Changeset 565:1b483194901c in documentViewer

Timestamp:

Oct 9, 2012, 5:01:18 PM (12 years ago)

Author:

casties

Branch:

default

Message:

more new MpiwgXmlTextServer?.

Files:

• MpiwgXmlTextServer.py (modified) (11 diffs)
• css/docuviewer.css (modified) (1 diff)
• documentViewer.py (modified) (1 diff)

MpiwgXmlTextServer.py

@@ -89 +89 @@
     
           
-    def getTextInfo(self, mode='', docinfo=None):
+    def getTextInfo(self, mode=None, docinfo=None):
         """reads document info, including page concordance, from text server"""
         logging.debug("getTextInfo mode=%s"%mode)
-        if mode not in ['toc', 'figures', '']:
-            mode = ''
+        
+        field = ''
+        if mode in ['pages', 'toc', 'figures']:
+            # translate mode to field param
+            field = '&field=%s'%mode
+        else:
+            mode = None
+
         # check cached info
         if mode:
@@ -101 +107 @@
             
         else:
-            # no toc-request
+            # cached but no toc-request?
             if 'numTextPages' in docinfo:
                 return docinfo
@@ -111 +117 @@
                 
         # fetch docinfo            
-        pagexml = self.getServerData("query/GetDocInfo","docId=%s&field=%s"%(docpath,mode))
+        pagexml = self.getServerData("query/GetDocInfo","docId=%s%s"%(docpath,field))
         dom = ET.fromstring(pagexml)
-        # all info in tag <document>
-        doc = dom.find("doc")
+        # all info in tag <doc>
+        doc = dom
         if doc is None:
             logging.error("getTextInfo: unable to find document-tag!")
         else:
-            # result is in list-tag
-            l = doc.find('list')
-            if l is not None:
-                lt = l.get('type')
-                # pageNumbers
-                if lt == 'pages':
-                    # contains tags with page numbers
-                    # <item n="14" o="2" o-norm="2" file="0014"/>
-                    # n=scan number, o=original page no, on=normalized original page no
-                    # pageNumbers is a dict indexed by scan number
-                    pages = {}
-                    for i in l:
-                        page = {}
-                        pn = getInt(i.get('n'))
-                        page['pn'] = pn
-                        no = getInt(i.get('o'))
-                        page['no'] = no
-                        non = getInt(i.get('o-norm'))
-                        page['non'] = non
-                                
-                        if pn > 0:
-                            pages[pn] = page
+            if mode is None:
+                # get general info from system-tag
+                cp = doc.find('system/countPages')
+                if cp is not None:
+                    docinfo['numTextPages'] = getInt(cp.text) 
+                    
+            else:
+                # result is in list-tag
+                l = doc.find('list')
+                if l is not None:
+                    lt = l.get('type')
+                    # pageNumbers
+                    if lt == 'pages':
+                        # contains tags with page numbers
+                        # <item n="14" o="2" o-norm="2" file="0014"/>
+                        # n=scan number, o=original page no, on=normalized original page no
+                        # pageNumbers is a dict indexed by scan number
+                        pages = {}
+                        for i in l:
+                            page = {}
+                            pn = getInt(i.get('n'))
+                            page['pn'] = pn
+                            no = getInt(i.get('o'))
+                            page['no'] = no
+                            non = getInt(i.get('o-norm'))
+                            page['non'] = non
+                                    
+                            if pn > 0:
+                                pages[pn] = page
+                            
+                        docinfo['pageNumbers'] = pages
+                        logging.debug("got pageNumbers=%s"%repr(pages))
+                                    
+                    # toc
+                    elif name == 'toc':
+                        # contains tags with table of contents/figures
+                        # <toc-entry><page>13</page><level>3</level><content>Chapter I</content><level-string>1.</level-string><real-level>1</real-level></toc-entry>
+                        tocs = []
+                        for te in tag:
+                            toc = {}
+                            for t in te:
+                                if t.tag == 'page':
+                                    toc['pn'] = getInt(t.text)
+                                elif t.tag == 'level':
+                                    toc['level'] = t.text
+                                elif t.tag == 'content':
+                                    toc['content'] = t.text
+                                elif t.tag == 'level-string':
+                                    toc['level-string'] = t.text
+                                elif t.tag == 'real-level':
+                                    toc['real-level'] = t.text
+                                    
+                            tocs.append(toc)
                         
