Mercurial > hg > documentViewer
diff documentViewer.py @ 453:beb7ccb92564 elementtree
first version using elementtree instead of 4suite xml
| author | casties |
|---|---|
| date | Thu, 14 Jul 2011 19:43:56 +0200 |
| parents | b8fb4c750d74 |
| children | 73e3273c7624 |
line wrap: on
line diff
--- a/documentViewer.py Wed Nov 24 17:01:26 2010 +0100 +++ b/documentViewer.py Thu Jul 14 19:43:56 2011 +0200 @@ -7,8 +7,11 @@ from Globals import package_home from Products.zogiLib.zogiLib import browserCheck -from Ft.Xml import EMPTY_NAMESPACE, Parse -import Ft.Xml.Domlette +#from Ft.Xml import EMPTY_NAMESPACE, Parse +#import Ft.Xml.Domlette + +import xml.etree.ElementTree as ET + import os.path import sys import urllib @@ -16,7 +19,6 @@ import logging import math import urlparse -import cStringIO import re import string @@ -32,23 +34,37 @@ except: return int(default) -def getTextFromNode(nodename): +def getText(node): """get the cdata content of a node""" - if nodename is None: + if node is None: return "" - nodelist=nodename.childNodes - rc = "" - for node in nodelist: - if node.nodeType == node.TEXT_NODE: - rc = rc + node.data - return rc + # ET: + text = node.text or "" + for e in node: + text += gettext(e) + if e.tail: + text += e.tail -def serializeNode(node, encoding='utf-8'): + # 4Suite: + #nodelist=node.childNodes + #text = "" + #for n in nodelist: + # if n.nodeType == node.TEXT_NODE: + # text = text + n.data + + return text + +getTextFromNode = getText + +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() + s = ET.tostring(node) + + # 4Suite: + # stream = cStringIO.StringIO() + # Ft.Xml.Domlette.Print(node, stream=stream, encoding=encoding) + # s = stream.getvalue() + # stream.close() return s def browserCheck(self): @@ -58,17 +74,51 @@ bt['ua'] = ua bt['isIE'] = False bt['isN4'] = False + bt['versFirefox']="" + bt['versIE']="" + bt['versSafariChrome']="" + bt['versOpera']="" + if string.find(ua, 'MSIE') > -1: bt['isIE'] = True else: bt['isN4'] = (string.find(ua, 'Mozilla/4.') > -1) - + # Safari oder Chrome identification + try: + nav = ua[string.find(ua, '('):] + nav1=ua[string.find(ua,')'):] + nav2=nav1[string.find(nav1,'('):] + nav3=nav2[string.find(nav2,')'):] + ie = string.split(nav, "; ")[1] + ie1 =string.split(nav1, " ")[2] + ie2 =string.split(nav3, " ")[1] + ie3 =string.split(nav3, " ")[2] + if string.find(ie3, "Safari") >-1: + bt['versSafariChrome']=string.split(ie2, "/")[1] + except: pass + # IE identification try: nav = ua[string.find(ua, '('):] ie = string.split(nav, "; ")[1] if string.find(ie, "MSIE") > -1: bt['versIE'] = string.split(ie, " ")[1] - except: pass + except:pass + # Firefox identification + try: + nav = ua[string.find(ua, '('):] + nav1=ua[string.find(ua,')'):] + if string.find(ie1, "Firefox") >-1: + nav5= string.split(ie1, "/")[1] + logging.debug("FIREFOX: %s"%(nav5)) + bt['versFirefox']=nav5[0:3] + except:pass + #Opera identification + try: + if string.find(ua,"Opera") >-1: + nav = ua[string.find(ua, '('):] + nav1=nav[string.find(nav,')'):] + bt['versOpera']=string.split(nav1,"/")[2] + except:pass bt['isMac'] = string.find(ua, 'Macintosh') > -1 bt['isWin'] = string.find(ua, 'Windows') > -1 @@ -127,8 +177,6 @@ raise IOError("ERROR fetching HTTP data from %s: %s"%(url,errmsg)) #return None - - ## ## documentViewer class ## @@ -147,6 +195,7 @@ 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_double = PageTemplateFile('zpt/page_main_double', 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()) @@ -196,26 +245,18 @@ """get page""" return self.template.fulltextclient.getTextPage(**args) - def getQuery(self, **args): - """get