From 6642c1c71c5c6ce6ef3401c8c9da84cf076b018b Mon Sep 17 00:00:00 2001
From: =?utf8?q?Marek=20St=C4=99pniowski?= <marek@stepniowski.com>
Date: Fri, 19 Mar 2010 17:00:19 +0100
Subject: [PATCH] Version 1.2.1.

---
 .gitignore                                    |   3 +
 MANIFEST.in                                   |   2 +-
 README.txt                                    |  49 +++
 librarian/__init__.py                         |  11 +
 librarian/__init__.pyc                        | Bin 151 -> 0 bytes
 librarian/book2txt.xslt                       |  19 +-
 librarian/dcparser.py                         | 323 ++++++++++++------
 librarian/dcparser.pyc                        | Bin 8696 -> 0 bytes
 librarian/html.py                             |  30 +-
 librarian/html.pyc                            | Bin 8882 -> 0 bytes
 librarian/parser.py                           |  70 ++++
 librarian/tests/__init__.py                   | 115 -------
 librarian/tests/files/dcparser/.DS_Store      | Bin 6148 -> 0 bytes
 librarian/text.py                             |  48 +--
 librarian/text.pyc                            | Bin 2993 -> 0 bytes
 scripts/book2html                             |  24 +-
 scripts/book2txt                              |  30 +-
 setup.py                                      |  12 +-
 tests/__init__.py                             |   0
 .../dcparser/andersen_brzydkie_kaczatko.out   |  17 +
 .../dcparser/andersen_brzydkie_kaczatko.xml   |   0
 tests/files/dcparser/biedrzycki_akslop.out    |  17 +
 .../files/dcparser/biedrzycki_akslop.xml      |   0
 tests/files/dcparser/kochanowski_piesn7.out   |  18 +
 .../files/dcparser/kochanowski_piesn7.xml     |   2 +-
 tests/files/dcparser/mickiewicz_rybka.out     |  18 +
 .../files/dcparser/mickiewicz_rybka.xml       |   0
 tests/files/dcparser/sofokles_antygona.out    |  19 ++
 .../files/dcparser/sofokles_antygona.xml      |   0
 .../andersen_brzydkie_kaczatko.out            |   0
 .../andersen_brzydkie_kaczatko.xml            |  24 ++
 tests/files/dcserialize/biedrzycki_akslop.out |   0
 tests/files/dcserialize/biedrzycki_akslop.xml |  25 ++
 .../files/dcserialize/kochanowski_piesn7.out  |   0
 .../files/dcserialize/kochanowski_piesn7.xml  |  27 ++
 tests/files/dcserialize/mickiewicz_rybka.out  |   0
 tests/files/dcserialize/mickiewicz_rybka.xml  |  28 ++
 tests/files/dcserialize/sofokles_antygona.out |   0
 tests/files/dcserialize/sofokles_antygona.xml |  25 ++
 .../files/erroneous/asnyk_miedzy_nami.html    |   0
 .../files/erroneous/asnyk_miedzy_nami.xml     |   0
 tests/files/text/asnyk_miedzy_nami.txt        |   0
 tests/files/text/asnyk_miedzy_nami.xml        |  25 ++
 tests/test_dcparser.py                        |  56 +++
 tests/test_text.py                            |  22 ++
 tests/utils.py                                |  62 ++++
 46 files changed, 836 insertions(+), 285 deletions(-)
 create mode 100644 README.txt
 delete mode 100644 librarian/__init__.pyc
 delete mode 100644 librarian/dcparser.pyc
 delete mode 100644 librarian/html.pyc
 create mode 100644 librarian/parser.py
 delete mode 100644 librarian/tests/__init__.py
 delete mode 100644 librarian/tests/files/dcparser/.DS_Store
 delete mode 100644 librarian/text.pyc
 mode change 100644 => 100755 setup.py
 create mode 100644 tests/__init__.py
 create mode 100644 tests/files/dcparser/andersen_brzydkie_kaczatko.out
 rename {librarian/tests => tests}/files/dcparser/andersen_brzydkie_kaczatko.xml (100%)
 create mode 100644 tests/files/dcparser/biedrzycki_akslop.out
 rename {librarian/tests => tests}/files/dcparser/biedrzycki_akslop.xml (100%)
 create mode 100644 tests/files/dcparser/kochanowski_piesn7.out
 rename {librarian/tests => tests}/files/dcparser/kochanowski_piesn7.xml (99%)
 create mode 100644 tests/files/dcparser/mickiewicz_rybka.out
 rename {librarian/tests => tests}/files/dcparser/mickiewicz_rybka.xml (100%)
 create mode 100644 tests/files/dcparser/sofokles_antygona.out
 rename {librarian/tests => tests}/files/dcparser/sofokles_antygona.xml (100%)
 create mode 100644 tests/files/dcserialize/andersen_brzydkie_kaczatko.out
 create mode 100644 tests/files/dcserialize/andersen_brzydkie_kaczatko.xml
 create mode 100644 tests/files/dcserialize/biedrzycki_akslop.out
 create mode 100644 tests/files/dcserialize/biedrzycki_akslop.xml
 create mode 100644 tests/files/dcserialize/kochanowski_piesn7.out
 create mode 100644 tests/files/dcserialize/kochanowski_piesn7.xml
 create mode 100644 tests/files/dcserialize/mickiewicz_rybka.out
 create mode 100644 tests/files/dcserialize/mickiewicz_rybka.xml
 create mode 100644 tests/files/dcserialize/sofokles_antygona.out
 create mode 100644 tests/files/dcserialize/sofokles_antygona.xml
 rename {librarian/tests => tests}/files/erroneous/asnyk_miedzy_nami.html (100%)
 rename {librarian/tests => tests}/files/erroneous/asnyk_miedzy_nami.xml (100%)
 create mode 100644 tests/files/text/asnyk_miedzy_nami.txt
 create mode 100644 tests/files/text/asnyk_miedzy_nami.xml
 create mode 100755 tests/test_dcparser.py
 create mode 100755 tests/test_text.py
 create mode 100644 tests/utils.py

diff --git a/.gitignore b/.gitignore
index 7189e7b..bfdc1af 100644
--- a/.gitignore
+++ b/.gitignore
@@ -2,3 +2,6 @@
 *.pyc
 MANIFEST
 dist
+build
+nbproject
+nbproject/*
diff --git a/MANIFEST.in b/MANIFEST.in
index 4c76fc3..9b7ec3d 100644
--- a/MANIFEST.in
+++ b/MANIFEST.in
@@ -1,2 +1,2 @@
 include librarian/*.xslt
-recursive-include librarian/tests/files/ *.xml
+recursive-include tests/files/ *.xml
diff --git a/README.txt b/README.txt
new file mode 100644
index 0000000..3debf76
--- /dev/null
+++ b/README.txt
@@ -0,0 +1,49 @@
+Librarian
+=========
+Librarian (*ang. bibliotekarz*) to biblioteka służąca do konwersji języka składu książek opartego na XML opracowanego przez Fundację Nowoczesna Polska na inne formaty.
+
+Obecnie obsługiwane są formaty:
+
+ * XML
+ * TXT
+ 
+Biblioteka librarian potrafi również parsować metadane opisane przez DublinCore oraz wyciągać fragmenty motywów z lektur.
+
+
+Wymagania
+---------
+
+ * [lxml 2.2](http://codespeak.net/lxml/)
+
+ 
+Instalacja
+----------
+Zainstaluj biblioteki z sekcji *Wymagania* powyżej. Następnie rozpakuj archiwum z biblioteką librarian, przejdź w terminalu do rozpakowanego katalogu i wpisz:
+
+<pre>python setup.py install</pre>
+
+Na Linuxie i OSX mogą być wymagane uprawnienia administratora. W takim wypadku wpisz:
+
+<pre>sudo python setup.py install</pre>
+
+Alternatywnie możesz zainstalować bibliotekę librarian w wybranym przez siebie katalogu. W takim wypadku należy użyć argumentu *prefix* do *setup.py*:
+
+<pre>python setup.py install --prefix=ŚCIEŻKA_DO_WYBRANEGO_KATALOGU</pre> 
+
+W takim wypadku będzie jednak potrzebne własnoręczne edytowanie zmiennych systemowych *PATH* i *PYTHONPATH*.
+
+
+Sposób użycia
+-------------
+Konwersja plików lektur do XHTML:
+
+<pre>book2html LEKTURA1 LEKTURA2...</pre>
+
+Konwersja plików lektur do TXT:
+
+<pre>book2txt LEKTURA1 LEKTURA2...</pre>
+
+Wyciągnięcie wszystkich fragmentów motywów z wygenerowanych plików XHTML:
+
+<pre>bookfragments PLIK1 PLIK2...</pre>
+
diff --git a/librarian/__init__.py b/librarian/__init__.py
index e69de29..9132f5c 100644
--- a/librarian/__init__.py
+++ b/librarian/__init__.py
@@ -0,0 +1,11 @@
+# -*- coding: utf-8 -*-
+# exception classes
+
+class ParseError(Exception):
+    pass
+
+class ValidationError(Exception):
+    pass
+
+class NoDublinCore(ValidationError):
+    pass
diff --git a/librarian/__init__.pyc b/librarian/__init__.pyc
deleted file mode 100644
index 3d4eb1301d13decc7aa425513c20d7c2e466f51b..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 151
zcmdn|iI+?LgQ{mT0~9a<X$K%KW&si@3=F{<AQ3+eAi;n}6gvRL^h1kNi;DHDN|REH
z^aG0Wvr@B5D)r0rbMjJifXvdOO5OC#68)UaBoHo2EXqvG(~pnO%*!l^kJl@xEam_j
SVw0PnQks)$2ePUdh#3I)UL$}2

