Changeset 511:551ca1641a5e in documentViewer for MpdlXmlTextServer.py

Timestamp:

Feb 28, 2012, 5:21:59 PM (12 years ago)

Author:

casties

Branch:

elementtree

Message:

more cleanup.
search really works now.

File:

• MpdlXmlTextServer.py (modified) (11 diffs)

MpdlXmlTextServer.py

@@ -7 +7 @@
 import logging
 import urllib
+import urlparse
 import base64
 
@@ -140 +141 @@
             
         # stuff for constructing full urls
-        characterNormalization = pageinfo.get('characterNormalization', None)
-        moreTextParam = ''
         selfurl = docinfo['viewerUrl']
+        textParams = {'document': docpath,
+                      'pn': pn}
+        if 'characterNormalization' in pageinfo:
+            textParams['characterNormalization'] = pageinfo['characterNormalization']
         
         if not mode:
@@ -157 +160 @@
             # add highlighting
             highlightQuery = pageinfo.get('highlightQuery', None)
-            sn = pageinfo.get('sn', None)
-            if highlightQuery and sn:
-                moreTextParam +="&highlightQuery=%s&sn=%s"%(urllib.quote(highlightQuery),sn)
-                
-            # remove mode
+            if highlightQuery:
+                textParams['highlightQuery'] = highlightQuery
+                textParams['highlightElement'] = pageinfo.get('highlightElement', '')
+                textParams['highlightElementPos'] = pageinfo.get('highlightElementPos', '')
+                
+            # ignore mode in the following
             modes.remove('search')
                             
@@ -175 +179 @@
             textmode = modes[0]
         
-        textParam = "document=%s&mode=%s&pn=%s&characterNormalization=%s"%(docpath,textmode,pn,characterNormalization)
-        textParam += moreTextParam
+        textParams['mode'] = textmode
         
         # fetch the page
-        pagexml = self.getServerData("page-fragment.xql",textParam)
+        pagexml = self.getServerData("page-fragment.xql",urllib.urlencode(textParams))
         dom = ET.fromstring(pagexml)
         # extract additional info
@@ -221 +224 @@
                     if href:
                         # is link with href
-                        if href.startswith('http://mpdl-proto.mpiwg-berlin.mpg.de/mpdl/interface/lt/wordInfo.xql'):
+                        linkurl = urlparse.urlparse(href)
+                        #logging.debug("getTextPage: linkurl=%s"%repr(linkurl))
+                        if linkurl.path.endswith('GetDictionaryEntries'):
+                            #TODO: replace wordInfo page
                             # is dictionary link - change href (keeping parameters)
-                            l.set('href', href.replace('http://mpdl-proto.mpiwg-berlin.mpg.de/mpdl/interface/lt/wordInfo.xql','%s/template/viewer_wordinfo'%viewerurl))
+                            #l.set('href', href.replace('http://mpdl-proto.mpiwg-berlin.mpg.de/mpdl/interface/lt/wordInfo.xql','%s/template/viewer_wordinfo'%viewerurl))
                             # add target to open new page
                             l.set('target', '_blank')
                                                           
                         # TODO: is this needed?
-                        if href.startswith('http://mpdl-proto.mpiwg-berlin.mpg.de/mpdl/lt/lemma.xql'):
-                            selfurl = self.absolute_url()
-                            l.set('href', href.replace('http://mpdl-proto.mpiwg-berlin.mpg.de/mpdl/lt/lemma.xql','%s/head_main_lemma'%selfurl))
-                            l.set('target', '_blank')
-                            l.set('onclick',"popupWin = window.open(this.href, 'InfoWindow', 'menubar=no, location,width=500,height=600,top=180, left=700, toolbar=no, scrollbars=1'); return false;")
-                            l.set('ondblclick', 'popupWin.focus();')   
+#                        if href.startswith('http://mpdl-proto.mpiwg-berlin.mpg.de/mpdl/lt/lemma.xql'):
+#                            selfurl = self.absolute_url()
+#                            l.set('href', href.replace('http://mpdl-proto.mpiwg-berlin.mpg.de/mpdl/lt/lemma.xql','%s/head_main_lemma'%selfurl))
+#                            l.set('target', '_blank')
+#                            l.set('onclick',"popupWin = window.open(this.href, 'InfoWindow', 'menubar=no, location,width=500,height=600,top=180, left=700, toolbar=no, scrollbars=1'); return false;")
+#                            l.set('ondblclick', 'popupWin.focus();')   
                     
