merged a conflict in dcparser fields

[librarian.git] / librarian / __init__.py

diff --git a/librarian/__init__.py b/librarian/__init__.py
old mode 100755 (executable)
new mode 100644 (file)
index d56440d..feb9974
--- a/librarian/__init__.py
+++ b/librarian/__init__.py
@@ -1,19 +1,32 @@
 # -*- coding: utf-8 -*-
 # -*- coding: utf-8 -*-

-# exception classes
+#
+# This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
+# Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
+#
+from __future__ import with_statement
+
+import os
+import re
+import shutil

 
 class ParseError(Exception):
 
 class ParseError(Exception):

-    
-    def __init__(self, cause, message=None):
-        self.cause = cause
-        try:
-            self.message = message or self.cause.message
-        except:
-            self.message = "No message."
+    def __str__(self):
+        """ Dirty workaround for Python Unicode handling problems. """
+        return self.message
+
+    def __unicode__(self):
+        """ Dirty workaround for Python Unicode handling problems. """
+        return self.message

 
 class ValidationError(Exception):
     pass
 
 class NoDublinCore(ValidationError):
 
 class ValidationError(Exception):
     pass
 
 class NoDublinCore(ValidationError):

+    """There's no DublinCore section, and it's required."""
+    pass
+
+class NoProvider(Exception):
+    """There's no DocProvider specified, and it's needed."""

     pass
 
 class XMLNamespace(object):
     pass
 
 class XMLNamespace(object):


@@ -26,7 +39,7 @@ class XMLNamespace(object):
         return '{%s}%s' % (self.uri, tag)
 
     def __contains__(self, tag):
         return '{%s}%s' % (self.uri, tag)
 
     def __contains__(self, tag):

-        return tag.startswith('{'+str(self)+'}')
+        return tag.startswith('{' + str(self) + '}')

 
     def __repr__(self):
         return 'XMLNamespace(%r)' % self.uri
 
     def __repr__(self):
         return 'XMLNamespace(%r)' % self.uri


@@ -46,20 +59,115 @@ RDFNS = XMLNamespace('http://www.w3.org/1999/02/22-rdf-syntax-ns#')
 DCNS = XMLNamespace('http://purl.org/dc/elements/1.1/')
 XINS = XMLNamespace("http://www.w3.org/2001/XInclude")
 XHTMLNS = XMLNamespace("http://www.w3.org/1999/xhtml")
 DCNS = XMLNamespace('http://purl.org/dc/elements/1.1/')
 XINS = XMLNamespace("http://www.w3.org/2001/XInclude")
 XHTMLNS = XMLNamespace("http://www.w3.org/1999/xhtml")

+NCXNS = XMLNamespace("http://www.daisy.org/z3986/2005/ncx/")
+OPFNS = XMLNamespace("http://www.idpf.org/2007/opf")

 
 WLNS = EmptyNamespace()
 
 
 WLNS = EmptyNamespace()
 

+
+class WLURI(object):
+    """Represents a WL URI. Extracts slug and language from it."""
+    DEFAULT_LANGUAGE = u'pol'
+
+    slug = None
+    language = None
+
+    example = 'http://wolnelektury.pl/katalog/lektura/template/'
+    _re_wl_uri = re.compile('http://wolnelektury.pl/katalog/lektura/'
+            '(?P<slug>[-a-z0-9]+)(/(?P<lang>[a-z]{3}))?/?$')
+
+    def __init__(self, uri=None):
+        if uri is not None:
+            uri = unicode(uri)
+            self.uri = uri
+            match = self._re_wl_uri.match(uri)
+            if not match:
+                raise ValueError('Supplied URI (%s) does not match '
+                        'the WL document URI template.' % uri)
+            self.slug = match.group('slug')
+            self.language = match.group('lang') or self.DEFAULT_LANGUAGE
+
+    @classmethod
+    def from_slug_and_lang(cls, slug, lang):
+        """Contructs an URI from slug and language code.
+
+        >>> WLURI.from_slug_and_lang('a-slug', WLURI.DEFAULT_LANGUAGE).uri
+        u'http://wolnelektury.pl/katalog/lektura/a-slug/'
+        >>> WLURI.from_slug_and_lang('a-slug', 'deu').uri
+        u'http://wolnelektury.pl/katalog/lektura/a-slug/deu/'
+
+        """
+        if lang is None:
+            lang = cls.DEFAULT_LANGUAGE
+        uri = 'http://wolnelektury.pl/katalog/lektura/%s/' % slug
+        if lang is not None and lang != cls.DEFAULT_LANGUAGE:
+            uri += lang + '/'
+        return cls(uri)
+
+    def __unicode__(self):
+        return self.uri
+
+    def __str__(self):
+        return self.uri
+
+    def __eq__(self, other):
+        return self.slug, self.language == other.slug, other.language
+
+    def filename_stem(self):
+        stem = self.slug
+        if self.language != self.DEFAULT_LANGUAGE:
+            stem += '_' + self.language
+        return stem
+
+    def validate_language(self, language):
+        if language != self.language:
+            raise ValidationError("Incorrect language definition in URI")
+
+
+class DocProvider(object):
+    """Base class for a repository of XML files.
+
+    Used for generating joined files, like EPUBs.
+    """
+
+    def by_slug_and_lang(self, slug, lang=None):
+        """Should return a file-like object with a WL document XML."""
+        raise NotImplementedError
+
+    def by_slug(self, slug):
+        """Should return a file-like object with a WL document XML."""
+        return self.by_slug_and_lang(slug)
+
+    def by_uri(self, uri, wluri=WLURI):
+        """Should return a file-like object with a WL document XML."""
+        wluri = wluri(uri)
+        return self.by_slug_and_lang(wluri.slug, wluri.language)
+
+
+class DirDocProvider(DocProvider):
+    """ Serve docs from a directory of files in form <slug>.xml """
+
+    def __init__(self, dir_):
+        self.dir = dir_
+        self.files = {}
+
+    def by_slug_and_lang(self, slug, lang=None):
+        fname = WLURI.from_slug_and_lang(slug, lang).filename_stem() + '.xml'
+        return open(os.path.join(self.dir, fname))
+
+

 import lxml.etree as etree
 import dcparser
 
 DEFAULT_BOOKINFO = dcparser.BookInfo(
 import lxml.etree as etree
 import dcparser
 
 DEFAULT_BOOKINFO = dcparser.BookInfo(

-        { RDFNS('about'): u'http://wiki.wolnepodreczniki.pl/Lektury:Template'},\
+        { RDFNS('about'): u'http://wiki.wolnepodreczniki.pl/Lektury:Template'},

         { DCNS('creator'): [u'Some, Author'],
           DCNS('title'): [u'Some Title'],
           DCNS('subject.period'): [u'Unknown'],
           DCNS('subject.type'): [u'Unknown'],
           DCNS('subject.genre'): [u'Unknown'],
           DCNS('date'): ['1970-01-01'],
         { DCNS('creator'): [u'Some, Author'],
           DCNS('title'): [u'Some Title'],
           DCNS('subject.period'): [u'Unknown'],
           DCNS('subject.type'): [u'Unknown'],
           DCNS('subject.genre'): [u'Unknown'],
           DCNS('date'): ['1970-01-01'],

+          DCNS('language'): [WLURI.DEFAULT_LANGUAGE],

           # DCNS('date'): [creation_date],
           DCNS('publisher'): [u"Fundacja Nowoczesna Polska"],
           DCNS('description'):
           # DCNS('date'): [creation_date],
           DCNS('publisher'): [u"Fundacja Nowoczesna Polska"],
           DCNS('description'):


@@ -67,43 +175,113 @@ DEFAULT_BOOKINFO = dcparser.BookInfo(
              Wolne Lektury (http://wolnelektury.pl). Reprodukcja cyfrowa
              wykonana przez Bibliotekę Narodową z egzemplarza
              pochodzącego ze zbiorów BN."""],
              Wolne Lektury (http://wolnelektury.pl). Reprodukcja cyfrowa
              wykonana przez Bibliotekę Narodową z egzemplarza
              pochodzącego ze zbiorów BN."""],

-          DCNS('identifier.url'):
-            [u"http://wolnelektury.pl/katalog/lektura/template"],
+          DCNS('identifier.url'): [WLURI.example],

           DCNS('rights'):
             [u"Domena publiczna - zm. [OPIS STANU PRAWNEGO TEKSTU]"] })
 
 def xinclude_forURI(uri):
           DCNS('rights'):
             [u"Domena publiczna - zm. [OPIS STANU PRAWNEGO TEKSTU]"] })
 
 def xinclude_forURI(uri):

-    e = etree.Element( XINS("include") )
+    e = etree.Element(XINS("include"))

     e.set("href", uri)
     return etree.tostring(e, encoding=unicode)
     e.set("href", uri)
     return etree.tostring(e, encoding=unicode)

-    
+

 def wrap_text(ocrtext, creation_date, bookinfo=DEFAULT_BOOKINFO):
     """Wrap the text within the minimal XML structure with a DC template."""
     bookinfo.created_at = creation_date
 def wrap_text(ocrtext, creation_date, bookinfo=DEFAULT_BOOKINFO):
     """Wrap the text within the minimal XML structure with a DC template."""
     bookinfo.created_at = creation_date

-    
-    dcstring = etree.tostring(bookinfo.to_etree(),\
+
+    dcstring = etree.tostring(bookinfo.to_etree(), \

         method='xml', encoding=unicode, pretty_print=True)
 
         method='xml', encoding=unicode, pretty_print=True)
 

-    return u'<utwor>\n' + dcstring + u'\n<plain-text>\n' + ocrtext +\
-        u'\n</plain-text>\n</utwor>';
+    return u'<utwor>\n' + dcstring + u'\n<plain-text>\n' + ocrtext + \
+        u'\n</plain-text>\n</utwor>'

 
 
 def serialize_raw(element):
     b = u'' + (element.text or '')
 
     for child in element.iterchildren():
 
 
 def serialize_raw(element):
     b = u'' + (element.text or '')
 
     for child in element.iterchildren():

-        e = etree.tostring(child, method='xml', encoding=unicode, pretty_print=True)
+        e = etree.tostring(child, method='xml', encoding=unicode,
+                pretty_print=True)

         b += e
 
     return b
 
         b += e
 
     return b
 

-from wl_light import serialize_nl
-
-

 SERIALIZERS = {
     'raw': serialize_raw,
 SERIALIZERS = {
     'raw': serialize_raw,

-    'nl': serialize_nl,

 }
 
 def serialize_children(element, format='raw'):
     return SERIALIZERS[format](element)
 }
 
 def serialize_children(element, format='raw'):
     return SERIALIZERS[format](element)

+
+def get_resource(path):
+    return os.path.join(os.path.dirname(__file__), path)
+
+
+class OutputFile(object):
+    """Represents a file returned by one of the converters."""
+
+    _string = None
+    _filename = None
+
+    def __del__(self):
+        if self._filename:
+            os.unlink(self._filename)
+
+    def __nonzero__(self):
+        return self._string is not None or self._filename is not None
+
+    @classmethod
+    def from_string(cls, string):
+        """Converter returns contents of a file as a string."""
+
+        instance = cls()
+        instance._string = string
+        return instance
+
+    @classmethod
+    def from_filename(cls, filename):
+        """Converter returns contents of a file as a named file."""
+
+        instance = cls()
+        instance._filename = filename
+        return instance
+
+    def get_string(self):
+        """Get file's contents as a string."""
+
+        if self._filename is not None:
+            with open(self._filename) as f:
+                return f.read()
+        else:
+            return self._string
+
+    def get_file(self):
+        """Get file as a file-like object."""
+
+        if self._string is not None:
+            from StringIO import StringIO
+            return StringIO(self._string)
+        elif self._filename is not None:
+            return open(self._filename)
+
+    def get_filename(self):
+        """Get file as a fs path."""
+
+        if self._filename is not None:
+            return self._filename
+        elif self._string is not None:
+            from tempfile import NamedTemporaryFile
+            temp = NamedTemporaryFile(prefix='librarian-', delete=False)
+            temp.write(self._string)
+            temp.close()
+            self._filename = temp.name
+            return self._filename
+        else:
+            return None
+
+    def save_as(self, path):
+        """Save file to a path. Create directories, if necessary."""
+
+        dirname = os.path.dirname(os.path.abspath(path))
+        if not os.path.isdir(dirname):
+            os.makedirs(dirname)
+        shutil.copy(self.get_filename(), path)