query""" - return self.template.fulltextclient.getQuery(**args) - - def getQueryResultHits(self, **args): - """get query""" - return self.template.fulltextclient.getQueryResultHits(**args) + def getOrigPages(self, **args): + """get page""" + return self.template.fulltextclient.getOrigPages(**args) - def getQueryResultHitsText(self, **args): - """get query""" - return self.template.fulltextclient.getQueryResultHitsText(**args) - - def getQueryResultHitsFigures(self, **args): - """get query""" - return self.template.fulltextclient.getQueryResultHitsFigures(**args) - - def getPDF(self, **args): - """get query""" - return self.template.fulltextclient.getPDF(**args) + def getOrigPagesNorm(self, **args): + """get page""" + return self.template.fulltextclient.getOrigPagesNorm(**args) + def getQuery(self, **args): + """get query in search""" + return self.template.fulltextclient.getQuery(**args) + def getSearch(self, **args): """get search""" return self.template.fulltextclient.getSearch(**args) @@ -227,19 +268,7 @@ 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) @@ -248,6 +277,14 @@ """get lemma""" return self.template.fulltextclient.getLemma(**args) + def getLemmaQuery(self, **args): + """get query""" + return self.template.fulltextclient.getLemmaQuery(**args) + + def getLex(self, **args): + """get lex""" + return self.template.fulltextclient.getLex(**args) + def getToc(self, **args): """get toc""" return self.template.fulltextclient.getToc(**args) @@ -283,7 +320,7 @@ pt = getattr(self.template, 'thumbs_main_rss') if viewMode=="auto": # automodus gewaehlt - if docinfo.has_key("textURL") or docinfo.has_key('textURLPath'): #texturl gesetzt und textViewer konfiguriert + if docinfo.has_key("textURL") or docinfo.get('textURLPath',None): #texturl gesetzt und textViewer konfiguriert viewMode="text" else: viewMode="images" @@ -319,14 +356,18 @@ docinfo = self.getToc(mode=tocMode, docinfo=docinfo) if viewMode=="auto": # automodus gewaehlt - if docinfo.has_key('textURL') or docinfo.has_key('textURLPath'): #texturl gesetzt und textViewer konfiguriert + if docinfo.has_key('textURL') or docinfo.get('textURLPath',None): #texturl gesetzt und textViewer konfiguriert viewMode="text_dict" else: viewMode="images" pageinfo = self.getPageinfo(start=start,current=pn, docinfo=docinfo,viewMode=viewMode,tocMode=tocMode) - pt = getattr(self.template, 'viewer_main') + if (docinfo.get('textURLPath',None)): + page = self.getTextPage(mode=viewMode, docinfo=docinfo, pageinfo=pageinfo) + pageinfo['textPage'] = page + tt = getattr(self, 'template') + pt = getattr(tt, 'viewer_main') return pt(docinfo=docinfo,pageinfo=pageinfo,viewMode=viewMode,mk=self.generateMarks(mk)) def generateMarks(self,mk): @@ -342,10 +383,9 @@ def getBrowser(self): """getBrowser the version of browser """ - names="" - names = browserCheck(self) - #logging.debug("XXXXXXXXXXXXXXXX: %s"%names) - return names + bt = browserCheck(self) + logging.debug("BROWSER VERSION: %s"%(bt)) + return bt def findDigilibUrl(self): """try to get the digilib URL from zogilib""" @@ -364,41 +404,48 @@ else: return style - def getLink(self,param=None,val=None): - """link to documentviewer with parameter param set to val""" - params=self.REQUEST.form.copy() + def getLink(self, param=None, val=None, params=None, baseUrl=None, paramSep='&'): + """returns URL to documentviewer with parameter param set to val or from dict params""" + # copy existing request params + urlParams=self.REQUEST.form.copy() + # change single param if param is not None: if val is None: - if params.has_key(param): - del params[param] + if urlParams.has_key(param): + del urlParams[param] else: - params[param] = str(val) + urlParams[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"]) + # change more params + if params is not None: + for k in params.keys(): + v = params[k] + if v is None: + # val=None removes param + if urlParams.has_key(k): + del urlParams[k] + + else: + urlParams[k] = v + + # FIXME: does this belong here? + if urlParams.get("mode", None) == "filepath": #wenn beim erst Aufruf filepath gesetzt wurde aendere das nun zu imagepath + urlParams["mode"] = "imagepath" + urlParams["url"] = getParentDir(urlParams["url"]) - # quote values and assemble into query string - #ps = "&".join(["%s=%s"%(k,urllib.quote(v)) for (k, v) in params.items()]) - ps = urllib.urlencode(params) - url=self.REQUEST['URL1']+"?"+ps + # quote values and assemble into query string (not escaping '/') + ps = paramSep.join(["%s=%s"%(k,urllib.quote_plus(v,'/')) for (k, v) in urlParams.items()]) + #ps = urllib.urlencode(urlParams) + if baseUrl is None: + baseUrl = self.REQUEST['URL1'] + + url = "%s?%s"%(baseUrl, ps) return url - def getLinkAmp(self,param=None,val=None): + + def getLinkAmp(self, param=None, val=None, params=None, baseUrl=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.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 + return self.getLink(param, val, params, baseUrl, '&') def getInfo_xml(self,url,mode): """returns info about the document as XML""" @@ -463,12 +510,14 @@ if txt is None: raise IOError("Unable to get dir-info from %s"%(infoUrl)) - dom = Parse(txt) - sizes=dom.xpath("//dir/size") - logging.debug("documentViewer (getparamfromdigilib) dirInfo:size"%sizes) + dom = ET.fromstring(txt) + #dom = Parse(txt) + size=getText(dom.find("size")) + #sizes=dom.xpath("//dir/size") + logging.debug("documentViewer (getparamfromdigilib) dirInfo:size=%s"%size) - if sizes: - docinfo['numPages'] = int(getTextFromNode(sizes[0])) + if size: + docinfo['numPages'] = int(size) else: docinfo['numPages'] = 0 @@ -513,7 +562,8 @@ if txt is None: raise IOError("Unable to read index meta from %s"%(url)) - dom = Parse(txt) + dom = ET.fromstring(txt) + #dom = Parse(txt) return dom def getPresentationInfoXML(self, url): @@ -532,7 +582,8 @@ if txt is None: raise IOError("Unable to read infoXMLfrom %s"%(url)) - dom = Parse(txt) + dom = ET.fromstring(txt) + #dom = Parse(txt) return dom @@ -550,11 +601,14 @@ path=getParentDir(path) dom = self.getDomFromIndexMeta(path) - acctype = dom.xpath("//access-conditions/access/@type") - if acctype and (len(acctype)>0): - access=acctype[0].value - if access in ['group', 'institution']: - access = getTextFromNode(dom.xpath("//access-conditions/access/name")[0]).lower() + acc = dom.find(".//access-conditions/access") + if acc is not None: + acctype = acc.get('type') + #acctype = dom.xpath("//access-conditions/access/@type") + if acctype: + access=acctype + if access in ['group', 'institution']: + access = dom.find(".//access-conditions/access/name").text.lower() docinfo['accessType'] = access return docinfo @@ -576,7 +630,8 @@ logging.debug("documentViewer (getbibinfofromindexmeta cutted) path: %s"%(path)) # put in all raw bib fields as dict "bib" - bib = dom.xpath("//bib/*") + bib = dom.find(".//bib/*") + #bib = dom.xpath("//bib/*") if bib and len(bib)>0: bibinfo = {} for e in bib: @@ -585,10 +640,10 @@ # 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: + bib = dom.find(".