br

[librarian.git] / src / librarian / parser.py

diff --git a/src/librarian/parser.py b/src/librarian/parser.py
index 2bb9509..1f18dbd 100644 (file)
--- a/src/librarian/parser.py
+++ b/src/librarian/parser.py
@@ -1,25 +1,41 @@
-# -*- coding: utf-8 -*-
-#

 # This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
 # This file is part of Librarian, licensed under GNU Affero GPLv3 or later.

-# Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
+# Copyright © Fundacja Wolne Lektury. See NOTICE for more information.

 #
 #

-from __future__ import unicode_literals
-
+from collections import Counter
+import io
+import os
+import re
+from xml.parsers.expat import ExpatError
+from lxml import etree
+from lxml.etree import XMLSyntaxError, XSLTApplyError

 from librarian import ValidationError, NoDublinCore,  ParseError, NoProvider
 from librarian import RDFNS
 from librarian.cover import make_cover
 from librarian import dcparser
 from librarian import ValidationError, NoDublinCore,  ParseError, NoProvider
 from librarian import RDFNS
 from librarian.cover import make_cover
 from librarian import dcparser

+from .elements import WL_ELEMENTS

 
 

-from xml.parsers.expat import ExpatError
-from lxml import etree
-from lxml.etree import XMLSyntaxError, XSLTApplyError

 
 

-import os
-import re
-import six
+class WLElementLookup(etree.CustomElementClassLookup):
+    def lookup(self, node_type, document, namespace, name):
+        if node_type != 'element':
+            return
+        if namespace:
+            return
+        try:
+            return WL_ELEMENTS[name]
+        except KeyError:
+            return
+

 
 

+parser = etree.XMLParser()
+parser.set_element_class_lookup(
+    WLElementLookup()
+)

 
 

-class WLDocument(object):
+
+
+class WLDocument:
+    """Legacy class, to be replaced with documents.WLDocument."""

     LINE_SWAP_EXPR = re.compile(r'/\s', re.MULTILINE | re.UNICODE)
     provider = None
 
     LINE_SWAP_EXPR = re.compile(r'/\s', re.MULTILINE | re.UNICODE)
     provider = None
 


@@ -53,13 +69,13 @@ class WLDocument(object):
 
     @classmethod
     def from_bytes(cls, xml, *args, **kwargs):
 
     @classmethod
     def from_bytes(cls, xml, *args, **kwargs):

-        return cls.from_file(six.BytesIO(xml), *args, **kwargs)
+        return cls.from_file(io.BytesIO(xml), *args, **kwargs)

 
     @classmethod
     def from_file(cls, xmlfile, *args, **kwargs):
 
         # first, prepare for parsing
 
     @classmethod
     def from_file(cls, xmlfile, *args, **kwargs):
 
         # first, prepare for parsing

-        if isinstance(xmlfile, six.text_type):
+        if isinstance(xmlfile, str):

             file = open(xmlfile, 'rb')
             try:
                 data = file.read()
             file = open(xmlfile, 'rb')
             try:
                 data = file.read()


@@ -68,14 +84,14 @@ class WLDocument(object):
         else:
             data = xmlfile.read()
 
         else:
             data = xmlfile.read()
 

-        if not isinstance(data, six.text_type):
+        if not isinstance(data, str):

             data = data.decode('utf-8')
 
             data = data.decode('utf-8')
 

-        data = data.replace(u'\ufeff', '')
+        data = data.replace('\ufeff', '')

 
         try:
             parser = etree.XMLParser(remove_blank_text=False)
 
         try:
             parser = etree.XMLParser(remove_blank_text=False)

-            tree = etree.parse(six.BytesIO(data.encode('utf-8')), parser)
+            tree = etree.parse(io.BytesIO(data.encode('utf-8')), parser)

 
             return cls(tree, *args, **kwargs)
         except (ExpatError, XMLSyntaxError, XSLTApplyError) as e:
 
             return cls(tree, *args, **kwargs)
         except (ExpatError, XMLSyntaxError, XSLTApplyError) as e:


