--- documentViewer/documentViewer.py 2010/06/17 17:35:24 1.65.2.1 +++ documentViewer/documentViewer.py 2010/06/14 10:49:48 1.69.2.1 @@ -2,20 +2,32 @@ from OFS.Folder import Folder from Products.PageTemplates.ZopePageTemplate import ZopePageTemplate from Products.PageTemplates.PageTemplateFile import PageTemplateFile +from Products.PythonScripts.standard import url_quote from AccessControl import ClassSecurityInfo from AccessControl import getSecurityManager from Globals import package_home -from Ft.Xml import EMPTY_NAMESPACE, Parse +from Ft.Xml.Domlette import NonvalidatingReader from Ft.Xml.Domlette import PrettyPrint, Print +from Ft.Xml import EMPTY_NAMESPACE, Parse + +from xml.dom.minidom import parse, parseString + +from extraFunction import * + + +import Ft.Xml.XPath +import cStringIO +import xmlrpclib import os.path import sys +import cgi import urllib -import urllib2 import logging import math + import urlparse -import cStringIO +from types import * def logger(txt,method,txt2): """logging""" @@ -54,56 +66,23 @@ def getParentDir(path): return '/'.join(path.split('/')[0:-1]) -def getHttpData(url, data=None, num_tries=3, timeout=10): - """returns result from url+data HTTP request""" - # we do GET (by appending data to url) - if isinstance(data, str) or isinstance(data, unicode): - # if data is string then append - url = "%s?%s"%(url,data) - elif isinstance(data, dict) or isinstance(data, list) or isinstance(data, tuple): - # urlencode - url = "%s?%s"%(url,urllib.urlencode(data)) - - response = None - errmsg = None - for cnt in range(num_tries): - try: - logging.debug("getHttpData(#%s %ss) url=%s"%(cnt+1,timeout,url)) - if sys.version_info < (2, 6): - # set timeout on socket -- ugly :-( - import socket - socket.setdefaulttimeout(float(timeout)) - response = urllib2.urlopen(url) - else: - response = urllib2.urlopen(url,timeout=float(timeout)) - # check result? - break - except urllib2.HTTPError, e: - logging.error("getHttpData: HTTP error(%s): %s"%(e.code,e)) - errmsg = str(e) - # stop trying - break - except urllib2.URLError, e: - logging.error("getHttpData: URLLIB error(%s): %s"%(e.reason,e)) - errmsg = str(e) - # stop trying - #break - - if response is not None: - data = response.read() - response.close() - return data - - raise IOError("ERROR fetching HTTP data from %s: %s"%(url,errmsg)) - #return None +import socket +def urlopen(url,timeout=2): + """urlopen mit timeout""" + socket.setdefaulttimeout(timeout) + ret=urllib.urlopen(url) + socket.setdefaulttimeout(5) + return ret ## ## documentViewer class ## -class documentViewer(Folder): +class documentViewer(Folder, extraFunction): """document viewer""" + #textViewerUrl="http://127.0.0.1:8080/HFQP/testXSLT/getPage?" + meta_type="Document viewer" security=ClassSecurityInfo() @@ -144,12 +123,12 @@ class documentViewer(Folder): #self['template'] = templateFolder # Zope-2.12 style self._setObject('template',templateFolder) # old style try: - import MpdlXmlTextServer - textServer = MpdlXmlTextServer.MpdlXmlTextServer(id='fulltextclient',serverName=textServerName) + from Products.XMLRpcTools.XMLRpcTools import XMLRpcServerProxy + xmlRpcClient = XMLRpcServerProxy(id='fulltextclient', serverUrl=textServerName, use_xmlrpc=False) #templateFolder['fulltextclient'] = xmlRpcClient - templateFolder._setObject('fulltextclient',textServer) + templateFolder._setObject('fulltextclient',xmlRpcClient) except Exception, e: - logging.error("Unable to create MpdlXmlTextServer for fulltextclient: "+str(e)) + logging.error("Unable to create XMLRpcTools for