diff --git a/librarian/book2txt.xslt b/librarian/book2txt.xslt
index cd98524..72f193d 100644
--- a/librarian/book2txt.xslt
+++ b/librarian/book2txt.xslt
@@ -3,22 +3,13 @@
 
 <xsl:output encoding="utf-8" method="text" />
 
+<xsl:param name="wrapping" select="0" />
+
 <!-- ============================================================================== -->
 <!-- = MASTER TAG                                                                 = -->
 <!-- = (can contain block tags, paragraph tags, standalone tags and special tags) = -->
 <!-- ============================================================================== -->
 <xsl:template match="powiesc|opowiadanie|liryka_l|liryka_lp|dramat_wierszowany_l|dramat_wierszowany_lp|dramat_wspolczesny">
-<xsl:text>Kodowanie znaków w dokumencie: UTF-8.
------
-Publikacja zrealizowana w ramach projektu Wolne Lektury (http://wolnelektury.pl/). Reprodukcja cyfrowa wykonana przez
-Bibliotekę Narodową z egzemplarza pochodzącego ze zbiorów BN. Ten utwór nie jest chroniony prawem autorskim i znajduje
-się w domenie publicznej, co oznacza, że możesz go swobodnie wykorzystywać, publikować i rozpowszechniać.
-
-Wersja lektury w opracowaniu merytorycznym i krytycznym (przypisy i motywy) dostępna jest na stronie %s.
------
-
-
-</xsl:text>
 <xsl:if test="nazwa_utworu"><xsl:apply-templates select="autor_utworu|dzielo_nadrzedne|nazwa_utworu|podtytul" mode="header" /></xsl:if>
 <xsl:text>
 
@@ -150,8 +141,8 @@ Wersja lektury w opracowaniu merytorycznym i krytycznym (przypisy i motywy) dost
     <xsl:apply-templates select="*|text()" mode="inline" />
 </xsl:variable>
 <xsl:text>
-
-/ </xsl:text><xsl:value-of select="wl:wrap_words(wl:strip($content))" /><xsl:text> /</xsl:text>
+    
+/ </xsl:text><xsl:value-of select="wl:wrap_words(wl:strip($content), $wrapping)" /><xsl:text> /</xsl:text>
 </xsl:template>
 
 <xsl:template match="lista_osoba">
@@ -167,7 +158,7 @@ Wersja lektury w opracowaniu merytorycznym i krytycznym (przypisy i motywy) dost
 <xsl:text>
 
 </xsl:text>
-<xsl:value-of select="wl:wrap_words(wl:strip($content))" />
+<xsl:value-of select="wl:wrap_words(wl:strip($content), $wrapping)" />
 </xsl:template>
 
 <xsl:template match="strofa">
diff --git a/librarian/dcparser.py b/librarian/dcparser.py
index 557509c..830b089 100644
--- a/librarian/dcparser.py
+++ b/librarian/dcparser.py
@@ -3,14 +3,10 @@ from xml.parsers.expat import ExpatError
 from datetime import date
 import time
 
-# Import ElementTree from anywhere
-try:
-    import xml.etree.ElementTree as etree # Python >= 2.5
-except ImportError:
-    try:
-        import elementtree.ElementTree as etree # effbot's pure Python module
-    except ImportError:
-        import lxml.etree as etree # ElementTree API using libxml2
+from librarian import ValidationError, NoDublinCore
+
+import lxml.etree as etree # ElementTree API using libxml2
+from lxml.etree import XMLSyntaxError
 
 
 # ==============
@@ -21,7 +17,22 @@ class Person(object):
     def __init__(self, last_name, *first_names):
         self.last_name = last_name
         self.first_names = first_names
-    
+        
+    @classmethod
+    def from_text(cls, text):
+        parts = [ token.strip() for token in text.split(',') ]
+        if len(parts) == 1:
+            surname = parts[0]
+            names = []
+        elif len(parts) != 2:
+            raise ValueError("Invalid person name. There should be at most one comma: \"%s\"." % text)
+        else:
+            surname = parts[0]
+            if len(parts[1]) == 0:
+                # there is no non-whitespace data after the comma
+                raise ValueError("Found a comma, but no names given: \"%s\" -> %r." % (text, parts))
+            names = [ name for name in parts[1].split() if len(name) ] # all non-whitespace tokens
+        return cls(surname, *names)
     
     def __eq__(self, right):
         return self.last_name == right.last_name and self.first_names == right.first_names
@@ -32,52 +43,71 @@ class Person(object):
             return '%s, %s' % (self.last_name, ' '.join(self.first_names))
         else:
             return self.last_name
-    
-    
+        
     def __repr__(self):
         return 'Person(last_name=%r, first_names=*%r)' % (self.last_name, self.first_names)
 
+def as_date(text):
+    try:
+        try:
+            t = time.strptime(text, '%Y-%m-%d')
+        except ValueError:
+            t = time.strptime(text, '%Y')
+        return date(t[0], t[1], t[2])
+    except ValueError, e:
+        raise ValueError("Unrecognized date format. Try YYYY-MM-DD or YYYY.")
 
-def str_to_unicode(value, previous):
-    return unicode(value)
-
+def as_person(text):
+    return Person.from_text(text)
 
-def str_to_unicode_list(value, previous):
-    if previous is None:
-        previous = []
-    previous.append(str_to_unicode(value, None))
-    return previous
+def as_unicode(text):
+    if isinstance(text, unicode):
+        return text
+    else:
+        return text.decode('utf-8')
 
+class Field(object):
+    def __init__(self, uri, attr_name, type=as_unicode, multiple=False, salias=None, **kwargs):
+        self.uri = uri
+        self.name = attr_name
+        self.validator = type
+        self.multiple = multiple
+        self.salias = salias
 
-def str_to_person(value, previous):
-    comma_count = value.count(',')
-    
-    if comma_count == 0:
-        last_name, first_names = value, []
-    elif comma_count == 1:
-        last_name, first_names = value.split(',')
-        first_names = [name for name in first_names.split(' ') if len(name)]
-    else:
-        raise ValueError("value contains more than one comma: %r" % value)
-    
-    return Person(last_name.strip(), *first_names)
+        self.required = kwargs.get('required', True) and not kwargs.has_key('default')
+        self.default = kwargs.get('default', [] if multiple else [None])
 