-                    docinfo['numTextPages'] = len(pages)
-                    docinfo['pageNumbers'] = pages
-                    logging.debug("got pageNumbers=%s"%repr(pages))
-                                
-                # toc
-                elif name == 'toc':
-                    # contains tags with table of contents/figures
-                    # <toc-entry><page>13</page><level>3</level><content>Chapter I</content><level-string>1.</level-string><real-level>1</real-level></toc-entry>
-                    tocs = []
-                    for te in tag:
-                        toc = {}
-                        for t in te:
-                            if t.tag == 'page':
-                                toc['pn'] = getInt(t.text)
-                            elif t.tag == 'level':
-                                toc['level'] = t.text
-                            elif t.tag == 'content':
-                                toc['content'] = t.text
-                            elif t.tag == 'level-string':
-                                toc['level-string'] = t.text
-                            elif t.tag == 'real-level':
-                                toc['real-level'] = t.text
-                                
-                        tocs.append(toc)
-                    
-                    # save as full_toc/full_figures
-                    docinfo['full_%s'%mode] = tocs
+                        # save as full_toc/full_figures
+                        docinfo['full_%s'%mode] = tocs
 
         return docinfo
@@ -221 +233 @@
         textParams = {'docId': docpath,
                       'page': pn}
+        
         if 'characterNormalization' in pageinfo:
-            textParams['normalization'] = pageinfo['characterNormalization']
+            cn = pageinfo['characterNormalization']
+            # TODO: change values in form
+            if cn == 'regPlusNorm':
+                cn = 'norm'
+                
+            textParams['normalization'] = cn
         
         if not mode:
@@ -273 +291 @@
         # fetch the page
         pagexml = self.getServerData("query/GetPage",urllib.urlencode(textParams))
-        dom = ET.fromstring(pagexml)
-        # extract additional info
-        #self.processPageInfo(dom, docinfo, pageinfo)
-        # page content is in <div class="pageContent">
+        try:
+            dom = ET.fromstring(pagexml)
+        except Exception, e:
+            logging.error("Error parsing page: %s"%e)
+            return None
+        
         pagediv = None
         body = dom.find('.//body')
@@ -283 +303 @@
             return None
         
-        # currently there's lots of divs...
-        textspan = body.find('span/span')
-        divs = textspan.findall('div')
-        logging.debug("textdivs: %s"%repr(divs))
-        pagediv = divs[0]
-        logging.debug("pagediv: %s"%serialize(pagediv))
+        # the text is in div@class=text
+        pagediv = body.find(".//div[@class='text']")
+        logging.debug("pagediv: %s"%repr(pagediv))
         
         # plain text mode
         if textmode == "text":
-            # get full url assuming documentViewer is parent
-            selfurl = self.getLink()
             if pagediv is not None:
+                # handle pb-tag
+                self._extractPbTag(pagediv, pageinfo)
+                # get full url assuming documentViewer is parent
+                selfurl = self.getLink()
                 if punditMode:
-                    pagediv = self.addPunditAttributes(pagediv, pageinfo, docinfo)
+                    self._addPunditAttributes(pagediv, pageinfo, docinfo)
                     
                 # fix empty div tags
-                divs = pagediv.findall('.//div')
-                for d in divs:
-                    if len(d) == 0 and not d.text:
-                        # make empty divs non-empty
-                        d.text = ' '
-                    
+                self._fixEmptyDivs(pagediv)
                 # check all a-tags
                 links = pagediv.findall('.//a')
                 for l in links:
                     href = l.get('href')
+                    # handle notes FIXME!
                     if href and href.startswith('#note-'):
                         href = href.replace('#note-',"%s#note-"%selfurl)
                         l.set('href', href)
-
+                        
                 return serialize(pagediv)
             