fulltextclient: "+str(e)) try: from Products.zogiLib.zogiLib import zogiLib zogilib = zogiLib(id="zogilib", title="zogilib for docuviewer", dlServerURL=imageScalerUrl, layout="book") @@ -158,41 +137,7 @@ class documentViewer(Folder): except Exception, e: logging.error("Unable to create zogiLib for zogilib: "+str(e)) - - # proxy text server methods to fulltextclient - def getTextPage(self, **args): - """get page""" - return self.template.fulltextclient.getTextPage(**args) - - def getQuery(self, **args): - """get query""" - return self.template.fulltextclient.getQuery(**args) - - def getSearch(self, **args): - """get search""" - return self.template.fulltextclient.getSearch(**args) - - def getNumPages(self, **args): - """get numpages""" - return self.template.fulltextclient.getNumPages(**args) - - def getTranslate(self, **args): - """get translate""" - return self.template.fulltextclient.getTranslate(**args) - - def getLemma(self, **args): - """get lemma""" - return self.template.fulltextclient.getLemma(**args) - - def getToc(self, **args): - """get toc""" - return self.template.fulltextclient.getToc(**args) - - def getTocPage(self, **args): - """get tocpage""" - return self.template.fulltextclient.getTocPage(**args) - security.declareProtected('View','thumbs_rss') def thumbs_rss(self,mode,url,viewMode="auto",start=None,pn=1): ''' @@ -217,7 +162,7 @@ class documentViewer(Folder): pt = getattr(self.template, 'thumbs_main_rss') if viewMode=="auto": # automodus gewaehlt - if docinfo.has_key("textURL") or docinfo.has_key('textURLPath'): #texturl gesetzt und textViewer konfiguriert + if docinfo.get("textURL",'') and self.textViewerUrl: #texturl gesetzt und textViewer konfiguriert viewMode="text" else: viewMode="images" @@ -243,16 +188,17 @@ class documentViewer(Folder): return "ERROR: template folder missing!" if not getattr(self, 'digilibBaseUrl', None): - self.digilibBaseUrl = self.findDigilibUrl() or "http://digilib.mpiwg-berlin.mpg.de/digitallibrary" + self.digilibBaseUrl = self.findDigilibUrl() or "http://nausikaa.mpiwg-berlin.mpg.de/digitallibrary" docinfo = self.getDocinfo(mode=mode,url=url) + if tocMode != "thumbs": # get table of contents docinfo = self.getToc(mode=tocMode, docinfo=docinfo) if viewMode=="auto": # automodus gewaehlt - if docinfo.has_key('textURL') or docinfo.has_key('textURLPath'): #texturl gesetzt und textViewer konfiguriert + if docinfo.get("textURL",''): #texturl gesetzt und textViewer konfiguriert viewMode="text_dict" else: viewMode="images" @@ -320,7 +266,7 @@ class documentViewer(Folder): params[param] = str(val) # quote values and assemble into query string - logging.debug("XYXXXXX: %s"%repr(params.items())) + logging.info("XYXXXXX: %s"%repr(params.items())) ps = "&".join(["%s=%s"%(k,urllib.quote(v)) for (k, v) in params.items()]) url=self.REQUEST['URL1']+"?"+ps return url @@ -339,9 +285,9 @@ class documentViewer(Folder): def isAccessible(self, docinfo): """returns if access to the resource is granted""" access = docinfo.get('accessType', None) - logging.debug("documentViewer (accessOK) access type %s"%access) + logger("documentViewer (accessOK)", logging.INFO, "access type %s"%access) if access is not None and access == 'free': - logging.debug("documentViewer (accessOK) access is free") + logger("documentViewer (accessOK)", logging.INFO, "access is free") return True elif access is None or access in self.authgroups: # only local access -- only logged in users @@ -352,12 +298,13 @@ class documentViewer(Folder): else: return False - logging.debug("documentViewer (accessOK) unknown