? + + #logging.debug("Hash:"+repr(hash)) +# +# if hasattr(self.cdliRoot,'v_tmpStore') and self.cdliRoot.v_tmpStore.has_key(hash): +# logging.debug("asking for storage") +# res=self.cdliRoot.v_tmpStore[hash] +# if res: +# if returnHash == True: +# return hash,res +# return res + #TODO: get rid of one of these.. - ids=[x.getObject().getId() for x in ret] - self.REQUEST.SESSION['fileIds']=ids#store fieldIds in session for further usage - self.REQUEST.SESSION['searchList']=self.REQUEST.SESSION['fileIds'] + #ids=[x.getObject().getId() for x in ret] + ret=[(self.getFileObject(x),self.getFileObjectLastVersion(x)) for x in ids] + #self.REQUEST.SESSION['fileIds']=ids#store fieldIds in session for further usage + #self.REQUEST.SESSION['searchList']=self.REQUEST.SESSION['fileIds'] + if display: pt=getattr(self,'filelist.html') return pt(search=ids) - else: + else: + #self.REQUEST.SESSION['hash'] = ret # store in session + + #logging.debug("HHHHHHNEU:"+repr(self.makelist(ids))) + #logging.debug("HHHHHHNEU:"+repr(hash)) + self.CDLICache.store(hash,ret) + + if returnHash == True: + return hash,ret return ret if start: RESPONSE.redirect("filelist.html?start:int="+str(start)) - security.declareProtected('Manage','createAllFilesAsSingleFile') def createAllFilesAsSingleFile(self,RESPONSE=None): @@ -2118,7 +727,162 @@ class CDLIRoot(Folder): splitter = {'words':cdliSplitter.wordSplitter(), 'graphemes':cdliSplitter.graphemeSplitter()} + def redirect(self,RESPONSE,url): + """mache ein redirect mit einem angehaengten time stamp um ein reload zu erzwingen""" + + timeStamp=time.time() + + if url.find("?")>-1: #giebt es schon parameter + addStr="&time=%s" + else: + addStr="?time=%s" + + RESPONSE.setHeader('Last-Modified',email.Utils.formatdate().split("-")[0]+'GMT') + logging.error(email.Utils.formatdate()+' GMT') + RESPONSE.redirect(url+addStr%timeStamp) + + def unicodify(self,txt): + return unicodify(txt) + def invalidateOldCacheVersion(self): + """loescht die alte Version des Cache""" + del self.v_tmpStore + return "done" + + def viewATF(self,id,RESPONSE): + """view an Object""" + ob = self.CDLICatalog({'title':id}) + logging.debug(ob[0].getObject().getLastVersion().absolute_url()+"/view") + if len(ob)>0: + RESPONSE.redirect(ob[0].getObject().getLastVersion().absolute_url()+"/view") + return "not found" + + def history(self,id,RESPONSE): + """view an Object""" + ob = self.CDLICatalog({'title':id}) + if len(ob)>0: + RESPONSE.redirect(ob[0].absolute_url+"/history") + return "not found" + + + def downloadLocked(self,id,RESPONSE): + """view an Object""" + ob = self.CDLICatalog({'title':id}) + if len(ob)>0: + RESPONSE.redirect(ob[0].absolute_url+"/downloadLocked") + return "not found" + + def download(self,id,RESPONSE): + """view an Object""" + ob = self.CDLICatalog({'title':id}) + if len(ob)>0: + logging.info("objekt:"+repr(ob[0])) + #RESPONSE.redirect(ob[0].getLastVersion().absolute_url()) + RESPONSE.redirect(ob[0].absolute_url+"/download") + return "not found" + def addCDLIFileObjectForm(self,id,RESPONSE): + """view an Object""" + ob = self.CDLICatalog({'title':id}) + if len(ob)>0: + RESPONSE.redirect(ob[0].absolute_url+"/addCDLIFileObjectForm") + return "not found" + + def addVersionedFileObjectForm(self,id,RESPONSE): + """view an Object""" + ob = self.CDLICatalog({'title':id}) + if len(ob)>0: + RESPONSE.redirect(ob[0].absolute_url+"/addVersionedFileObjectForm") + return "not found" + + def unlock(self,id,RESPONSE): + """view an Object""" + ob = self.CDLICatalog({'title':id}) + if