Mercurial > hg > documentViewer
changeset 465:224aad394350 elementtree
really works with new getDocinfo
| author | casties |
|---|---|
| date | Fri, 29 Jul 2011 20:36:04 +0200 |
| parents | 19bd41d95f62 |
| children | 1641be8dc6b5 |
| files | MpdlXmlTextServer.py documentViewer.py |
| diffstat | 2 files changed, 50 insertions(+), 363 deletions(-) [+] |
line wrap: on
line diff
--- a/MpdlXmlTextServer.py Fri Jul 29 18:27:24 2011 +0200 +++ b/MpdlXmlTextServer.py Fri Jul 29 20:36:04 2011 +0200 @@ -260,8 +260,8 @@ if np > 0: docinfo['numTextPages'] = np if docinfo.get('numPages', 0) == 0: - # seems to be text-only - docinfo['numTextPages'] = np + # seems to be text-only - update page count + docinfo['numPages'] = np pageinfo['end'] = min(pageinfo['end'], np) pageinfo['numgroups'] = int(np / pageinfo['groupsize']) if np % pageinfo['groupsize'] > 0:
--- a/documentViewer.py Fri Jul 29 18:27:24 2011 +0200 +++ b/documentViewer.py Fri Jul 29 20:36:04 2011 +0200 @@ -263,6 +263,7 @@ viewMode="images" return pt(docinfo=docinfo,pageinfo=pageinfo,viewMode=viewMode) + security.declareProtected('View','index_html') def index_html(self,url,mode="texttool",viewMode="auto",tocMode="thumbs",start=None,pn=1,mk=None): @@ -293,7 +294,6 @@ # 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: viewMode="images" @@ -389,7 +389,6 @@ 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" @@ -397,28 +396,14 @@ 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) logging.debug("documentViewer (accessOK) access type %s"%access) - if access is not None and access == 'free': + if access == '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() @@ -432,343 +417,6 @@ logging.error("documentViewer (accessOK) unknown access type %s"%access) return False - - def getDirinfoFromDigilib(self,path,docinfo=None,cut=0): - """gibt param von dlInfo aus""" - if docinfo is None: - docinfo = {} - - for x in range(cut): - path=getParentPath(path) - - infoUrl=self.digilibBaseUrl+"/dirInfo-xml.jsp?mo=dir&fn="+path - - logging.debug("documentViewer (getparamfromdigilib) dirInfo from %s"%(infoUrl)) - - 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) - - if size: - docinfo['numPages'] = int(size) - else: - docinfo['numPages'] = 0 - - # TODO: produce and keep list of image names and numbers - - return docinfo - - 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 - metaUrl = url - else: - # online path - server=self.digilibBaseUrl+"/servlet/Texter?fn=" - metaUrl=server+url.replace("/mpiwg/online","") - if not metaUrl.endswith("index.meta"): - metaUrl += "/index.meta" - - return metaUrl - - def getDomFromIndexMeta(self, url): - """get dom from index meta""" - dom = None - metaUrl = self.getIndexMetaUrl(url) - - 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""" - dom = None - metaUrl = None - if url.startswith("http://"): - # real URL - metaUrl = url - else: - # online path - server=self.digilibBaseUrl+"/servlet/Texter?fn=" - metaUrl=server+url.replace("/mpiwg/online","") - - 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,cut=0): - """gets authorization info from the index.meta file at path or given by dom""" - logging.debug("documentViewer (getauthinfofromindexmeta) path: %s"%(path)) - - access = None - - if docinfo is None: - docinfo = {} - - if dom is None: - for x in range(cut): - path=getParentPath(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() - - docinfo['accessType'] = access - return docinfo - - - 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)) - - if docinfo is None: - docinfo = {} - - if dom is None: - 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)) - 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!") - 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 = {} - - if dom is None: - for x in range(cut): - path=getParentPath(path) - dom = self.getDomFromIndexMeta(path) - - 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) - - texttool = self.metadata.getTexttoolData(dom=dom) - - 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 = texttool.get('image', None) - - 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 = texttool.get('digiliburlprefix', None) - if viewerUrl: - docinfo['viewerURL'] = viewerUrl - - # old style text URL - textUrl = texttool.get('text', None) - 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 = texttool.get('text-url-path', None) - 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) - - # get bib info - 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) - - # TODO: what to do with presentation? - presentationUrl = texttool.get('presentation', None) - 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) - - # get authorization - 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']=getText(dom.find(".