access type %s"%access) + logger("documentViewer (accessOK)", logging.INFO, "unknown access type %s"%access) return False def getDirinfoFromDigilib(self,path,docinfo=None,cut=0): """gibt param von dlInfo aus""" + num_retries = 3 if docinfo is None: docinfo = {} @@ -367,15 +314,21 @@ class documentViewer(Folder): infoUrl=self.digilibBaseUrl+"/dirInfo-xml.jsp?mo=dir&fn="+path - logging.debug("documentViewer (getparamfromdigilib) dirInfo from %s"%(infoUrl)) + logger("documentViewer (getparamfromdigilib)", logging.INFO, "dirInfo from %s"%(infoUrl)) - txt = getHttpData(infoUrl) - if txt is None: + for cnt in range(num_retries): + try: + # dom = NonvalidatingReader.parseUri(imageUrl) + txt=urllib.urlopen(infoUrl).read() + dom = Parse(txt) + break + except: + logger("documentViewer (getdirinfofromdigilib)", logging.ERROR, "error reading %s (try %d)"%(infoUrl,cnt)) + else: raise IOError("Unable to get dir-info from %s"%(infoUrl)) - - dom = Parse(txt) + sizes=dom.xpath("//dir/size") - logging.debug("documentViewer (getparamfromdigilib) dirInfo:size"%sizes) + logger("documentViewer (getparamfromdigilib)", logging.INFO, "dirInfo:size"%sizes) if sizes: docinfo['numPages'] = int(getTextFromNode(sizes[0])) @@ -389,6 +342,7 @@ class documentViewer(Folder): def getIndexMeta(self, url): """returns dom of index.meta document at url""" + num_retries = 3 dom = None metaUrl = None if url.startswith("http://"): @@ -400,17 +354,25 @@ class documentViewer(Folder): metaUrl=server+url.replace("/mpiwg/online","") if not metaUrl.endswith("index.meta"): metaUrl += "/index.meta" + logging.debug("METAURL: %s"%metaUrl) + for cnt in range(num_retries): + try: + # patch dirk encoding fehler treten dann nicht mehr auf + # dom = NonvalidatingReader.parseUri(metaUrl) + txt=urllib.urlopen(metaUrl).read() + dom = Parse(txt) + break + except: + logger("ERROR documentViewer (getIndexMeta)", logging.INFO,"%s (%s)"%sys.exc_info()[0:2]) - logging.debug("(getIndexMeta): METAURL: %s"%metaUrl) - txt=getHttpData(metaUrl) - if txt is None: + if dom is None: raise IOError("Unable to read index meta from %s"%(url)) - - dom = Parse(txt) + return dom def getPresentationInfoXML(self, url): """returns dom of info.xml document at url""" + num_retries = 3 dom = None metaUrl = None if url.startswith("http://"): @@ -421,17 +383,25 @@ class documentViewer(Folder): server=self.digilibBaseUrl+"/servlet/Texter?fn=" metaUrl=server+url.replace("/mpiwg/online","") - txt=getHttpData(metaUrl) - if txt is None: + for cnt in range(num_retries): + try: + # patch dirk encoding fehler treten dann nicht mehr auf + # dom = NonvalidatingReader.parseUri(metaUrl) + txt=urllib.urlopen(metaUrl).read() + dom = Parse(txt) + break + except: + logger("ERROR documentViewer (getPresentationInfoXML)", logging.INFO,"%s (%s)"%sys.exc_info()[0:2]) + + if dom is None: raise IOError("Unable to read infoXMLfrom %s"%(url)) - - dom = Parse(txt) + return dom def getAuthinfoFromIndexMeta(self,path,docinfo=None,dom=None,cut=0): """gets authorization info from the index.meta file at path or given by dom""" - logging.debug("documentViewer (getauthinfofromindexmeta) path: %s"%(path)) + logger("documentViewer (getauthinfofromindexmeta)", logging.INFO,"path: %s"%(path)) access = None @@ -507,7 +477,7 @@ class documentViewer(Folder): def getDocinfoFromTextTool(self, url, dom=None, docinfo=None): """parse texttool tag in index meta""" - logging.debug("documentViewer (getdocinfofromtexttool) url: %s" % (url)) + logger("documentViewer (getdocinfofromtexttool)", logging.INFO, "url: %s" % (url)) if