--- documentViewer/documentViewer.py 2006/04/10 11:23:18 1.6 +++ documentViewer/documentViewer.py 2010/06/08 16:57:36 1.67 @@ -1,32 +1,49 @@ -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 Products.PythonScripts.standard import url_quote 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 + +from xml.dom.minidom import parse, parseString + -import Ft.Xml.XPath +import Ft.Xml.XPath +import cStringIO +import xmlrpclib import os.path +import sys import cgi import urllib -import zLOG +import logging +import math + +import urlparse +from types import * +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 - + 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: @@ -34,16 +51,28 @@ def getTextFromNode(nodename): rc = rc + node.data return rc +def serializeNode(node, encoding='utf-8'): + """returns a string containing node as XML""" + buf = cStringIO.StringIO() + Print(node, stream=buf, encoding=encoding) + s = buf.getvalue() + buf.close() + return s + + +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 - - ## @@ -51,7 +80,8 @@ def urlopen(url): ## class documentViewer(Folder): """document viewer""" - + #textViewerUrl="http://127.0.0.1:8080/HFQP/testXSLT/getPage?" + meta_type="Document viewer" security=ClassSecurityInfo() @@ -61,43 +91,65 @@ class documentViewer(Folder): # templates and forms viewer_main = PageTemplateFile('zpt/viewer_main', globals()) - thumbs_main = PageTemplateFile('zpt/thumbs_main', globals()) - image_main = PageTemplateFile('zpt/image_main', globals()) + toc_thumbs = PageTemplateFile('zpt/toc_thumbs', globals()) + toc_text = PageTemplateFile('zpt/toc_text', globals()) + toc_figures = PageTemplateFile('zpt/toc_figures', globals()) + page_main_images = PageTemplateFile('zpt/page_main_images', globals()) + page_main_text = PageTemplateFile('zpt/page_main_text', globals()) + page_main_text_dict = PageTemplateFile('zpt/page_main_text_dict', globals()) + page_main_xml = PageTemplateFile('zpt/page_main_xml', globals()) head_main = PageTemplateFile('zpt/head_main', globals()) docuviewer_css = PageTemplateFile('css/docuviewer.css', globals()) + info_xml = PageTemplateFile('zpt/info_xml', globals()) + thumbs_main_rss = PageTemplateFile('zpt/thumbs_main_rss', globals()) security.declareProtected('View management screens','changeDocumentViewerForm') changeDocumentViewerForm = PageTemplateFile('zpt/changeDocumentViewer', globals()) - def __init__(self,id,imageViewerUrl,title="",digilibBaseUrl=None,thumbcols=2,thumbrows=10): + def __init__(self,id,imageScalerUrl=None,textServerName=None,title="",digilibBaseUrl=None,thumbcols=2,thumbrows=5,authgroups="mpiwg"): """init document viewer""" self.id=id self.title=title - self.imageViewerUrl=imageViewerUrl - if not digilibBaseUrl: - self.digilibBaseUrl = self.findDigilibUrl() - else: - self.digilibBaseUrl = digilibBaseUrl self.thumbcols = thumbcols self.thumbrows = thumbrows - # add template folder so we can always use template.something - self.manage_addFolder('template') - + # authgroups is list of authorized groups (delimited by ,) + self.authgroups = [s.strip().lower() for s in authgroups.split(',')] + # create template folder so we can always use template.something + + templateFolder = Folder('template') + #self['template'] = templateFolder # Zope-2.12 style + self._setObject('template',templateFolder) # old style + try: + from Products.XMLRpcTools.XMLRpcTools import XMLRpcServerProxy + xmlRpcClient = XMLRpcServerProxy(id='fulltextclient', serverUrl=textServerName, use_xmlrpc=False) + #templateFolder['fulltextclient'] = xmlRpcClient + templateFolder._setObject('fulltextclient',xmlRpcClient) + except Exception, 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") + #templateFolder['zogilib'] = zogilib + templateFolder._setObject('zogilib',zogilib) + except Exception, e: + logging.error("Unable