--- documentViewer/documentViewer.py 2011/07/29 16:27:24 1.175.2.11 +++ documentViewer/documentViewer.py 2012/01/04 07:38:17 1.178 @@ -1,40 +1,56 @@ + 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 import EMPTY_NAMESPACE, Parse -#import Ft.Xml.Domlette - -import xml.etree.ElementTree as ET - +from Ft.Xml import EMPTY_NAMESPACE, Parse +import Ft.Xml.Domlette import os.path import sys import urllib +import urllib2 import logging import math import urlparse +import cStringIO 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)""" + try: + return int(number) + except: + return int(default) + +def getTextFromNode(nodename): + """get the cdata content of a node""" + if nodename is None: + return "" + nodelist=nodename.childNodes + rc = "" + for node in nodelist: + if node.nodeType == node.TEXT_NODE: + rc = rc + node.data + return rc + 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() + stream = cStringIO.StringIO() + #logging.debug("BUF: %s"%(stream)) + Ft.Xml.Domlette.Print(node, stream=stream, encoding=encoding) + s = stream.getvalue() + #logging.debug("BUF: %s"%(s)) + stream.close() return s def browserCheck(self): @@ -98,13 +114,54 @@ def browserCheck(self): return bt -def getParentPath(path, cnt=1): - """returns pathname shortened by cnt""" - # make sure path doesn't end with / - path = path.rstrip('/') - # split by /, shorten, and reassemble - return '/'.join(path.split('/')[0:-cnt]) - + +def getParentDir(path): + """returns pathname shortened by one""" + return '/'.join(path.split('/')[0:-1]) + + +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 ## ## documentViewer class @@ -117,9 +174,6 @@ class documentViewer(Folder): manage_options=Folder.manage_options+( {'label':'main config','action':'changeDocumentViewerForm'}, ) - - metadataService = None - """MetaDataFolder instance""" # templates and forms viewer_main = PageTemplateFile('zpt/viewer_main', globals()) @@ -139,6 +193,8 @@ class documentViewer(Folder): thumbs_main_rss = PageTemplateFile('zpt/thumbs_main_rss', globals()) + security.declareProtected('View management screens','changeDocumentViewerForm') + changeDocumentViewerForm = PageTemplateFile('zpt/changeDocumentViewer', globals()) def __init__(self,id,imageScalerUrl=None,textServerName=None,title="",digilibBaseUrl=None,thumbcols=2,thumbrows=5,authgroups="mpiwg"): @@ -161,7 +217,6 @@ class documentViewer(Folder): 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") @@ -169,13 +224,7 @@ class documentViewer(Folder): templateFolder._setObject('zogilib',zogilib) except Exception, e: logging.error("Unable to create zogiLib for zogilib: "+str(e)) - - try: - # assume MetaDataFolder instance is called metadata - self.metadataService = getattr(self, 'metadata') - except Exception, e: - logging.error("Unable to find MetaDataFolder 'metadata': "+str(e)) - + # proxy text server methods to fulltextclient def getTextPage(self, **args): @@ -241,7 +290,7 @@ class documentViewer(Folder): ''' logging.debug("HHHHHHHHHHHHHH:load the rss") - logging.debug("documentViewer (index) mode: %s url:%s start:%s pn:%s"%(mode,url,start,pn)) + logger("documentViewer (index)", logging.INFO, "mode: %s url:%s start:%s pn:%s"%(mode,url,start,pn)) if not hasattr(self, 'template'): # create template folder if it doesn't exist @@ -266,13 +315,15 @@ class documentViewer(Folder): 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)) @@ -289,26 +340,21 @@ class documentViewer(Folder): 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 + + if viewMode=="auto": # automodus gewaehlt + 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) + 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) + if (docinfo.get('textURLPath',None)): + page = self.getTextPage(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)) + 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): ret="" @@ -371,7 +417,7 @@ class documentViewer(Folder): # 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"] = getParentPath(urlParams["url"]) + 