+    def validate_value(self, val):
+        try:
+            if self.multiple:
+                if self.validator is None:
+                    return val
+                return [ self.validator(v) if v is not None else v for v in val ]
+            elif len(val) > 1:
+                raise ValidationError("Mulitply values not allowed for field '%s'" % self.uri)
+            elif len(val) == 0:
+                raise ValidationError("Field %s has no value to assign. Check your defaults." % self.uri)
+            else:
+                if self.validator is None or val[0] is None:
+                    return val[0]
+                return self.validator(val[0])
+        except ValueError, e:
+            raise ValidationError("Field '%s' - invald value: %s" % (self.uri, e.message))
 
-def str_to_date(value, previous):
-    try:
-        t = time.strptime(value, '%Y-%m-%d')
-    except ValueError:
-        t = time.strptime(value, '%Y')
-    return date(t[0], t[1], t[2])
+    def validate(self, fdict):
+        if not fdict.has_key(self.uri):
+            if not self.required:
+                f = self.default
+            else:
+                raise ValidationError("Required field %s not found" % self.uri)
+        else:
+            f = fdict[self.uri]
 
+        return self.validate_value(f)
 
 # ==========
 # = Parser =
 # ==========
-class ParseError(Exception):
-    def __init__(self, message):
-        super(ParseError, self).__init__(message)
-
 
 class XMLNamespace(object):
     '''Represents XML namespace.'''
@@ -102,96 +132,175 @@ class BookInfo(object):
     RDF = XMLNamespace('http://www.w3.org/1999/02/22-rdf-syntax-ns#')
     DC = XMLNamespace('http://purl.org/dc/elements/1.1/')
     
-    mapping = {
-        DC('creator')        : ('author', str_to_person),
-        DC('title')          : ('title', str_to_unicode),
-        DC('subject.period') : ('epoch', str_to_unicode),
-        DC('subject.type')   : ('kind', str_to_unicode),
-        DC('subject.genre')  : ('genre', str_to_unicode),
-        DC('date')           : ('created_at', str_to_date),
-        DC('date.pd')        : ('released_to_public_domain_at', str_to_date),
-        DC('contributor.translator') : ('translator', str_to_person),
-        DC('contributor.technical_editor') : ('technical_editor', str_to_person),
-        DC('publisher')      : ('publisher', str_to_unicode),
-        DC('source')         : ('source_name', str_to_unicode),
-        DC('source.URL')     : ('source_url', str_to_unicode),
-        DC('identifier.url') : ('url', str_to_unicode),
-        DC('relation.hasPart') : ('parts', str_to_unicode_list),
-        DC('rights.license') : ('license', str_to_unicode),
-        DC('rights')         : ('license_description', str_to_unicode), 
-    }
+    FIELDS = (
+        Field( DC('creator'), 'author', as_person),
+        Field( DC('title'), 'title'),
+        Field( DC('subject.period'), 'epoches', salias='epoch', multiple=True),
+        Field( DC('subject.type'), 'kinds', salias='kind', multiple=True),
+        Field( DC('subject.genre'), 'genres', salias='genre', multiple=True),
+        Field( DC('date'), 'created_at', as_date),
+        Field( DC('date.pd'), 'released_to_public_domain_at', as_date, required=False),
+        Field( DC('contributor.editor'), 'editors', \
+            as_person, salias='editor', multiple=True, default=[]),
+        Field( DC('contributor.translator'), 'translators', \
+            as_person,  salias='translator', multiple=True, default=[]),
+        Field( DC('contributor.technical_editor'), 'technical_editors',
+            as_person, salias='technical_editor', multiple=True, default=[]),
+        Field( DC('publisher'), 'publisher'),
+        Field( DC('source'), 'source_name', required=False),
+        Field( DC('source.URL'), 'source_url', required=False),
+        Field( DC('identifier.url'), 'url'),
+        Field( DC('relation.hasPart'), 'parts', multiple=True, required=False),
+        Field( DC('rights.license'), 'license', required=False),
+        Field( DC('rights'), 'license_description'), 
+    )
 
     @classmethod
     def from_string(cls, xml):
         from StringIO import StringIO
         return cls.from_file(StringIO(xml))
-    
+   
     @classmethod
-    def from_file(cls, xml_file):
-        book_info = cls()
-        
+    def from_file(cls, xmlfile):
+        desc_tag = None        
         try:
-            tree = etree.parse(xml_file)
+            iter = etree.iterparse(xmlfile, ['start', 'end'])            
+            for (event, element) in iter:
+                if element.tag == cls.RDF('RDF') and event == 'start':
+                    desc_tag = element
+                    break
+
+            if desc_tag is None:
+                raise NoDublinCore("DublinCore section not found. \
+                    Check if there are rdf:RDF and rdf:Description tags.")
+
+            # continue 'till the end of RDF section
+            for (event, element) in iter:
+                if element.tag == cls.RDF('RDF') and event == 'end':
+                    break
+
+            # if there is no end, Expat should yell at us with an ExpatError
+            
+            # extract data from the element and make the info
+            return cls.from_element(desc_tag)
+        except XMLSyntaxError, e:
+            raise ParseError(e)
         except ExpatError, e:
             raise ParseError(e)
 
-        description = tree.find('//' + book_info.RDF('Description'))
-        book_info.wiki_url = description.get(cls.RDF('about'), None)
-        
-        if description is None:
-            raise ParseError('no Description tag found in document')
-        
-        for element in description.findall('*'):
-            book_info.parse_element(element) 
+    @classmethod
+    def from_element(cls, rdf_tag):
+        # the tree is already parsed, so we don't need to worry about Expat errors
+        field_dict = {}
+        desc = rdf_tag.find(".//" + cls.RDF('Description') )        
         
-        return book_info
+        if desc is None:
+            raise NoDublinCore("No DublinCore section found.")
+
+        for e in desc.getchildren():
+            fv = field_dict.get(e.tag, [])
+            fv.append(e.text)
+            field_dict[e.tag] = fv
+                
+        return cls( desc.attrib, field_dict )        
+
+    def __init__(self, rdf_attrs, dc_fields):
+        """rdf_attrs should be a dictionary-like object with any attributes of the RDF:Description.
+        dc_fields - dictionary mapping DC fields (with namespace) to list of text values for the 
+        given field. """
+
+        self.about = rdf_attrs.get(self.RDF('about'))
+        self.fmap = {}
+
+        for field in self.FIELDS:
+            value = field.validate( dc_fields )
+            setattr(self, 'prop_' + field.name, value)
+            self.fmap[field.name] = field
+            if field.salias: self.fmap[field.salias] = field
+
+    def __getattribute__(self, name):
+        try:
+            field = object.__getattribute__(self, 'fmap')[name]
+            value = object.__getattribute__(self, 'prop_'+field.name)
+            if field.name == name:
+                return value
+            else: # singular alias
+                if not field.multiple:
+                    raise "OUCH!! for field %s" % name
+                
+                return value[0]
+        except (KeyError, AttributeError):
+            return object.__getattribute__(self, name)
 
-    def parse_element(self, element):
+    def __setattr__(self, name, newvalue):
         try:
-            attribute, converter = self.mapping[element.tag]
-            setattr(self, attribute, converter(element.text, getattr(self, attribute, None)))
-        except KeyError:
-            pass
+            field = object.__getattribute__(self, 'fmap')[name]
+            if field.name == name:
+                object.__setattr__(self, 'prop_'+field.name, newvalue)
+            else: # singular alias
+                if not field.multiple:
+                    raise "OUCH! while setting field %s" % name
 
-    def to_xml(self):
+                object.__setattr__(self, 'prop_'+field.name, [newvalue])
+        except (KeyError, AttributeError):
+            return object.__setattr__(self, name, newvalue)
+
+    def update(self, field_dict):
+        """Update using field_dict. Verify correctness, but don't check if all 
+        required fields are present."""
+        for field in self.FIELDS:
+            if field_dict.has_key(field.name):
+                setattr(self, field.name, field_dict[field.name])
+
+    def to_etree(self, parent = None):
         """XML representation of this object."""
-        etree._namespace_map[str(self.RDF)] = 'rdf'
-        etree._namespace_map[str(self.DC)] = 'dc'
+        #etree._namespace_map[str(self.RDF)] = 'rdf'
+        #etree._namespace_map[str(self.DC)] = 'dc'
         
