--- documentViewer/documentViewer.py 2005/12/23 14:01:29 1.2 +++ documentViewer/documentViewer.py 2007/05/04 13:17:02 1.26 @@ -1,23 +1,42 @@ -genericDigilib="http://nausikaa2.rz-berlin.mpg.de/digitallibrary/" from OFS.Folder import Folder from Products.PageTemplates.ZopePageTemplate import ZopePageTemplate -from Products.PageTemplates.PageTemplateFile import PageTemplateFile +from Products.PageTemplates.PageTemplateFile import PageTemplateFile from AccessControl import ClassSecurityInfo +from AccessControl import getSecurityManager from Globals import package_home from Ft.Xml.Domlette import NonvalidatingReader from Ft.Xml.Domlette import PrettyPrint, Print -from Ft.Xml import EMPTY_NAMESPACE +from Ft.Xml import EMPTY_NAMESPACE, Parse import Ft.Xml.XPath import os.path +import sys import cgi import urllib +import logging + +import urlparse + +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 default def getTextFromNode(nodename): + """get the cdata content of a node""" + if nodename is None: + return "" nodelist=nodename.childNodes rc = "" for node in nodelist: @@ -25,214 +44,465 @@ def getTextFromNode(nodename): rc = rc + node.data return rc + +def getParentDir(path): + """returns pathname shortened by one""" + return '/'.join(path.split('/')[0:-1]) + + import socket -def urlopen(url): +def urlopen(url,timeout=2): """urlopen mit timeout""" - socket.setdefaulttimeout(2) + socket.setdefaulttimeout(timeout) ret=urllib.urlopen(url) socket.setdefaulttimeout(5) return ret - -def getParamFromDigilib(path,param): - """gibt param von dlInfo aus""" - imageUrl=genericDigilib+"/dlInfo-xml.jsp?fn="+path - try: - dom = NonvalidatingReader.parseUri(imageUrl) - except: - return None - - - params=dom.xpath("//document-parameters/parameter[@name='%s']/@value"%param) - - if params: - return params[0].value - -def parseUrlTextTool(url): - """parse index meta""" - - try: - dom = NonvalidatingReader.parseUri(url) - except: - zLOG.LOG("documentViewer (parseUrlTexttool)", zLOG.INFO,"%s (%s)"%sys.exc_info()[0:2]) - return (None,None,None) - - archivePaths=dom.xpath("//resource/archive-path") - - if archivePaths and (len(archivePaths)>0): - archivePath=getTextFromNode(archivePaths[0]) - else: - archivePath=None - - - images=dom.xpath("//texttool/image") - - if images and (len(images)>0): - image=getTextFromNode(images[0]) - else: - image=None - - if image and archivePath: - image=os.path.join(archivePath,image) - image=image.replace("/mpiwg/online",'') - pt=getParamFromDigilib(image,'pt') - else: - image=None - - viewerUrls=dom.xpath("//texttool/digiliburlprefix") - - if viewerUrls and (len(viewerUrls)>0): - viewerUrl=getTextFromNode(viewerUrls[0]) - else: - viewerUrl=None - - - textUrls=dom.xpath("//texttool/text") - - if textUrls and (len(textUrls)>0): - textUrl=getTextFromNode(textUrls[0]) - else: - textUrl=None - return viewerUrl,(image,pt),textUrl - - -class documentViewer(ZopePageTemplate): +## +## documentViewer class +## +class documentViewer(Folder): """document viewer""" - + #textViewerUrl="http://127.0.0.1:8080/HFQP/testXSLT/getPage?" + meta_type="Document viewer" security=ClassSecurityInfo() - manage_options=ZopePageTemplate.manage_options+( + manage_options=Folder.manage_options+( {'label':'main config','action':'changeDocumentViewerForm'}, ) - _default_content_fn = os.path.join(package_home(globals()),'zpt','documentViewer_template.zpt') + # templates and forms + viewer_main = PageTemplateFile('zpt/viewer_main', globals()) + thumbs_main = PageTemplateFile('zpt/thumbs_main', globals()) + image_main = PageTemplateFile('zpt/image_main', globals()) + head_main = PageTemplateFile('zpt/head_main', globals()) + docuviewer_css = PageTemplateFile('css/docuviewer.css', globals()) + info_xml = PageTemplateFile('zpt/info_xml', globals()) + + security.declareProtected('View