--- documentViewer/documentViewer.py 2007/05/03 17:28:23 1.24 +++ documentViewer/documentViewer.py 2011/07/20 08:22:36 1.175.2.6 @@ -1,5 +1,3 @@ - - from OFS.Folder import Folder from Products.PageTemplates.ZopePageTemplate import ZopePageTemplate from Products.PageTemplates.PageTemplateFile import PageTemplateFile @@ -7,66 +5,122 @@ from AccessControl import ClassSecurityI 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 Ft.Xml import EMPTY_NAMESPACE, Parse +#import Ft.Xml.Domlette -import Ft.Xml.XPath +import xml.etree.ElementTree as ET import os.path import sys -import cgi import urllib import logging - +import math import urlparse +import re +import string + +from SrvTxtUtils import getInt, getText, getHttpData def logger(txt,method,txt2): """logging""" logging.info(txt+ txt2) -def getInt(number, default=0): - """returns always an int (0 in case of problems)""" +def serializeNode(node, encoding="utf-8"): + """returns a string containing node as XML""" + 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): + """check the browsers request to find out the browser type""" + bt = {} + ua = self.REQUEST.get_header("HTTP_USER_AGENT") + 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 + # 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: - return int(number) - except: - return default - -def getTextFromNode(nodename): - """get the cdata content of a node""" - if nodename is None: - return "" - nodelist=nodename.childNodes - rc = "" - for node in nodelist: - if node.nodeType == node.TEXT_NODE: - rc = rc + node.data - return rc + 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 + bt['isIEWin'] = bt['isIE'] and bt['isWin'] + bt['isIEMac'] = bt['isIE'] and bt['isMac'] + bt['staticHTML'] = False + + return bt - def getParentDir(path): """returns pathname shortened by one""" return '/'.join(path.split('/')[0:-1]) - -import socket - -def urlopen(url,timeout=2): - """urlopen mit timeout""" - socket.setdefaulttimeout(timeout) - ret=urllib.urlopen(url) - socket.setdefaulttimeout(5) - return ret - +def getBibdataFromDom(dom): + """returns dict with all elements from bib-tag""" + bibinfo = {} + bib = dom.find(".//meta/bib") + if bib is not None: + # put type in @type + type = bib.get('type') + bibinfo['@type'] = type + # put all subelements in dict + for e in bib: + bibinfo[e.tag] = getText(e) + + return bibinfo ## ## documentViewer class ## class documentViewer(Folder): """document viewer""" - #textViewerUrl="http://127.0.0.1:8080/HFQP/testXSLT/getPage?" - meta_type="Document viewer" security=ClassSecurityInfo() @@ -76,45 +130,120 @@ 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_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()) + page_main_xml = PageTemplateFile('zpt/page_main_xml', globals()) + page_main_pureXml = PageTemplateFile('zpt/page_main_pureXml', 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 getOrigPages(self, **args): + """get page""" + return self.template.fulltextclient.getOrigPages(**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) + + 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 getTranslate(self, **args): + """get translate""" + return self.template.fulltextclient.getTranslate(**args) + def getLemma(self, **args): + """get lemma""" + return self.template.fulltextclient.getLemma(**args) - security.declareProtected('View','index_html') - def index_html(self,mode,url,viewMode="auto",start=None,pn=1): + 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) + + 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): ''' view it - @param mode: defines which type of document is behind url (text,images or auto) + @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 + @param viewMode: if images display images, if text display text, default is images (text,images or auto) ''' - - logger("documentViewer (index)", logging.INFO, "mode: %s url:%s start:%s pn:%s"%(mode,url,start,pn)) + logging.debug("HHHHHHHHHHHHHH:load the rss") + 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 @@ -124,34 +253,92 @@ 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) - pt = getattr(self.template, 