                         if href.startswith('#note-'):
@@ -273 +279 @@
     def getSearchResults(self, mode, query=None, pageinfo=None, docinfo=None):
         """loads list of search results and stores XML in docinfo"""
+        
         logging.debug("getSearchResults mode=%s query=%s"%(mode, query))
         if mode == "none":
             return docinfo
               
-        if 'resultSize_%s_%s'%(mode,query) in docinfo:
-            # cached result
-            return docinfo
-        
+        cachedQuery = docinfo.get('cachedQuery', None)
+        if cachedQuery is not None:
+            # cached search result
+            if cachedQuery == '%s_%s'%(mode,query):
+                # same query
+                return docinfo
+            
+            else:
+                # different query
+                del docinfo['resultSize']
+                del docinfo['resultXML']
+        
+        # cache query
+        docinfo['cachedQuery'] = '%s_%s'%(mode,query)
+        
+        # fetch full results
         docpath = docinfo['textURLPath']
-        
-        # fetch full results
         params = {'document': docpath,
                   'mode': 'text',
@@ -305 +322 @@
                 
             elif dc == 'queryResultHits':
-                docinfo['resultSize_%s_%s'%(mode,query)] = getInt(div.text)
+                docinfo['resultSize'] = getInt(div.text)
 
         if pagediv is not None:
             # store XML in docinfo
-            docinfo['resultXML_%s_%s'%(mode,query)] = ET.tostring(pagediv, 'UTF-8')
+            docinfo['resultXML'] = ET.tostring(pagediv, 'UTF-8')
 
         return docinfo
@@ -317 +334 @@
         """returns single page from the table of contents"""
         logging.debug("getResultsPage mode=%s, pn=%s"%(mode,pn))
-        # check for cached TOC
-        #TODO: cache only one search
-        if not docinfo.has_key('resultXML_%s_%s'%(mode,query)):
+        # check for cached result
+        if not 'resultXML' in docinfo:
             self.getSearchResults(mode=mode, query=query, pageinfo=pageinfo, docinfo=docinfo)
             
-        resultxml = docinfo.get('resultXML_%s_%s'%(mode,query), None)
+        resultxml = docinfo.get('resultXML', None)
         if not resultxml:
             logging.error("getResultPage: unable to find resultXML")
@@ -328 +344 @@
         
         if size is None:
-            size = pageinfo.get('searchResultPageSize', 20)
+            size = pageinfo.get('resultPageSize', 10)
             
         if start is None:
@@ -337 +353 @@
         if fullresult is not None:
             # paginate
-            first = start
+            first = start-1
             len = size
             del fullresult[:first]
@@ -348 +364 @@
                 href = l.get('href')
                 if href:
-                    # take pn from href
-                    m = re.match(r'page-fragment\.xql.*pn=(\d+)', href)
-                    if m is not None:
-                        # and create new url (assuming parent is documentViewer)
-                        #TODO: add highlighting params
-                        url = self.getLink('pn', m.group(1))
-                        l.set('href', url)
-                    else:
-                        logging.warning("getResultPage: Problem with link=%s"%href)
+                    # assume all links go to pages
+                    linkUrl = urlparse.urlparse(href)
+                    linkParams = urlparse.parse_qs(linkUrl.query)
+                    # take some parameters
+                    params = {'pn': linkParams['pn'],
+                              'highlightQuery': linkParams.get('highlightQuery',''),
+                              'highlightElement': linkParams.get('highlightElement',''),
+                              'highlightElementPos': linkParams.get('highlightElementPos','')
+                              }
+                    url = self.getLink(params=params)
+                    l.set('href', url)
                         
-            # fix two-divs-per-row with containing div
-#            newtoc = ET.Element('div', {'class':'queryResultPage'})
-#            for (d1,d2) in zip(tocdivs[::2],tocdivs[1::2]):
-#                e = ET.Element('div',{'class':'tocline'})
-#                e.append(d1)
-#                e.append(d2)
-#                newtoc.append(e)
-                
             return serialize(tocdivs)