X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/3a0c83394d5783715fab2be29fa1a9cfc3574e28..18db48c927f52d497aea34b5d0daf855902367a3:/src/librarian/document.py diff --git a/src/librarian/document.py b/src/librarian/document.py index aa6f37f..2e9a4a5 100644 --- a/src/librarian/document.py +++ b/src/librarian/document.py @@ -1,13 +1,14 @@ # 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 from .parser import parser -from . import dcparser, DCNS, DirDocProvider +from . import dcparser, DCNS, RDFNS, DirDocProvider from .functions import lang_code_3to2 @@ -36,8 +37,12 @@ class WLDocument: def children(self): for part_uri in self.meta.parts or []: with self.provider.by_slug(part_uri.slug) as f: - yield type(self)(filename=f, provider=self.provider) - + 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) @@ -67,7 +72,7 @@ class WLDocument: 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 @@ -121,3 +126,64 @@ class WLDocument: 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