'viewer_main') + #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.get('textURLPath',None): #texturl gesetzt und textViewer konfiguriert viewMode="text" else: viewMode="images" - return pt(docinfo=docinfo,pageinfo=pageinfo,viewMode=viewMode) - - 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] + security.declareProtected('View','index_html') + def index_html(self,url,mode="texttool",viewMode="auto",tocMode="thumbs",start=None,pn=1,mk=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) + ''' + + logging.debug("documentViewer (index) mode: %s url:%s start:%s pn:%s"%(mode,url,start,pn)) + + 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://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) + + # auto viewMode: text_dict if text else images + if viewMode=="auto": + if docinfo.get('textURL', None) or docinfo.get('textURLPath', None): + #texturl gesetzt und textViewer konfiguriert + viewMode="text_dict" else: - params[param] = str(val) + viewMode="images" - # 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 + pageinfo = self.getPageinfo(start=start, current=pn, docinfo=docinfo, viewMode=viewMode, tocMode=tocMode) + + if viewMode != 'images' and docinfo.get('textURLPath', None): + # get full text page + page = self.getTextPage(mode=viewMode, pn=pn, docinfo=docinfo, pageinfo=pageinfo) + pageinfo['textPage'] = page + + # get template /template/viewer_main + pt = getattr(self.template, 'viewer_main') + # and execute with parameters + return pt(docinfo=docinfo, pageinfo=pageinfo, viewMode=viewMode, mk=self.generateMarks(mk)) + + def generateMarks(self,mk): + ret="" + if mk is None: + return "" + if not isinstance(mk, list): + mk=[mk] + for m in mk: + ret+="mk=%s"%m + return ret + + + def getBrowser(self): + """getBrowser the version of browser """ + bt = browserCheck(self) + logging.debug("BROWSER VERSION: %s"%(bt)) + return bt + + 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.""" @@ -160,64 +347,141 @@ class documentViewer(Folder): return style + 'sel' else: return style + + 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 urlParams.has_key(param): + del urlParams[param] + else: + urlParams[param] = str(val) + + # 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 (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, params=None, baseUrl=None): + """link to documentviewer with parameter param set to val""" + return self.getLink(param, val, params, baseUrl, '&') + + 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 getOptionToggle(self, newState=None, optionName='text_options_open', initialState=True): + """returns new option state""" + if not self.REQUEST.SESSION.has_key(optionName): + # not in session -- initial + opt = {'lastState': newState, 'state': initialState} + else: + opt = self.REQUEST.SESSION.get(optionName) + if opt['lastState'] != newState: + # state in session has changed -- toggle + opt['state'] = not opt['state'] + opt['lastState'] = newState + self.REQUEST.SESSION[optionName] = opt + return opt['state'] + 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): + def getDirinfoFromDigilib(self,path,docinfo=None,cut=0): """gibt param von dlInfo aus""" - 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 - 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 = ET.fromstring(txt) + #dom = Parse(txt) + size=getText(dom.find("size")) + #sizes=dom.xpath("//dir/size") + logging.debug("documentViewer (getparamfromdigilib) dirInfo:size=%s"%size) - sizes=dom.xpath("//dir/size") - logger("documentViewer (getparamfromdigilib)", logging.INFO, "dirInfo:size"%sizes) - - if sizes: - docinfo['numPages'] = int(getTextFromNode(sizes[0])) + if size: + docinfo['numPages'] = int(size) 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 @@ -228,25 +492,25 @@ class documentViewer(Folder): metaUrl=server+url.replace("/mpiwg/online","") if not metaUrl.endswith("index.meta"): metaUrl += "/index.meta" - print metaUrl - for cnt in range(num_retries): - try: - # patch dirk encoding fehler treten dann nicht mehr auf - # dom = NonvalidatingReader.parseUri(metaUrl) - txt=urllib.urlopen(metaUrl).read() - dom = Parse(txt) - break - except: - logger("ERROR