management screens','changeDocumentViewerForm') + changeDocumentViewerForm = PageTemplateFile('zpt/changeDocumentViewer', globals()) + - def __init__(self,id,imageViewerUrl,title=""): + def __init__(self,id,imageViewerUrl,textViewerUrl=None,title="",digilibBaseUrl=None,thumbcols=2,thumbrows=10,authgroups="mpiwg"): """init document viewer""" self.id=id self.title=title self.imageViewerUrl=imageViewerUrl + self.textViewerUrl=textViewerUrl - security.declareProtected('View management screens','changeDocumentViewerForm') - def changeDocumentViewerForm(self): - """change it""" - pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','changeDocumentViewer.zpt')).__of__(self) - return pt() + if not digilibBaseUrl: + self.digilibBaseUrl = self.findDigilibUrl() + else: + self.digilibBaseUrl = digilibBaseUrl + self.thumbcols = thumbcols + self.thumbrows = thumbrows + # authgroups is list of authorized groups (delimited by ,) + self.authgroups = [s.strip().lower() for s in authgroups.split(',')] + # add template folder so we can always use template.something + self.manage_addFolder('template') + + + security.declareProtected('View','index_html') + def index_html(self,mode,url,viewMode="auto",start=None,pn=1): + ''' + view it + @param mode: defines how to access the document behind url + @param url: url which contains display information + @param viewMode: if images display images, if text display text, default is images (text,images or auto) + + ''' + + 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 + self.manage_addFolder('template') + + if not self.digilibBaseUrl: + self.digilibBaseUrl = self.findDigilibUrl() or "http://nausikaa.mpiwg-berlin.mpg.de/digitallibrary" + + docinfo = self.getDocinfo(mode=mode,url=url) + pageinfo = self.getPageinfo(start=start,current=pn,docinfo=docinfo) + pt = getattr(self.template, 'viewer_main') + + if viewMode=="auto": # automodus gewaehlt + if docinfo.get("textURL",'') and self.textViewerUrl: #texturl gesetzt und textViewer konfiguriert + viewMode="text" + else: + viewMode="images" + + return pt(docinfo=docinfo,pageinfo=pageinfo,viewMode=viewMode) - def changeDocumentViewer(self,imageViewerUrl,title="",RESPONSE=None): - """init document viewer""" - self.title=title - self.imageViewerUrl=imageViewerUrl - - if RESPONSE is not None: - RESPONSE.redirect('manage_main') - - - def imageLink(self,nr): - """link hinter den images""" - paramsTmp=cgi.parse_qs(self.REQUEST['QUERY_STRING']) - params={} - for x in paramsTmp.iteritems(): - params[x[0]]=x[1][0] - - params['pn']=nr - newUrl=self.REQUEST['URL']+"?"+urllib.urlencode(params) - return newUrl - - - def thumbruler(self,cols,rows,start,maximum): - """ruler for thumbs""" - ret="" - paramsTmp=cgi.parse_qs(self.REQUEST['QUERY_STRING']) - params={} - for x in paramsTmp.iteritems(): - - if not x[0]=="start": - params[x[0]]=x[1][0] - - newUrlSelect=self.REQUEST['URL']+"?"+urllib.urlencode(params) - if start>0: - newStart=max(start-cols*rows,0) - params['start']=newStart - newUrl=self.REQUEST['URL']+"?"+urllib.urlencode(params) - ret+="""prev"""%newUrl - - - ret+="""" + params[param] = str(val) + + # quote values and assemble into query string + ps = "&".join(["%s=%s"%(k,urllib.quote(v)) for (k, v) in params.items()]) + url=self.REQUEST['URL1']+"?"