--- documentViewer/documentViewer.py 2009/03/09 08:37:19 1.39 +++ documentViewer/documentViewer.py 2010/10/28 15:44:20 1.147 @@ -1,27 +1,25 @@ - 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.Domlette import NonvalidatingReader -from Ft.Xml.Domlette import PrettyPrint, Print from Ft.Xml import EMPTY_NAMESPACE, Parse - -import Ft.Xml.XPath - +import Ft.Xml.Domlette import os.path import sys -import cgi import urllib +import urllib2 import logging import math - import urlparse -from types import * +import cStringIO +import re + + def logger(txt,method,txt2): """logging""" logging.info(txt+ txt2) @@ -45,20 +43,73 @@ 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() + Ft.Xml.Domlette.Print(node, stream=buf, encoding=encoding) + s = buf.getvalue() + buf.close() + return s + +def getBrowserType(self): + """get browser type object""" + if self.REQUEST.SESSION.has_key('browserType'): + return self.REQUEST.SESSION['browserType'] + else: + bt = browserCheck(self) + self.REQUEST.SESSION.set('browserType', bt) + logging.debug("documentViewer (BROWSER TYPE) bt %s"%bt) + return bt + + def getParentDir(path): """returns pathname shortened by one""" return '/'.join(path.split('/')[0:-1]) -import socket +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 -def urlopen(url,timeout=2): - """urlopen mit timeout""" - socket.setdefaulttimeout(timeout) - ret=urllib.urlopen(url) - socket.setdefaulttimeout(5) - return ret ## @@ -66,8 +117,6 @@ def urlopen(url,timeout=2): ## class documentViewer(Folder): """document viewer""" - #textViewerUrl="http://127.0.0.1:8080/HFQP/testXSLT/getPage?" - meta_type="Document viewer" security=ClassSecurityInfo() @@ -77,36 +126,107 @@ 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_gis =PageTemplateFile ('zpt/page_main_gis', 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,textViewerUrl=None,title="",digilibBaseUrl=None,thumbcols=2,thumbrows=10,authgroups="mpiwg"): + 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 - self.textViewerUrl=textViewerUrl - - 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') - + # 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: + import MpdlXmlTextServer + textServer = MpdlXmlTextServer.MpdlXmlTextServer(id='fulltextclient',serverName=textServerName) + #templateFolder['fulltextclient'] = xmlRpcClient + 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") + #templateFolder['zogilib'] = zogilib + templateFolder._setObject('zogilib',zogilib) + except Exception, e: + logging.error("Unable to create zogiLib for zogilib: "+str(e)) + + + # proxy text server methods to fulltextclient + def getTextPage(self, **args): + """get page""" + return self.template.fulltextclient.getTextPage(**args) + + def getQuery(self, **args): + """get query""" + return self.template.fulltextclient.getQuery(**args) + + def getPDF(self, **args): + """get query""" + return self.template.fulltextclient.getPDF(**args) + + def getSearch(self, **args): + """get search""" + return self.template.fulltextclient.getSearch(**args) + + def getGisPlaces(self, **args): + """get gis places""" + return self.template.fulltextclient.getGisPlaces(**args) + + def getAllGisPlaces(self, **args): + """get all gis places """ + return self.template.fulltextclient.getAllGisPlaces(**args) + + def getOrigPages(self, **args): + """get original page number """ + return self.template.fulltextclient.getOrigPages(**args) + + def getNumPages(self, docinfo): + """get numpages""" + return self.template.fulltextclient.getNumPages(docinfo) + + def getNumTextPages(self, docinfo): + """get numpages text""" + return self.template.fulltextclient.getNumTextPages(docinfo) + + def getTranslate(self, **args): + """get translate""" + return self.template.fulltextclient.getTranslate(**args) + + def getLemma(self, **args): + """get lemma""" + return self.template.fulltextclient.getLemma(**args) + + def getToc(self, **args): + """get toc""" + return self.template.fulltextclient.getToc(**args) + + def getTocPage(self, **args): + """get tocpage""" + return self.template.fulltextclient.getTocPage(**args) + security.declareProtected('View','thumbs_rss') def thumbs_rss(self,mode,url,viewMode="auto",start=None,pn=1): ''' @@ -116,7 +236,7 @@ class documentViewer(Folder): @param viewMode: if images display images, if text display text, default is images (text,images or auto) ''' - logging.info("HHHHHHHHHHHHHH:load the rss") + 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'): @@ -127,11 +247,13 @@ class documentViewer(Folder): 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) + #pageinfo = self.getPageinfo(start=start,current=pn,docinfo=docinfo) + pageinfo = self.getPageinfo(start=start,current=pn, docinfo=docinfo) + ''' ZDES ''' pt = getattr(self.template, 'thumbs_main_rss') if viewMode=="auto": # automodus gewaehlt - if docinfo.get("textURL",'') and self.textViewerUrl: #texturl gesetzt und textViewer konfiguriert + if docinfo.has_key("textURL") or docinfo.has_key('textURLPath'): #texturl gesetzt und textViewer konfiguriert viewMode="text" else: viewMode="images" @@ -139,46 +261,71 @@ class documentViewer(Folder): return pt(docinfo=docinfo,pageinfo=pageinfo,viewMode=viewMode) security.declareProtected('View','index_html') - def index_html(self,mode,url,viewMode="auto",start=None,pn=1,mk=None): + def index_html(self,url,mode="texttool",viewMode="auto",tocMode="thumbs",start=None,pn=1,mk=None, query=None, querySearch=None, characterNormalization=""): ''' 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) - + @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 characterNormalization type of text display (reg, norm, none) + @param querySearch: type of different search modes (fulltext, fulltextMorph, xpath, xquery, ftIndex, ftIndexMorph, fulltextMorphLemma) ''' - 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 - self.manage_addFolder('template') + # this won't work + logging.error("template folder missing!") + return "ERROR: template folder missing!" - if not self.digilibBaseUrl: - self.digilibBaseUrl = self.findDigilibUrl() or "http://nausikaa.mpiwg-berlin.mpg.de/digitallibrary" + if not getattr(self, 'digilibBaseUrl', None): + self.digilibBaseUrl = self.findDigilibUrl() or "http://digilib.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 tocMode != "thumbs": + # get table of contents + docinfo = self.getToc(mode=tocMode, docinfo=docinfo) + if viewMode=="auto": # automodus gewaehlt - if docinfo.get("textURL",'') and self.textViewerUrl: #texturl gesetzt und textViewer konfiguriert - viewMode="text" + if docinfo.has_key('textURL') or docinfo.has_key('textURLPath'): #texturl gesetzt und textViewer konfiguriert + viewMode="text_dict" else: viewMode="images" - + + pageinfo = self.getPageinfo(start=start,current=pn, docinfo=docinfo,viewMode=viewMode,tocMode=tocMode) + + 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] + if mk is None: + return "" + if not isinstance(mk, list): + 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.""" + #logger("documentViewer (getstyle)", logging.INFO, "idx: %s selected: %s style: %s"%(idx,selected,style)) + if idx == selected: + return style + 'sel' + else: + return style def getLink(self,param=None,val=None): """link to documentviewer with parameter param set to val""" @@ -189,12 +336,14 @@ class documentViewer(Folder): del params[param] else: params[param] = str(val) - if params["mode"] == "filepath": #wenn beim erst Aufruf filepath gesetzt wurde aendere das nun zu imagepath + + 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()]) + #ps = "&".join(["%s=%s"%(k,urllib.quote(v)) for (k, v) in params.items()]) + ps = urllib.urlencode(params) url=self.REQUEST['URL1']+"?"+ps return url @@ -209,10 +358,11 @@ class documentViewer(Folder): params[param] = str(val) # quote values and assemble into query string - logging.info("XYXXXXX: %s"%repr(params.items())) + logging.debug("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""" @@ -224,38 +374,29 @@ class documentViewer(Folder): 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 - - 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) + logging.debug("documentViewer (accessOK) access type %s"%access) if access is not None and access == 'free': - logger("documentViewer (accessOK)", logging.INFO, "access is free") + logging.debug("documentViewer (accessOK) access is free") return True elif access is None or access in self.authgroups: # only local access -- only logged in users user = getSecurityManager().getUser() + logging.debug("documentViewer (accessOK) user=%s ip=%s"%(user,self.REQUEST.getClientAddr())) if user is not None: #print "user: ", user return (user.getUserName() != "Anonymous User") else: return False - logger("documentViewer (accessOK)", logging.INFO, "unknown access type %s"%access) + logging.error("documentViewer (accessOK) unknown access type %s"%access) return False def getDirinfoFromDigilib(self,path,docinfo=None,cut=0): """gibt param von dlInfo aus""" - num_retries = 3 if docinfo is None: docinfo = {} @@ -265,34 +406,39 @@ class documentViewer(Folder): infoUrl=self.digilibBaseUrl+"/dirInfo-xml.jsp?mo=dir&fn="+path - logger("documentViewer (getparamfromdigilib)", logging.INFO, "dirInfo from %s"%(infoUrl)) + logging.debug("documentViewer (getparamfromdigilib) dirInfo from %s"%(infoUrl)) - 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: + txt = getHttpData(infoUrl) + if txt is None: raise IOError("Unable to get dir-info from %s"%(infoUrl)) - + + dom = Parse(txt) sizes=dom.xpath("//dir/size") - logger("documentViewer (getparamfromdigilib)", logging.INFO, "dirInfo:size"%sizes) + logging.debug("documentViewer (getparamfromdigilib) dirInfo:size"%sizes) 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 + def getIndexMetaPath(self,url): + """gib nur den Pfad zurueck""" + regexp = re.compile(r".*(experimental|permanent)/(.*)") + regpath = regexp.match(url) + if (regpath==None): + return "" + logging.debug("(getDomFromIndexMeta): URLXAXA: %s"%regpath.group(2)) + return ("/mpiwg/online/"+regpath.group(1)+"/"+regpath.group(2)) + + + + def getIndexMetaUrl(self,url): + """returns utr of index.meta document at url""" + metaUrl = None if url.startswith("http://"): # real URL @@ -303,25 +449,24 @@ class documentViewer(Folder): metaUrl=server+url.replace("/mpiwg/online","") if not metaUrl.endswith("index.meta"): metaUrl += "/index.meta" - logging.debug("METAURL: %s"%metaUrl) - for cnt in range(num_retries): - try: - # patch dirk encoding fehler treten dann nicht mehr auf - # dom = NonvalidatingReader.parseUri(metaUrl) - txt=urllib.urlopen(metaUrl).read() - dom = Parse(txt) - break - except: - logger("ERROR documentViewer (getIndexMata)", logging.INFO,"%s (%s)"%sys.exc_info()[0:2]) + + return metaUrl + + def getDomFromIndexMeta(self, url): + """get dom from index meta""" + dom = None + metaUrl = self.getIndexMetaUrl(url) - if dom is None: + logging.debug("(getDomFromIndexMeta): METAURL: %s"%metaUrl) + txt=getHttpData(metaUrl) + if txt is None: raise IOError("Unable to read index meta from %s"%(url)) - + + dom = Parse(txt) return dom def getPresentationInfoXML(self, url): """returns dom of info.xml document at url""" - num_retries = 3 dom = None metaUrl = None if url.startswith("http://"): @@ -331,27 +476,18 @@ class documentViewer(Folder): # 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: + txt=getHttpData(metaUrl) + if txt is None: raise IOError("Unable to read infoXMLfrom %s"%(url)) - + + dom = Parse(txt) return dom def getAuthinfoFromIndexMeta(self,path,docinfo=None,dom=None,cut=0): """gets authorization info from the index.meta file at path or given by dom""" - logger("documentViewer (getauthinfofromindexmeta)", logging.INFO,"path: %s"%(path)) + logging.debug("documentViewer (getauthinfofromindexmeta) path: %s"%(path)) access = None @@ -361,7 +497,7 @@ class documentViewer(Folder): if dom is None: for x in range(cut): path=getParentDir(path) - dom = self.getIndexMeta(path) + dom = self.getDomFromIndexMeta(path) acctype = dom.xpath("//access-conditions/access/@type") if acctype and (len(acctype)>0): @@ -383,7 +519,9 @@ class documentViewer(Folder): if dom is None: for x in range(cut): path=getParentDir(path) - dom = self.getIndexMeta(path) + dom = self.getDomFromIndexMeta(path) + + docinfo['indexMetaPath']=self.getIndexMetaPath(path); logging.debug("documentViewer (getbibinfofromindexmeta cutted) path: %s"%(path)) # put in all raw bib fields as dict "bib" @@ -405,6 +543,8 @@ class documentViewer(Folder): 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: @@ -423,94 +563,125 @@ class documentViewer(Folder): docinfo['lang']='' return docinfo - + + + 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: + 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) - # fix URLs starting with /mpiwg/online - if textUrl.startswith("/mpiwg/online"): - textUrl = textUrl.replace("/mpiwg/online",'',1) - - 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 fuer 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) + if dom is None: + for x in range(cut): + path=getParentDir(path) + dom = self.getDomFromIndexMeta(path) - docinfo = self.getAuthinfoFromIndexMeta(url,docinfo=docinfo,dom=dom) # get access info - return docinfo + docinfo['name']=getTextFromNode(dom.xpath("/resource/name")[0]) + logging.debug("documentViewer docinfo[name] %s"%docinfo['name']) + return docinfo + + def getDocinfoFromTextTool(self, url, dom=None, docinfo=None): + """parse texttool tag in index meta""" + logging.debug("documentViewer (getdocinfofromtexttool) 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.getDomFromIndexMeta(url) + + archivePath = None + archiveName = None + + 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)) + + 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 + logging.warning("documentViewer (getdocinfofromtexttool) 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 + + # old style text URL + 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 + + # new style text-url-path + 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.getNumTextPages(docinfo) + + presentationUrls = dom.xpath("//texttool/presentation") + docinfo = self.getBibinfoFromIndexMeta(url, docinfo=docinfo, dom=dom) # get info von bib tag + docinfo = self.getNameFromIndexMeta(url, docinfo=docinfo, dom=dom) + + + 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 getBibinfoFromTextToolPresentation(self,url,docinfo=None,dom=None): @@ -533,7 +704,7 @@ class documentViewer(Folder): def getDocinfoFromImagePath(self,path,docinfo=None,cut=0): """path ist the path to the images it assumes that the index.meta file is one level higher.""" - logger("documentViewer (getdocinfofromimagepath)", logging.INFO,"path: %s"%(path)) + logging.debug("documentViewer (getdocinfofromimagepath) path: %s"%(path)) if docinfo is None: docinfo = {} path=path.replace("/mpiwg/online","") @@ -543,7 +714,7 @@ class documentViewer(Folder): pathorig=path for x in range(cut): path=getParentDir(path) - logging.error("PATH:"+path) + logging.debug("documentViewer (getdocinfofromimagepath) PATH:"+path) imageUrl=self.digilibBaseUrl+"/servlet/Scaler?fn="+path docinfo['imageURL'] = imageUrl @@ -555,13 +726,13 @@ class documentViewer(Folder): def getDocinfo(self, mode, url): """returns docinfo depending on mode""" - logger("documentViewer (getdocinfo)", logging.INFO,"mode: %s, url: %s"%(mode,url)) + logging.debug("documentViewer (getdocinfo) 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) + logging.debug("documentViewer (getdocinfo) docinfo in session: %s"%docinfo) return docinfo # new docinfo docinfo = {'mode': mode, 'url': url} @@ -572,18 +743,19 @@ class documentViewer(Folder): elif mode=="filepath": docinfo = self.getDocinfoFromImagePath(url, docinfo=docinfo,cut=1) else: - logger("documentViewer (getdocinfo)", logging.ERROR,"unknown mode!") - raise ValueError("Unknown mode %s"%(mode)) + logging.error("documentViewer (getdocinfo) unknown mode: %s!"%mode) + raise ValueError("Unknown mode %s! Has to be one of 'texttool','imagepath','filepath'."%(mode)) - logger("documentViewer (getdocinfo)", logging.INFO,"docinfo: %s"%docinfo) + logging.debug("documentViewer (getdocinfo) docinfo: %s"%docinfo) + #logging.debug("documentViewer (getdocinfo) docinfo: %s"%) 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,characterNormalization=""): """returns pageinfo with the given parameters""" pageinfo = {} current = getInt(current) + pageinfo['current'] = current rows = int(rows or self.thumbrows) pageinfo['rows'] = rows @@ -595,120 +767,66 @@ class documentViewer(Folder): # 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 - - 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 - try: - dom = NonvalidatingReader.parseUri(textpath) - except: - return None - - list=[] - nodes=dom.xpath("//pb") - - node=nodes[int(pn)-1] - - p=node - - while p.tagName!="p": - p=p.parentNode - - - endNode=nodes[int(pn)] - + pageinfo['numgroups'] += 1 + pageinfo['viewMode'] = viewMode + pageinfo['tocMode'] = tocMode + #pageinfo ['originalPage'] = self.getOrigPages(docinfo=None, pageinfo=None) + pageinfo['characterNormalization'] = self.REQUEST.get('characterNormalization','') + 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) - 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 findDigilibUrl(self): - """try to get the digilib URL from zogilib""" - url = self.imageViewerUrl[:-1] + "/getScalerUrl" - #print urlparse.urlparse(url)[0] - #print urlparse.urljoin(self.absolute_url(),url) - logging.info("finddigiliburl: %s"%urlparse.urlparse(url)[0]) - logging.info("finddigiliburl: %s"%urlparse.urljoin(self.absolute_url(),url)) - - try: - if urlparse.urlparse(url)[0]=='': #relative path - url=urlparse.urljoin(self.absolute_url()+"/",url) - - scaler = urlopen(url).read() - return scaler.replace("/servlet/Scaler?", "") - except: - return None + 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 changeDocumentViewer(self,imageViewerUrl,textViewerUrl,title="",digilibBaseUrl=None,thumbrows=2,thumbcols=10,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.imageViewerUrl=imageViewerUrl - self.textViewerUrl=textViewerUrl 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="",textViewerUrl="",title="",RESPONSE=None): +def manage_AddDocumentViewer(self,id,imageScalerUrl="",textServerName="",title="",RESPONSE=None): """add the viewer""" - newObj=documentViewer(id,imageViewerUrl,title=title,textViewerUrl=textViewerUrl) + newObj=documentViewer(id,imageScalerUrl=imageScalerUrl,title=title,textServerName=textServerName) self._setObject(id,newObj) if RESPONSE is not None: RESPONSE.redirect('manage_main') - -## ## DocumentViewerTemplate class -## class DocumentViewerTemplate(ZopePageTemplate): """Template for document viewer""" meta_type="DocumentViewer Template"