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()]) @@ -439,7 +485,8 @@ class documentViewer(Folder): docinfo = {} for x in range(cut): - path=getParentPath(path) + + path=getParentDir(path) infoUrl=self.digilibBaseUrl+"/dirInfo-xml.jsp?mo=dir&fn="+path @@ -449,14 +496,12 @@ class documentViewer(Folder): 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) + dom = Parse(txt) + sizes=dom.xpath("//dir/size") + logging.debug("documentViewer (getparamfromdigilib) dirInfo:size"%sizes) - if size: - docinfo['numPages'] = int(size) + if sizes: + docinfo['numPages'] = int(getTextFromNode(sizes[0])) else: docinfo['numPages'] = 0 @@ -501,8 +546,7 @@ class documentViewer(Folder): if txt is None: raise IOError("Unable to read index meta from %s"%(url)) - dom = ET.fromstring(txt) - #dom = Parse(txt) + dom = Parse(txt) return dom def getPresentationInfoXML(self, url): @@ -521,8 +565,7 @@ class documentViewer(Folder): if txt is None: raise IOError("Unable to read infoXMLfrom %s"%(url)) - dom = ET.fromstring(txt) - #dom = Parse(txt) + dom = Parse(txt) return dom @@ -537,17 +580,14 @@ class documentViewer(Folder): if dom is None: for x in range(cut): - path=getParentPath(path) + path=getParentDir(path) dom = self.getDomFromIndexMeta(path) - 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() + acctype = dom.xpath("//access-conditions/access/@type") + if acctype and (len(acctype)>0): + access=acctype[0].value + if access in ['group', 'institution']: + access = getTextFromNode(dom.xpath("//access-conditions/access/name")[0]).lower() docinfo['accessType'] = access return docinfo @@ -555,7 +595,7 @@ class documentViewer(Folder): def getBibinfoFromIndexMeta(self,path,docinfo=None,dom=None,cut=0): """gets bibliographical info from the index.meta file at path or given by dom""" - logging.debug("documentViewer (getbibinfofromindexmeta) path: %s"%(path)) + #logging.debug("documentViewer (getbibinfofromindexmeta) path: %s"%(path)) if docinfo is None: docinfo = {} @@ -567,24 +607,88 @@ class documentViewer(Folder): docinfo['indexMetaPath']=self.getIndexMetaPath(path); - logging.debug("documentViewer (getbibinfofromindexmeta cutted) path: %s"%(path)) - if self.metadataService is not None: - # put all raw bib fields in dict "bib" - bib = self.metadataService.getBibData(dom=dom) - docinfo['bib'] = bib - bibtype = bib.get('@type', None) - docinfo['bib_type'] = bibtype - # also store DC metadata for convenience - dc = self.metadataService.getDCMappedData(bib) - docinfo['creator'] = dc.get('creator',None) - docinfo['title'] = dc.get('title',None) - docinfo['date'] = dc.get('date',None) - else: - logging.error("MetadataService not found!") + #logging.debug("documentViewer (getbibinfofromindexmeta cutted) path: %s"%(path)) + # put in all raw bib fields as dict "bib" + bib = dom.xpath("//bib/*") + if bib and len(bib)>0: + bibinfo = {} + for e in bib: + bibinfo[e.localName] = getTextFromNode(e) + docinfo['bib'] = bibinfo + + # extract some fields (author, title, year) according to their mapping + metaData=self.metadata.main.meta.bib + bibtype=dom.xpath("//bib/@type") + if bibtype and (len(bibtype)>0): + bibtype=bibtype[0].value + else: + bibtype="generic" + + bibtype=bibtype.replace("-"," ") # wrong typesiin index meta "-" instead of " " (not wrong! ROC) + docinfo['bib_type'] = bibtype + bibmap=metaData.generateMappingForType(bibtype) + #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: + docinfo['author']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['author'][0])[0]) + except: pass + try: + docinfo['title']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['title'][0])[0]) + except: pass + try: + docinfo['year']=getTextFromNode(dom.xpath("//bib/%s"%bibmap['year'][0])[0]) + except: pass + #logging.debug("documentViewer (getbibinfofromindexmeta) using mapping for %s"%bibtype) + try: + docinfo['lang']=getTextFromNode(dom.xpath("//bib/lang")[0]) + except: + docinfo['lang']='' + try: + docinfo['city']=getTextFromNode(dom.xpath("//bib/city")[0]) + except: + docinfo['city']='' + try: + docinfo['number_of_pages']=getTextFromNode(dom.xpath("//bib/number_of_pages")[0]) + except: + docinfo['number_of_pages']='' + try: + docinfo['series_volume']=getTextFromNode(dom.xpath("//bib/series_volume")[0]) + except: + docinfo['series_volume']='' + try: + docinfo['number_of_volumes']=getTextFromNode(dom.xpath("//bib/number_of_volumes")[0]) + except: + docinfo['number_of_volumes']='' + try: + docinfo['translator']=getTextFromNode(dom.xpath("//bib/translator")[0]) + except: + docinfo['translator']='' + try: + docinfo['edition']=getTextFromNode(dom.xpath("//bib/edition")[0]) + except: + docinfo['edition']='' + try: + docinfo['series_author']=getTextFromNode(dom.xpath("//bib/series_author")[0]) + except: + docinfo['series_author']='' + try: + docinfo['publisher']=getTextFromNode(dom.xpath("//bib/publisher")[0]) + except: + docinfo['publisher']='' + try: + docinfo['series_title']=getTextFromNode(dom.xpath("//bib/series_title")[0]) + except: + docinfo['series_title']='' + try: + docinfo['isbn_issn']=getTextFromNode(dom.xpath("//bib/isbn_issn")[0]) + except: + docinfo['isbn_issn']='' + #logging.debug("I NEED BIBTEX %s"%docinfo) 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: @@ -592,13 +696,12 @@ class documentViewer(Folder): if dom is None: for x in range(cut): - path=getParentPath(path) + path=getParentDir(path) dom = self.getDomFromIndexMeta(path) - docinfo['name']=getText(dom.find("name")) + 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""" @@ -609,18 +712,19 @@ class documentViewer(Folder): docinfo['lang'] = '' # default keine Sprache gesetzt if dom is None: dom = self.getDomFromIndexMeta(url) - - texttool = self.metadata.getTexttoolData(dom=dom) archivePath = None archiveName = None - archiveName = getText(dom.find("name")) - if not archiveName: + 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)) - archivePath = getText(dom.find("archive-path")) - if archivePath: + 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 @@ -636,9 +740,11 @@ class documentViewer(Folder): # we balk without archive-path raise IOError("Missing archive-path (for text-tool) in %s" % (url)) - imageDir = texttool.get('image', None) + imageDirs = dom.xpath("//texttool/image") + if imageDirs and (len(imageDirs) > 0): + imageDir = getTextFromNode(imageDirs[0]) - if not imageDir: + 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 = "" @@ -655,13 +761,15 @@ class documentViewer(Folder): docinfo['imageURL'] = self.digilibBaseUrl + "/servlet/Scaler?fn=" + imageDir - viewerUrl = texttool.get('digiliburlprefix', None) - if viewerUrl: + viewerUrls = dom.xpath("//texttool/digiliburlprefix") + if viewerUrls and (len(viewerUrls) > 0): + viewerUrl = getTextFromNode(viewerUrls[0]) docinfo['viewerURL'] = viewerUrl # old style text URL - textUrl = texttool.get('text', None) - if textUrl: + 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 @@ -669,10 +777,24 @@ class documentViewer(Folder): textUrl = textUrl.replace("/mpiwg/online", '', 1) docinfo['textURL'] = textUrl - + + + #TODO: hack-DW for annalen + if (textUrl is not None) and (textUrl.startswith("/permanent/einstein/annalen")): + textUrl=textUrl.replace("/permanent/einstein/annalen/","/diverse/de/") + splitted=textUrl.split("/fulltext") + textUrl=splitted[0]+".xml" + textUrlkurz = string.split(textUrl, ".")[0] + docinfo['textURLPathkurz'] = textUrlkurz + docinfo['textURLPath'] = textUrl + logging.debug("hack") + logging.debug(textUrl) + + # new style text-url-path - textUrl = texttool.get('text-url-path', None) - if textUrl: + textUrls = dom.xpath("//texttool/text-url-path") + if textUrls and (len(textUrls) > 0): + textUrl = getTextFromNode(textUrls[0]) docinfo['textURLPath'] = textUrl textUrlkurz = string.split(textUrl, ".")[0] docinfo['textURLPathkurz'] = textUrlkurz @@ -680,17 +802,17 @@ class documentViewer(Folder): # text-only, no page images #docinfo = self.getNumTextPages(docinfo) - # get bib info + + presentationUrls = dom.xpath("//texttool/presentation") docinfo = self.getBibinfoFromIndexMeta(url, docinfo=docinfo, dom=dom) # get info von bib tag - # TODO: is this needed here? + #docinfo = self.getDownloadfromDocinfoToBibtex(url, docinfo=docinfo, dom=dom) docinfo = self.getNameFromIndexMeta(url, docinfo=docinfo, dom=dom) - # TODO: what to do with presentation? - presentationUrl = texttool.get('presentation', None) - if presentationUrl: # ueberschreibe diese durch presentation informationen + + 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 = presentationUrl + presentationPath = getTextFromNode(presentationUrls[0]) if url.endswith("index.meta"): presentationUrl = url.replace('index.meta', presentationPath) else: @@ -698,7 +820,6 @@ class documentViewer(Folder): docinfo = self.getBibinfoFromTextToolPresentation(presentationUrl, docinfo=docinfo, dom=dom) - # get authorization docinfo = self.getAuthinfoFromIndexMeta(url, docinfo=docinfo, dom=dom) # get access info return docinfo @@ -708,9 +829,18 @@ class documentViewer(Folder): """gets the bibliographical information from the preseantion entry in texttools """ dom=self.getPresentationInfoXML(url) - docinfo['author']=getText(dom.find(".//author")) - docinfo['title']=getText(dom.find(".//title")) - docinfo['year']=getText(dom.find(".//date")) + try: + docinfo['author']=getTextFromNode(dom.xpath("//author")[0]) + except: + pass + try: + docinfo['title']=getTextFromNode(dom.xpath("//title")[0]) + except: + pass + try: + docinfo['year']=getTextFromNode(dom.xpath("//date")[0]) + except: + pass return docinfo def getDocinfoFromImagePath(self,path,docinfo=None,cut=0): @@ -724,19 +854,19 @@ class documentViewer(Folder): pathorig=path for x in range(cut): - path=getParentPath(path) + path=getParentDir(path) logging.debug("documentViewer (getdocinfofromimagepath) PATH:"+path) imageUrl=self.digilibBaseUrl+"/servlet/Scaler?fn="+path docinfo['imageURL'] = imageUrl - #TODO: use getDocinfoFromIndexMeta #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.getDownloadfromDocinfoToBibtex(pathorig,docinfo=docinfo,cut=cut+1) docinfo = self.getAuthinfoFromIndexMeta(pathorig,docinfo=docinfo,cut=cut+1) return docinfo - def OLDgetDocinfo(self, mode, url): + def getDocinfo(self, mode, url): """returns docinfo depending on mode""" logging.debug("documentViewer (getdocinfo) mode: %s, url: %s"%(mode,url)) # look for cached docinfo in session @@ -744,221 +874,29 @@ class documentViewer(Folder): 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: - logging.debug("documentViewer (getdocinfo) docinfo in session. keys=%s"%docinfo.keys()) + logging.debug("documentViewer (getdocinfo) docinfo in session: %s"%docinfo) return docinfo - # new docinfo docinfo = {'mode': mode, 'url': url} - # add self url - docinfo['viewerUrl'] = self.getDocumentViewerURL() - if mode=="texttool": - # index.meta with texttool information + 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: logging.error("documentViewer (getdocinfo) unknown mode: %s!"%mode) raise ValueError("Unknown mode %s! Has to be one of 'texttool','imagepath','filepath'."%(mode)) - logging.debug("documentViewer (getdocinfo) docinfo: keys=%s"%docinfo.keys()) - #logging.debug("documentViewer (getdocinfo) docinfo: %s"%docinfo) - # store in session - self.REQUEST.SESSION['docinfo'] = docinfo - return docinfo - - - def getDocinfo(self, mode, url): - """returns docinfo depending on mode""" - logging.debug("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', None) == mode and docinfo.get('url', None) == url: - logging.debug("getDocinfo: docinfo in session. keys=%s"%docinfo.keys()) - return docinfo - - # new docinfo - docinfo = {'mode': mode, 'url': url} - # add self url - docinfo['viewerUrl'] = self.getDocumentViewerURL() - # get index.meta DOM - docUrl = None - metaDom = None - if mode=="texttool": - # url points to document