X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/52de56522d8d29146b3be2266f57ccef0abe417a..86becb1ae4d215dd782dfc054471d724f6dcd1d7:/src/librarian/parser.py?ds=inline

diff --git a/src/librarian/parser.py b/src/librarian/parser.py
index 8adde33..f4288f8 100644
--- a/src/librarian/parser.py
+++ b/src/librarian/parser.py
@@ -30,7 +30,10 @@ class WLElementLookup(etree.CustomElementClassLookup):
             return
         if namespace:
             return
-        return WL_ELEMENTS[name]
+        try:
+            return WL_ELEMENTS[name]
+        except KeyError:
+            return
 
 
 parser = etree.XMLParser()
@@ -74,30 +77,42 @@ class WLDocument(object):
             self.book_info = None
 
     def get_statistics(self):
-        def count_text(text, counter, in_fn=False):
+        def count_text(text, counter, in_fn=False, stanza=False):
             if text:
                 text = re.sub(r'\s+', ' ', text)
 
                 chars = len(text) if text.strip() else 0
                 words = len(text.split()) if text.strip() else 0
                 
-                counter['chars'] += chars
-                counter['words'] += words
+                counter['chars_with_fn'] += chars
+                counter['words_with_fn'] += words
                 if not in_fn:
-                    counter['chars_with_fn'] += chars
-                    counter['words_with_fn'] += words
+                    counter['chars'] += chars
+                    counter['words'] += words
+                if not stanza:
+                    counter['chars_out_verse_with_fn'] += chars
+                    if not in_fn:
+                        counter['chars_out_verse'] += chars
                 
-        def count(elem, counter, in_fn=False):
+        def count(elem, counter, in_fn=False, stanza=False):
             if elem.tag in (RDFNS('RDF'), 'nota_red', 'abstrakt', 'uwaga', 'ekstra'):
                 return
             if not in_fn and elem.tag in ('pa', 'pe', 'pr', 'pt', 'motyw'):
                 in_fn = True
-            count_text(elem.text, counter, in_fn=in_fn)
+            if elem.tag == 'strofa':
+                # count verses now
+                verses = len(elem.findall('.//br')) + 1
+                counter['verses_with_fn'] += verses
+                if not in_fn:
+                    counter['verses'] += verses
+                stanza = True
+            count_text(elem.text, counter, in_fn=in_fn, stanza=stanza)
             for child in elem:
-                count(child, counter, in_fn=in_fn)
-                count_text(child.tail, counter, in_fn=in_fn)
-            
-            
+                count(child, counter, in_fn=in_fn, stanza=stanza)
+                count_text(child.tail, counter, in_fn=in_fn, stanza=stanza)
+
+        self.swap_endlines()
+
         data = {
             "self": Counter(),
             "parts": [],
@@ -115,7 +130,7 @@ class WLDocument(object):
             else:
                 data['parts'].append((part, part.get_statistics()))
                 for k, v in data['parts'][-1][1]['total'].items():
-                    data['total'][k] += v
+                    data['total'][k] = data['total'].get(k, 0) + v
             
         return data
 
@@ -178,7 +193,7 @@ class WLDocument(object):
         for part_uri in self.book_info.parts:
             try:
                 yield self.from_file(
-                    self.provider.by_uri(part_uri), provider=self.provider
+                    self.provider.by_slug(part_uri.slug), provider=self.provider
                 )
             except Exception as e:
                 if pass_part_errors: