--- documentViewer/documentViewer.py 2011/07/20 08:22:36 1.175.2.6 +++ documentViewer/documentViewer.py 2012/01/04 07:38:17 1.178 @@ -1,40 +1,56 @@ + from OFS.Folder import Folder from Products.PageTemplates.ZopePageTemplate import ZopePageTemplate from Products.PageTemplates.PageTemplateFile import PageTemplateFile from AccessControl import ClassSecurityInfo from AccessControl import getSecurityManager from Globals import package_home +from Products.zogiLib.zogiLib import browserCheck -#from Ft.Xml import EMPTY_NAMESPACE, Parse -#import Ft.Xml.Domlette - -import xml.etree.ElementTree as ET - +from Ft.Xml import EMPTY_NAMESPACE, Parse +import Ft.Xml.Domlette import os.path import sys import urllib +import urllib2 import logging import math import urlparse +import cStringIO import re import string -from SrvTxtUtils import getInt, getText, getHttpData - def logger(txt,method,txt2): """logging""" logging.info(txt+ txt2) +def getInt(number, default=0): + """returns always an int (0 in case of problems)""" + try: + return int(number) + except: + return int(default) + +def getTextFromNode(nodename): + """get the cdata content of a node""" + if nodename is None: + return "" + nodelist=nodename.childNodes + rc = "" + for node in nodelist: + if node.nodeType == node.TEXT_NODE: + rc = rc + node.data + return rc + def serializeNode(node, encoding="utf-8"): """returns a string containing node as XML""" - s = ET.tostring(node) - - # 4Suite: - # stream = cStringIO.StringIO() - # Ft.Xml.Domlette.Print(node, stream=stream, encoding=encoding) - # s = stream.getvalue() - # stream.close() + stream = cStringIO.StringIO() + #logging.debug("BUF: %s"%(stream)) + Ft.Xml.Domlette.Print(node, stream=stream, encoding=encoding) + s = stream.getvalue() + #logging.debug("BUF: %s"%(s)) + stream.close() return s def browserCheck(self): @@ -98,23 +114,54 @@ def browserCheck(self): return bt + def getParentDir(path): """returns pathname shortened by one""" return '/'.join(path.split('/')[0:-1]) -def getBibdataFromDom(dom): - """returns dict with all elements from bib-tag""" - bibinfo = {} - bib = dom.find(".//meta/bib") - if bib is not None: - # put type in @type - type = bib.get('type') - bibinfo['@type'] = type - # put all subelements in dict - for e in bib: - bibinfo[e.tag] = getText(e) - - return bibinfo + +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 ## ## documentViewer class @@ -243,7 +290,7 @@ class documentViewer(Folder): ''' logging.debug("HHHHHHHHHHHHHH:load the rss") - logging.debug("documentViewer (index) mode: %s url:%s start:%s pn:%s"%(mode,url,start,pn)) + logger("documentViewer (index)", logging.INFO, "mode: %s url:%s start:%s pn:%s"%(mode,url,start,pn)) if not hasattr(self, 'template'): # create template folder if it doesn't exist @@ -293,26 +340,21 @@ class documentViewer(Folder): if tocMode != "thumbs": # get table of contents docinfo = self.getToc(mode=tocMode, docinfo=docinfo) - - # auto viewMode: text_dict if text else images - if viewMode=="auto": - if docinfo.get('textURL', None) or docinfo.get('textURLPath', None): - #texturl gesetzt und textViewer konfiguriert + + if viewMode=="auto": # automodus gewaehlt + if docinfo.has_key('textURL') or docinfo.get('textURLPath',None): #texturl gesetzt und textViewer konfiguriert viewMode="text_dict" else: viewMode="images" - pageinfo = self.getPageinfo(start=start, current=pn, docinfo=docinfo, viewMode=viewMode, tocMode=tocMode) + pageinfo = self.getPageinfo(start=start,current=pn, docinfo=docinfo,viewMode=viewMode,tocMode=tocMode) - if viewMode != 'images' and docinfo.get('textURLPath', None): - # get full text page - page = self.getTextPage(mode=viewMode, pn=pn, docinfo=docinfo, pageinfo=pageinfo) + if (docinfo.get('textURLPath',None)): + page = self.getTextPage(docinfo=docinfo, pageinfo=pageinfo) pageinfo['textPage'] = page - - # get template /template/viewer_main - pt = getattr(self.template, 'viewer_main') - # and execute with parameters - return pt(docinfo=docinfo, pageinfo=pageinfo, viewMode=viewMode, mk=self.generateMarks(mk)) + tt = getattr(self, 'template') + pt = getattr(tt, 'viewer_main') + return pt(docinfo=docinfo,pageinfo=pageinfo,viewMode=viewMode,mk=self.generateMarks(mk)) def generateMarks(self,mk): ret="" @@ -443,7 +485,8 @@ class documentViewer(Folder): docinfo = {} for x in range(cut): - path=getParentDir(path) + + path=getParentDir(path) infoUrl=self.digilibBaseUrl+"/dirInfo-xml.jsp?mo=dir&fn="+path @@ -453,14 +496,12 @@ class documentViewer(Folder): if txt is None: raise IOError("Unable to get dir-info from %s"%(infoUrl)) - dom = ET.fromstring(txt) - #dom = Parse(txt) - size=getText(dom.find("size")) - #sizes=dom.xpath("//dir/size") - logging.debug("documentViewer (getparamfromdigilib) dirInfo:size=%s"%size) + dom = Parse(txt) + sizes=dom.xpath("//dir/size") + logging.debug("documentViewer (getparamfromdigilib) dirInfo:size"%sizes) - if size: - docinfo['numPages'] = int(size) + if sizes: + docinfo['numPages'] = int(getTextFromNode(sizes[0])) else: docinfo['numPages'] = 0 @@ -505,8 +546,7 @@ class documentViewer(Folder): if txt is None: raise IOError("Unable to read index meta from %s"%(url)) - dom = ET.fromstring(txt) - #dom = Parse(txt) + dom = Parse(txt) return dom def getPresentationInfoXML(self, url): @@ -525,8 +565,7 @@ class documentViewer(Folder): if txt is None: raise IOError("Unable to read infoXMLfrom %s"%(url)) - dom = ET.fromstring(txt) - #dom = Parse(txt) + dom = Parse(txt) return dom @@ -544,14 +583,11 @@ class documentViewer(Folder): path=getParentDir(path) dom = self.getDomFromIndexMeta(path) - acc = dom.find(".//access-conditions/access") - if acc is not None: - acctype = acc.get('type') - #acctype = dom.xpath("//access-conditions/access/@type") - if acctype: - access=acctype - if access in ['group', 'institution']: - access = dom.find(".//access-conditions/access/name").text.lower() + acctype = dom.xpath("//access-conditions/access/@type") + if acctype and (len(acctype)>0): + access=acctype[0].value + if access in ['group', 'institution']: + access = getTextFromNode(dom.xpath("//access-conditions/access/name")[0]).lower() docinfo['accessType'] = access return docinfo @@ -559,7 +595,7 @@ class documentViewer(Folder): def getBibinfoFromIndexMeta(self,path,docinfo=None,dom=None,cut=0): """gets bibliographical info from the index.meta file at path or given by dom""" - logging.debug("documentViewer (getbibinfofromindexmeta) path: %s"%(path)) + #logging.debug("documentViewer (getbibinfofromindexmeta) path: %s"%(path)) if docinfo is None: docinfo = {} @@ -571,85 +607,88 @@ class documentViewer(Folder): docinfo['indexMetaPath']=self.getIndexMetaPath(path); - logging.debug("documentViewer (getbibinfofromindexmeta cutted) path: %s"%(path)) - # put all raw bib fields in dict "bib" - bib = getBibdataFromDom(dom) - docinfo['bib'] = bib + #logging.debug("documentViewer (getbibinfofromindexmeta cutted) path: %s"%(path)) + # put in all raw bib fields as dict "bib" + bib = dom.xpath("//bib/*") + if bib and len(bib)>0: + bibinfo = {} + for e in bib: + bibinfo[e.localName] = getTextFromNode(e) + docinfo['bib'] = bibinfo # extract some fields (author, title, year) according to their mapping metaData=self.metadata.main.meta.bib - bibtype=bib.get("@type") - #bibtype=dom.xpath("//bib/@type") - if not bibtype: + bibtype=dom.xpath("//bib/@type") + if bibtype and (len(bibtype)>0): + bibtype=bibtype[0].value + else: bibtype="generic" - bibtype=bibtype.replace("-"," ") # wrong types in index meta "-" instead of " " (not wrong! ROC) + bibtype=bibtype.replace("-"," ") # wrong typesiin index meta "-" instead of " " (not wrong! ROC) docinfo['bib_type'] = bibtype bibmap=metaData.generateMappingForType(bibtype) - logging.debug("documentViewer (getbibinfofromindexmeta) bibmap:"+repr(bibmap)) - logging.debug("documentViewer (getbibinfofromindexmeta) bibtype:"+repr(bibtype)) + #logging.debug("documentViewer (getbibinfofromindexmeta) bibmap:"+repr(bibmap)) + #logging.debug("documentViewer (getbibinfofromindexmeta) bibtype:"+repr(bibtype)) # if there is no mapping bibmap is empty (mapping sometimes has empty fields) - if len(bibmap) > 0 and bibmap.get('author',None) or bibmap.get('title',None): + if len(bibmap) > 0 and len(bibmap['author'][0]) > 0: try: - docinfo['author']=bib.get(bibmap['author'][0]) + docinfo['author']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['author'][0])[0]) except: pass try: - docinfo['title']=bib.get(bibmap['title'][0]) + docinfo['title']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['title'][0])[0]) except: pass try: - docinfo['year']=bib.get(bibmap['year'][0]) + docinfo['year']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['year'][0])[0]) except: pass - - # ROC: why is this here? - # logging.debug("documentViewer (getbibinfofromindexmeta) using mapping for %s"%bibtype) - # try: - # docinfo['lang']=getTextFromNode(dom.find(".//bib/lang")[0]) - # except: - # docinfo['lang']='' - # try: - # docinfo['city']=getTextFromNode(dom.find(".//bib/city")[0]) - # except: - # docinfo['city']='' - # try: - # docinfo['number_of_pages']=getTextFromNode(dom.find(".//bib/number_of_pages")[0]) - # except: - # docinfo['number_of_pages']='' - # try: - # docinfo['series_volume']=getTextFromNode(dom.find(".//bib/series_volume")[0]) - # except: - # docinfo['series_volume']='' - # try: - # docinfo['number_of_volumes']=getTextFromNode(dom.find(".//bib/number_of_volumes")[0]) - # except: - # docinfo['number_of_volumes']='' - # try: - # docinfo['translator']=getTextFromNode(dom.find(".//bib/translator")[0]) - # except: - # docinfo['translator']='' - # try: - # docinfo['edition']=getTextFromNode(dom.find(".//bib/edition")[0]) - # except: - # docinfo['edition']='' - # try: - # docinfo['series_author']=getTextFromNode(dom.find(".//bib/series_author")[0]) - # except: - # docinfo['series_author']='' - # try: - # docinfo['publisher']=getTextFromNode(dom.find(".//bib/publisher")[0]) - # except: - # docinfo['publisher']='' - # try: - # docinfo['series_title']=getTextFromNode(dom.find(".//bib/series_title")[0]) - # except: - # docinfo['series_title']='' - # try: - # docinfo['isbn_issn']=getTextFromNode(dom.find(".//bib/isbn_issn")[0]) - # except: - # docinfo['isbn_issn']='' + #logging.debug("documentViewer (getbibinfofromindexmeta) using mapping for %s"%bibtype) + try: + docinfo['lang']=getTextFromNode(dom.xpath("//bib/lang")[0]) + except: + docinfo['lang']='' + try: + docinfo['city']=getTextFromNode(dom.xpath("//bib/city")[0]) + except: + docinfo['city']='' + try: + docinfo['number_of_pages']=getTextFromNode(dom.xpath("//bib/number_of_pages")[0]) + except: + docinfo['number_of_pages']='' + try: + docinfo['series_volume']=getTextFromNode(dom.xpath("//bib/series_volume")[0]) + except: + docinfo['series_volume']='' + try: + docinfo['number_of_volumes']=getTextFromNode(dom.xpath("//bib/number_of_volumes")[0]) + except: + docinfo['number_of_volumes']='' + try: + docinfo['translator']=getTextFromNode(dom.xpath("//bib/translator")[0]) + except: + docinfo['translator']='' + try: + docinfo['edition']=getTextFromNode(dom.xpath("//bib/edition")[0]) + except: + docinfo['edition']='' + try: + docinfo['series_author']=getTextFromNode(dom.xpath("//bib/series_author")[0]) + except: + docinfo['series_author']='' + try: + docinfo['publisher']=getTextFromNode(dom.xpath("//bib/publisher")[0]) + except: + docinfo['publisher']='' + try: + docinfo['series_title']=getTextFromNode(dom.xpath("//bib/series_title")[0]) + except: + docinfo['series_title']='' + try: + docinfo['isbn_issn']=getTextFromNode(dom.xpath("//bib/isbn_issn")[0]) + except: + docinfo['isbn_issn']='' + #logging.debug("I NEED BIBTEX %s"%docinfo) return docinfo - # TODO: is this needed? def getNameFromIndexMeta(self,path,docinfo=None,dom=None,cut=0): """gets name info from the index.meta file at path or given by dom""" if docinfo is None: @@ -660,7 +699,7 @@ class documentViewer(Folder): path=getParentDir(path) dom = self.getDomFromIndexMeta(path) - docinfo['name']=getText(dom.find("name")) + docinfo['name']=getTextFromNode(dom.xpath("/resource/name")[0]) logging.debug("documentViewer docinfo[name] %s"%docinfo['name']) return docinfo @@ -677,12 +716,15 @@ class documentViewer(Folder): archivePath = None archiveName = None - archiveName = getText(dom.find("name")) - if not archiveName: + archiveNames = dom.xpath("//resource/name") + if archiveNames and (len(archiveNames) > 0): + archiveName = getTextFromNode(archiveNames[0]) + else: logging.warning("documentViewer (getdocinfofromtexttool) resource/name missing in: %s" % (url)) - archivePath = getText(dom.find("archive-path")) - if archivePath: + archivePaths = dom.xpath("//resource/archive-path") + if archivePaths and (len(archivePaths) > 0): + archivePath = getTextFromNode(archivePaths[0]) # clean up archive path if archivePath[0] != '/': archivePath = '/' + archivePath @@ -698,9 +740,11 @@ class documentViewer(Folder): # we balk without archive-path raise IOError("Missing archive-path (for text-tool) in %s" % (url)) - imageDir = getText(dom.find(".//texttool/image")) + imageDirs = dom.xpath("//texttool/image") + if imageDirs and (len(imageDirs) > 0): + imageDir = getTextFromNode(imageDirs[0]) - if not imageDir: + else: # we balk with no image tag / not necessary anymore because textmode is now standard #raise IOError("No text-tool info in %s"%(url)) imageDir = "" @@ -717,13 +761,15 @@ class documentViewer(Folder): docinfo['imageURL'] = self.digilibBaseUrl + "/servlet/Scaler?fn=" + imageDir - viewerUrl = getText(dom.find(".//texttool/digiliburlprefix")) - if viewerUrl: + viewerUrls = dom.xpath("//texttool/digiliburlprefix") + if viewerUrls and (len(viewerUrls) > 0): + viewerUrl = getTextFromNode(viewerUrls[0]) docinfo['viewerURL'] = viewerUrl # old style text URL - textUrl = getText(dom.find(".//texttool/text")) - if textUrl: + textUrls = dom.xpath("//texttool/text") + if textUrls and (len(textUrls) > 0): + textUrl = getTextFromNode(textUrls[0]) if urlparse.urlparse(textUrl)[0] == "": #keine url textUrl = os.path.join(archivePath, textUrl) # fix URLs starting with /mpiwg/online @@ -731,10 +777,24 @@ class documentViewer(Folder): textUrl = textUrl.replace("/mpiwg/online", '', 1) docinfo['textURL'] = textUrl - + + + #TODO: hack-DW for annalen + if (textUrl is not None) and (textUrl.startswith("/permanent/einstein/annalen")): + textUrl=textUrl.replace("/permanent/einstein/annalen/","/diverse/de/") + splitted=textUrl.split("/fulltext") + textUrl=splitted[0]+".xml" + textUrlkurz = string.split(textUrl, ".")[0] + docinfo['textURLPathkurz'] = textUrlkurz + docinfo['textURLPath'] = textUrl + logging.debug("hack") + logging.debug(textUrl) + + # new style text-url-path - textUrl = getText(dom.find(".//texttool/text-url-path")) - if textUrl: + textUrls = dom.xpath("//texttool/text-url-path") + if textUrls and (len(textUrls) > 0): + textUrl = getTextFromNode(textUrls[0]) docinfo['textURLPath'] = textUrl textUrlkurz = string.split(textUrl, ".")[0] docinfo['textURLPathkurz'] = textUrlkurz @@ -743,16 +803,16 @@ class documentViewer(Folder): #docinfo = self.getNumTextPages(docinfo) - presentationUrl = getText(dom.find(".