--- documentViewer/documentViewer.py 2011/07/20 08:22:36 1.175.2.6 +++ documentViewer/documentViewer.py 2011/07/29 16:27:24 1.175.2.11 @@ -98,23 +98,13 @@ def browserCheck(self): return bt -def getParentDir(path): - """returns pathname shortened by one""" - return '/'.join(path.split('/')[0:-1]) - -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 +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]) + ## ## documentViewer class @@ -127,6 +117,9 @@ 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()) @@ -146,8 +139,6 @@ 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"): @@ -170,6 +161,7 @@ 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") @@ -177,7 +169,13 @@ 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): @@ -268,15 +266,13 @@ 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)) @@ -375,7 +371,7 @@ 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"] = getParentDir(urlParams["url"]) + urlParams["url"] = getParentPath(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()]) @@ -443,7 +439,7 @@ class documentViewer(Folder): docinfo = {} for x in range(cut): - path=getParentDir(path) + path=getParentPath(path) infoUrl=self.digilibBaseUrl+"/dirInfo-xml.jsp?mo=dir&fn="+path @@ -541,7 +537,7 @@ class documentViewer(Folder): if dom is None: for x in range(cut): - path=getParentDir(path) + path=getParentPath(path) dom = self.getDomFromIndexMeta(path) acc = dom.find(".//access-conditions/access") @@ -572,80 +568,19 @@ class documentViewer(Folder): 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=bib.get("@type") - #bibtype=dom.xpath("//bib/@type") - if not bibtype: - bibtype="generic" - - bibtype=bibtype.replace("-"," ") # wrong types in 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 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['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']='' + 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 @@ -657,12 +592,13 @@ class documentViewer(Folder): if dom is None: for x in range(cut): - path=getParentDir(path) + 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""" @@ -673,6 +609,8 @@ 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 @@ -698,7 +636,7 @@ class documentViewer(Folder): # we balk without archive-path raise IOError("Missing archive-path (for text-tool) in %s" % (url)) - imageDir = getText(dom.find(".//texttool/image")) + imageDir = texttool.get('image', None) if not imageDir: # we balk with no image tag / not necessary anymore because textmode is now standard @@ -717,12 +655,12 @@ class documentViewer(Folder): docinfo['imageURL'] = self.digilibBaseUrl + "/servlet/Scaler?fn=" + imageDir - viewerUrl = getText(dom.find(".//texttool/digiliburlprefix")) + viewerUrl = texttool.get('digiliburlprefix', None) if viewerUrl: docinfo['viewerURL'] = viewerUrl # old style text URL - textUrl = getText(dom.find(".//texttool/text")) + textUrl = texttool.get('text', None) if textUrl: if urlparse.urlparse(textUrl)[0] == "": #keine url textUrl = os.path.join(archivePath, textUrl) @@ -733,7 +671,7 @@ class documentViewer(Folder): docinfo['textURL'] = textUrl # new style text-url-path - textUrl = getText(dom.find(".//texttool/text-url-path")) + textUrl = texttool.get('text-url-path', None) if textUrl: docinfo['textURLPath'] = textUrl textUrlkurz = string.split(textUrl, ".")[0] @@ -742,13 +680,13 @@ class documentViewer(Folder): # text-only, no page images #docinfo = self.getNumTextPages(docinfo) - - presentationUrl = getText(dom.find(".//texttool/presentation")) + # 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 @@ -760,6 +698,7 @@ 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 @@ -785,18 +724,19 @@ class documentViewer(Folder): pathorig=path for x in range(cut): - path=getParentDir(path) + 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 getDocinfo(self, mode, url): + 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 @@ -824,15 +764,201 @@ class documentViewer(Folder): 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) + # 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)) + 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 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 = {} @@ -845,6 +971,7 @@ 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 @@ -854,7 +981,8 @@ 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') @@ -866,10 +994,12 @@ class documentViewer(Folder): 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) + pageinfo['tocPN'] = self.REQUEST.get('tocPN', '1') + # WTF?: + 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']) @@ -878,18 +1008,30 @@ 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','') 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')