aaa

[librarian.git] / librarian / parser.py

diff --git a/librarian/parser.py b/librarian/parser.py
index 5ae06e2..f02b64c 100644 (file)
--- a/librarian/parser.py
+++ b/librarian/parser.py
@@ -4,7 +4,7 @@
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #
 from librarian import ValidationError, NoDublinCore,  ParseError, NoProvider
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #
 from librarian import ValidationError, NoDublinCore,  ParseError, NoProvider

-from librarian import RDFNS
+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


@@ -19,56 +19,68 @@ class WLDocument(object):
     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
 

-    def __init__(self, edoc, parse_dublincore=True, provider=None):
-        self.edoc = edoc
-        self.provider = provider
-
-        root_elem = edoc.getroot()
-
-        dc_path = './/' + RDFNS('RDF')
+    _edoc = None
+    @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(remove_blank_text=False)
+                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

 
 

-        if root_elem.tag != 'utwor':
+    _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
+        if self.edoc.getroot().tag != 'utwor':

             raise ValidationError("Invalid root element. Found '%s', should be 'utwor'" % root_elem.tag)
             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)
-        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
 
     @classmethod

-    def from_file(cls, xmlfile, parse_dublincore=True, provider=None):
-
-        # first, prepare for parsing
+    def from_file(cls, xmlfile, *args, **kwargs):

         if isinstance(xmlfile, basestring):
         if isinstance(xmlfile, basestring):

-            file = open(xmlfile, 'rb')
-            try:
-                data = file.read()
-            finally:
-                file.close()
+            iofile = IOFile.from_filename(xmlfile)

         else:
         else:

-            data = xmlfile.read()
-
-        if not isinstance(data, unicode):
-            data = data.decode('utf-8')
-
-        data = data.replace(u'\ufeff', '')
+            iofile = IOFile.from_file(xmlfile)
+        return cls(iofile, *args, **kwargs)

 
 

-        try:
-            parser = etree.XMLParser(remove_blank_text=False)
-            tree = etree.parse(StringIO(data.encode('utf-8')), parser)
-
-            return cls(tree, parse_dublincore=parse_dublincore, provider=provider)
-        except (ExpatError, XMLSyntaxError, XSLTApplyError), e:
-            raise ParseError(e)

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


@@ -92,10 +104,10 @@ class WLDocument(object):
                 elem.text = chunks.pop(0)
 
     def parts(self):
                 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.')
         if self.book_info is None:
             raise NoDublinCore('No Dublin Core in document.')

+        if self.book_info.parts and self.provider is None:
+            raise NoProvider('No document provider supplied.')

         for part_uri in self.book_info.parts:
             yield self.from_file(self.provider.by_uri(part_uri),
                     provider=self.provider)
         for part_uri in self.book_info.parts:
             yield self.from_file(self.provider.by_uri(part_uri),
                     provider=self.provider)


@@ -146,7 +158,7 @@ class WLDocument(object):
                 xpath = self.path_to_xpath(key)
                 node = self.edoc.xpath(xpath)[0]
                 repl = etree.fromstring(u"<%s>%s</%s>" %(node.tag, data, node.tag) )
                 xpath = self.path_to_xpath(key)
                 node = self.edoc.xpath(xpath)[0]
                 repl = etree.fromstring(u"<%s>%s</%s>" %(node.tag, data, node.tag) )

-                node.getparent().replace(node, repl);
+                node.getparent().replace(node, repl)

             except Exception, e:
                 unmerged.append( repr( (key, xpath, e) ) )
 
             except Exception, e:
                 unmerged.append( repr( (key, xpath, e) ) )
 


@@ -162,10 +174,25 @@ class WLDocument(object):
             node.tag = 'span'
             node.tail = tail
 
             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):
     # Converters
 
     def as_html(self, *args, **kwargs):

-        from librarian import html
+        from librarian import pyhtml as html

         return html.transform(self, *args, **kwargs)
 
     def as_text(self, *args, **kwargs):
         return html.transform(self, *args, **kwargs)
 
     def as_text(self, *args, **kwargs):


@@ -184,6 +211,16 @@ class WLDocument(object):
         from librarian import mobi
         return mobi.transform(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_cover(self, cover_class=None, *args, **kwargs):
+        if cover_class is None:
+            from librarian.cover import WLCover
+            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):
         if output_dir_path:
     def save_output_file(self, output_file, output_path=None,
             output_dir_path=None, make_author_dir=False, ext=None):
         if output_dir_path:


@@ -192,7 +229,7 @@ class WLDocument(object):
                 save_path = os.path.join(save_path,
                         unicode(self.book_info.author).encode('utf-8'))
             save_path = os.path.join(save_path,
                 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.filename_stem())
+                                self.book_info.uri.slug)

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