//texttool/presentation")) + presentationUrls = dom.xpath("//texttool/presentation") docinfo = self.getBibinfoFromIndexMeta(url, docinfo=docinfo, dom=dom) # get info von bib tag - # TODO: is this needed here? + #docinfo = self.getDownloadfromDocinfoToBibtex(url, docinfo=docinfo, dom=dom) docinfo = self.getNameFromIndexMeta(url, docinfo=docinfo, dom=dom) - if presentationUrl: # ueberschreibe diese durch presentation informationen + if presentationUrls and (len(presentationUrls) > 0): # ueberschreibe diese durch presentation informationen # presentation url ergiebt sich ersetzen von index.meta in der url der fuer die Metadaten # durch den relativen Pfad auf die presentation infos - presentationPath = presentationUrl + presentationPath = getTextFromNode(presentationUrls[0]) if url.endswith("index.meta"): presentationUrl = url.replace('index.meta', presentationPath) else: @@ -769,9 +829,18 @@ class documentViewer(Folder): """gets the bibliographical information from the preseantion entry in texttools """ dom=self.getPresentationInfoXML(url) - docinfo['author']=getText(dom.find(".//author")) - docinfo['title']=getText(dom.find(".//title")) - docinfo['year']=getText(dom.find(".//date")) + try: + docinfo['author']=getTextFromNode(dom.xpath("//author")[0]) + except: + pass + try: + docinfo['title']=getTextFromNode(dom.xpath("//title")[0]) + except: + pass + try: + docinfo['year']=getTextFromNode(dom.xpath("//date")[0]) + except: + pass return docinfo def getDocinfoFromImagePath(self,path,docinfo=None,cut=0): @@ -792,6 +861,7 @@ class documentViewer(Folder): #path ist the path to the images it assumes that the index.meta file is one level higher. docinfo = self.getBibinfoFromIndexMeta(pathorig,docinfo=docinfo,cut=cut+1) + #docinfo = self.getDownloadfromDocinfoToBibtex(pathorig,docinfo=docinfo,cut=cut+1) docinfo = self.getAuthinfoFromIndexMeta(pathorig,docinfo=docinfo,cut=cut+1) return docinfo @@ -804,21 +874,15 @@ class documentViewer(Folder): 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. keys=%s"%docinfo.keys()) + logging.debug("documentViewer (getdocinfo) docinfo in session: %s"%docinfo) return docinfo - # new docinfo docinfo = {'mode': mode, 'url': url} - # add self url - docinfo['viewerUrl'] = self.getDocumentViewerURL() - if mode=="texttool": - # index.meta with texttool information + if mode=="texttool": #index.meta with texttool information docinfo = self.getDocinfoFromTextTool(url, docinfo=docinfo) elif mode=="imagepath": - # folder with images, index.meta optional docinfo = self.getDocinfoFromImagePath(url, docinfo=docinfo) elif mode=="filepath": - # filename docinfo = self.getDocinfoFromImagePath(url, docinfo=docinfo,cut=1) else: logging.error("documentViewer (getdocinfo) unknown mode: %s!"%mode) @@ -828,8 +892,8 @@ class documentViewer(Folder): if not docinfo.has_key('textURLPath'): docinfo['textURLPath'] = None - logging.debug("documentViewer (getdocinfo) docinfo: keys=%s"%docinfo.keys()) - #logging.debug("documentViewer (getdocinfo) docinfo: %s"%docinfo) + logging.debug("documentViewer (getdocinfo) docinfo: %s"%docinfo) + #logging.debug("documentViewer (getdocinfo) docinfo: %s"%) self.REQUEST.SESSION['docinfo'] = docinfo return docinfo @@ -864,6 +928,12 @@ class documentViewer(Folder): pageinfo['querySearch'] =self.REQUEST.get('querySearch', 'fulltext') pageinfo['textPN'] = self.REQUEST.get('textPN','1') pageinfo['highlightQuery'] = self.REQUEST.get('highlightQuery','') + + pageinfo ['highlightElementPos'] = self.REQUEST.get('highlightElementPos','') + pageinfo ['highlightElement'] = self.REQUEST.get('highlightElement','') + + pageinfo ['xpointer'] = self.REQUEST.get('xpointer','') + pageinfo['tocPageSize'] = self.REQUEST.get('tocPageSize', '30') pageinfo['queryPageSize'] =self.REQUEST.get('queryPageSize', '10') pageinfo['tocPN'] = self.REQUEST.get('tocPN', '1') @@ -880,7 +950,8 @@ class documentViewer(Folder): tocPages=tocSize/tocPageSize pageinfo['tocPN'] = min (tocPages,toc) pageinfo['searchPN'] =self.REQUEST.get('searchPN','1') - pageinfo['sn'] =self.REQUEST.get('sn','') + #pageinfo['sn'] =self.REQUEST.get('sn','') + pageinfo['s'] =self.REQUEST.get('s','') return pageinfo def changeDocumentViewer(self,title="",digilibBaseUrl=None,thumbrows=2,thumbcols=5,authgroups='mpiwg',RESPONSE=None):