//author")) - docinfo['title']=getText(dom.find(".//title")) - docinfo['year']=getText(dom.find(".//date")) - return docinfo - - 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.""" - 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,cut=cut) - - pathorig=path - for x in range(cut): - path=getParentPath(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.getAuthinfoFromIndexMeta(pathorig,docinfo=docinfo,cut=cut+1) - return docinfo - - - def OLDgetDocinfo(self, mode, url): - """returns docinfo depending on mode""" - 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: - logging.debug("documentViewer (getdocinfo) docinfo in session. keys=%s"%docinfo.keys()) - return docinfo - - # new docinfo - docinfo = {'mode': mode, 'url': url} - # 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: - 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): @@ -829,6 +477,9 @@ bib = self.metadataService.getBibData(dom=metaDom) if bib: docinfo = self.getDocinfoFromBib(docinfo, bib) + else: + # no bib - try info.xml + docinfo = self.getDocinfoFromPresentationInfoXml(docinfo) # auth info access = self.metadataService.getAccessData(dom=metaDom) @@ -838,7 +489,7 @@ # image path if mode != 'texttool': # override image path from texttool - docinfo['imagePath'] = url + docinfo['imagePath'] = url.replace('/mpiwg/online/', '', 1) # number of images from digilib if docinfo.get('imagePath', None): @@ -869,7 +520,10 @@ docUrl = docinfo['documentURL'] if not docUrl.startswith('http:'): docPath = docUrl - + if docPath: + # fix URLs starting with /mpiwg/online + docPath = docPath.replace('/mpiwg/online', '', 1) + docinfo['documentPath'] = docPath return docinfo @@ -889,8 +543,6 @@ 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 @@ -905,12 +557,16 @@ # old presentation stuff presentation = texttool.get('presentation', None) if presentation and docPath: - docinfo['presentationPath'] = os.path.join(docPath, presentation) + if presentation.startswith('http:'): + docinfo['presentationUrl'] = presentation + else: + docinfo['presentationUrl'] = os.path.join(docPath, presentation) return docinfo def getDocinfoFromBib(self, docinfo, bib): """reads contents of bib element into docinfo""" + logging.debug("getDocinfoFromBib bib=%s"%repr(bib)) # put all raw bib fields in dict "bib" docinfo['bib'] = bib bibtype = bib.get('@type', None) @@ -925,8 +581,9 @@ def getDocinfoFromAccess(self, docinfo, acc): """reads contents of access element into docinfo""" #TODO: also read resource type + logging.debug("getDocinfoFromAccess acc=%s"%repr(acc)) try: - acctype = accc['@attr']['type'] + acctype = acc['@attr']['type'] if acctype: access=acctype if access in ['group', 'institution']: @@ -959,6 +616,36 @@ return docinfo + def getDocinfoFromPresentationInfoXml(self,docinfo): + """gets DC-like bibliographical information from the presentation entry in texttools""" + url = docinfo.get('presentationUrl', None) + if not url: + logging.error("getDocinfoFromPresentation: no URL!") + return docinfo + + dom = None + metaUrl = None + if url.startswith("http://"): + # real URL + metaUrl = url + else: + # online path + + server=self.digilibBaseUrl+"/servlet/Texter?fn=" + metaUrl=server+url + + txt=getHttpData(metaUrl) + if txt is None: + logging.error("Unable to read info.xml from %s"%(url)) + return docinfo + + dom = ET.fromstring(txt) + docinfo['creator']=getText(dom.find(".//author")) + docinfo['title']=getText(dom.find(".//title")) + docinfo['date']=getText(dom.find(".//date")) + 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 = {}