dir or index.meta - metaDom = self.metadataService.getDomFromPathOrUrl(url) - docUrl = url.replace('/index.meta', '') - if metaDom is None: - raise IOError("Unable to find index.meta for mode=texttool!") - - elif mode=="imagepath": - # url points to folder with images, index.meta optional - # asssume index.meta in parent dir - docUrl = getParentPath(url) - metaDom = self.metadataService.getDomFromPathOrUrl(docUrl) - - elif mode=="filepath": - # url points to image file, index.meta optional - # asssume index.meta is two path segments up - docUrl = getParentPath(url, 2) - metaDom = self.metadataService.getDomFromPathOrUrl(docUrl) - - 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 - docinfo['documentUrl'] = docUrl - # process index.meta contents - if metaDom is not None: - # document directory name and path - resource = self.metadataService.getResourceData(dom=metaDom) - if resource: - docinfo = self.getDocinfoFromResource(docinfo, resource) - - # texttool info - texttool = self.metadataService.getTexttoolData(dom=metaDom) - if texttool: - docinfo = self.getDocinfoFromTexttool(docinfo, texttool) - - # bib info - bib = self.metadataService.getBibData(dom=metaDom) - if bib: - docinfo = self.getDocinfoFromBib(docinfo, bib) - - # auth info - access = self.metadataService.getAccessData(dom=metaDom) - if access: - docinfo = self.getDocinfoFromAccess(docinfo, access) - - # image path - if mode != 'texttool': - # override image path from texttool - docinfo['imagePath'] = url - - # number of images from digilib - if docinfo.get('imagePath', None): - docinfo['imageURL'] = self.digilibBaseUrl + "/servlet/Scaler?fn=" + docinfo['imagePath'] - docinfo = self.getDocinfoFromDigilib(docinfo, docinfo['imagePath']) - - logging.debug("documentViewer (getdocinfo) docinfo: keys=%s"%docinfo.keys()) - #logging.debug("documentViewer (getdocinfo) docinfo: %s"%docinfo) - # store in session + logging.debug("documentViewer (getdocinfo) docinfo: %s"%docinfo) + #logging.debug("documentViewer (getdocinfo) docinfo: %s"%) self.REQUEST.SESSION['docinfo'] = docinfo return docinfo - - def getDocinfoFromResource(self, docinfo, resource): - """reads contents of resource element into docinfo""" - docName = resource.get('name', None) - docinfo['documentName'] = docName - docPath = resource.get('archive-path', None) - if docPath: - # clean up document path - if docPath[0] != '/': - docPath = '/' + docPath - - if docName and (not docPath.endswith(docName)): - docPath += "/" + docName - - else: - # use docUrl as docPath - docUrl = docinfo['documentURL'] - if not docUrl.startswith('http:'): - docPath = docUrl - - docinfo['documentPath'] = docPath - return docinfo - - def getDocinfoFromTexttool(self, docinfo, texttool): - """reads contents of texttool element into docinfo""" - # image dir - imageDir = texttool.get('image', None) - docPath = docinfo.get('documentPath', None) - if imageDir and docPath: - #print "image: ", imageDir, " archivepath: ", archivePath - imageDir = os.path.join(docPath, imageDir) - imageDir = imageDir.replace('/mpiwg/online', '', 1) - docinfo['imagePath'] = imageDir - - # old style text URL - textUrl = texttool.get('text', None) - if textUrl and docPath: - if urlparse.urlparse(textUrl)[0] == "": #keine url - textUrl = os.path.join(docPath, textUrl) - # fix URLs starting with /mpiwg/online - textUrl = textUrl.replace('/mpiwg/online', '', 1) - - docinfo['textURL'] = textUrl - - # new style text-url-path - textUrl = texttool.get('text-url-path', None) - if textUrl: - docinfo['textURLPath'] = textUrl - #TODO: ugly: - #textUrlkurz = string.split(textUrl, ".")