@@ -102,15 +118,20 @@ class WLDocument(object):
                     elem.insert(0, ins)
                 elem.text = chunks.pop(0)
 
                     elem.insert(0, ins)
                 elem.text = chunks.pop(0)
 

-    def parts(self):
+    def parts(self, pass_part_errors=False):

         if self.provider is None:
             raise NoProvider('No document provider supplied.')
         if self.book_info is None:
             raise NoDublinCore('No Dublin Core in document.')
         for part_uri in self.book_info.parts:
         if self.provider is None:
             raise NoProvider('No document provider supplied.')
         if self.book_info is None:
             raise NoDublinCore('No Dublin Core in document.')
         for part_uri in self.book_info.parts:

-            yield self.from_file(
-                self.provider.by_uri(part_uri), provider=self.provider
-            )
+            try:
+                with self.provider.by_slug(part_uri.slug) as f:
+                    yield self.from_file(f, provider=self.provider)
+            except Exception as e:
+                if pass_part_errors:
+                    yield e
+                else:
+                    raise

 
     def chunk(self, path):
         # convert the path to XPath
 
     def chunk(self, path):
         # convert the path to XPath


@@ -171,12 +192,26 @@ class WLDocument(object):
 
         for node in self.edoc.xpath('|'.join(
                 '//%s//%s' % (note_tag, tag) for tag in
 
         for node in self.edoc.xpath('|'.join(
                 '//%s//%s' % (note_tag, tag) for tag in

-                ('pa', 'pe', 'pr', 'pt', 'begin', 'end', 'motyw'))):
+                ('pa', 'pe', 'pr', 'pt', 'ptrad', 'begin', 'end', 'motyw'))):

             tail = node.tail
             node.clear()
             node.tag = 'span'
             node.tail = tail
 
             tail = node.tail
             node.clear()
             node.tag = 'span'
             node.tail = tail
 

+    def fix_pa_akap(self):
+        for pa in ('pa','pe','pr','pt', 'ptrad'):
+            for akap in self.edoc.findall(f'//{pa}/akap'):
+                akap.getparent().set('blocks', 'true')
+                if not akap.getparent().index(akap):
+                    akap.set('inline', 'true')
+            
+    def hebr_protect(self):
+        for s in self.edoc.findall('//slowo_obce'):
+            if not s.text and len(s) == 1 and s[0].tag == 'slowo_obce':
+                continue
+            if re.match(r'^[\s\u0590-\u05ff]+$', s.text or ''):
+                s.attrib['protect'] = 'true'
+

     def editors(self):
         """Returns a set of all editors for book and its children.
 
     def editors(self):
         """Returns a set of all editors for book and its children.
 


@@ -202,18 +237,10 @@ class WLDocument(object):
         from librarian import text
         return text.transform(self, *args, **kwargs)
 
         from librarian import text
         return text.transform(self, *args, **kwargs)
 

-    def as_epub(self, *args, **kwargs):
-        from librarian import epub
-        return epub.transform(self, *args, **kwargs)
-

     def as_pdf(self, *args, **kwargs):
         from librarian import pdf
         return pdf.transform(self, *args, **kwargs)
 
     def as_pdf(self, *args, **kwargs):
         from librarian import pdf
         return pdf.transform(self, *args, **kwargs)
 

-    def as_mobi(self, *args, **kwargs):
-        from librarian import mobi
-        return mobi.transform(self, *args, **kwargs)
-

     def as_fb2(self, *args, **kwargs):
         from librarian import fb2
         return fb2.transform(self, *args, **kwargs)
     def as_fb2(self, *args, **kwargs):
         from librarian import fb2
         return fb2.transform(self, *args, **kwargs)


@@ -237,7 +264,7 @@ class WLDocument(object):
             if make_author_dir:
                 save_path = os.path.join(
                     save_path,
             if make_author_dir:
                 save_path = os.path.join(
                     save_path,

-                    six.text_type(self.book_info.author).encode('utf-8')
+                    str(self.book_info.author).encode('utf-8')

                 )
             save_path = os.path.join(save_path, self.book_info.url.slug)
             if ext:
                 )
             save_path = os.path.join(save_path, self.book_info.url.slug)
             if ext: