X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/9d05775ad3576582aeb43bdd95224c58f17e1673..9c9a55d7dda5595a236f11977d7f21f804d37872:/src/librarian/document.py?ds=sidebyside

diff --git a/src/librarian/document.py b/src/librarian/document.py
index fbaf6ca..7780b61 100644
--- a/src/librarian/document.py
+++ b/src/librarian/document.py
@@ -1,16 +1,20 @@
+# This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
+# Copyright © Fundacja Wolne Lektury. See NOTICE for more information.
+#
+from collections import Counter
 import gettext
 import os
 import re
+import urllib.request
 from lxml import etree
-import six
 from .parser import parser
-from . import dcparser, DCNS, DirDocProvider
+from . import dcparser, DCNS, RDFNS, DirDocProvider
 from .functions import lang_code_3to2
 
 
 class WLDocument:
     def __init__(self, filename=None, url=None, provider=None):
-        source = filename or six.moves.urllib.request.urlopen(url)
+        source = filename or urllib.request.urlopen(url)
         tree = etree.parse(source, parser=parser)
         self.tree = tree
         tree.getroot().document = self
@@ -32,15 +36,43 @@ class WLDocument:
     @property
     def children(self):
         for part_uri in self.meta.parts or []:
-            yield type(self)(
-                filename=self.provider.by_slug(part_uri.slug),
-                provider=self.provider
-            )
-            
-    
+            with self.provider.by_slug(part_uri.slug) as f:
+                try:
+                    yield type(self)(filename=f, provider=self.provider)
+                except Exception as e:
+
+                    yield e
+
     def build(self, builder, base_url=None, **kwargs):
         return builder(base_url=base_url).build(self, **kwargs)
 
+    def assign_ids(self, existing=None):
+        # Find all existing IDs.
+        existing = existing or set()
+        que = [self.tree.getroot()]
+        while que:
+            item = que.pop(0)
+            try:
+                item.normalize_insides()
+            except AttributeError:
+                pass
+            existing.add(item.attrib.get('id'))
+            que.extend(item)
+
+        i = 1
+        que = [self.tree.getroot()]
+        while que:
+            item = que.pop(0)
+            que.extend(item)
+            if item.attrib.get('id'):
+                continue
+            if not getattr(item, 'SHOULD_HAVE_ID', False):
+                continue
+            while f'e{i}' in existing:
+                i += 1
+            item.attrib['id'] = f'e{i}'
+            i += 1
+
     def _compat_assign_ordered_ids(self):
         """
         Compatibility: ids in document order, to be roughly compatible with legacy
@@ -49,6 +81,7 @@ class WLDocument:
         """
         EXPR = re.compile(r'/\s', re.MULTILINE | re.UNICODE)
         def _compat_assign_ordered_ids_in_elem(elem, i):
+            if isinstance(elem, etree._Comment): return i
             elem.attrib['_compat_ordered_id'] = str(i)
             i += 1
             if getattr(elem, 'HTML_CLASS', None) == 'stanza':
@@ -75,7 +108,10 @@ class WLDocument:
         def _compat_assigns_section_ids_in_elem(elem, prefix='sec'):
             for i, child in enumerate(elem):
                 idfier = '{}{}'.format(prefix, i + 1)
-                child.attrib['_compat_section_id'] = idfier
+                try:
+                    child.attrib['_compat_section_id'] = idfier
+                except:
+                    pass
                 _compat_assigns_section_ids_in_elem(child, idfier + '-')
         _compat_assigns_section_ids_in_elem(self.tree.getroot().master)
 
@@ -89,3 +125,66 @@ class WLDocument:
             persons.remove(None)
         return persons
 
+    def references(self):
+        return self.tree.findall('.//ref')
+
+    def get_statistics(self):
+        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_with_fn'] += chars
+                counter['words_with_fn'] += words
+                if not in_fn:
+                    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, 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
+            if elem.tag == 'strofa':
+                # count verses now
+                #verses = len(elem.findall('.//br')) + 1
+                verses = list(elem.get_verses())
+                counter['verses_with_fn'] += len(verses)
+                if not in_fn:
+                    counter['verses'] += len(verses)
+                stanza = True
+
+                for child in verses:
+                    count(child, counter, in_fn=in_fn, stanza=True)
+            else:
+                count_text(elem.text, counter, in_fn=in_fn, stanza=stanza)
+                for child in elem:
+                    count(child, counter, in_fn=in_fn, stanza=stanza)
+                    count_text(child.tail, counter, in_fn=in_fn, stanza=stanza)
+
+        data = {
+            "self": Counter(),
+            "parts": [],
+            "total": {
+            }
+        }
+
+        count(self.tree.getroot(), data['self'])
+        for k, v in data['self'].items():
+            data['total'][k] = v
+
+        for part in self.children:
+            if isinstance(part, Exception):
+                data['parts'].append((None, {'error': part}))
+            else:
+                data['parts'].append((part, part.get_statistics()))
+                for k, v in data['parts'][-1][1]['total'].items():
+                    data['total'][k] = data['total'].get(k, 0) + v
+
+        return data