--- documentViewer/documentViewer.py 2010/06/14 10:49:48 1.69.2.1 +++ documentViewer/documentViewer.py 2010/08/12 10:08:24 1.78 @@ -2,32 +2,20 @@ from OFS.Folder import Folder from Products.PageTemplates.ZopePageTemplate import ZopePageTemplate 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, Parse - -from xml.dom.minidom import parse, parseString - -from extraFunction import * - - -import Ft.Xml.XPath -import cStringIO -import xmlrpclib +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 def logger(txt,method,txt2): """logging""" @@ -55,7 +43,7 @@ def getTextFromNode(nodename): def serializeNode(node, encoding='utf-8'): """returns a string containing node as XML""" buf = cStringIO.StringIO() - Print(node, stream=buf, encoding=encoding) + Ft.Xml.Domlette.Print(node, stream=buf, encoding=encoding) s = buf.getvalue() buf.close() return s @@ -66,23 +54,56 @@ def getParentDir(path): 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 ## ## documentViewer class ## -class documentViewer(Folder, extraFunction): +class documentViewer(Folder): """document viewer""" - #textViewerUrl="http://127.0.0.1:8080/HFQP/testXSLT/getPage?" - meta_type="Document viewer" security=ClassSecurityInfo() @@ -98,6 +119,7 @@ class documentViewer(Folder, extraFuncti 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()) @@ -123,12 +145,12 @@ class documentViewer(Folder, extraFuncti #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) + import MpdlXmlTextServer + textServer = MpdlXmlTextServer.MpdlXmlTextServer(id='fulltextclient',serverName=textServerName) #templateFolder['fulltextclient'] = xmlRpcClient - templateFolder._setObject('fulltextclient',xmlRpcClient) + templateFolder._setObject('fulltextclient',textServer) except Exception, e: - logging.error("Unable to create XMLRpcTools for fulltextclient: "+str(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") @@ -137,7 +159,41 @@ class documentViewer(Folder, extraFuncti 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 getSearch(self, **args): + """get search""" + return self.template.fulltextclient.getSearch(**args) + + def getNumPages(self, docinfo): + """get numpages""" + return self.template.fulltextclient.getNumPages(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): ''' @@ -162,7 +218,7 @@ class documentViewer(Folder, extraFuncti 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" @@ -170,13 +226,14 @@ class documentViewer(Folder, extraFuncti 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): + def index_html(self,url,mode="texttool",viewMode="auto",tocMode="thumbs",start=None,pn=1,mk=None, query=None, querySearch=None, characterNormalization=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 characterNormalization type of text display (reg, norm, none) @param querySearch: type of different search modes (fulltext, fulltextMorph, xpath, xquery, ftIndex, ftIndexMorph, fulltextMorphLemma) ''' @@ -188,17 +245,16 @@ class documentViewer(Folder, extraFuncti return "ERROR: template folder missing!" if not getattr(self, 'digilibBaseUrl', None): - self.digilibBaseUrl = self.findDigilibUrl() or "http://nausikaa.mpiwg-berlin.mpg.de/digitallibrary" + self.digilibBaseUrl = self.findDigilibUrl() or "http://digilib.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) if viewMode=="auto": # automodus gewaehlt - if docinfo.get("textURL",''): #texturl gesetzt und textViewer konfiguriert + if docinfo.has_key('textURL') or docinfo.has_key('textURLPath'): #texturl gesetzt und textViewer konfiguriert viewMode="text_dict" else: viewMode="images" @@ -212,8 +268,8 @@ class documentViewer(Folder, extraFuncti ret="" if mk is None: return "" - if type(mk) is not ListType: - mk=[mk] + if not isinstance(mk, list): + mk=[mk] for m in mk: ret+="mk=%s"%m return ret @@ -251,7 +307,8 @@ class documentViewer(Folder, extraFuncti 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 @@ -266,7 +323,7 @@ class documentViewer(Folder, extraFuncti 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 @@ -285,26 +342,26 @@ class documentViewer(Folder, extraFuncti 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 = {} @@ -314,21 +371,15 @@ class documentViewer(Folder, extraFuncti 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])) @@ -342,7 +393,6 @@ class documentViewer(Folder, extraFuncti def getIndexMeta(self, url): """returns dom of index.meta document at url""" - num_retries = 3 dom = None metaUrl = None if url.startswith("http://"): @@ -354,25 +404,17 @@ class documentViewer(Folder, extraFuncti 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: + logging.debug("(getIndexMeta): 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://"): @@ -383,25 +425,17 @@ class documentViewer(Folder, extraFuncti 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 @@ -477,7 +511,7 @@ class documentViewer(Folder, extraFuncti def getDocinfoFromTextTool(self, url, dom=None, docinfo=None): """parse texttool tag in index meta""" - logger("documentViewer (getdocinfofromtexttool)", logging.INFO, "url: %s" % (url)) + logging.debug("documentViewer (getdocinfofromtexttool) url: %s" % (url)) if docinfo is None: docinfo = {} if docinfo.get('lang', None) is None: @@ -492,7 +526,7 @@ class documentViewer(Folder, extraFuncti if archiveNames and (len(archiveNames) > 0): archiveName = getTextFromNode(archiveNames[0]) else: - logger("documentViewer (getdocinfofromtexttool)", logging.WARNING, "resource/name missing in: %s" % (url)) + logging.warning("documentViewer (getdocinfofromtexttool) resource/name missing in: %s" % (url)) archivePaths = dom.xpath("//resource/archive-path") if archivePaths and (len(archivePaths) > 0): @@ -504,7 +538,7 @@ class documentViewer(Folder, extraFuncti archivePath += "/" + archiveName else: # try to get archive-path from url - logger("documentViewer (getdocinfofromtexttool)", logging.WARNING, "resource/archive-path missing in: %s" % (url)) + logging.warning("documentViewer (getdocinfofromtexttool) resource/archive-path missing in: %s" % (url)) if (not url.startswith('http')): archivePath = url.replace('index.meta', '') @@ -537,7 +571,8 @@ class documentViewer(Folder, extraFuncti 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]) @@ -549,13 +584,14 @@ class documentViewer(Folder, extraFuncti 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.getNumPages(docinfo) #im moment einfach auf eins setzen, navigation ueber die thumbs geht natuerlich nicht + docinfo = self.getNumPages(docinfo) presentationUrls = dom.xpath("//texttool/presentation") docinfo = self.getBibinfoFromIndexMeta(url, docinfo=docinfo, dom=dom) # get info von bib tag @@ -596,7 +632,7 @@ class documentViewer(Folder, extraFuncti 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","") @@ -606,7 +642,7 @@ class documentViewer(Folder, extraFuncti 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 @@ -618,13 +654,13 @@ class documentViewer(Folder, extraFuncti 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} @@ -635,14 +671,14 @@ class documentViewer(Folder, extraFuncti elif mode=="filepath": docinfo = self.getDocinfoFromImagePath(url, docinfo=docinfo,cut=1) else: - logger("documentViewer (getdocinfo)", logging.ERROR,"unknown 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) self.REQUEST.SESSION['docinfo'] = docinfo return docinfo - def getPageinfo(self, current, start=None, rows=None, cols=None, docinfo=None, viewMode=None, tocMode=None): + def getPageinfo(self, current, start=None, rows=None, cols=None, docinfo=None, viewMode=None, tocMode=None, characterNormalization=None): """returns pageinfo with the given parameters""" pageinfo = {} current = getInt(current) @@ -665,6 +701,7 @@ class documentViewer(Folder, extraFuncti pageinfo['numgroups'] += 1 pageinfo['viewMode'] = viewMode pageinfo['tocMode'] = tocMode + pageinfo['characterNormalization'] =characterNormalization pageinfo['query'] = self.REQUEST.get('query',' ') pageinfo['queryType'] = self.REQUEST.get('queryType',' ') pageinfo['querySearch'] =self.REQUEST.get('querySearch', 'fulltext')