--- documentViewer/documentViewer.py 2011/06/14 09:57:11 1.175 +++ documentViewer/documentViewer.py 2011/07/28 13:00:07 1.175.2.10 @@ -1,56 +1,40 @@ - 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 +#from Ft.Xml import EMPTY_NAMESPACE, Parse +#import Ft.Xml.Domlette + +import xml.etree.ElementTree as ET + 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""" - 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() + s = ET.tostring(node) + + # 4Suite: + # stream = cStringIO.StringIO() + # Ft.Xml.Domlette.Print(node, stream=stream, encoding=encoding) + # s = stream.getvalue() + # stream.close() return s def browserCheck(self): @@ -114,55 +98,11 @@ def browserCheck(self): return bt - def getParentDir(path): """returns pathname shortened by one""" 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 - ## ## documentViewer class ## @@ -174,6 +114,9 @@ class documentViewer(Folder): manage_options=Folder.manage_options+( {'label':'main config','action':'changeDocumentViewerForm'}, ) + + metadataService = None + """MetaDataFolder instance""" # templates and forms viewer_main = PageTemplateFile('zpt/viewer_main', globals()) @@ -193,8 +136,6 @@ class documentViewer(Folder): thumbs_main_rss = PageTemplateFile('zpt/thumbs_main_rss', globals()) - security.declareProtected('View management screens','changeDocumentViewerForm') - changeDocumentViewerForm = PageTemplateFile('zpt/changeDocumentViewer', globals()) def __init__(self,id,imageScalerUrl=None,textServerName=None,title="",digilibBaseUrl=None,thumbcols=2,thumbrows=5,authgroups="mpiwg"): @@ -217,6 +158,7 @@ class documentViewer(Folder): templateFolder._setObject('fulltextclient',textServer) except Exception, e: logging.error("Unable to create MpdlXmlTextServer for fulltextclient: "+str(e)) + try: from Products.zogiLib.zogiLib import zogiLib zogilib = zogiLib(id="zogilib", title="zogilib for docuviewer", dlServerURL=imageScalerUrl, layout="book") @@ -224,7 +166,13 @@ class documentViewer(Folder): templateFolder._setObject('zogilib',zogilib) except Exception, e: logging.error("Unable to create zogiLib for zogilib: "+str(e)) - + + try: + # assume MetaDataFolder instance is called metadata + self.metadataService = getattr(self, 'metadata') + except Exception, e: + logging.error("Unable to find MetaDataFolder 'metadata': "+str(e)) + # proxy text server methods to fulltextclient def getTextPage(self, **args): @@ -290,7 +238,7 @@ class documentViewer(Folder): ''' logging.debug("HHHHHHHHHHHHHH:load the rss") - logger("documentViewer (index)", logging.INFO, "mode: %s url:%s start:%s pn:%s"%(mode,url,start,pn)) + logging.debug("documentViewer (index) 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 @@ -340,21 +288,26 @@ class documentViewer(Folder): 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.get('textURLPath',None): #texturl gesetzt und textViewer konfiguriert + + # 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 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 (docinfo.get('textURLPath',None)): - page = self.getTextPage(docinfo=docinfo, pageinfo=pageinfo) + if viewMode != 'images' and docinfo.get('textURLPath', None): + # get full text page + page = self.getTextPage(mode=viewMode, pn=pn, docinfo=docinfo, pageinfo=pageinfo) pageinfo['textPage'] = page - tt = getattr(self, 'template') - pt = getattr(tt, 'viewer_main') - return pt(docinfo=docinfo,pageinfo=pageinfo,viewMode=viewMode,mk=self.generateMarks(mk)) + + # 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)) def generateMarks(self,mk): ret="" @@ -485,8 +438,7 @@ 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 @@ -496,12 +448,14 @@ class documentViewer(Folder): if txt is None: 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) + 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) - if sizes: - docinfo['numPages'] = int(getTextFromNode(sizes[0])) + if size: + docinfo['numPages'] = int(size) else: docinfo['numPages'] = 0 @@ -546,7 +500,8 @@ class documentViewer(Folder): if txt is None: raise IOError("Unable to read index meta from %s"%(url)) - dom = Parse(txt) + dom = ET.fromstring(txt) + #dom = Parse(txt) return dom def getPresentationInfoXML(self, url): @@ -565,7 +520,8 @@ class documentViewer(Folder): if txt is None: raise IOError("Unable to read infoXMLfrom %s"%(url)) - dom = Parse(txt) + dom = ET.fromstring(txt) + #dom = Parse(txt) return dom @@ -583,11 +539,14 @@ class documentViewer(Folder): path=getParentDir(path) dom = self.getDomFromIndexMeta(path) - 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() + 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() docinfo['accessType'] = access return docinfo @@ -608,86 +567,23 @@ class