add option to save html dir (debug html to pdf)

[librarian.git] / librarian / parser.py

diff --git a/librarian/parser.py b/librarian/parser.py
index 9068fc0..d515954 100644 (file)
--- a/librarian/parser.py
+++ b/librarian/parser.py
@@ -3,9 +3,8 @@
 # This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #
 # This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #

-from librarian import ValidationError, NoDublinCore,  ParseError, NoProvider
-from librarian import RDFNS
-from librarian.cover import WLCover
+from librarian import ValidationError, NoDublinCore,  ParseError
+from librarian import RDFNS, IOFile

 from librarian import dcparser
 
 from xml.parsers.expat import ExpatError
 from librarian import dcparser
 
 from xml.parsers.expat import ExpatError


@@ -16,62 +15,71 @@ import os
 import re
 from StringIO import StringIO
 
 import re
 from StringIO import StringIO
 

+

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

-    def __init__(self, edoc, parse_dublincore=True, provider=None, 
-                    strict=False, meta_fallbacks=None):
-        self.edoc = edoc
-        self.provider = provider
+    _edoc = None

 
 

-        root_elem = edoc.getroot()
+    @property
+    def edoc(self):
+        if self._edoc is None:
+            data = self.source.get_string()
+            if not isinstance(data, unicode):
+                data = data.decode('utf-8')
+            data = data.replace(u'\ufeff', '')
+            try:
+                parser = etree.XMLParser()
+                self._edoc = etree.parse(StringIO(data.encode('utf-8')), parser)
+            except (ExpatError, XMLSyntaxError, XSLTApplyError), e:
+                raise ParseError(e)
+        return self._edoc
+
+    _rdf_elem = None
+
+    @property
+    def rdf_elem(self):
+        if self._rdf_elem is None:
+            dc_path = './/' + RDFNS('RDF')
+            self._rdf_elem = self.edoc.getroot().find(dc_path)
+            if self._rdf_elem is None:
+                raise NoDublinCore('Document has no DublinCore - which is required.')
+        return self._rdf_elem

 
 

-        dc_path = './/' + RDFNS('RDF')
+    _book_info = None

 
 

+    @property
+    def book_info(self):
+        if not self.parse_dublincore:
+            return None
+        if self._book_info is None:
+            self._book_info = dcparser.BookInfo.from_element(
+                    self.rdf_elem, fallbacks=self.meta_fallbacks, strict=self.strict)
+        return self._book_info
+
+    def __init__(self, iofile, provider=None, parse_dublincore=True,  # shouldn't it be in a subclass?
+                 strict=False,  # ?
+                 meta_fallbacks=None):  # ?
+        self.source = iofile
+        self.provider = provider
+        self.parse_dublincore = parse_dublincore
+        self.strict = strict
+        self.meta_fallbacks = meta_fallbacks
+        root_elem = self.edoc.getroot()

         if root_elem.tag != 'utwor':
             raise ValidationError("Invalid root element. Found '%s', should be 'utwor'" % root_elem.tag)
         if root_elem.tag != 'utwor':
             raise ValidationError("Invalid root element. Found '%s', should be 'utwor'" % root_elem.tag)

-

         if parse_dublincore:
         if parse_dublincore:

-            self.rdf_elem = root_elem.find(dc_path)
-
-            if self.rdf_elem is None:
-                raise NoDublinCore('Document has no DublinCore - which is required.')
-
-            self.book_info = dcparser.BookInfo.from_element(
-                    self.rdf_elem, fallbacks=meta_fallbacks, strict=strict)
-        else:
-            self.book_info = None
+            self.book_info

 
     @classmethod
     def from_string(cls, xml, *args, **kwargs):
 
     @classmethod
     def from_string(cls, xml, *args, **kwargs):

-        return cls.from_file(StringIO(xml), *args, **kwargs)
+        return cls(IOFile.from_string(xml), *args, **kwargs)

 
     @classmethod
     def from_file(cls, xmlfile, *args, **kwargs):
 
     @classmethod
     def from_file(cls, xmlfile, *args, **kwargs):

-
-        # first, prepare for parsing
-        if isinstance(xmlfile, basestring):
-            file = open(xmlfile, 'rb')
-            try:
-                data = file.read()
-            finally:
-                file.close()
-        else:
-            data = xmlfile.read()
-
-        if not isinstance(data, unicode):
-            data = data.decode('utf-8')
-
-        data = data.replace(u'\ufeff', '')
-
-        try:
-            parser = etree.XMLParser(remove_blank_text=False)
-            tree = etree.parse(StringIO(data.encode('utf-8')), parser)
-
-            return cls(tree, *args, **kwargs)
-        except (ExpatError, XMLSyntaxError, XSLTApplyError), e:
-            raise ParseError(e)
+        iofile = IOFile.from_filename(xmlfile)
+        return cls(iofile, *args, **kwargs)

 
     def swap_endlines(self):
         """Converts line breaks in stanzas into <br/> tags."""
 
     def swap_endlines(self):
         """Converts line breaks in stanzas into <br/> tags."""