len(ob)>0: + RESPONSE.redirect(ob[0].absolute_url+"/unlock") + return "not found" + + + def getFileObject(self,fileId): + """get an object""" + logging.debug("getFileObj:"+repr(fileId)) + if isinstance(fileId,CDLIFileObject): # support for old baskets + return fileId + x=self.v_files.get(fileId) + logging.debug("obj: "+repr(x)) + if x is None: + logging.debug("fileId"+repr(fileId)) + folder=fileId[0:3] + f2=fileId[0:5] + fObj = getattr(self.cdliRoot.cdli_main,folder); + f2Obj = getattr(fObj,f2) + + o = getattr(f2Obj,fileId) + logging.debug(o); + self.updateOrAddToFileBTree(o) + return o + return x + + def getFileObjectLastVersion(self,fileId): + """get an object""" + x=self.v_files_lastVersion.get(fileId) + logging.debug("lastVersion: "+repr(x)) + if x==None: + folder=fileId[0:3] + f2=fileId[0:5] + fObj = getattr(self.cdliRoot.cdli_main,folder); + f2Obj = getattr(fObj,f2) + + o =getattr(f2Obj,fileId) + logging.debug(o); + return o.getLastVersion() + + + return x + + def showFileIds(self): + """showIds""" + return self.v_file_ids + + def generateFileBTree(self): + """erzeuge einen Btree aus allen Files""" + self.v_files = OOBTree() + self.v_files_lastVersion = OOBTree() + self.v_file_ids = Set() + + for x in self.CDLICatalog.searchResults(): + + self.v_files.update({x.getId:x.getObject()}) + self.v_files_lastVersion.update({x.getId:x.getObject().getLastVersion()}) + self.v_file_ids.add(x.getId) + logging.debug("add:"+x.getId+"XXX"+repr(x.getObject())) + + return True + + def updateOrAddToFileBTree(self,obj): + """update a BTree""" + self.v_files.update({obj.getId():obj}) + self.v_files_lastVersion.update({obj.getId():obj.getLastVersion()}) + + self.v_file_ids.add(obj.getId()) + #change everthing around to make it persistent... + tmp = self.v_files + self.v_files=tmp + + tmp2=self.v_file_ids + self.v_file_ids=tmp2 + + self.CDLICache.cleanCache() #be sure that the cache is clean + logging.debug("update:"+obj.getId()+"XXX"+repr(obj)) + + + def deleteFromBTree(self,objId): + """delete an obj""" + self.v_files.pop(objId) + self.v_files_lastVersion.pop(objId) + self.v_file_ids.remove(objId) + + + def deleteFiles(self,ids): """delete files""" for id in ids: @@ -2131,8 +895,9 @@ class CDLIRoot(Folder): - def searchText(self, query, index='graphemes'): - """searches query in the fulltext index and returns a list of file ids/P-numbers""" + def searchText(self, query, index='graphemes', resultFilter=None): + """searches query in the fulltext index and returns a list of file IDs/P-numbers + resultFilter is matched against the beginning of the file ID""" # 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 @@ -2142,8 +907,11 @@ class CDLIRoot(Folder): 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] + # put only the P-Number in the result + if resultFilter is None: + results = [res.getId[:7] for res in resultset] + else: + results = [res.getId[:7] for res in resultset if res.getId.startswith(resultFilter)] logging.debug("searchtext: found %d texts"%len(results)) return results @@ -2157,7 +925,6 @@ class CDLIRoot(Folder): return f[0].getObject().getData() - def showFile(self,fileId,wholePage=False): """show a file @param fileId: P-Number of the document to be displayed @@ -2244,7 +1011,7 @@ class CDLIRoot(Folder): if line.lstrip().startswith('#lem:'): continue # ignore p-num line - if line.startswith('&P'): + if line.startswith('&P') or line.startswith('&S'): continue # ignore version lines if line.startswith('#version'): @@ -2296,6 +1063,15 @@ class CDLIRoot(Folder): 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""" return urllib.quote(str) @@ -2312,19 +1088,68 @@ class CDLIRoot(Folder): """unquote url""" return urllib.unquote_plus(str) + def changeUserForPObjectFromFile(self,user,fname): + """aendere user fuer alle p in fiele""" + pns = file(os.path.join(package_home(globals()),'inputs',fname),"r") + for p in pns.readlines(): + p=p.lstrip().rstrip() + logging.info(str(p+".atf")) + pObj=self.getFileObject(p+".atf") + if pObj is not None: + logging.info(pObj) + lv=pObj.getContentObject() + logging.info("author:"+lv.author) + lv.author=user + lv.versionComment="XXXXXXX" - def forceunlock(self): + def forceunlock(self,REQUEST=None,user=None,fid=None): "break all locks" + if fid is not None: + self.getFileObject(fid).forceunlock() + return fid ret=[] + for f in self.ZopeFind(self,obj_metatypes="CDLI file",search_sub=1): - un=f[1].forceunlock() - + un=f[1].forceunlock(user=user) + logging.info("check:"+f[0]) if un and un !="": ret.append((f[0],un)) + if REQUEST is not None: + pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','breakLockResponse.zpt')).__of__(self) + + return pt(ret=ret) + return ret + def getLockedFiles(self,REQUEST=None,user=None): + """hole alle gesperrten files""" + ret={} + + for nm,f in self.v_files.items(): + lb = str(f.lockedBy) + add=False + if (lb is not None) and (lb!=""): + + if user is None: + add=True + else: + if str(lb)==user: + add=True + if add==True: + if not ret.has_key(lb): + ret[lb]=[] + ret[lb].append(nm) + + + if REQUEST is not None: + pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','showlockResponse.zpt')).__of__(self) + + return pt(ret=ret) + + return ret + def getChangesByAuthor(self,author,n=100): """getChangesByAuthor""" zcat=self.CDLIObjectsCatalog @@ -2368,6 +1193,11 @@ class CDLIRoot(Folder): return """ 2;url=%s?repeat=%s """%(self.absolute_url()+txt,threadName) + def refreshTxtBasket(self,txt="",threadName=None): + """txt fuer refresh""" + + return """ 2;url=%s?repeat=%s """%(txt,threadName) + def getResult(self,threadName=None): """result of thread""" @@ -2414,7 +1244,39 @@ class CDLIRoot(Folder): return generateXMLReturn(stObj.returnValue) + + def uploadATFAsync(self,repeat=None,upload=None,basketId=0,RESPONSE=None): + """upload an atf file / basket file""" + #self._v_uploadATF.returnValue=None + + + idTmp=str(randint(0,1000000000)) + + if upload is None: + return "ERROR NO FILE!" + + tmpFile = File("/tmp/idTmp","w") + # sicher basket in file + for x in upload.read(): + tmpFile.write(x) + + tmpFile.close(); + uploadATFAsync(baskerId,idTmp,basketId,self.REQUEST['AUTHENTICATED_USER'],idTmp,serverport=self.REQUEST['SERVER_PORT']) + + return idTMP + + def viewTicketAsync(self,ticketNr=1): + """viewticket""" + tmp = pickle.load(file("/tmp/"+str(ticketNr)+".result")) + + pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','uploadCheckAsync.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'], + basketNameFromId=tmp['basketNameFromId'],basketNameFromFile=tmp['basketNameFromFile'],basketId=tmp['basketId']) + + + def uploadATF(self,repeat=None,upload=None,basketId=0,RESPONSE=None): """upload an atf file / basket file""" #self._v_uploadATF.returnValue=None @@ -2486,6 +1348,7 @@ class CDLIRoot(Folder): return pt(changed=tmp['changed'],lockerrors=tmp['lockerrors'],errors=tmp['errors'],dir=tmp['dir'],newPs=tmp['newPs'],basketLen=tmp['basketLen'],numberOfFiles=tmp['numberOfFiles'], basketNameFromId=tmp['basketNameFromId'],basketNameFromFile=tmp['basketNameFromFile'],basketId=tmp['basketId']) + def redoUpload(self,threadName): """redo the upload""" tmp=self.cdli_main.tmpStore2[threadName] @@ -2495,9 +1358,6 @@ class CDLIRoot(Folder): def uploadATFfinally(self,procedure='',comment="",basketname='',unlock=None,repeat=None,RESPONSE=None): """nowupload the files""" - - - threadName=repeat if not threadName or threadName=="": thread=uploadATFfinallyThread() @@ -2544,7 +1404,24 @@ class CDLIRoot(Folder): pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','uploadATFWait.zpt')).__of__(self) return pt(txt='/uploadATFfinally',threadName=threadName) else: + + + idTmp=self.REQUEST.SESSION['idTmp'] + stObj=getattr(self.temp_folder,idTmp) self.REQUEST.SESSION['idTmp']=None + + #update changed + logging.debug("dir:"+repr(stObj.returnValue['changed'])) + for x in stObj.returnValue['changed']: + ob=self.CDLICatalog.search({'title':x[0]}) + try: + self.cdliRoot.updateOrAddToFileBTree(ob[0].getObject()) + except: + + logging.error("uploadATFfinally - cannot update Object %s Error: %s %s"%(ob[1],sys.exc_info()[0],sys.exc_info()[1])) + for x in stObj.returnValue['newPs']: + obj=self.getFileObject(x) #updates the object in the cache + logging.debug("Got:"+repr(obj)) if RESPONSE is not None: RESPONSE.redirect(self.absolute_url()) @@ -2559,12 +1436,19 @@ class CDLIRoot(Folder): for f in files: folder=f[0:3] f2=f[0:5] + + #check if main folder PXX already exists 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 type(ext.result) is types.FileType: + ext.result.write("

adding: %s

\n"%f) + else: + ext.result="

adding: %s

"%f+ext.result - if not obj: + + if not obj: # if not create it manage_addCDLIFileFolder(root,folder,folder) fobj=getattr(root,folder) #transaction.get().commit() @@ -2572,27 +1456,37 @@ class CDLIRoot(Folder): else: fobj=obj[0][1] + #check IF PYYYYY already exist obj2=fobj.ZopeFind(fobj,obj_ids=[f2]) logging.debug("importFiles: fobj=%s obj2=%s"%(fobj,obj2)) - if not obj2: + if not obj2:# if not create it manage_addCDLIFileFolder(fobj,f2,f2) fobj2=getattr(fobj,f2) else: fobj2=obj2[0][1] + # not add the file file2=os.path.join(folderName,f) id=f 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) + logging.debug("importfiles: fobj2.add") count+=1 - + + #now add the file to the storage + ob = getattr(fobj2,f) + logging.debug("importfiles: btree_start") + #self.cdliRoot.updateOrAddToFileBTree(ob) + logging.debug("importfiles: btree_end") if count%100==0: logging.debug("importfiles: committing") transaction.get().commit() + logging.debug("importfiles: committing") transaction.get().commit() + logging.debug("importfiles: committing done") return "ok" @@ -2631,3 +1525,37 @@ def manage_addCDLIRoot(self, id, title=' if REQUEST is not None: return self.manage_main(self, REQUEST, update_menu=1) + +import cdli_basket + + +# Die folgenden Klassen sollte nicht mehr aus diesem Paket benutzt werden, sondern direkt aus +# cdli_basket importiert werden. +class uploadATFfinallyThread(cdli_basket.uploadATFfinallyThread): + """depricates""" + pass + +class tmpStore(cdli_basket.tmpStore): + """depricated""" + pass + +class uploadATFThread(cdli_basket.uploadATFThread): + """depricated""" + pass + +class CDLIBasketContainer(cdli_basket.CDLIBasketContainer): + """depricated""" + pass + +class CDLIBasket(cdli_basket.CDLIBasket): + """depricated""" + pass + +class CDLIBasketVersion(cdli_basket.CDLIBasketVersion): + """depricated""" + pass + +class BasketContent(cdli_basket.BasketContent): + """depricated""" + pass +