Add unprefixed term for path components

src/filters/rclpdf.py

@@ -470,39 +470,6 @@ class PDFExtractor:
             self.em.setfield("pdfannot", annotsfield)
         return html
     
-    def _patch_meta(self, html):
-        '''This fixes https://gitlab.freedesktop.org/poppler/poppler/-/issues/136'''
-
-        if not _mswindows:
-            pdfinfo = rclexecm.which("pdfinfo")
-        if not pdfinfo:
-            pdfinfo = rclexecm.which("poppler/pdfinfo")
-        
-        if not pdfinfo:
-            return html
-
-        info = subprocess.check_output([pdfinfo, '-isodates', self.filename])
-
-        meta = {}
-        for line in info.split(b'\n'):
-            try:
-                key, value = line.strip().split(b':', 1)
-                meta[key.strip()] = value.strip().replace(b'"', b'\"')
-            except ValueError:
-                pass
-
-        if b'CreationDate' in meta:
-            meta[b'date'] = meta[b'CreationDate']
-
-        title = meta.get('Title')
-        head = [b'<title>%s</title>' % title] if title else []
-        head += [ b'<meta name="%s" content="%s"/>' % x for x in meta.items() ]
-
-        start = html.index(b'<head>') + 6
-        end = html.index(b'</head>')
-        return html[:start] + b"".join(head) + html[end:]
-
-
     def _selfdoc(self):
         '''Extract the text from the pdf doc (as opposed to attachment)'''
         self.em.setmimetype('text/html')
@@ -516,8 +483,6 @@ class PDFExtractor:
                                         "UTF-8", "-eol", "unix", "-q",
                                         self.filename, "-"])
 
-        html = self._patch_meta(html)
-
         html, isempty = self._fixhtml(html)
         #self.em.rclog("ISEMPTY: %d : data: \n%s" % (isempty, html))
 

src/rcldb/rcldb.cpp

@@ -1598,7 +1598,9 @@ bool Db::addOrUpdate(const string &udi, const string &parent_udi, Doc &doc)
                     elt = elt.substr(0, 230);
                 }
                 newdocument.add_posting(wrap_prefix(pathelt_prefix) + elt, 
-                                        splitter.basepos + splitter.curpos++);
+                                        splitter.basepos + splitter.curpos);
+		newdocument.add_posting(elt,
+					splitter.basepos + splitter.curpos++);
             }
             splitter.basepos += splitter.curpos + 100;
         }