-        root = etree.Element(self.RDF('RDF'))
+        if parent is None:
+            root = etree.Element(self.RDF('RDF'))
+        else:
+            root = parent.makeelement(self.RDF('RDF'))
+
         description = etree.SubElement(root, self.RDF('Description'))
         
-        if self.wiki_url:
-            description.set(self.RDF('about'), self.wiki_url)
+        if self.about:
+            description.set(self.RDF('about'), self.about)
         
-        for tag, (attribute, converter) in self.mapping.iteritems():
-            if hasattr(self, attribute):
-                e = etree.Element(tag)
-                e.text = unicode(getattr(self, attribute))
-                description.append(e)
+        for field in self.FIELDS:
+            v = getattr(self, field.name, None)
+            if v is not None:
+                if field.multiple:
+                    if len(v) == 0: continue
+                    for x in v:
+                        e = etree.Element(field.uri)
+                        e.text = unicode(x)
+                        description.append(e)
+                else:
+                    e = etree.Element(field.uri)
+                    e.text = unicode(v)
+                    description.append(e)
         
-        return unicode(etree.tostring(root, 'utf-8'), 'utf-8')
+        return root
 
     def to_dict(self):
-        etree._namespace_map[str(self.RDF)] = 'rdf'
-        etree._namespace_map[str(self.DC)] = 'dc'
-        
-        result = {'about': self.wiki_url}
-        for tag, (attribute, converter) in self.mapping.iteritems():
-            if hasattr(self, attribute):
-                result[attribute] = unicode(getattr(self, attribute))
+        result = {'about': self.about}
+        for field in self.FIELDS:
+            v = getattr(self, field.name, None)
+
+            if v is not None:
+                if field.multiple:
+                    if len(v) == 0: continue
+                    v = [ unicode(x) for x in v ]
+                else:
+                    v = unicode(v)
+                result[field.name] = v
+
+            if field.salias:
+                v = getattr(self, field.salias)
+                if v is not None: result[field.salias] = v
         
         return result
 
-
 def parse(file_name):
     return BookInfo.from_file(file_name)
-
-
-if __name__ == '__main__':
-    import sys
-    
-    info = parse(sys.argv[1])
-    for attribute, _ in BookInfo.mapping.values():
-        print '%s: %r' % (attribute, getattr(info, attribute, None))
-
diff --git a/librarian/dcparser.pyc b/librarian/dcparser.pyc
deleted file mode 100644
index 0e911b8d97be8a856f2b1d6fcbfb500ab12c811a..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 8696
zcmcIpTXPi074Df`t+YZ{1S@1?vKTgz9kgpms$x4hq%hdV7a5OiV`T7Tw9}GCTFore
zy+}$RRS7)iAuoAwRo?UHO64&Rsmf2vA4tCM^y~#(NySxb!Swc=zMu1*)90M|+rMU~
z{}Eq*zpaYDdHntok9q1T7306B_Lb^*YSU95-+i_3t4>L6mfU+u?U&VNSsmA;J6<PL
zQc=mI+N`KM=?AK;r_|oG%C0F7f5%JxZq<ID2Y(#&sYxF$gL$So4<<po_eJWXh|5Z!
zP}`o0C)87=o+x#9b4JA#zHH8_cv32JDxQ+cyo#$*Sy1t`(u*?k1SvA|q>5+y^UV&Z
z*xy{AIzOPEQt^UJby{g0bW!d3D*I}{fn~M1tTft|RdPnfC)LUmaBM{#T~+#&N?uUO
zi(K=@{U51&zS=yiHeXWllG10DKBx2wD>(6!lG^rF?cMhteB#|n$?uZdSUm-4B>%?W
zDkP%pG>V%?-N@d|^DK9?GH7vR^}4N~c3bBrO9v9=sOxyl6%-nb0{c+<6fOByqD)yS
za}TUcsJJYZiV8MXCph+JC^nLGr=#m#nZACQ*xh<3GPa&Z2f7}maXqSc61->I_3b1t
z8ceh8V{8czf8WJpYA7Cg>XA|*D!x=?QyX%e^;Pz?RTP{*<oxyo-egvJHLIM^mQfhp
z*$(&?@NE^3E6g?2TAvy0w)N}Ymd;zZ^K4J=+vC<@)=6~-)n0zw>~>n6WQ$K8<w=yb
z;&wO6F|64=wo{lR43jjmVR#yK!!?vw_2f|8CZD(Qm<uRyqC}2{k^osy9!WT=#}cH`
z6OQbfvjk+cWsxU4yLQYcAuz&FKM2DYP%~Umd2M2Y5053MxQ>c-2k{gz42<lRRdya&
z6gee;O-UssIjZy^VmHz$cTyZM22fTx-#VCpmAG)jT&g#W<w|vFvlLiIry?`qgUE6(
zOHzS=@M6pyAcDfMmnQ8j)?s)Kud+Gi)x7imfEC=Cu!4Gr;}<d&G0Is5o$IE7W~XVZ
z1C+ki$S*nUJCwflY9oI|`j5z(T=+iD=dhaVZXSl*Tp{DtNz@A<Q<K6_Oh_0WWO1*<
zS9S=)IBSPtKxqY>DVSj~i()7Ssf|Og0>W<zkD)cIconZYQ7u;|zpYhlcX$LC`^rNO
z%B~AxI(tFQc^GwiS`ORI^}{6Vnc?oov2i<xMU2fuoAr_O3Mqv;@rTbqoi5`Ax(}V$
zn^M`d@^+MZ>?zuIXA-Pv722QpR5w!lrz8+XE7r4Ai(!wtU7f}Of$2bX619MQ9P=qf
zG|!weq-9^jN@fa$@@BnhsEx>r@FDL-Y5t0e(-^1@4)hr3gURs494?BH01N!bCH1(h
z(i{XaC^VLzN-NBy8;N_;^+Nyh;|cYsOcRQW3A5g~e@;EFC^E0Y_v{45qDw`ce9%@N
zp5IQ$0$35I2HpbpR71z#dt~V&Sg;B81UoIOKak->{koW^OJcOJ5ExrUVSb20P*rbd
zsg07<)DN;;*X?eU)*%>>4-TR?>y2Dy5&PZFda0F~>2?wu2#8sHPI$__EgImu*gWZq
zk_H2D%55hThcegDdI%uGFE8Jh3iWx+qNm0e0CyMtPf1unp}cv2)?0+=PkYz=1z&_u
zkP;Ch{7qEg20~xKKyaLbgcN<afelrLwCDz|h>JJu1Kdj=aqu842Z$B^l$HEUz?-5Y
zBvKOmHSVr84%Qm6=w0LPP-eKEpbS($;g3z=IdFS8QI{{li+BkrU2z~rI4(A2%)I1{
zQ}B}B*>%1W2FfczdFH%jPfYP`x*LLuHXHfJ-AMOER5V;Nu^~@U@#&JWS=i6xQUt<&
zUhLp2a17pqsbB`o2~bbJ{TZu5bu~T6H)x<HM&B6}Hi%_C(8feNV+NNjw9d7;Z{;}|
z(4+<oJKR3lGF~~zAwR;jvTVh(yxP9_=)-krkLgBj{cX1Y96bhX|36)27n5vGQb2^N
zSwRutI@U&~*3_}0xXQAm^NPs^J|aG99^?ht!7Cn-m>`ov?&Zl)N&)YkG4qML{tVCa
z*BKX=EfS4Gm&lW7WkdF~F0RyX8s>4sI1f(t$HO_lmUGx>XS_Qu+?saO>4f1ISmD3!
zoHrV8+<|(CV)T`Q+Z0taHp;DGXd@mfly`*VucFtO_fI5-u<`C@3Bxb(-aiB7&SwA+
znv((dYkj`cG>?LLl`bd8IfO`!YCYbai)Z*1ru~k)<zb#d$oI6r!l|Y84Wr>USEo=9
zh8Bm+6SEWin6)Mg>S54AG31!k#^FB(oGHL#D0w)(ezoNO36!sA+5W9`J8Pp+{wTQX
zc+5B8lQT#f7bDC&(hDWUx&Rrm`!3uh_*r^_TIp*E=F#hF8P}+;42a@CfjBWr6ekMA
z$x-4I7Xh(4N}Mi;CzPHX%sEpKXGe*11#y0KOA7^Yag=znAR-?!+~-n3JT*!@T@Y)d
z#N~o`W|X*65MNNpKJ^)b9E_X@S(o8LFBL@OVTQyX2yqifpH^|5xeziOhsNvq{wp$1
zFSE|f(*>y@^K?-v^SlHa6dZB2Yi;+<R_pNauzC2CW|r@?F2C`{8?D!_w60uP%j4}e
za|{D;w3eC+ratW7?d2Wm9k*LZR~+ybX<cq!Zn<EfXLqw)aNf>!WKkztY+^fFYPy@X
zcMY*_dRu$CZJW68lPng`Xg^6~!C=459(Uca9i8IF5;w3ICKNG`vXG9$$chhT&tmc%
zTZ;^c2#LDAtxnQLD0u)!8NCc+1G+xVZY&enJW5T6o0TLG4PTyY_3+*tHd$(=)$Lt`
z>QN`ual-yw>zV$~b&!0KY{2a5Ts&x#_42lM`6IcJ<=q*2*8~oWN9KAqKMOt-yH7G1
zm&D+CvYj9q1d?zSxgl%BAI8BWn`CKoH!^VHtcU`~*QwDmTGBYCS&%HRprRhe+O*;J
zbGQi8brqM67iAfbT(}odfrdk`_c0i;BeI}y5a?d_ojU+?Lma8~ktE?1#%+ML>CUZ>
z-3@jF6k*YJo*jhSNk_XB8kcK#j2!#upcAm4m^&%PVe?$-j;q2NV@KG3kw^0vh7lx@
zB1irwD(Lb=4vTDY5h;k7LqK>ZJwFH~P9u~+7|9S1ksKI-@J_-A26!-l!X-p{U>_Qn
z@i2~*^fZ}>2IO4Dvxc_}hK5sYja{&g+#TkP2lK*>KDsEQs|+YH?#BHkMxhv(Oc#Nu
z_uwz;4ha2Vh1XrHB^<ahB62Efw3YR&xrkbt)khlZ@RI7=h#=#7lGc%x=~3||IDK`M
zE`U>Rn`@`9lC0#FqQuV??cRoVi$sGP@5t|twgK~(a?2bh`$<SVljrOT_sLS>WvZ|+
zH7+E5r#}eT5Yz~Uws21&E4rPzD2rR+J?^r!quAyO{rOOJ$QlQzf}5D_cX$k=9_3g4
zlDFh7c#Hm7?;OI^1%Cn8>!J%%%jm)TR`7KXmE-G#7@R@6gq~WtClM&n<<c<_AP~eh
zP>?P_KX>T@E`Sm~eB4QwfWnFdw38Chy6+RiIEyD21|J}61Jnof9))1EjjYWb>-3Sc
z8xRnQLm_>jk6nVFn->>4xXT%;DA<!$Q_CBS;zJlbosU;-gAJ_oCp;q|>m>xOZAzM_
zqeT^tE`S1EbBgLX2*yD{u0rvip*rV-B+1C2m(xJ>MOKV-sFQ?1Y?9t!ICDs7np9a`
z@<j>M6z`JP!ARGJejd|mNBcs+$0fKb{Nl37FD?)*%Q6iBw2>0U>_|Gj4lQ@vksULs
zG^3ylgC1O<qQW^*^Z9|Ae~m2D*Xj;CWB3Ahjn7b+(UlvbJ?3Swh+!X-vJwT#g~1Ob
zb~iEg%w5Z(p>PAhIxw;we2kyq7K`^$7#fS7-ClcrmEl+LAqq)Dx)d|x*AN&B*4W6Q
zGR=**1#35*+aOooMsKTl4L)JF5(;23VAe)}lmJ6`4vY-NttN<+Men0eaE}GQ-G~*j
znKS3kPKfJ4sODK_1BPM(_yD1-3Vy}n4vO(2RBldq9WYIQ#^i-1sChO2jQ65H?bTre
zPQoV4duP1UF2T=j@)6t;+}{PaySSKv3t-m=Y_9^bN4EuR^0iaIc1>XB7mx$qz$!vX
z3`h17z*c3TIUo^T_>b8htOHT&xIB2BlacE|gfiG-K^%zf7wHFW(gyMlV!l$OBFX^U
zQ~_+83I?K+Z+I>ar`7_Cy^eiOKx3AT*uN}{leYZ^OBNYK3I`$)v7Gl;CH$hJ<?=t=
zf<wO?sm|9^EH1KGW+7nZE!f6K7lZ8HZWagcvE?QU+N=P6k_u83!y`#;Ecec9m{;R5
ztW>-iZ^kbpA^N}mEtlpe=1)~8t4nR#PCnFh!62_)S`aZ~K*MN8BH$8oE{W$wN9nkU
z;t$U@mScT{rT&hm2<Jt^@rj6Te2C4Vfxu}PNp#mywFDP5uPIJLZkn3kq{R=B!Swni
z<I2Iod5)k1!V#LjOS7;=pQ7$;8l^fy6sPIdK^Jk2T!fbL*M&WMZUV7yox$8<7@TpH
zJi7yt;HQpB1Wv-un@4RePPog1lBbU8WQU{)oRvH_a$86F&O@ot^>I)nSLE}65mP5_
zP#B20<Z*{m8XE#70e;ney!!=lBuHF<otr7Sv@$)RId7#>tMQv!b*_4%Hc?xueO@ik
L0<G|p)QtBZbHfps