documentViewer(Folder): docinfo['indexMetaPath']=self.getIndexMetaPath(path); 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=dom.xpath("//bib/@type") - if bibtype and (len(bibtype)>0): - bibtype=bibtype[0].value + if self.metadataService is not None: + # put all raw bib fields in dict "bib" + bib = self.metadataService.getBibData(dom=dom) + docinfo['bib'] = bib + bibtype = bib.get('@type', None) + docinfo['bib_type'] = bibtype + # also store DC metadata for convenience + dc = self.metadataService.getDCMappedData(bib) + docinfo['creator'] = dc.get('creator',None) + docinfo['title'] = dc.get('title',None) + docinfo['date'] = dc.get('date',None) else: - bibtype="generic" - - 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)) - # if there is no mapping bibmap is empty (mapping sometimes has empty fields) - if len(bibmap) > 0 and len(bibmap['author'][0]) > 0: - try: - docinfo['author']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['author'][0])[0]) - except: pass - try: - docinfo['title']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['title'][0])[0]) - except: pass - try: - docinfo['year']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['year'][0])[0]) - except: pass - 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.error("MetadataService not found!") 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: @@ -698,9 +594,10 @@ class documentViewer(Folder): path=getParentDir(path) dom = self.getDomFromIndexMeta(path) - docinfo['name']=getTextFromNode(dom.xpath("/resource/name")[0]) + docinfo['name']=getText(dom.find("name")) logging.debug("documentViewer docinfo[name] %s"%docinfo['name']) return docinfo + def getDocinfoFromTextTool(self, url, dom=None, docinfo=None): """parse texttool tag in index meta""" @@ -711,19 +608,18 @@ class documentViewer(Folder): docinfo['lang'] = '' # default keine Sprache gesetzt if dom is None: dom = self.getDomFromIndexMeta(url) + + texttool = self.metadata.getTexttoolData(dom=dom) archivePath = None archiveName = None - archiveNames = dom.xpath("//resource/name") - if archiveNames and (len(archiveNames) > 0): - archiveName = getTextFromNode(archiveNames[0]) - else: + archiveName = getText(dom.find("name")) + if not archiveName: logging.warning("documentViewer (getdocinfofromtexttool) resource/name missing in: %s" % (url)) - archivePaths = dom.xpath("//resource/archive-path") - if archivePaths and (len(archivePaths) > 0): - archivePath = getTextFromNode(archivePaths[0]) + archivePath = getText(dom.find("archive-path")) + if archivePath: # clean up archive path if archivePath[0] != '/': archivePath = '/' + archivePath @@ -739,11 +635,9 @@ class documentViewer(Folder): # we balk without archive-path raise IOError("Missing archive-path (for text-tool) in %s" % (url)) - imageDirs = dom.xpath("//texttool/image") - if imageDirs and (len(imageDirs) > 0): - imageDir = getTextFromNode(imageDirs[0]) + imageDir = texttool.get('image', None) - else: + if not imageDir: # we balk with no image tag / not necessary anymore because textmode is now standard #raise IOError("No text-tool info in %s"%(url)) imageDir = "" @@ -760,15 +654,13 @@ class documentViewer(Folder): docinfo['imageURL'] = self.digilibBaseUrl + "/servlet/Scaler?fn=" + imageDir - viewerUrls = dom.xpath("//texttool/digiliburlprefix") - if viewerUrls and (len(viewerUrls) > 0): - viewerUrl = getTextFromNode(viewerUrls[0]) + viewerUrl = texttool.get('digiliburlprefix', None) + if viewerUrl: docinfo['viewerURL'] = viewerUrl # old style text URL - textUrls = dom.xpath("//texttool/text") - if textUrls and (len(textUrls) > 0): - textUrl = getTextFromNode(textUrls[0]) + textUrl = texttool.get('text', None) + if textUrl: if urlparse.urlparse(textUrl)[0] == "": #keine url textUrl = os.path.join(archivePath, textUrl) # fix URLs starting with /mpiwg/online @@ -778,9 +670,8 @@ 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]) + textUrl = texttool.get('text-url-path', None) + if textUrl: docinfo['textURLPath'] = textUrl textUrlkurz = string.split(textUrl, ".")