@@ -318 +333 @@
         elif textmode == "dict":
             if pagediv is not None:
+                # handle pb-div
+                self._extractPbTag(pagediv, pageinfo)
                 viewerurl = docinfo['viewerUrl']
                 selfurl = self.getLink()
@@ -324 +341 @@
                     
                 # fix empty div tags
-                divs = pagediv.findall('.//div')
-                for d in divs:
-                    if len(d) == 0 and not d.text:
-                        # make empty divs non-empty
-                        d.text = ' '
-                    
+                self._fixEmptyDivs(pagediv)   
                 # check all a-tags
                 links = pagediv.findall(".//a")
                 for l in links:
                     href = l.get('href')
-                    
                     if href:
                         # is link with href
@@ -366 +377 @@
             if pagediv is not None:
                 # fix empty div tags
-                divs = pagediv.findall('.//div')
-                for d in divs:
-                    if len(d) == 0 and not d.text:
-                        # make empty divs non-empty
-                        d.text = ' '
-                    
+                self._fixEmptyDivs(pagediv)
                 # check all a-tags
                 links = pagediv.findall(".//a")
@@ -387 +393 @@
                     
         return None
+
+    def _extractPbTag(self, pagediv, pageinfo):
+        """extracts information from pb-tag and removes it from pagediv"""
+        pbdiv = pagediv.find(".//span[@class='pb']")
+        if pbdiv is None:
+            logging.warning("getTextPage: no pb-span!")
+            return pagediv
+        
+        # extract running head
+        rh = pbdiv.find(".//span[@class='rhead']")
+        if rh is not None:
+            pageinfo['pageHeaderTitle'] = getText(rh)
+            
+        # remove pb-div from parent
+        ppdiv = pagediv.find(".//span[@class='pb']/..")
+        ppdiv.remove(pbdiv)        
+        return pagediv
     
-    def addPunditAttributes(self, pagediv, pageinfo, docinfo):
+    def _addPunditAttributes(self, pagediv, pageinfo, docinfo):
         """add about attributes for pundit annotation tool"""
         textid = docinfo.get('DRI', "fn=%s"%docinfo.get('documentPath', '???'))
@@ -404 +427 @@
 
         return pagediv
+
+    def _fixEmptyDivs(self, pagediv):
+        """fixes empty div-tags by inserting a space"""
+        divs = pagediv.findall('.//div')
+        for d in divs:
+            if len(d) == 0 and not d.text:
+                # make empty divs non-empty
+                d.text = ' '
+  
+        return pagediv
+
 
     def getSearchResults(self, mode, query=None, pageinfo=None, docinfo=None):

css/docuviewer.css

@@ -257 +257 @@
     font-size: 12px;
 }
+div.col.main div.content.text .bf {
+    font-weight: bold;
+}
+div.col.main div.content.text .head {
+    margin-top: 0.5em;
+    margin-bottom: 0.25em;
+}
+/* running head */
+div.col.main div.content.text div.pageHeaderTitle {
+    text-align: center;
+    margin-bottom: 1em;
+}
+/* figures */
+div.col.main div.content.text span.figure {
+    display: block;
+    width: 200px;
+    margin-top: 0.5em;
+    margin-bottom: 0.5em;
+    padding: 5px;
+    border: 1px dashed silver;
+    /* float: right; */
+    /* text-align: center; */
+}
+div.col.main div.content.text span.figure>a,
+div.col.main div.content.text span.figure span.figureNumber,
+div.col.main div.content.text span.figure span.caption,
+div.col.main div.content.text span.figure span.description {
+    display:block;
+}
+div.col.main div.content.text span.figure span.figureNum {
+    display: none;
+}
 /*
  * search results

documentViewer.py

@@ -587 +587 @@
                 # document info (including toc) from full text
                 if docinfo.get('textURLPath', None):
-                    docinfo = self.getTextInfo(mode='pages', docinfo=docinfo)
+                    docinfo = self.getTextInfo(mode=None, docinfo=docinfo)
             
             # bib info