documentViewer (getIndexMata)", logging.INFO,"%s (%s)"%sys.exc_info()[0:2]) + + 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 = ET.fromstring(txt) + #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://"): @@ -256,27 +520,19 @@ 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 = ET.fromstring(txt) + #dom = Parse(txt) return dom - def getAuthinfoFromIndexMeta(self,path,docinfo=None,dom=None): + 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 @@ -284,189 +540,304 @@ class documentViewer(Folder): docinfo = {} if dom is None: - dom = self.getIndexMeta(getParentDir(path)) + for x in range(cut): + 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 - def getBibinfoFromIndexMeta(self,path,docinfo=None,dom=None): + def getBibinfoFromIndexMeta(self,path,docinfo=None,dom=None,cut=0): """gets bibliographical info from the index.meta file at path or given by dom""" - logger("documentViewer (getbibinfofromindexmeta)", logging.INFO,"path: %s"%(path)) + logging.debug("documentViewer (getbibinfofromindexmeta) path: %s"%(path)) if docinfo is None: docinfo = {} - + if dom is None: - dom = self.getIndexMeta(getParentDir(path)) - + for x in range(cut): + path=getParentDir(path) + dom = self.getDomFromIndexMeta(path) + + docinfo['indexMetaPath']=self.getIndexMetaPath(path); + + logging.debug("documentViewer (getbibinfofromindexmeta cutted) path: %s"%(path)) + # put all raw bib fields in dict "bib" + bib = getBibdataFromDom(dom) + docinfo['bib'] = bib + + # 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=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) + + bibtype=bibtype.replace("-"," ") # wrong types in index meta "-" instead of " " (not wrong! ROC) + docinfo['bib_type'] = bibtype bibmap=metaData.generateMappingForType(bibtype) - #print "bibmap: ", bibmap, " for: ", 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: - docinfo['author']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['author'][0])[0]) - docinfo['title']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['title'][0])[0]) - docinfo['year']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['year'][0])[0]) - - logging.info("bla") + if len(bibmap) > 0 and bibmap.get('author',None) or bibmap.get('title',None): + try: + docinfo['author']=bib.get(bibmap['author'][0]) + except: pass + try: + docinfo['title']=bib.get(bibmap['title'][0]) + except: pass try: - docinfo['lang']=getTextFromNode(dom.xpath("//bib/lang")[0]) - except: - docinfo['lang']='' + docinfo['year']=bib.get(bibmap['year'][0]) + except: pass + + # 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 - + + + # TODO: is this needed? + 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 dom is None: + for x in range(cut): + path=getParentDir(path) + dom = self.getDomFromIndexMeta(path) - 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 + docinfo['name']=getText(dom.find("name")) + 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 + + archiveName = getText(dom.find("name")) + if not archiveName: + logging.warning("documentViewer (getdocinfofromtexttool) resource/name missing in: %s" % (url)) + + archivePath = getText(dom.find("archive-path")) + if archivePath: + # 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)) + + imageDir = getText(dom.find(".//texttool/image")) + + 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 = "" + #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 + + viewerUrl = getText(dom.find(".//texttool/digiliburlprefix")) + if viewerUrl: + docinfo['viewerURL'] = viewerUrl + + # old style text URL + 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 + if textUrl.startswith("/mpiwg/online"): + textUrl = textUrl.replace("/mpiwg/online", '', 1) + + docinfo['textURL'] = textUrl + + # new style text-url-path + textUrl = getText(dom.find(".//texttool/text-url-path")) + if textUrl: + docinfo['textURLPath'] = textUrl + textUrlkurz = string.split(textUrl, ".")