docinfo is None: docinfo = {} if docinfo.get('lang', None) is None: @@ -522,7 +492,7 @@ class documentViewer(Folder): if archiveNames and (len(archiveNames) > 0): archiveName = getTextFromNode(archiveNames[0]) else: - logging.warning("documentViewer (getdocinfofromtexttool) resource/name missing in: %s" % (url)) + logger("documentViewer (getdocinfofromtexttool)", logging.WARNING, "resource/name missing in: %s" % (url)) archivePaths = dom.xpath("//resource/archive-path") if archivePaths and (len(archivePaths) > 0): @@ -534,7 +504,7 @@ class documentViewer(Folder): archivePath += "/" + archiveName else: # try to get archive-path from url - logging.warning("documentViewer (getdocinfofromtexttool) resource/archive-path missing in: %s" % (url)) + logger("documentViewer (getdocinfofromtexttool)", logging.WARNING, "resource/archive-path missing in: %s" % (url)) if (not url.startswith('http')): archivePath = url.replace('index.meta', '') @@ -567,8 +537,7 @@ class documentViewer(Folder): if viewerUrls and (len(viewerUrls) > 0): viewerUrl = getTextFromNode(viewerUrls[0]) docinfo['viewerURL'] = viewerUrl - - # old style text URL + textUrls = dom.xpath("//texttool/text") if textUrls and (len(textUrls) > 0): textUrl = getTextFromNode(textUrls[0]) @@ -580,7 +549,6 @@ class documentViewer(Folder): docinfo['textURL'] = textUrl - # new style text-url-path textUrls = dom.xpath("//texttool/text-url-path") if textUrls and (len(textUrls) > 0): textUrl = getTextFromNode(textUrls[0]) @@ -628,7 +596,7 @@ class documentViewer(Folder): def getDocinfoFromImagePath(self,path,docinfo=None,cut=0): """path ist the path to the images it assumes that the index.meta file is one level higher.""" - logging.debug("documentViewer (getdocinfofromimagepath) path: %s"%(path)) + logger("documentViewer (getdocinfofromimagepath)", logging.INFO,"path: %s"%(path)) if docinfo is None: docinfo = {} path=path.replace("/mpiwg/online","") @@ -638,7 +606,7 @@ class documentViewer(Folder): pathorig=path for x in range(cut): path=getParentDir(path) - logging.debug("documentViewer (getdocinfofromimagepath) PATH:"+path) + logging.error("PATH:"+path) imageUrl=self.digilibBaseUrl+"/servlet/Scaler?fn="+path docinfo['imageURL'] = imageUrl @@ -650,13 +618,13 @@ class documentViewer(Folder): def getDocinfo(self, mode, url): """returns docinfo depending on mode""" - logging.debug("documentViewer (getdocinfo) mode: %s, url: %s"%(mode,url)) + logger("documentViewer (getdocinfo)", logging.INFO,"mode: %s, url: %s"%(mode,url)) # look for cached docinfo in session if self.REQUEST.SESSION.has_key('docinfo'): docinfo = self.REQUEST.SESSION['docinfo'] # check if its still current if docinfo is not None and docinfo.get('mode') == mode and docinfo.get('url') == url: - logging.debug("documentViewer (getdocinfo) docinfo in session: %s"%docinfo) + logger("documentViewer (getdocinfo)", logging.INFO,"docinfo in session: %s"%docinfo) return docinfo # new docinfo docinfo = {'mode': mode, 'url': url} @@ -667,10 +635,10 @@ class documentViewer(Folder): elif mode=="filepath": docinfo = self.getDocinfoFromImagePath(url, docinfo=docinfo,cut=1) else: - logging.error("documentViewer (getdocinfo) unknown mode: %s!"%mode) + logger("documentViewer (getdocinfo)", logging.ERROR,"unknown mode!") raise ValueError("Unknown mode %s! Has to be one of 'texttool','imagepath','filepath'."%(mode)) - logging.debug("documentViewer (getdocinfo) docinfo: %s"%docinfo) + logger("documentViewer (getdocinfo)", logging.INFO,"docinfo: %s"%docinfo) self.REQUEST.SESSION['docinfo'] = docinfo return docinfo