+ps + return url + + + def getInfo_xml(self,url,mode): + """returns info about the document as XML""" + + if not self.digilibBaseUrl: + self.digilibBaseUrl = self.findDigilibUrl() or "http://nausikaa.mpiwg-berlin.mpg.de/digitallibrary" + + docinfo = self.getDocinfo(mode=mode,url=url) + pt = getattr(self.template, 'info_xml') + return pt(docinfo=docinfo) + + + def getStyle(self, idx, selected, style=""): + """returns a string with the given style and append 'sel' if path == selected.""" + #logger("documentViewer (getstyle)", logging.INFO, "idx: %s selected: %s style: %s"%(idx,selected,style)) + if idx == selected: + return style + 'sel' + else: + return style + - if startnext"""%newUrl + def isAccessible(self, docinfo): + """returns if access to the resource is granted""" + access = docinfo.get('accessType', None) + logger("documentViewer (accessOK)", logging.INFO, "access type %s"%access) + if access is not None and access == '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 + user = getSecurityManager().getUser() + if user is not None: + #print "user: ", user + return (user.getUserName() != "Anonymous User") + else: + return False - return ret + logger("documentViewer (accessOK)", logging.INFO, "unknown access type %s"%access) + return False + + + def getDirinfoFromDigilib(self,path,docinfo=None): + """gibt param von dlInfo aus""" + num_retries = 3 + if docinfo is None: + docinfo = {} + + infoUrl=self.digilibBaseUrl+"/dirInfo-xml.jsp?mo=dir&fn="+path + + logger("documentViewer (getparamfromdigilib)", logging.INFO, "dirInfo from %s"%(infoUrl)) - def textToolThumb(self,url,start=0): - """understands the texttool format - @param url: url to index.meta with texttool tag - """ - (viewerUrl,imagepath,textpath)=parseUrlTextTool(url) + 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)) - imageUrl=genericDigilib+"/servlet/Scaler?fn=%s"%imagepath[0] + sizes=dom.xpath("//dir/size") + logger("documentViewer (getparamfromdigilib)", logging.INFO, "dirInfo:size"%sizes) - pt=PageTemplateFile(os.path.join(package_home(globals()),'zpt','thumbs.zpt')).__of__(self) - return pt(imageUrl=imageUrl,pt=imagepath[1],start=start) + if sizes: + docinfo['numPages'] = int(getTextFromNode(sizes[0])) + else: + docinfo['numPages'] = 0 + + return docinfo - def imagePathThumb(self,path,start=0): - """path ist the path to the images it is assumes that the index.meta file is one level higher.""" + + def getIndexMeta(self, url): + """returns dom of index.meta document at url""" + num_retries = 3 + dom = None + metaUrl = None + if url.startswith("http://"): + # real URL + metaUrl = url + else: + # online path + server=self.digilibBaseUrl+"/servlet/Texter?fn=" + metaUrl=server+url.replace("/mpiwg/online","") + if not metaUrl.endswith("index.meta"): + metaUrl += "/index.meta" + print 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 (getIndexMata)", logging.INFO,"%s (%s)"%sys.exc_info()[0:2]) + + if dom is None: + raise IOError("Unable to read index meta from %s"%(url)) + + 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://"): + # real URL + metaUrl = url + else: + # online path + server=self.digilibBaseUrl+"/servlet/Texter?fn=" + metaUrl=server+url.replace("/mpiwg/online","") + - path=path.replace("/mpiwg/online","") - pt=getParamFromDigilib(path,'pt') - imageUrl=genericDigilib+"/servlet/Scaler?fn=%s"%path + 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)) + + return dom + + + def getAuthinfoFromIndexMeta(self,path,docinfo=None,dom=None): + """gets authorization info from the index.meta file at path or given by dom""" + logger("documentViewer (getauthinfofromindexmeta)", logging.INFO,"path: %s"%(path)) - pageT=PageTemplateFile(os.path.join(package_home(globals()),'zpt','thumbs.zpt')).