//bib") + bibtype=bib.get("type") + #bibtype=dom.xpath("//bib/@type") + if not bibtype: bibtype="generic" bibtype=bibtype.replace("-"," ") # wrong typesiin index meta "-" instead of " " (not wrong! ROC) @@ -597,22 +652,63 @@ 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: + if len(bibmap) > 0 and len(bibmap['author'][0]) > 0 or len(bibmap['title'][0]) > 0: try: - docinfo['author']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['author'][0])[0]) + docinfo['author']=getText(bib.find(bibmap['author'][0])) except: pass try: - docinfo['title']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['title'][0])[0]) + docinfo['title']=getText(bib.find(bibmap['title'][0])) except: pass try: - docinfo['year']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['year'][0])[0]) + docinfo['year']=getText(bib.find(bibmap['year'][0])) except: pass - logging.debug("documentViewer (getbibinfofromindexmeta) using mapping for %s"%bibtype) - try: - docinfo['lang']=getTextFromNode(dom.xpath("//bib/lang")[0]) - except: - docinfo['lang']='' - + + # ROC: why is this here? + # logging.debug("documentViewer (getbibinfofromindexmeta) using mapping for %s"%bibtype) + # try: + # docinfo['lang']=getTextFromNode(dom.find(".//bib/lang")[0]) + # except: + # docinfo['lang']='' + # try: + # docinfo['city']=getTextFromNode(dom.find(".//bib/city")[0]) + # except: + # docinfo['city']='' + # try: + # docinfo['number_of_pages']=getTextFromNode(dom.find(".//bib/number_of_pages")[0]) + # except: + # docinfo['number_of_pages']='' + # try: + # docinfo['series_volume']=getTextFromNode(dom.find(".//bib/series_volume")[0]) + # except: + # docinfo['series_volume']='' + # try: + # docinfo['number_of_volumes']=getTextFromNode(dom.find(".//bib/number_of_volumes")[0]) + # except: + # docinfo['number_of_volumes']='' + # try: + # docinfo['translator']=getTextFromNode(dom.find(".//bib/translator")[0]) + # except: + # docinfo['translator']='' + # try: + # docinfo['edition']=getTextFromNode(dom.find(".//bib/edition")[0]) + # except: + # docinfo['edition']='' + # try: + # docinfo['series_author']=getTextFromNode(dom.find(".//bib/series_author")[0]) + # except: + # docinfo['series_author']='' + # try: + # docinfo['publisher']=getTextFromNode(dom.find(".//bib/publisher")[0]) + # except: + # docinfo['publisher']='' + # try: + # docinfo['series_title']=getTextFromNode(dom.find(".//bib/series_title")[0]) + # except: + # docinfo['series_title']='' + # try: + # docinfo['isbn_issn']=getTextFromNode(dom.find(".//bib/isbn_issn")[0]) + # except: + # docinfo['isbn_issn']='' return docinfo @@ -626,7 +722,7 @@ path=getParentDir(path) dom = self.getDomFromIndexMeta(path) - docinfo['name']=getTextFromNode(dom.xpath("/resource/name")[0]) + docinfo['name']=getText(dom.find("name")) logging.debug("documentViewer docinfo[name] %s"%docinfo['name']) return docinfo @@ -643,15 +739,12 @@ archivePath = None archiveName = None - archiveNames = dom.xpath("//resource/name") - if archiveNames and (len(archiveNames) > 0): - archiveName = getTextFromNode(archiveNames[0]) - else: + archiveName = getTextFromNode(dom.find("name")) + if not archiveName: logging.warning("documentViewer (getdocinfofromtexttool) resource/name missing in: %s" % (url)) - archivePaths = dom.xpath("//resource/archive-path") - if archivePaths and (len(archivePaths) > 0): - archivePath = getTextFromNode(archivePaths[0]) + archivePath = getTextFromNode(dom.find("archive-path")) + if archivePath: # clean up archive path if archivePath[0] != '/': archivePath = '/' + archivePath @@ -667,11 +760,9 @@ # we balk without archive-path raise IOError("Missing archive-path (for text-tool) in %s" % (url)) - imageDirs = dom.xpath("//texttool/image") - if imageDirs and (len(imageDirs) > 0): - imageDir = getTextFromNode(imageDirs[0]) + imageDir = getText(dom.find(".//texttool/image")) - else: + if not imageDir: # we balk with no image tag / not necessary anymore because textmode is now standard #raise IOError("No text-tool info in %s"%(url)) imageDir = "" @@ -688,15 +779,13 @@ docinfo['imageURL'] = self.digilibBaseUrl + "/servlet/Scaler?fn=" + imageDir - viewerUrls = dom.xpath("//texttool/digiliburlprefix") - if viewerUrls and (len(viewerUrls) > 0): - viewerUrl = getTextFromNode(viewerUrls[0]) + viewerUrl = getText(dom.find(".//texttool/digiliburlprefix")) + if viewerUrl: docinfo['viewerURL'] = viewerUrl # old style text URL - textUrls = dom.xpath("//texttool/text") - if textUrls and (len(textUrls) > 0): - textUrl = getTextFromNode(textUrls[0]) + textUrl = getText(dom.find(".//texttool/text")) + if textUrl: if urlparse.urlparse(textUrl)[0] == "": #keine url textUrl = os.path.join(archivePath, textUrl) # fix URLs starting with /mpiwg/online @@ -706,23 +795,25 @@ docinfo['textURL'] = textUrl # new style text-url-path - textUrls = dom.xpath("//texttool/text-url-path") - if textUrls and (len(textUrls) > 0): - textUrl = getTextFromNode(textUrls[0]) + textUrl = getText(dom.find(".//texttool/text-url-path")) + if textUrl: docinfo['textURLPath'] = textUrl - if not docinfo['imagePath']: + textUrlkurz = string.split(textUrl, ".")[0] + docinfo['textURLPathkurz'] = textUrlkurz + #if not docinfo['imagePath']: # text-only, no page images - docinfo = self.getNumTextPages(docinfo) + #docinfo = self.getNumTextPages(docinfo) + - presentationUrls = dom.xpath("//texttool/presentation") + presentationUrl = getText(dom.find(".//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 + if presentationUrl: # ueberschreibe diese durch presentation informationen # presentation url ergiebt sich ersetzen von index.meta in der url der fuer die Metadaten # durch den relativen Pfad auf die presentation infos - presentationPath = getTextFromNode(presentationUrls[0]) + presentationPath = presentationUrl if url.endswith("index.meta"): presentationUrl = url.replace('index.meta', presentationPath) else: @@ -740,15 +831,15 @@ """ dom=self.getPresentationInfoXML(url) try: - docinfo['author']=getTextFromNode(dom.xpath("//author")[0]) + docinfo['author']=getText(dom.find(".//author")) except: pass try: - docinfo['title']=getTextFromNode(dom.xpath("//title")[0]) + docinfo['title']=getText(dom.find(".//title")) except: pass try: - docinfo['year']=getTextFromNode(dom.xpath("//date")[0]) + docinfo['year']=getText(dom.find(".//date")) except: pass return docinfo @@ -796,7 +887,11 @@ else: logging.error("documentViewer (getdocinfo) unknown mode: %s!"%mode) raise ValueError("Unknown mode %s! Has to be one of 'texttool','imagepath','filepath'."%(mode)) - + + # FIXME: fake texturlpath + if not docinfo.has_key('textURLPath'): + docinfo['textURLPath'] = None + logging.debug("documentViewer (getdocinfo) docinfo: %s"%docinfo) #logging.debug("documentViewer (getdocinfo) docinfo: %s"%) self.REQUEST.SESSION['docinfo'] = docinfo @@ -826,8 +921,8 @@ pageinfo['numgroups'] += 1 pageinfo['viewMode'] = viewMode pageinfo['tocMode'] = tocMode - pageinfo['characterNormalization'] = self.REQUEST.get('characterNormalization','regPlusNorm') - pageinfo['optionToggle'] = self.REQUEST.get('optionToggle','') + pageinfo['characterNormalization'] = self.REQUEST.get('characterNormalization','reg') + #pageinfo['optionToggle'] = self.REQUEST.get('optionToggle','1') pageinfo['query'] = self.REQUEST.get('query','') pageinfo['queryType'] = self.REQUEST.get('queryType','') pageinfo['querySearch'] =self.REQUEST.get('querySearch', 'fulltext') @@ -835,12 +930,10 @@ 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') + 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'])