[0] + docinfo['textURLPathkurz'] = textUrlkurz + #if not docinfo['imagePath']: + # text-only, no page images + #docinfo = self.getNumTextPages(docinfo) - textUrls=dom.xpath("//texttool/text") - if textUrls and (len(textUrls)>0): - textUrl=getTextFromNode(textUrls[0]) - if urlparse.urlparse(textUrl)[0]=="": #keine url - textUrl=os.path.join(archivePath,textUrl) - - docinfo['textURL'] = textUrl - - presentationUrls=dom.xpath("//texttool/presentation") - docinfo = self.getBibinfoFromIndexMeta(url,docinfo=docinfo,dom=dom) # get info von bib tag - - if presentationUrls and (len(presentationUrls)>0): # ueberschreibe diese durch presentation informationen - # presentation url ergiebt sich ersetzen von index.meta in der url der fŸr die Metadaten - # durch den relativen Pfad auf die presentation infos - presentationUrl=url.replace('index.meta',getTextFromNode(presentationUrls[0])) - docinfo = self.getBibinfoFromTextToolPresentation(presentationUrl,docinfo=docinfo,dom=dom) - - docinfo = self.getAuthinfoFromIndexMeta(path,docinfo=docinfo) - return docinfo + + presentationUrl = getText(dom.find(".//texttool/presentation")) + docinfo = self.getBibinfoFromIndexMeta(url, docinfo=docinfo, dom=dom) # get info von bib tag + # TODO: is this needed here? + docinfo = self.getNameFromIndexMeta(url, docinfo=docinfo, dom=dom) + + + 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 = presentationUrl + 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): """gets the bibliographical information from the preseantion entry in texttools """ dom=self.getPresentationInfoXML(url) - docinfo['author']=getTextFromNode(dom.xpath("//author")[0]) - docinfo['title']=getTextFromNode(dom.xpath("//title")[0]) - docinfo['year']=getTextFromNode(dom.xpath("//date")[0]) + docinfo['author']=getText(dom.find(".//author")) + docinfo['title']=getText(dom.find(".//title")) + docinfo['year']=getText(dom.find(".//date")) return docinfo - def getDocinfoFromImagePath(self,path,docinfo=None): + 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","") docinfo['imagePath'] = path - docinfo=self.getDirinfoFromDigilib(path,docinfo=docinfo) + docinfo=self.getDirinfoFromDigilib(path,docinfo=docinfo,cut=cut) + + pathorig=path + for x in range(cut): + path=getParentDir(path) + logging.debug("documentViewer (getdocinfofromimagepath) PATH:"+path) imageUrl=self.digilibBaseUrl+"/servlet/Scaler?fn="+path docinfo['imageURL'] = imageUrl - docinfo = self.getBibinfoFromIndexMeta(path,docinfo=docinfo) - docinfo = self.getAuthinfoFromIndexMeta(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""" - 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. keys=%s"%docinfo.keys()) return docinfo + # new docinfo docinfo = {'mode': mode, 'url': url} - if mode=="texttool": #index.meta with texttool information + # add self url + docinfo['viewerUrl'] = self.getDocumentViewerURL() + if mode=="texttool": + # index.meta with texttool information docinfo = self.getDocinfoFromTextTool(url, docinfo=docinfo) elif mode=="imagepath": + # folder with images, index.meta optional docinfo = self.getDocinfoFromImagePath(url, docinfo=docinfo) + elif mode=="filepath": + # filename + docinfo = self.getDocinfoFromImagePath(url, docinfo=docinfo,cut=1) else: - logger("documentViewer (getdocinfo)", logging.ERROR,"unknown mode!") - raise ValueError("Unknown mode %s"%(mode)) - - logger("documentViewer (getdocinfo)", logging.INFO,"docinfo: %s"%docinfo) + 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: keys=%s"%docinfo.keys()) + #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): + + 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) + pageinfo['current'] = current rows = int(rows or self.thumbrows) pageinfo['rows'] = rows @@ -474,123 +845,68 @@ class documentViewer(Folder): pageinfo['cols'] = cols grpsize = cols * rows pageinfo['groupsize'] = grpsize - start = getInt(start, default=(int(current / grpsize) * grpsize +1)) + 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['numgroups'] += 1 + pageinfo['viewMode'] = viewMode + pageinfo['tocMode'] = tocMode + 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') + 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) - - #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)] - - - 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 - 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"