diff --git a/librarian/html.py b/librarian/html.py
index b279e5d..4edbf33 100644
--- a/librarian/html.py
+++ b/librarian/html.py
@@ -3,9 +3,9 @@ import os
 import cStringIO
 import re
 import copy
-import pkgutil
 
 from lxml import etree
+from librarian.parser import WLDocument
 
 
 ENTITY_SUBSTITUTIONS = [
@@ -31,32 +31,28 @@ ns = etree.FunctionNamespace('http://wolnelektury.pl/functions')
 ns['substitute_entities'] = substitute_entities
 
 
-def transform(input_filename, output_filename):
+def transform(input, output_filename=None, is_file=True):
     """Transforms file input_filename in XML to output_filename in XHTML."""
     # Parse XSLT
     style_filename = os.path.join(os.path.dirname(__file__), 'book2html.xslt')
     style = etree.parse(style_filename)
 
-    doc_file = cStringIO.StringIO()
-    expr = re.compile(r'/\s', re.MULTILINE | re.UNICODE);
-
-    f = open(input_filename, 'r')
-    for line in f:
-        line = line.decode('utf-8')
-        line = expr.sub(u'<br/>\n', line)
-        doc_file.write(line.encode('utf-8'))
-    f.close()
-
-    doc_file.seek(0);
+    if is_file:
+        document = WLDocument.from_file(input, True)
+    else:
+        document = WLDocument.from_string(input, True)
 
-    parser = etree.XMLParser(remove_blank_text=True)
-    doc = etree.parse(doc_file, parser)
+    result = document.transform(style)
+    del document # no longer needed large object :)
 
-    result = doc.xslt(style)
     if result.find('//p') is not None:
         add_anchors(result.getroot())
         add_table_of_contents(result.getroot())
-        result.write(output_filename, xml_declaration=False, pretty_print=True, encoding='utf-8')
+        
+        if output_filename is not None:
+            result.write(output_filename, xml_declaration=False, pretty_print=True, encoding='utf-8')
+        else:
+            return result
         return True
     else:
         return False
diff --git a/librarian/html.pyc b/librarian/html.pyc
deleted file mode 100644
index dfb837db2bd1077d904719c9d591cb02dd42d5bc..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 8882
zcmb_iU2GiJb-pu8F1ag`qDY!DEj!vsp}n$2GF4IAP-(@sC`&>lN}ZKTnH95H?hMHx
zmov+`vl5v~K@hqr`c@zT`cNcq$#c=X1n5%>B#$jx6etP=C{UmeeJarRqW!+_?k*|C
zE)bZa=JNjEbMHClJLlZXpa1u0`O&|<dAn)4|3>ipLwxqvNJ9Mg%#Jau{k&&Zy<WX<
zR{dVSU{(vg{D4^<P~JB?gQit9t3~65ubObigs++KoC#kyn}cR`$b_TjGt540qN0gQ
zCLA+|SfFeo)Rawl+AN0S>T$}1XG}C=Hn8{vYYvV1bam8(lWHC_(P{n%wc(Wc48)u<
z;VWvIFyXW^lO{Z?%oMxTs&8O}@&DrAa@9jNA4C%6Srl~&$j!~o>G!{X??)Y9<!<qZ
z`aM5C-!1=4&6h8Ci$CdbvJ2hPkDDAj(EYCCx4n-fH|CjVa?c$4CY{9DcqaB)h21~*
zOyU`D|4ow@%+lId%wyj?$NvJ`;{o#w9Sa<Yj(cyL71Y+MT*ZC^$)o#?e5(@e=TTzg
zG^uQKl4j0KGfkdESso{wl}4*oiIP0dBU_1+O1oig6jpK!ng?OVRCyA48A)v8#O95p
z8RgtVE4H~Z+i9HSQ>eMI{NVP3PwTaZAJ>q1_~7=v<yt<3hAe8g8qKK6VXDCTTtUN?
zDWXMG9OPrj+RnNKNge!v+$w5}61#}p!b2NncHujn^(b4opQYQ;PJXcPG;JkO3)xO~
zFt-`!3$1vaU)IRtMzXM#@3!XK2l?dj*`vDFj3Z0(Ev4P~<KCn<WJbJc&sI?0%JcTs
zg%_ruZ?_hDM`?po=p@E6!*dd2zXitQ{9^I4n<*G<7)%9&$uXZ&78l7wavH5eCLX3N
zKOZo8$;6O(^g-s7(h>79hYgy1)I5U<I=&ap{tff22;um^JRdU8Fvpk)wGN9|0g5ol
zdJaa-fQiS|4N9SISlru7AUn{y2`=xk4-}+Gx;zpIB^fgLD`@nd5BJE1^nqGTS~AZ-
z<g_Fa8!4M-WfPw@AzHE75wl$~=}k;DqlVYGgeE@sVws~r8HSWh{<;avTrq^MlgfgJ
zLM<7@G_RW)CaqORan|-@BoDGiVmH!k*H$*-Rs=C_ck(*3Nn@AA%A>n?DtTH-JNa+4
z+<I{L&b+12t*7bE70JoIZRNIrs)ffEdI<rx)PhdFG53ClqPn=AEnF+d$eH{Mk}TRy
zpG5WbRwLP|Q=r}Gg?4@#t^2#JdKfiZjjTa&&M9Z@EXwnPdOL&e<in_ll4cr0A(sV3
zIdwR-Qjj*3IOwZR7-!s-#?-aRdObjA2U+b#W+R9UM6#A+?#<ix)U#jUyjYrA*-Ura
zI3xAG`|!?#+jnj+-%!iL<=fZqE#1)SX&XAiJ!1!H7y*W)o&w0H8DK+=qCQe(vz3Ab
z(rY8ABWc0$-zQ<2svl$>*Un>B%|@Jr$}Fczlv9ryVOR&!Y^9kMMVr9lG|eFct}Sn@
zx1xHwQ5TXVxz*C2G+H*Qa=cLFML7^FUXlO@Hu(YO>WPacvaHg<g=tewA}`u+XOf2v
z6<aZ=<<aQWGaU{3t163Zr<I>1a`7PZo}^zu!M=^e6u#=6_D8(e{nJ3LqIcSx@Lu!E
z{-{6UO`vYt8^_OiZ`jv%KFJ!JyI^ROL-~)=O&zkoha`gah58v8Mi_HX8CY^(8JO{c
zGK18~&(OJOf|?9OlVZ{Pu5-RCNPv4dE?^Z<0w(Q{(4ul3_6wHFnEHU(g!w}Ta~dc$
z=~=GwP#q&W?I;W0XOT)$ug6K8*XvS`IMhk!ThVT0ofJO7)@+q@h(R`LZ3MrMQm}~R
z1ep<B#e6z8Qe|K<O8EKBLwF61*c^ZYhJai5&4$q4JGNP>WbgqJ?V-_z!iCyHaE+Dn
zM7S|20!$qFcanRdr*I3OB_YOZk_7|!$ON(SFOUJ3;9~3_bhop#mK<|MBr4nRUe1ca
z*efIkeo1$)d*i4^AK$Rg?-=i3Ozw-5h?TwXngQs}cG;xph<}i>n;3ob7FxVV9F32z
z4ICcHkWLikvo*OC%UzvqWVv(@=O=@H{m#AGjit)+z2&)k_irp$9(;83s-#t?9dO_3
z-%i_daBe4Iw0|SZ(o9F4Mevb;Cp?QIR=@*l?gSsQ?js~8$OxEqfHKiDtYc|N@;6ao
zUqxcPqF07bGU}bhSMbL0Gwqj;ctt!Hul@s>F0cNf%PSn&(%RpC9=jgB$S$ZTjy{II
zOS{1W9C$p)Ghc$i#(yw$nx>mgMUzhT#=A0G_Wl_koLKUEQ2c(k*9rav`yHEN=%p#(
zqa2?C42R0V&0*!FS%D7%;=p%Q)T-3H<$y}q;qDh_ZDrP86FgU%l;HwC^3L+Rgk=8$
zPh{<!jslJfNR;dCv~v0ny}cq9>3h(eb=H1xVb)$)c5LR1f}Z)WYn>gXPfkz<EKz=O
zGfHS`uC1X!>P>JRNry<VxPVOt+@1~~xW(4KB<GDy=P6Kk8#z3Hb?FAo^sWtfV#39|
z5x0)SI(V0Je8}XaC0aHsty{SwpJD||M5Hv2N}AvG%HFGx=qdj8&-mgSH^+~Awsfc+
zf*u~32et-(fJEj(x_N>;kONY$caoS64@?f}aM)O-Omr;~y)alJ5?HVAreOz8L|ukK
z{BWEhkpbPaBl?v4(t(U@BH5)ADte`2sZbg#jq3z?>}0l1@XtZ|me8B2Z}y-1E*`<L
zLf8>WA9~)>+PD43aODe(J?Q>JH(>6e`^Zd~Z7BPXJgPka0wCTVf<|6YX!>&JqLlt0
z;iseXaBr>>brv2ydlSTcgfYwuT)5aUgLf`=%4`=*`aOU~uXz)yiYVtBjB?=k&wSya
za~Ome1~Ja@j#${+`<3w&-zdED2n39nrWFt&zQNrC43<J9Z*SULVa$!ti545JC9{~U
z!D^jVg*zN=Xa>7jN#23^D$9JGcjimsS@&GjkFa_3Rx$%?##ynHo3%gx2kd>yq?Zxu
zU*alDYi}4tTI?{49fLlYO!dJTd3%5Dt?Yf@Cks486Wyg0=b;%JZeyu#ZJ&$dVhrN2
zJ{MC$^KZBCJAQxR{dXLp7hfb4v%`?ZI0F}4lH5PSs0*;**tl#+tvW@?okMbCA3kj}
zw-vc@m5r`{ibyU?cVCL+=AGRp<O>i5Z3wHZ6(CwU{`3at)}u|_RS11lhDRVjQ09Ik
z4&@K3ok8CmBQ{<B*z=fs)?Rg1xRRuKh4D|HtjfjNV=$^XjQZb5&{MIUyL%x}!%xUu
z7)W=H@bUlCw_Ifm>LNK34#>`2G*>_g@sBj#VL01SgJyUM!CM?h6A*mDg#NA|>YLGl
zoS8tEOq>M)vGBi{whyYDB)H7ZT$tOe_fJ0`M^)Dq=~o_^2u<Ism#^eVabCE~3q^Z<
z>TxL8Q}U&;Pa!v0M!`9vc^4$xTS$yw^ostp8S!Vlsltr+)xvp%MQ<WRDl0^Sp&2N^
zks9;f@JGFJ!PQ<YUO^p0sEa<+S-(j(67cz3Q0l?(K#0>HNN?agY!92Xi~<m9Kxhbu
zfM_NM7*L4@_)8wa4jM1V%3-PpJ~#zI17aVVxIvRrZIxp!Knxy}LL=c38h``3(;*0`
z4p12b1Ny51MsXX{T@|ANhi=s}!PNu0qU!V3L+ux?hkKyzdf0NU%Ik||c|HwE(^eL3
z=)Qx>@eqwJj#2d9L6PAPL~R9$%;>BW6X*wISQn%f=gUZt*;gWU!$mq$SzIWnu4|K}
z1x~E#=s?wRwIJIp4@T?bf@LG#*Y(5GINz!xSl^LMs^k_|D;soY6Tm(ayQt5+d8oTi
zCn>lBi~iNdPqCE!8WQ6T!3ipRGYHjB<2N1sDSt@iQn%bBgiwk3NgBV746TI@_>fNi
zIofw{A2JFfkPb^SYI~!A;8C*z&$Sm{a}Mp4D8n0Fo8zV>AT7Z=NL=_y_rm#$XqV3S
zx}M~jP@Edc0k3;R#^0bUr^diZgrkVlXwu;$&Ha${KT&{|F(m!<@e}xaU90K@+@%F1
zH>I;O>6{haLpUEM{WkPLH|SImTt7(6>FLl$LW4bdI%iyA$V|gWVf+d&gUSmncbZ18
zh8Mb?(?|_`kJs_M%7?yT$(QDz{lXdCvc1uYdHNcMSO}QRgdZ`fhR<F^;zR(-39e%s
z*h3pUXD!N#!fH-Cq@4p80ej_c>vnOemg+`Qxq3Sd+ql@LpoUQxH{FdTi|ehlxwF^7
zKo>gfqb@jL@*O6Bj6`NGSY`1pizHM)5uK#K3}e>CR%3TPY+U;#@&c7<f*@|<<P#`v
z7Zk}c0J3&LNHDcA8wPu*k@mw_H8!(Gd&|9Z@Vb^d5k|HNyWKz#9y~>z_TzTOfb8gs
zNw<<T63w)OH;wG1L#Ti!_jmaE_W)FiNq@+{<bB0E2O~BEvn8>h&87=<oxR@zTx`ng
zr<%a)|DWS(@?UyELNE^bn38w@uaHS#hm6H%y0DYl%s91iJ>~kRkI}Oig{SwNb_|*s
zEHl%+;kkNLS%mW{Wx>o`3=uqx9r7?XdLDqpg3Jy1ie9pyk-x<VSOwAyO@xch7X!XX
zAW)GFeh>TJgGt4=0%r3)<rGz<8{*()VT5mvG;VB`BaJPf41-OF8%B7v`}6{1jR#0J
zAr$!qHGXOBugv2CMi6k+@dko-kM1nk+qk;o-3Ihw9nePrIYasoPT{%a(D!wWu**Q|
z^*#mY?3NiZ`*-<1vb6S5Z<$doGuB%M#JgoqYZ<uzpTErBuMjQNJrCasF#*p*lX5=j
z5D;Eofxs*aUUa#!b;bSqT}c(Sil#?ka|xLjz-B;T!}}g-7*EeZP7(7Gt(`kcUvmTZ
z_k<niBgh%X-L6fOIld&iNd8aJPr1);CbGDW&zZyLk}vI12H(Y0|AJ4qEeZty+Mu4h
zx7-Iqgz6>XhXsD5tYVKaesUC9<Y}`*b5;X{R`7WI%fD>KI?J@BU#&Q3qgm)9r^o^3
zpRn!+NUDUN;QLI7pALC=3FO>+0=MvOUr$1$Rn}TTL$J-_4!aPoYn}D31Ld40ddF2F
zL~zJ1#84aQG1vyscX#y@{2ALg96PCRMU4=C6}z&i_g<byB3G<C+65$p9lqIiXVg<m
zw@~liSi8D@A2%JBIvOVgujdmlRQAT7B)8wj3jf4a_^`m+DBedE;e6rc*T3Rj_2+@9
zd_5V*Yf1^ZA%oWxg$ol15tzT^i+6h2KlaxFu9EuF|BHbi9(6qScLGNpkN#M|)$xe?
ze=#844EV$tOf%tSaPTc8;z4_76A$Dq`DiV?GCDz^wbQG#;-pWyZ~n`;uCi^s!b=hp
z8@$I^DB9`zHu&bk%ZOvY#^fy~5;5679)G}dJK}?KUt0C-FeY7e_h;95aF+9q!0XeG
fKR!^vg*Sgn1En&&<3j2D^ySj~6Vs)k()j-YS#S+m