__of__(self) - return pageT(imageUrl=imageUrl,pt=pt,start=start) + access = None + + if docinfo is None: + docinfo = {} + + if dom is None: + dom = self.getIndexMeta(getParentDir(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() + + docinfo['accessType'] = access + return docinfo - def headerFromIndexMeta(self,path): - """gibt header from index meta out""" + def getBibinfoFromIndexMeta(self,path,docinfo=None,dom=None): + """gets bibliographical info from the index.meta file at path or given by dom""" + logger("documentViewer (getbibinfofromindexmeta)", logging.INFO,"path: %s"%(path)) + + if docinfo is None: + docinfo = {} + + if dom is None: + dom = self.getIndexMeta(getParentDir(path)) + metaData=self.metadata.main.meta.bib - server="http://foxridge.mpiwg-berlin.mpg.de" - path="/".join(path.split("/")[0:-1]) - metaUrl=server+path+"/index.meta" - - dom = NonvalidatingReader.parseUri(metaUrl) - type=dom.xpath("//bib/@type") - if type and (len(type)>0): - type=type[0].value + bibtype=dom.xpath("//bib/@type") + if bibtype and (len(bibtype)>0): + bibtype=bibtype[0].value else: - type="generic" - type=type.replace("-"," ")# wrong typesiin index meta "-" instead of " " - hash=metaData.generateMappingForType(type) - - author=getTextFromNode(dom.xpath("//bib/%s"%hash['author'][0])[0]) - title=getTextFromNode(dom.xpath("//bib/%s"%hash['title'][0])[0]) - year=getTextFromNode(dom.xpath("//bib/%s"%hash['year'][0])[0]) - - return author,title,year + bibtype="generic" + bibtype=bibtype.replace("-"," ") # wrong typesiin index meta "-" instead of " " (not wrong! ROC) + bibmap=metaData.generateMappingForType(bibtype) + #print "bibmap: ", bibmap, " for: ", bibtype + # if there is no mapping bibmap is empty (mapping sometimes has empty fields) + if len(bibmap) > 0 and len(bibmap['author'][0]) > 0: + docinfo['author']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['author'][0])[0]) + docinfo['title']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['title'][0])[0]) + docinfo['year']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['year'][0])[0]) + + logging.info("bla") + try: + docinfo['lang']=getTextFromNode(dom.xpath("//bib/lang")[0]) + except: + docinfo['lang']='' + return docinfo + + def getDocinfoFromTextTool(self,url,dom=None,docinfo=None): + """parse texttool tag in index meta""" + logger("documentViewer (getdocinfofromtexttool)", logging.INFO,"url: %s"%(url)) + if docinfo is None: + docinfo = {} + + if docinfo.get('lang',None) is None: + docinfo['lang']='' # default keine Sprache gesetzt + if dom is None: + dom = self.getIndexMeta(url) + + archivePath = None + archiveName = None + + archiveNames=dom.xpath("//resource/name") + if archiveNames and (len(archiveNames)>0): + archiveName=getTextFromNode(archiveNames[0]) + else: + logger("documentViewer (getdocinfofromtexttool)", logging.WARNING,"resource/name missing in: %s"%(url)) + + 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 + if archiveName and (not archivePath.endswith(archiveName)): + archivePath += "/" + archiveName + else: + # try to get archive-path from url + logger("documentViewer (getdocinfofromtexttool)", logging.WARNING,"resource/archive-path missing in: %s"%(url)) + if (not url.startswith('http')): + archivePath = url.replace('index.meta', '') + + if archivePath is None: + # 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]) + 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="" + docinfo['numPages']=1 # im moment einfach auf eins setzen, navigation ueber die thumbs geht natuerlich nicht + + docinfo['imagePath'] = "" # keine Bilder + docinfo['imageURL'] = "" + + if imageDir and archivePath: + #print "image: ", imageDir, " archivepath: ", archivePath + imageDir=os.path.join(archivePath,imageDir) + imageDir=imageDir.replace("/mpiwg/online",'') + docinfo=self.getDirinfoFromDigilib(imageDir,docinfo=docinfo) + docinfo['imagePath'] = imageDir + docinfo['imageURL'] = self.digilibBaseUrl+"/servlet/Scaler?fn="+imageDir + + viewerUrls=dom.xpath("//texttool/digiliburlprefix") + if viewerUrls and (len(viewerUrls)>0): + viewerUrl=getTextFromNode(viewerUrls[0]) + docinfo['viewerURL'] = viewerUrl + + 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) + + docinfo['textURL'] = textUrl + + presentationUrls=dom.xpath("//texttool/presentation") + docinfo = self.getBibinfoFromIndexMeta(url,docinfo=docinfo,dom=dom) # get info von bib tag + + if presentationUrls and (len(presentationUrls)>0): # ueberschreibe diese durch presentation informationen + # presentation url ergiebt sich ersetzen von index.meta in der url der fŸr die Metadaten + # durch den relativen Pfad auf die presentation infos + presentationUrl=url.replace('index.meta',getTextFromNode(presentationUrls[0])) + docinfo = self.getBibinfoFromTextToolPresentation(presentationUrl,docinfo=docinfo,dom=dom) + + docinfo = self.getAuthinfoFromIndexMeta(url,docinfo=docinfo,dom=dom) # get access info + return docinfo + + + def getBibinfoFromTextToolPresentation(self,url,docinfo=None,dom=None): + """gets the bibliographical information from the preseantion entry in texttools + """ + dom=self.getPresentationInfoXML(url) + docinfo['author']=getTextFromNode(dom.xpath("//author")[0]) + docinfo['title']=getTextFromNode(dom.xpath("//title")[0]) + docinfo['year']=getTextFromNode(dom.xpath("//date")[0]) + return docinfo + + def getDocinfoFromImagePath(self,path,docinfo=None): + """path ist the path to the images it assumes that the index.meta file is one level higher.""" + logger("documentViewer (getdocinfofromimagepath)", logging.INFO,"path: %s"%(path)) + if docinfo is None: + docinfo = {} + path=path.replace("/mpiwg/online","") + docinfo['imagePath'] = path + docinfo=self.getDirinfoFromDigilib(path,docinfo=docinfo) + imageUrl=self.digilibBaseUrl+"/servlet/Scaler?fn="+path + docinfo['imageURL'] = imageUrl + + docinfo = self.getBibinfoFromIndexMeta(path,docinfo=docinfo) + docinfo = self.getAuthinfoFromIndexMeta(path,docinfo=docinfo) + return docinfo + + + def getDocinfo(self, mode, url): + """returns docinfo depending on mode""" + 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: + logger("documentViewer (getdocinfo)", logging.INFO,"docinfo in session: %s"%docinfo) + return docinfo + # new docinfo + docinfo = {'mode': mode, 'url': url} + if mode=="texttool": #index.meta with texttool information + docinfo = self.getDocinfoFromTextTool(url, docinfo=docinfo) + elif mode=="imagepath": + docinfo = self.getDocinfoFromImagePath(url, docinfo=docinfo) + else: + logger("documentViewer (getdocinfo)", logging.ERROR,"unknown mode!") + raise ValueError("Unknown mode %s"%(mode)) + + logger("documentViewer (getdocinfo)", logging.INFO,"docinfo: %s"%docinfo) + self.REQUEST.SESSION['docinfo'] = docinfo + return docinfo + + + def getPageinfo(self, current, start=None, rows=None, cols=None, docinfo=None): + """returns pageinfo with the given parameters""" + pageinfo = {} + current = getInt(current) + pageinfo['current'] = current + rows = int(rows or self.thumbrows) + pageinfo['rows'] = rows + cols = int(cols or self.thumbcols) + pageinfo['cols'] = cols + grpsize = cols * rows + pageinfo['groupsize'] = grpsize + start = getInt(start, default=(int(current / grpsize) * grpsize +1)) + pageinfo['start'] = start + pageinfo['end'] = start + grpsize + if docinfo is not None: + np = int(docinfo['numPages']) + pageinfo['end'] = min(pageinfo['end'], np) + pageinfo['numgroups'] = int(np / grpsize) + if np % grpsize > 0: + pageinfo['numgroups'] += 1 + + return pageinfo + def text(self,mode,url,pn): """give text""" if mode=="texttool": #index.meta with texttool information (viewerUrl,imagepath,textpath)=parseUrlTextTool(url) - print textpath + #print textpath try: dom = NonvalidatingReader.parseUri(textpath) except: @@ -279,58 +549,88 @@ class documentViewer(ZopePageTemplate): # PrettyPrint(x) # # return list -# - def image(self,mode,url,pn): - """give image out""" - if mode=="texttool": #index.meta with texttool information - (viewerUrl,imagepath,textpath)=parseUrlTextTool(url) - if not viewerUrl: - viewerUrl=self.imageViewerUrl - url=viewerUrl+"pn=%s&fn=%s"%(pn,imagepath[0]) - ret="""