@@ -94,15 +102,6 @@ class WLDocument(object):
                     elem.insert(0, ins)
                 elem.text = chunks.pop(0)
 
                     elem.insert(0, ins)
                 elem.text = chunks.pop(0)
 

-    def parts(self):
-        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)
-

     def chunk(self, path):
         # convert the path to XPath
         expr = self.path_to_xpath(path)
     def chunk(self, path):
         # convert the path to XPath
         expr = self.path_to_xpath(path)


@@ -122,7 +121,7 @@ class WLDocument(object):
                 parts.append(part)
             else:
                 tag, n = match.groups()
                 parts.append(part)
             else:
                 tag, n = match.groups()

-                parts.append("*[%d][name() = '%s']" % (int(n)+1, tag) )
+                parts.append("*[%d][name() = '%s']" % (int(n)+1, tag))

 
         if parts[0] == '.':
             parts[0] = ''
 
         if parts[0] == '.':
             parts[0] = ''


@@ -135,7 +134,7 @@ class WLDocument(object):
     def update_dc(self):
         if self.book_info:
             parent = self.rdf_elem.getparent()
     def update_dc(self):
         if self.book_info:
             parent = self.rdf_elem.getparent()

-            parent.replace( self.rdf_elem, self.book_info.to_etree(parent) )
+            parent.replace(self.rdf_elem, self.book_info.to_etree(parent))

 
     def serialize(self):
         self.update_dc()
 
     def serialize(self):
         self.update_dc()


@@ -148,44 +147,34 @@ class WLDocument(object):
             try:
                 xpath = self.path_to_xpath(key)
                 node = self.edoc.xpath(xpath)[0]
             try:
                 xpath = self.path_to_xpath(key)
                 node = self.edoc.xpath(xpath)[0]

-                repl = etree.fromstring(u"<%s>%s</%s>" %(node.tag, data, node.tag) )
+                repl = etree.fromstring(u"<%s>%s</%s>" % (node.tag, data, node.tag))

                 node.getparent().replace(node, repl)
             except Exception, e:
                 node.getparent().replace(node, repl)
             except Exception, e:

-                unmerged.append( repr( (key, xpath, e) ) )
+                # WTF xpath may be unused; also: too broad except
+                unmerged.append(repr((key, xpath, e)))

 
         return unmerged
 
     def clean_ed_note(self):
         """ deletes forbidden tags from nota_red """
 
 
         return unmerged
 
     def clean_ed_note(self):
         """ deletes forbidden tags from nota_red """
 

-        for node in self.edoc.xpath('|'.join('//nota_red//%s' % tag for tag in
-                    ('pa', 'pe', 'pr', 'pt', 'begin', 'end', 'motyw'))):
+        forbidden_tags = ('pa', 'pe', 'pr', 'pt', 'begin', 'end', 'motyw')
+        for node in self.edoc.xpath('|'.join('//nota_red//%s' % tag for tag in forbidden_tags)):

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

-    def editors(self):
-        """Returns a set of all editors for book and its children.
-
-        :returns: set of dcparser.Person objects
-        """
-        if self.book_info is None:
-            raise NoDublinCore('No Dublin Core in document.')
-        persons = set(self.book_info.editors +
-                        self.book_info.technical_editors)
-        for child in self.parts():
-            persons.update(child.editors())
-        if None in persons:
-            persons.remove(None)
-        return persons
-

     # Converters
 
     def as_html(self, *args, **kwargs):
         from librarian import pyhtml as html
         return html.transform(self, *args, **kwargs)
 
     # Converters
 
     def as_html(self, *args, **kwargs):
         from librarian import pyhtml as html
         return html.transform(self, *args, **kwargs)
 

+    def as_weasy(self, *args, **kwargs):
+        from librarian import weasy
+        return weasy.transform(self, *args, **kwargs)
+

     def as_text(self, *args, **kwargs):
         from librarian import text
         return text.transform(self, *args, **kwargs)
     def as_text(self, *args, **kwargs):
         from librarian import text
         return text.transform(self, *args, **kwargs)


@@ -195,8 +184,8 @@ class WLDocument(object):
         return epub.transform(self, *args, **kwargs)
 
     def as_pdf(self, *args, **kwargs):
         return epub.transform(self, *args, **kwargs)
 
     def as_pdf(self, *args, **kwargs):

-        from librarian import pdf
-        return pdf.transform(self, *args, **kwargs)
+        from librarian import pypdf
+        return pypdf.EduModulePDFFormat(self).build(*args, **kwargs)

 
     def as_mobi(self, *args, **kwargs):
         from librarian import mobi
 
     def as_mobi(self, *args, **kwargs):
         from librarian import mobi


@@ -208,18 +197,16 @@ class WLDocument(object):
 
     def as_cover(self, cover_class=None, *args, **kwargs):
         if cover_class is None:
 
     def as_cover(self, cover_class=None, *args, **kwargs):
         if cover_class is None:

+            from librarian.styles.wolnelektury.cover import WLCover

             cover_class = WLCover
         return cover_class(self.book_info, *args, **kwargs).output_file()
 
             cover_class = WLCover
         return cover_class(self.book_info, *args, **kwargs).output_file()
 

-    def save_output_file(self, output_file, output_path=None,
-            output_dir_path=None, make_author_dir=False, ext=None):
+    def save_output_file(self, output_file, output_path=None, output_dir_path=None, make_author_dir=False, ext=None):

         if output_dir_path:
             save_path = output_dir_path
             if make_author_dir:
         if output_dir_path:
             save_path = output_dir_path
             if make_author_dir:

-                save_path = os.path.join(save_path,
-                        unicode(self.book_info.author).encode('utf-8'))
-            save_path = os.path.join(save_path,
-                                self.book_info.uri.slug)
+                save_path = os.path.join(save_path, unicode(self.book_info.author).encode('utf-8'))
+            save_path = os.path.join(save_path, self.book_info.uri.slug)

             if ext:
                 save_path += '.%s' % ext
         else:
             if ext:
                 save_path += '.%s' % ext
         else: