Changeset 464:19bd41d95f62 in documentViewer

Timestamp:

Jul 29, 2011, 4:27:24 PM (14 years ago)

Author:

casties

Branch:

elementtree

Message:

first version with new getdocinfo

Files:

• SrvTxtUtils.py (modified) (1 diff)
• documentViewer.py (modified) (10 diffs)

SrvTxtUtils.py

@@ -23 +23 @@
     text = node.text or ""
     for e in node:
-        text += gettext(e)
+        text += getText(e)
         if e.tail:
             text += e.tail

documentViewer.py

@@ -99 +99 @@
     return bt
 
-def getParentDir(path):
-    """returns pathname shortened by one"""
-    return '/'.join(path.split('/')[0:-1])
-        
+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])
+
 
 ##
@@ -264 +267 @@
     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 
@@ -270 +273 @@
         @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))
@@ -371 +372 @@
         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 '/')
@@ -439 +440 @@
         
         for x in range(cut):
-            path=getParentDir(path)
+            path=getParentPath(path)
        
         infoUrl=self.digilibBaseUrl+"/dirInfo-xml.jsp?mo=dir&fn="+path
@@ -537 +538 @@
         if dom is None:
             for x in range(cut):
-                path=getParentDir(path)
+                path=getParentPath(path)
             dom = self.getDomFromIndexMeta(path)
        
@@ -592 +593 @@
         if dom is None:
             for x in range(cut):
-                path=getParentDir(path)
+                path=getParentPath(path)
             dom = self.getDomFromIndexMeta(path)
 
@@ -724 +725 @@
         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
@@ -736 +737 @@
     
     
-    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))
@@ -769 +770 @@
         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"""