[0] docinfo['textURLPathkurz'] = textUrlkurz @@ -788,16 +679,17 @@ class documentViewer(Folder): # text-only, no page images #docinfo = self.getNumTextPages(docinfo) - - presentationUrls = dom.xpath("//texttool/presentation") + # get bib info docinfo = self.getBibinfoFromIndexMeta(url, docinfo=docinfo, dom=dom) # get info von bib tag + # TODO: is this needed here? docinfo = self.getNameFromIndexMeta(url, docinfo=docinfo, dom=dom) - - if presentationUrls and (len(presentationUrls) > 0): # ueberschreibe diese durch presentation informationen + # TODO: what to do with presentation? + presentationUrl = texttool.get('presentation', None) + if presentationUrl: # 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 = getTextFromNode(presentationUrls[0]) + presentationPath = presentationUrl if url.endswith("index.meta"): presentationUrl = url.replace('index.meta', presentationPath) else: @@ -805,6 +697,7 @@ class documentViewer(Folder): docinfo = self.getBibinfoFromTextToolPresentation(presentationUrl, docinfo=docinfo, dom=dom) + # get authorization docinfo = self.getAuthinfoFromIndexMeta(url, docinfo=docinfo, dom=dom) # get access info return docinfo @@ -814,18 +707,9 @@ class documentViewer(Folder): """gets the bibliographical information from the preseantion entry in texttools """ dom=self.getPresentationInfoXML(url) - 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 + docinfo['author']=getText(dom.find(".//author")) + docinfo['title']=getText(dom.find(".//title")) + docinfo['year']=getText(dom.find(".//date")) return docinfo def getDocinfoFromImagePath(self,path,docinfo=None,cut=0): @@ -844,6 +728,7 @@ class documentViewer(Folder): imageUrl=self.digilibBaseUrl+"/servlet/Scaler?fn="+path docinfo['imageURL'] = imageUrl + #TODO: use getDocinfoFromIndexMeta #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.getAuthinfoFromIndexMeta(pathorig,docinfo=docinfo,cut=cut+1) @@ -858,26 +743,29 @@ 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: %s"%docinfo) + logging.debug("documentViewer (getdocinfo) docinfo in session. keys=%s"%docinfo.keys()) return docinfo + # new docinfo docinfo = {'mode': mode, 'url': url} - if mode=="texttool": #index.meta with texttool information + # add self url + docinfo['viewerUrl'] = self.getDocumentViewerURL() + 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) raise ValueError("Unknown mode %s! Has to be one of 'texttool','imagepath','filepath'."%(mode)) - # FIXME: fake texturlpath - if not docinfo.has_key('textURLPath'): - docinfo['textURLPath'] = None - - logging.debug("documentViewer (getdocinfo) docinfo: %s"%docinfo) - #logging.debug("documentViewer (getdocinfo) docinfo: %s"%) + logging.debug("documentViewer (getdocinfo) docinfo: keys=%s"%docinfo.keys()) + #logging.debug("documentViewer (getdocinfo) docinfo: %s"%docinfo) + # store in session self.REQUEST.SESSION['docinfo'] = docinfo return docinfo @@ -893,6 +781,7 @@ class documentViewer(Folder): pageinfo['cols'] = cols grpsize = cols * rows pageinfo['groupsize'] = grpsize + # what does this do? start = getInt(start, default=(math.ceil(float(current)/float(grpsize))*grpsize-(grpsize-1))) # int(current / grpsize) * grpsize +1)) pageinfo['start'] = start @@ -902,7 +791,8 @@ class documentViewer(Folder): pageinfo['end'] = min(pageinfo['end'], np) pageinfo['numgroups'] = int(np / grpsize) if np % grpsize > 0: - pageinfo['numgroups'] += 1 + pageinfo['numgroups'] += 1 + pageinfo['viewMode'] = viewMode pageinfo['tocMode'] = tocMode pageinfo['characterNormalization'] = self.REQUEST.get('characterNormalization','reg') @@ -914,10 +804,12 @@ class documentViewer(Folder): pageinfo['highlightQuery'] = self.REQUEST.get('highlightQuery','') pageinfo['tocPageSize'] = self.REQUEST.get('tocPageSize', '30') pageinfo['queryPageSize'] =self.REQUEST.get('queryPageSize', '10') - pageinfo['tocPN'] = self.REQUEST.get('tocPN', '1') - toc = int (pageinfo['tocPN']) - pageinfo['textPages'] =int (toc) + pageinfo['tocPN'] = self.REQUEST.get('tocPN', '1') + # WTF?: + toc = int(pageinfo['tocPN']) + pageinfo['textPages'] =int(toc) + # What does this do? if 'tocSize_%s'%tocMode in docinfo: tocSize = int(docinfo['tocSize_%s'%tocMode]) tocPageSize = int(pageinfo['tocPageSize']) @@ -926,18 +818,30 @@ class documentViewer(Folder): tocPages=tocSize/tocPageSize+1 else: tocPages=tocSize/tocPageSize - pageinfo['tocPN'] = min (tocPages,toc) + + pageinfo['tocPN'] = min(tocPages,toc) + pageinfo['searchPN'] =self.REQUEST.get('searchPN','1') pageinfo['sn'] =self.REQUEST.get('sn','') return pageinfo + + + security.declareProtected('View management screens','changeDocumentViewerForm') + changeDocumentViewerForm = PageTemplateFile('zpt/changeDocumentViewer', globals()) -def changeDocumentViewer(self,title="",digilibBaseUrl=None,thumbrows=2,thumbcols=5,authgroups='mpiwg',RESPONSE=None): + def changeDocumentViewer(self,title="",digilibBaseUrl=None,thumbrows=2,thumbcols=5,authgroups='mpiwg',RESPONSE=None): """init document viewer""" self.title=title self.digilibBaseUrl = digilibBaseUrl self.thumbrows = thumbrows self.thumbcols = thumbcols self.authgroups = [s.strip().lower() for s in authgroups.split(',')] + try: + # assume MetaDataFolder instance is called metadata + self.metadataService = getattr(self, 'metadata') + except Exception, e: + logging.error("Unable to find MetaDataFolder 'metadata': "+str(e)) + if RESPONSE is not None: RESPONSE.redirect('manage_main')