to create zogiLib for zogilib: "+str(e)) + - security.declareProtected('View','index_html') - def index_html(self,mode,url,start=None,pn=1): + security.declareProtected('View','thumbs_rss') + def thumbs_rss(self,mode,url,viewMode="auto",start=None,pn=1): ''' view it - @param mode: defines which type of document is behind url + @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) - zLOG.LOG("documentViewer (index)", zLOG.INFO, "mode: %s url:%s start:%s pn:%s"%(mode,url,start,pn)) + ''' + logging.debug("HHHHHHHHHHHHHH:load the rss") + 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 - print "no template folder -- creating" self.manage_addFolder('template') if not self.digilibBaseUrl: @@ -105,156 +157,472 @@ class documentViewer(Folder): docinfo = self.getDocinfo(mode=mode,url=url) pageinfo = self.getPageinfo(start=start,current=pn,docinfo=docinfo) - pt = getattr(self.template, 'viewer_main') - return pt(docinfo=docinfo,pageinfo=pageinfo) - - - def getLink(self,param=None,val=None): - """link to documentviewer with parameter param set to val""" - params=cgi.parse_qs(self.REQUEST['QUERY_STRING']) - if param is not None: - if val is None and params.has_key(param): - del params[param] + pt = getattr(self.template, 'thumbs_main_rss') + + if viewMode=="auto": # automodus gewaehlt + if docinfo.get("textURL",'') and self.textViewerUrl: #texturl gesetzt und textViewer konfiguriert + viewMode="text" else: - params[param] = val + viewMode="images" + + return pt(docinfo=docinfo,pageinfo=pageinfo,viewMode=viewMode) + + security.declareProtected('View','index_html') + def index_html(self,url,mode="texttool",viewMode="auto",tocMode="thumbs",start=None,pn=1,mk=None, query=None, querySearch=None): + ''' + 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 auto (text,images or auto) + @param tocMode: type of 'table of contents' for navigation (thumbs, text, figures, none) + @param querySearch: type of different search modes (fulltext, fulltextMorph, xpath, xquery, ftIndex, ftIndexMorph, fulltextMorphLemma) + ''' + + logging.debug("documentViewer (index) mode: %s url:%s start:%s pn:%s"%(mode,url,start,pn)) - url=self.REQUEST['URL']+"?"+urllib.urlencode(params, doseq=True) + if not hasattr(self, 'template'): + # this won't work + logging.error("template folder missing!") + return "ERROR: template folder missing!" + + if not getattr(self, 'digilibBaseUrl', None): + 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) + + pageinfo = self.getPageinfo(start=start,current=pn,docinfo=docinfo,viewMode=viewMode,tocMode=tocMode) + + if viewMode=="auto": # automodus gewaehlt + if docinfo.get("textURL",''): #texturl gesetzt und textViewer konfiguriert + viewMode="text" + else: + viewMode="images" + + pt = getattr(self.template, 'viewer_main') + return pt(docinfo=docinfo,pageinfo=pageinfo,viewMode=viewMode,mk=self.generateMarks(mk)) + + def generateMarks(self,mk): + ret="" + if mk is None: + return "" + if type(mk) is not ListType: + mk=[mk] + for m in mk: + ret+="mk=%s"%m + return ret + + + def findDigilibUrl(self): + """try to get the digilib URL from zogilib""" + url = self.template.zogilib.getDLBaseUrl() return url + def getDocumentViewerURL(self): + """returns the URL of this instance""" + return self.absolute_url() def getStyle(self, idx, selected, style=""): """returns a string with the given style and append 'sel' if path == selected.""" - #zLOG.LOG("documentViewer (getstyle)", zLOG.INFO, "idx: %s selected: %s style: %s"%(idx,selected,style)) + #logger("documentViewer (getstyle)", logging.INFO, "idx: %s selected: %s style: %s"%(idx,selected,style)) if idx == selected: return style + 'sel' else: - return style + return style + + def getLink(self,param=None,val=None): + """link to documentviewer with parameter param set to val""" + params=self.REQUEST.form.copy() + if param is not None: + if val is None: + if params.has_key(param): + del params[param] + else: + params[param] = str(val) + + if params.get("mode", None) == "filepath": #wenn beim erst Aufruf filepath gesetzt wurde aendere das nun zu imagepath + params["mode"] = "imagepath" + params["url"] = getParentDir(params["url"]) + + # 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 getLinkAmp(self,param=None,val=None): + """link to documentviewer with parameter param set to val""" + params=self.REQUEST.form.copy() + if param is not None: + if val is None: + if params.has_key(param): + del params[param] + else: + params[param] = str(val) + + # quote values and assemble into query string + 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 + + 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 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 - def getParamFromDigilib(self,path,param): + 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""" - imageUrl=self.digilibBaseUrl+"/dirInfo-xml.jsp?mo=dir&fn="+path + num_retries = 3 + if docinfo is None: + docinfo = {} + + for x in range(cut): + + path=getParentDir(path) + + infoUrl=self.digilibBaseUrl+"/dirInfo-xml.jsp?mo=dir&fn="+path - zLOG.LOG("documentViewer (getparamfromdigilib)", zLOG.INFO, "dirInfo (%s) from %s"%(param,imageUrl)) + logger("documentViewer (getparamfromdigilib)", logging.INFO, "dirInfo from %s"%(infoUrl)) - try: - dom = NonvalidatingReader.parseUri(imageUrl) - except: - return 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)) - params=dom.xpath("//dir/%s"%param) - zLOG.LOG("documentViewer (getparamfromdigilib)", zLOG.INFO, "dirInfo:%s"%params) + sizes=dom.xpath("//dir/size") + logger("documentViewer (getparamfromdigilib)", logging.INFO, "dirInfo:size"%sizes) - if params: - return getTextFromNode(params[0]) + if sizes: + docinfo['numPages'] = int(getTextFromNode(sizes[0])) + else: + docinfo['numPages'] = 0 + + # TODO: produce and keep list of image names and numbers + + return docinfo + + 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" + 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]) + + if dom is None: + raise IOError("Unable to read index meta from %s"%(url)) + + return dom - def getBibinfoFromIndexMeta(self,path,docinfo=None,dom=None): - """gets bibliographical info from the index.meta file at url or given by dom""" - zLOG.LOG("documentViewer (getbibinfofromindexmeta)", zLOG.INFO,"path: %s"%(path)) + 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","") + + 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,cut=0): + """gets authorization info from the index.meta file at path or given by dom""" + logger("documentViewer (getauthinfofromindexmeta)", logging.INFO,"path: %s"%(path)) + + access = None if docinfo is None: docinfo = {} if dom is None: - server="http://nausikaa.mpiwg-berlin.mpg.de/digitallibrary/servlet/Texter?fn=" - path="/".join(path.split("/")[0:-1]) - metaUrl=server+path+"/index.meta" - try: - dom = NonvalidatingReader.parseUri(metaUrl) - except: - return docinfo + for x in range(cut): + path=getParentDir(path) + dom = self.getIndexMeta(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 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)) + + if docinfo is None: + docinfo = {} + + if dom is None: + for x in range(cut): + path=getParentDir(path) + dom = self.getIndexMeta(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 else: bibtype="generic" + bibtype=bibtype.replace("-"," ") # wrong typesiin index meta "-" instead of " " (not wrong! ROC) + docinfo['bib_type'] = bibtype bibmap=metaData.generateMappingForType(bibtype) - if len(bibmap) > 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]) - - return docinfo + # 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']='' - - def getDocinfoFromTextTool(self,url,docinfo=None): - """parse texttool tag in index meta""" - zLOG.LOG("documentViewer (getdocinfofromtexttool)", zLOG.INFO,"url: %s"%(url)) - if docinfo is None: + 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 = {} - - try: - dom = NonvalidatingReader.parseUri(url) - except: - zLOG.LOG("documentViewer (parseUrlTexttool)", zLOG.INFO,"%s (%s)"%sys.exc_info()[0:2]) - return docinfo - - 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=self.getParamFromDigilib(image,'size') - docinfo['imagePath'] = image - docinfo['numPages'] = pt - - viewerUrls=dom.xpath("//texttool/digiliburlprefix") - - if viewerUrls and (len(viewerUrls)>0): - viewerUrl=getTextFromNode(viewerUrls[0]) - docinfo['imageURL'] = viewerURL - - textUrls=dom.xpath("//texttool/text") - - if textUrls and (len(textUrls)>0): - textUrl=getTextFromNode(textUrls[0]) - docinfo['textURL'] = textURL - - docinfo = self.getBibinfoFromIndexMeta(url,docinfo=docinfo,dom=dom) - return 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 = "" + #xquery="//pb" + 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) + # fix URLs starting with /mpiwg/online + if textUrl.startswith("/mpiwg/online"): + textUrl = textUrl.replace("/mpiwg/online", '', 1) + + docinfo['textURL'] = textUrl + + textUrls = dom.xpath("//texttool/text-url-path") + if textUrls and (len(textUrls) > 0): + textUrl = getTextFromNode(textUrls[0]) + docinfo['textURLPath'] = textUrl + if not docinfo['imagePath']: + # text-only, no page images + docinfo = self.getNumPages(docinfo) #im moment einfach auf eins setzen, navigation ueber die thumbs geht natuerlich nicht + + 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 fuer die Metadaten + # durch den relativen Pfad auf die presentation infos + presentationPath = getTextFromNode(presentationUrls[0]) + if url.endswith("index.meta"): + presentationUrl = url.replace('index.meta', presentationPath) + else: + presentationUrl = url + "/" + presentationPath + + docinfo = self.getBibinfoFromTextToolPresentation(presentationUrl, docinfo=docinfo, dom=dom) + + docinfo = self.getAuthinfoFromIndexMeta(url, docinfo=docinfo, dom=dom) # get access info + + return docinfo - - def getDocinfoFromImagePath(self,path,docinfo=None): + + def getBibinfoFromTextToolPresentation(self,url,docinfo=None,dom=None): + """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 + return docinfo + + 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.""" - zLOG.LOG("documentViewer (getdocinfofromimagepath)", zLOG.INFO,"path: %s"%(path)) + logger("documentViewer (getdocinfofromimagepath)", logging.INFO,"path: %s"%(path)) if docinfo is None: docinfo = {} path=path.replace("/mpiwg/online","") docinfo['imagePath'] = path - pt=self.getParamFromDigilib(path,'size') - docinfo['numPages'] = pt - imageUrl=self.digilibBaseUrl+"/servlet/Scaler?fn=%s"%path + docinfo=self.getDirinfoFromDigilib(path,docinfo=docinfo,cut=cut) + + pathorig=path + for x in range(cut): + path=getParentDir(path) + logging.error("PATH:"+path) + imageUrl=self.digilibBaseUrl+"/servlet/Scaler?fn="+path docinfo['imageURL'] = imageUrl - docinfo = self.getBibinfoFromIndexMeta(path,docinfo=docinfo) + #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) return docinfo def getDocinfo(self, mode, url): """returns docinfo depending on mode""" - zLOG.LOG("documentViewer (getdocinfo)", zLOG.INFO,"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: - zLOG.LOG("documentViewer (getdocinfo)", zLOG.INFO,"docinfo in session: %s"%docinfo) + logger("documentViewer (getdocinfo)", logging.INFO,"docinfo in session: %s"%docinfo) return docinfo # new docinfo docinfo = {'mode': mode, 'url': url} @@ -262,14 +630,18 @@ class documentViewer(Folder): docinfo = self.getDocinfoFromTextTool(url, docinfo=docinfo) elif mode=="imagepath": docinfo = self.getDocinfoFromImagePath(url, docinfo=docinfo) + elif mode=="filepath": + docinfo = self.getDocinfoFromImagePath(url, docinfo=docinfo,cut=1) else: - zLOG.LOG("documentViewer (getdocinfo)", zLOG.ERROR,"unknown mode!") - zLOG.LOG("documentViewer (getdocinfo)", zLOG.INFO,"docinfo: %s"%docinfo) + logger("documentViewer (getdocinfo)", logging.ERROR,"unknown mode!") + raise ValueError("Unknown mode %s! Has to be one of 'texttool','imagepath','filepath'."%(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): + def getPageinfo(self, current, start=None, rows=None, cols=None, docinfo=None, viewMode=None, tocMode=None): """returns pageinfo with the given parameters""" pageinfo = {} current = getInt(current) @@ -280,106 +652,344 @@ class documentViewer(Folder): pageinfo['cols'] = cols grpsize = cols * rows pageinfo['groupsize'] = grpsize - start = getInt(start, default=(int(current / grpsize) * grpsize +1)) - print "start3:", start + start = getInt(start, default=(math.ceil(float(current)/float(grpsize))*grpsize-(grpsize-1))) + # int(current / grpsize) * grpsize +1)) pageinfo['start'] = start pageinfo['end'] = start + grpsize - if docinfo is not None: + if (docinfo is not None) and ('numPages' in docinfo): np = int(docinfo['numPages']) pageinfo['end'] = min(pageinfo['end'], np) pageinfo['numgroups'] = int(np / grpsize) if np % grpsize > 0: pageinfo['numgroups'] += 1 - + + + pageinfo['viewMode'] = viewMode + pageinfo['tocMode'] = tocMode + pageinfo['query'] = self.REQUEST.get('query',' ') + pageinfo['queryType'] = self.REQUEST.get('queryType',' ') + pageinfo['querySearch'] =self.REQUEST.get('querySearch', 'fulltext') + pageinfo['textPN'] = self.REQUEST.get('textPN','1') + 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) + + if 'tocSize_%s'%tocMode in docinfo: + tocSize = int(docinfo['tocSize_%s'%tocMode]) + tocPageSize = int(pageinfo['tocPageSize']) + # cached toc + + if tocSize%tocPageSize>0: + tocPages=tocSize/tocPageSize+1 + else: + tocPages=tocSize/tocPageSize + pageinfo['tocPN'] = min (tocPages,toc) + + pageinfo['searchPN'] =self.REQUEST.get('searchPN','1') + pageinfo['sn'] =self.REQUEST.get('sn','') + return pageinfo - def text(self,mode,url,pn): - """give text""" - if mode=="texttool": #index.meta with texttool information - (viewerUrl,imagepath,textpath)=parseUrlTextTool(url) + def getSearch(self, pn=1, pageinfo=None, docinfo=None, query=None, queryType=None): + """get search list""" + docpath = docinfo['textURLPath'] + url = docinfo['url'] + logging.debug("documentViewer (gettoc) docpath: %s"%(docpath)) + logging.debug("documentViewer (gettoc) url: %s"%(url)) + pagesize = pageinfo['queryPageSize'] + pn = pageinfo['searchPN'] + sn = pageinfo['sn'] + highlightQuery = pageinfo['highlightQuery'] + query =pageinfo['query'] + queryType =pageinfo['queryType'] + viewMode= pageinfo['viewMode'] + tocMode = pageinfo['tocMode'] + tocPN = pageinfo['tocPN'] + selfurl = self.absolute_url() + page=self.template.fulltextclient.eval("/mpdl/interface/doc-query.xql","document=%s&mode=%s&queryType=%s&query=%s&queryResultPageSize=%s&queryResultPN=%s&sn=%s&viewMode=%s&highlightQuery=%s"%(docpath, 'text', queryType, query, pagesize, pn, sn, viewMode,highlightQuery) ,outputUnicode=False) + pagexml = page.replace('?document=%s'%str(docpath),'?url=%s'%url) + pagedom = Parse(pagexml) + if (queryType=="fulltext")or(queryType=="xpath")or(queryType=="xquery")or(queryType=="fulltextMorphLemma"): + pagedivs = pagedom.xpath("//div[@class='queryResultPage']") + if len(pagedivs)>0: + pagenode=pagedivs[0] + links=pagenode.xpath("//a") + for l in links: + hrefNode = l.getAttributeNodeNS(None, u"href") + if hrefNode: + href = hrefNode.nodeValue + if href.startswith('page-fragment.xql'): + selfurl = self.absolute_url() + pagexml=href.replace('mode=text','mode=texttool&viewMode=%s&queryType=%s&query=%s&queryResultPageSize=%s&queryResultPN=%s&tocMode=%s&searchPN=%s&tocPN=%s'%(viewMode,queryType,query,pagesize,pn,tocMode,pn,tocPN)) + hrefNode.nodeValue = pagexml.replace('page-fragment.xql','%s'%selfurl) + return serializeNode(pagenode) + + if (queryType=="fulltextMorph"): + pagedivs = pagedom.xpath("//div[@class='queryResult']") + if len(pagedivs)>0: + pagenode=pagedivs[0] + links=pagenode.xpath("//a") + for l in links: + hrefNode = l.getAttributeNodeNS(None, u"href") + if hrefNode: + href = hrefNode.nodeValue + if href.startswith('page-fragment.xql'): + selfurl = self.absolute_url() + pagexml=href.replace('mode=text','mode=texttool&viewMode=%s&queryType=%s&query=%s&queryResultPageSize=%s&queryResultPN=%s&tocMode=%s&searchPN=%s&tocPN=%s'%(viewMode,queryType,query,pagesize,pn,tocMode,pn,tocPN)) + hrefNode.nodeValue = pagexml.replace('page-fragment.xql','%s'%selfurl) + if href.startswith('../lt/lemma.xql'): + + hrefNode.nodeValue = href.replace('../lt/lemma.xql','%s/template/head_main_lemma'%selfurl) + l.setAttributeNS(None, 'target', '_blank') + l.setAttributeNS(None, 'onClick',"popupWin = window.open(this.href, 'contacts', 'location,width=500,height=600,top=180, left=400, scrollbars=1'); return false;") + l.setAttributeNS(None, 'onDblclick', 'popupWin.focus();') + pagedivs = pagedom.xpath("//div[@class='queryResultMorphExpansion']") + return serializeNode(pagenode) + + if (queryType=="ftIndex")or(queryType=="ftIndexMorph"): + pagedivs= pagedom.xpath("//div[@class='queryResultPage']") + if len(pagedivs)>0: + pagenode=pagedivs[0] + links=pagenode.xpath("//a") + for l in links: + hrefNode = l.getAttributeNodeNS(None, u"href") + if hrefNode: + href = hrefNode.nodeValue + hrefNode.nodeValue=href.replace('mode=text','mode=texttool&viewMode=%s&tocMode=%s&tocPN=%s&pn=%s'%(viewMode,tocMode,tocPN,pn)) + + if href.startswith('../lt/lex.xql'): + hrefNode.nodeValue = href.replace('../lt/lex.xql','%s/template/head_main_voc'%selfurl) + l.setAttributeNS(None, 'target', '_blank') + l.setAttributeNS(None, 'onClick',"popupWin = window.open(this.href, 'contacts', 'location,width=500,height=600,top=180, left=400, scrollbars=1'); return false;") + l.setAttributeNS(None, 'onDblclick', 'popupWin.focus();') + if href.startswith('../lt/lemma.xql'): + hrefNode.nodeValue = href.replace('../lt/lemma.xql','%s/template/head_main_lemma'%selfurl) + l.setAttributeNS(None, 'target', '_blank') + l.setAttributeNS(None, 'onClick',"popupWin = window.open(this.href, 'contacts', 'location,width=500,height=600,top=180, left=400, scrollbars=1'); return false;") + l.setAttributeNS(None, 'onDblclick', 'popupWin.focus();') + return serializeNode(pagenode) + return "no text here" + + def getNumPages(self,docinfo=None): + """get list of pages from fulltext and put in docinfo""" + xquery = '//pb' + text = self.template.fulltextclient.eval("/mpdl/interface/xquery.xql", "document=%s&xquery=%s"%(docinfo['textURLPath'],xquery)) + # TODO: better processing of the page list. do we need the info somewhere else also? + docinfo['numPages'] = text.count(" 0: + pagenode = pagedivs[0] + links = pagenode.xpath("//a") + for l in links: + hrefNode = l.getAttributeNodeNS(None, u"href") + + if hrefNode: + href= hrefNode.nodeValue + if href.startswith('#note-'): + hrefNode.nodeValue = href.replace('#note-',"?url=%s&viewMode=%s&tocMode=%s&tocPN=%s&pn=%s#note-"%(url,viewMode,tocMode,tocPN,pn)) + + return serializeNode(pagenode) + if mode == "xml": + # first div contains text + pagedivs = pagedom.xpath("/div") + if len(pagedivs) > 0: + pagenode = pagedivs[0] + return serializeNode(pagenode) + if mode == "pureXml": + # first div contains text + pagedivs = pagedom.xpath("/div") + if len(pagedivs) > 0: + pagenode = pagedivs[0] + return serializeNode(pagenode) + # text-with-links mode + if mode == "text_dict": + # first div contains text + pagedivs = pagedom.xpath("/div") + if