[0] - #docinfo['textURLPathkurz'] = textUrlkurz - - # old presentation stuff - presentation = texttool.get('presentation', None) - if presentation and docPath: - docinfo['presentationPath'] = os.path.join(docPath, presentation) - - return docinfo - - def getDocinfoFromBib(self, docinfo, bib): - """reads contents of bib element into docinfo""" - # put all raw bib fields in dict "bib" - docinfo['bib'] = bib - bibtype = bib.get('@type', None) - docinfo['bibType'] = bibtype - # also store DC metadata for convenience - dc = self.metadataService.getDCMappedData(bib) - docinfo['creator'] = dc.get('creator',None) - docinfo['title'] = dc.get('title',None) - docinfo['date'] = dc.get('date',None) - return docinfo - - def getDocinfoFromAccess(self, docinfo, acc): - """reads contents of access element into docinfo""" - #TODO: also read resource type - try: - acctype = accc['@attr']['type'] - if acctype: - access=acctype - if access in ['group', 'institution']: - access = acc['name'].lower() - - docinfo['accessType'] = access - - except: - pass - - return docinfo - - def getDocinfoFromDigilib(self, docinfo, path): - infoUrl=self.digilibBaseUrl+"/dirInfo-xml.jsp?mo=dir&fn="+path - # fetch data - txt = getHttpData(infoUrl) - if not txt: - logging.error("Unable to get dir-info from %s"%(infoUrl)) - return docinfo - - dom = ET.fromstring(txt) - size = getText(dom.find("size")) - logging.debug("getDocinfoFromDigilib: size=%s"%size) - if size: - docinfo['numPages'] = int(size) - else: - docinfo['numPages'] = 0 - - # TODO: produce and keep list of image names and numbers - return docinfo - - + def getPageinfo(self, current, start=None, rows=None, cols=None, docinfo=None, viewMode=None, tocMode=None): """returns pageinfo with the given parameters""" pageinfo = {} @@ -971,7 +909,6 @@ class documentViewer(Folder): pageinfo['cols'] = cols grpsize = cols * rows pageinfo['groupsize'] = grpsize - # what does this do? start = getInt(start, default=(math.ceil(float(current)/float(grpsize))*grpsize-(grpsize-1))) # int(current / grpsize) * grpsize +1)) pageinfo['start'] = start @@ -981,8 +918,7 @@ class documentViewer(Folder): 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') @@ -992,14 +928,18 @@ class documentViewer(Folder): pageinfo['querySearch'] =self.REQUEST.get('querySearch', 'fulltext') pageinfo['textPN'] = self.REQUEST.get('textPN','1') pageinfo['highlightQuery'] = self.REQUEST.get('highlightQuery','') + + pageinfo ['highlightElementPos'] = self.REQUEST.get('highlightElementPos','') + pageinfo ['highlightElement'] = self.REQUEST.get('highlightElement','') + + pageinfo ['xpointer'] = self.REQUEST.get('xpointer','') + pageinfo['tocPageSize'] = self.REQUEST.get('tocPageSize', '30') pageinfo['queryPageSize'] =self.REQUEST.get('queryPageSize', '10') - pageinfo['tocPN'] = self.REQUEST.get('tocPN', '1') - # WTF?: - toc = int(pageinfo['tocPN']) - pageinfo['textPages'] =int(toc) + pageinfo['tocPN'] = self.REQUEST.get('tocPN', '1') + toc = int (pageinfo['tocPN']) + pageinfo['textPages'] =int (toc) - # What does this do? if 'tocSize_%s'%tocMode in docinfo: tocSize = int(docinfo['tocSize_%s'%tocMode]) tocPageSize = int(pageinfo['tocPageSize']) @@ -1008,30 +948,19 @@ class documentViewer(Folder): tocPages=tocSize/tocPageSize+1 else: tocPages=tocSize/tocPageSize - - pageinfo['tocPN'] = min(tocPages,toc) - + pageinfo['tocPN'] = min (tocPages,toc) pageinfo['searchPN'] =self.REQUEST.get('searchPN','1') - pageinfo['sn'] =self.REQUEST.get('sn','') + #pageinfo['sn'] =self.REQUEST.get('sn','') + pageinfo['s'] =self.REQUEST.get('s','') return pageinfo - - - security.declareProtected('View management screens','changeDocumentViewerForm') - changeDocumentViewerForm = PageTemplateFile('zpt/changeDocumentViewer', globals()) - def changeDocumentViewer(self,title="",digilibBaseUrl=None,thumbrows=2,thumbcols=5,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.digilibBaseUrl = digilibBaseUrl self.thumbrows = thumbrows self.thumbcols = thumbcols self.authgroups = [s.strip().lower() for s in authgroups.split(',')] - try: - # assume MetaDataFolder instance is called metadata - self.metadataService = getattr(self, 'metadata') - except Exception, e: - logging.error("Unable to find MetaDataFolder 'metadata': "+str(e)) - if RESPONSE is not None: RESPONSE.redirect('manage_main')