len(pagedivs) > 0: + pagenode = pagedivs[0] + # check all a-tags + links = pagenode.xpath("//a") + for l in links: + hrefNode = l.getAttributeNodeNS(None, u"href") + if hrefNode: + # is link with href + href = hrefNode.nodeValue + if href.startswith('lt/lex.xql'): + # is pollux link + selfurl = self.absolute_url() + # change href + hrefNode.nodeValue = href.replace('lt/lex.xql','%s/template/head_main_voc'%selfurl) + # add target + l.setAttributeNS(None, 'target', '_blank') + l.setAttributeNS(None, 'onClick',"popupWin = window.open(this.href, 'contacts', 'location,width=500,height=600,top=180, left=700, scrollbars=1'); return false;") + l.setAttributeNS(None, 'onDblclick', 'popupWin.focus();') + + if href.startswith('lt/lemma.xql'): + selfurl = self.absolute_url() + hrefNode.nodeValue = href.replace('lt/lemma.xql','%s/template/head_main_lemma'%selfurl) + l.setAttributeNS(None, 'target', '_blank') + l.setAttributeNS(None, 'onClick',"popupWin = window.open(this.href, 'contacts', 'location,width=500,height=600,top=180, left=700, scrollbars=1'); return false;") + l.setAttributeNS(None, 'onDblclick', 'popupWin.focus();') + + if href.startswith('#note-'): + hrefNode.nodeValue = href.replace('#note-',"?url=%s&viewMode=%s&tocMode=%s&tocPN=%s&pn=%s#note-"%(url,viewMode,tocMode,tocPN,pn)) + + + return serializeNode(pagenode) + + return "no text here" + + def getTranslate(self, query=None, language=None): + """translate into another languages""" + pagexml=self.template.fulltextclient.eval("/mpdl/interface/lt/lex.xql","document=&language="+str(language)+"&query="+url_quote(str(query))) + return pagexml - list=[] - nodes=dom.xpath("//pb") - - node=nodes[int(pn)-1] - - p=node - - while p.tagName!="p": - p=p.parentNode - - - endNode=nodes[int(pn)] - - - e=endNode - - while e.tagName!="p": - e=e.parentNode - - - next=node.parentNode - - #sammle s - while next and (next!=endNode.parentNode): - list.append(next) - next=next.nextSibling - list.append(endNode.parentNode) - - if p==e:# beide im selben paragraphen - pass -# else: -# next=p -# while next!=e: -# print next,e -# list.append(next) -# next=next.nextSibling -# -# for x in list: -# PrettyPrint(x) -# -# return list -# + def getLemma(self, lemma=None, language=None): + """simular words lemma """ + pagexml=self.template.fulltextclient.eval("/mpdl/interface/lt/lemma.xql","document=&language="+str(language)+"&lemma="+url_quote(str(lemma))) + #pagexml=self.template.fulltextclient.eval("/mpdl/interface/lt/lemma.xql","lemma=%s&language=%s"%(lemma,language),outputUnicode=False) + return pagexml + + def getQuery (self, docinfo=None, pageinfo=None, query=None, queryType=None, pn=1): + """number of""" + docpath = docinfo['textURLPath'] + pagesize = pageinfo['queryPageSize'] + pn = pageinfo['searchPN'] + query =pageinfo['query'] + queryType =pageinfo['queryType'] + + tocSearch = 0 + tocDiv = None + pagexml=self.template.fulltextclient.eval("/mpdl/interface/doc-query.xql","document=%s&mode=%s&queryType=%s&query=%s&queryResultPageSize=%s&queryResultPN=%s"%(docpath, 'text', queryType, query, pagesize, pn) ,outputUnicode=False) + + pagedom = Parse(pagexml) + numdivs = pagedom.xpath("//div[@class='queryResultHits']") + tocSearch = int(getTextFromNode(numdivs[0])) + tc=int((tocSearch/10)+1) + logging.debug("documentViewer (gettoc) tc: %s"%(tc)) + return tc + + def getToc(self, mode="text", docinfo=None): + """loads table of contents and stores in docinfo""" + logging.debug("documentViewer (gettoc) mode: %s"%(mode)) + if mode == "none": + return docinfo + + if 'tocSize_%s'%mode in docinfo: + # cached toc + return docinfo + + docpath = docinfo['textURLPath'] + # we need to set a result set size + pagesize = 1000 + pn = 1 + if mode == "text": + queryType = "toc" + else: + queryType = mode + # number of entries in toc + tocSize = 0 + tocDiv = None + pagexml=self.template.fulltextclient.eval("/mpdl/interface/doc-query.xql", "document=%s&queryType=%s&queryResultPageSize=%s&queryResultPN=%s"%(docpath,queryType,pagesize,pn), outputUnicode=False) + # post-processing downloaded xml + pagedom = Parse(pagexml) + # get number of entries + numdivs = pagedom.xpath("//div[@class='queryResultHits']") + if len(numdivs) > 0: + tocSize = int(getTextFromNode(numdivs[0])) + # div contains text + #pagedivs = pagedom.xpath("//div[@class='queryResultPage']") + #if len(pagedivs) > 0: + # tocDiv = pagedivs[0] - def findDigilibUrl(self): - """try to get the digilib URL from zogilib""" - url = self.imageViewerUrl[:-1] + "/getScalerUrl" - try: - scaler = urlopen(url).read() - return scaler.replace("/servlet/Scaler?", "") - except: - return None + docinfo['tocSize_%s'%mode] = tocSize + #docinfo['tocDiv_%s'%mode] = tocDiv + return docinfo - def changeDocumentViewer(self,imageViewerUrl,title="",digilibBaseUrl=None,thumbrows=2,thumbcols=10,RESPONSE=None): + def getTocPage(self, mode="text", pn=1, pageinfo=None, docinfo=None): + """returns single page from the table of contents""" + # TODO: this should use the cached TOC + if mode == "text": + queryType = "toc" + else: + queryType = mode + docpath = docinfo['textURLPath'] + path = docinfo['textURLPath'] + #logging.debug("documentViewer (gettoc) pathNomer: %s"%(pathNomer)) + pagesize = pageinfo['tocPageSize'] + pn = pageinfo['tocPN'] + url = docinfo['url'] + selfurl = self.absolute_url() + viewMode= pageinfo['viewMode'] + tocMode = pageinfo['tocMode'] + tocPN = pageinfo['tocPN'] + + pagexml=self.template.fulltextclient.eval("/mpdl/interface/doc-query.xql", "document=%s&queryType=%s&queryResultPageSize=%s&queryResultPN=%s"%(docpath,queryType, pagesize, pn), outputUnicode=False) + page = pagexml.replace('page-fragment.xql?document=%s'%str(path),'%s?url=%s&viewMode=%s&tocMode=%s&tocPN=%s'%(selfurl,url, viewMode, tocMode, tocPN)) + text = page.replace('mode=image','mode=texttool') + return text + # post-processing downloaded xml + #pagedom = Parse(text) + # div contains text + #pagedivs = pagedom.xpath("//div[@class='queryResultPage']") + #if len(pagedivs) > 0: + # pagenode = pagedivs[0] + # return serializeNode(pagenode) + #else: + # return "No TOC!" + + def changeDocumentViewer(self,title="",digilibBaseUrl=None,thumbrows=2,thumbcols=5,authgroups='mpiwg',RESPONSE=None): """init document viewer""" self.title=title - self.imageViewerUrl=imageViewerUrl self.digilibBaseUrl = digilibBaseUrl self.thumbrows = thumbrows self.thumbcols = thumbcols - + self.authgroups = [s.strip().lower() for s in authgroups.split(',')] if RESPONSE is not None: RESPONSE.redirect('manage_main') - -# security.declareProtected('View management screens','renameImageForm') - def manage_AddDocumentViewerForm(self): """add the viewer form""" pt=PageTemplateFile('zpt/addDocumentViewer', globals()).__of__(self) return pt() -def manage_AddDocumentViewer(self,id,imageViewerUrl="",title="",RESPONSE=None): +def manage_AddDocumentViewer(self,id,imageScalerUrl="",textServerName="",title="",RESPONSE=None): """add the viewer""" - newObj=documentViewer(id,imageViewerUrl,title) + newObj=documentViewer(id,imageScalerUrl=imageScalerUrl,title=title,textServerName=textServerName) self._setObject(id,newObj) if RESPONSE is not None: @@ -405,7 +1015,9 @@ def manage_addDocumentViewerTemplate(sel self._setObject(id, DocumentViewerTemplate(id)) ob = getattr(self, id) - ob.pt_edit(open(os.path.join(package_home(globals()),'zpt/viewer_main.zpt')).read(),None) + txt=file(os.path.join(package_home(globals()),'zpt/viewer_main.zpt'),'r').read() + logging.info("txt %s:"%txt) + ob.pt_edit(txt,"text/html") if title: ob.pt_setTitle(title) try: @@ -418,4 +1030,4 @@ def manage_addDocumentViewerTemplate(sel return '' - \ No newline at end of file +