diff --git a/librarian/parser.py b/librarian/parser.py
new file mode 100644
index 0000000..595dd97
--- /dev/null
+++ b/librarian/parser.py
@@ -0,0 +1,70 @@
+# -*- coding: utf-8 -*-
+from librarian import ValidationError, NoDublinCore, dcparser, ParseError
+from xml.parsers.expat import ExpatError
+from lxml import etree
+from lxml.etree import XMLSyntaxError
+
+import re
+from StringIO import StringIO
+
+class WLDocument(object):
+    LINE_SWAP_EXPR = re.compile(r'/\s', re.MULTILINE | re.UNICODE);
+
+    def __init__(self, edoc):
+        self.edoc = edoc
+
+        root_elem = edoc.getroot()
+        rdf_ns = dcparser.BookInfo.RDF
+        dc_path = './/' + rdf_ns('RDF')
+        
+        if root_elem.tag != 'utwor':
+            raise ValidationError("Invalid root element. Found '%s', should be 'utwor'" % root_elem.tag)
+
+        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)
+
+    @classmethod
+    def from_string(cls, xml, swap_endlines=False):
+        return cls.from_file(StringIO(xml), swap_endlines)
+
+    @classmethod
+    def from_file(cls, xmlfile, swap_endlines=False):
+
+        # 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')
+
+        if swap_endlines:
+            data = cls.LINE_SWAP_EXPR.sub(u'<br />\n', data)
+    
+        try:
+            parser = etree.XMLParser(remove_blank_text=True)
+            return cls( etree.parse(StringIO(data), parser) )
+        except XMLSyntaxError, e:
+             raise ParseError(e.message)            
+        except ExpatError, e:
+            raise ParseError(e.message)            
+
+    def transform(self, stylesheet, **options):
+        return self.edoc.xslt(stylesheet, **options)
+
+    def update_dc(self):
+        parent = self.rdf_elem.getparent()
+        parent.replace( self.rdf_elem, self.book_info.to_etree(parent) )
+
+    def serialize(self):
+        self.update_dc()
+        return etree.tostring(self.edoc, encoding=unicode, pretty_print=True)
diff --git a/librarian/tests/__init__.py b/librarian/tests/__init__.py
deleted file mode 100644
index 3f02541..0000000
--- a/librarian/tests/__init__.py
+++ /dev/null
@@ -1,115 +0,0 @@
-# -*- coding: utf-8 -*-
-import unittest
-from os.path import dirname, join, realpath
-
-from lxml import etree
-from librarian import dcparser, html
-
-
-def test_file_path(dir_name, file_name):
-    return realpath(join(dirname(__file__), 'files', dir_name, file_name))
-
-
-class TestDCParser(unittest.TestCase):
-    KNOWN_RESULTS = (
-        ('dcparser', 'andersen_brzydkie_kaczatko.xml', {
-            'publisher': u'Fundacja Nowoczesna Polska',
-            'about': u'http://wiki.wolnepodreczniki.pl/Lektury:Andersen/Brzydkie_kaczątko',
-            'source_name': u'Andersen, Hans Christian (1805-1875), Baśnie, Gebethner i Wolff, wyd. 7, Kraków, 1925',
-            'author': u'Andersen, Hans Christian',
-            'url': u'http://wolnelektury.pl/katalog/lektura/brzydkie-kaczatko',
-            'created_at': u'2007-08-14',
-            'title': u'Brzydkie kaczątko',
-            'kind': u'Epika',
-            'source_url': u'http://www.polona.pl/dlibra/doccontent2?id=3563&dirids=4',
-            'translator': u'Niewiadomska, Cecylia',
-            'released_to_public_domain_at': u'1925-01-01',
-            'epoch': u'Romantyzm',
-            'genre': u'Baśń',
-            'technical_editor': u'Gałecki, Dariusz',
-            'license_description': u'Domena publiczna - tłumacz Cecylia Niewiadomska zm. 1925',
-        }),
-        ('dcparser', 'kochanowski_piesn7.xml', {
-            'publisher': u'Fundacja Nowoczesna Polska',
-            'about': u'http://wiki.wolnepodreczniki.pl/Lektury:Kochanowski/Pieśni/Pieśń_VII_(1)',
-            'source_name': u'Kochanowski, Jan (1530-1584), Dzieła polskie, tom 1, oprac. Julian Krzyżanowski, wyd. 8, Państwowy Instytut Wydawniczy, Warszawa, 1976',
-            'author': u'Kochanowski, Jan',
-            'url': u'http://wolnelektury.pl/katalog/lektura/piesni-ksiegi-pierwsze-piesn-vii-trudna-rada-w-tej-mierze-pr',
-            'created_at': u'2007-08-31',
-            'title': u'Pieśń VII (Trudna rada w tej mierze: przyjdzie się rozjechać...)',
-            'kind': u'Liryka',
-            'source_url': u'http://www.polona.pl/Content/1499',
-            'released_to_public_domain_at': u'1584-01-01',
-            'epoch': u'Renesans',
-            'genre': u'Pieśń',
-            'technical_editor': u'Gałecki, Dariusz',
-            'license_description': u'Domena publiczna - Jan Kochanowski zm. 1584 ',
-        }),
-        ('dcparser', 'mickiewicz_rybka.xml', {
-            'publisher': u'Fundacja Nowoczesna Polska',
-            'about': 'http://wiki.wolnepodreczniki.pl/Lektury:Mickiewicz/Ballady/Rybka',
-            'source_name': u'Mickiewicz, Adam (1798-1855), Poezje, tom 1 (Wiersze młodzieńcze - Ballady i romanse - Wiersze do r. 1824), Krakowska Spółdzielnia Wydawnicza, wyd. 2 zwiększone, Kraków, 1922',
-            'author': u'Mickiewicz, Adam',
-            'url': u'http://wolnelektury.pl/katalog/lektura/ballady-i-romanse-rybka',
-            'created_at': u'2007-09-06',
-            'title': u'Rybka',
-            'kind': u'Liryka',
-            'source_url': u'http://www.polona.pl/Content/2222',
-            'released_to_public_domain_at': u'1855-01-01',
-            'epoch': u'Romantyzm',
-            'genre': u'Ballada',
-            'technical_editor': u'Sutkowska, Olga',
-            'license_description': u'Domena publiczna - Adam Mickiewicz zm. 1855',
-        }),
-        ('dcparser', 'sofokles_antygona.xml', {
-            'publisher': u'Fundacja Nowoczesna Polska',
-            'about': 'http://wiki.wolnepodreczniki.pl/Lektury:Sofokles/Antygona',
-            'source_name': u'Sofokles (496-406 a.C.), Antygona, Zakład Narodowy im. Ossolińskich, wyd. 7, Lwów, 1939',
-            'author': u'Sofokles',
-            'url': u'http://wolnelektury.pl/katalog/lektura/antygona',
-            'created_at': u'2007-08-30',
-            'title': u'Antygona',
-            'kind': u'Dramat',
-            'source_url': u'http://www.polona.pl/Content/3768',
-            'translator': u'Morawski, Kazimierz',
-            'released_to_public_domain_at': u'1925-01-01',
-            'epoch': u'Starożytność',
-            'genre': u'Tragedia',
-            'technical_editor': u'Gałecki, Dariusz',
-            'license_description': u'Domena publiczna - tłumacz Kazimierz Morawski zm. 1925',
-        }),
-        ('dcparser', 'biedrzycki_akslop.xml', {
-            'publisher': u'Fundacja Nowoczesna Polska',
-            'about': 'http://wiki.wolnepodreczniki.pl/Lektury:Biedrzycki/Akslop',
-            'source_name': u'Miłosz Biedrzycki, * ("Gwiazdka"), Fundacja "brulion", Kraków-Warszawa, 1993',
-            'author': u'Biedrzycki, Miłosz',
-            'url': u'http://wolnelektury.pl/katalog/lektura/akslop',
-            'created_at': u'2009-06-04',
-            'title': u'Akslop',
-            'kind': u'Liryka',
-            'source_url': u'http://free.art.pl/mlb/gwiazdka.html#t1',
-            'epoch': u'Współczesność',
-            'genre': u'Wiersz',
-            'technical_editor': u'Sutkowska, Olga',
-            'license': u'http://creativecommons.org/licenses/by-sa/3.0/',
-            'license_description': u'Creative Commons Uznanie Autorstwa - Na Tych Samych Warunkach 3.0.PL'
-        }),
-    )
-    
-    def test_parse(self):
-        for dir_name, file_name, result in self.KNOWN_RESULTS:
-            self.assertEqual(dcparser.parse(test_file_path(dir_name, file_name)).to_dict(), result)
-
-
-class TestParserErrors(unittest.TestCase):
-    def test_error(self):
-        try:
-            html.transform(test_file_path('erroneous', 'asnyk_miedzy_nami.xml'),
-                           test_file_path('erroneous', 'asnyk_miedzy_nami.html'))
-            self.fail()
-        except etree.XMLSyntaxError, e:
-            self.assertEqual(e.position, (25, 13))
-
-
-if __name__ == '__main__':
-    unittest.main()
\ No newline at end of file
diff --git a/librarian/tests/files/dcparser/.DS_Store b/librarian/tests/files/dcparser/.DS_Store
deleted file mode 100644
index 8817fe608d38202e1a6a534b520570045b510a0f..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 6148
zcmeHK!AiqG5Pe%KRJ`<H#gpt82%fx#QV*W}fS9ILOJX9eqU|Yvu5V_SVyHq+Qpya>
zzU^e^?d;o-?E-M)_t`Zt1Tdfr_D<ORV)9&k$9gexOmxbz#0qm%NHN_RZGnAMK=$qu
z8TV|)o%^|eOKh-aw7`N-QeuXju?%nQ*SNwbUQphZbv`X?+3(?3%XGOTz6p{)Be(G7
z%9oO*;JZ?i=d~l#xz3v7S*xw_!F;JLC4<bc;u>x`a-FgAMXY5`Ch2d;v}A=-vSuB5
zMx3?ABfa2Sqy|}8MTORQVzlbaw2$wG)np2o0*9*r&up>5X+UdD0aL&f*eW332TvDF
zA~pf_)4@is0K`7SYV6BhLUMe>Bw`bgPiQWr5<{wsR}2@@8TWBq60r#w(&2Pw^2C{4
zyrDRqopG<j;gW#XngXUkTY;nSb13Kk?cewRc9Q)v1x$g1Qo!{m_meS~6wlV3#mQOg
s=?`=<8P^2drm&+&F~`bLd_h-Z+)Ia;L~H^wLbE>tDuXqqz`rW+4Xmw^kpKVy

diff --git a/librarian/text.py b/librarian/text.py
index db0d2b2..0754a99 100644
--- a/librarian/text.py
+++ b/librarian/text.py
@@ -18,7 +18,20 @@ ENTITY_SUBSTITUTIONS = [
 ]
 
 
-MAX_LINE_LENGTH = 80
+TEMPLATE = u"""\
+Kodowanie znaków w dokumencie: UTF-8.
+-----
+Publikacja zrealizowana w ramach projektu Wolne Lektury (http://wolnelektury.pl/). Reprodukcja cyfrowa wykonana przez
+Bibliotekę Narodową z egzemplarza pochodzącego ze zbiorów BN. Ten utwór nie jest chroniony prawem autorskim i znajduje
+się w domenie publicznej, co oznacza, że możesz go swobodnie wykorzystywać, publikować i rozpowszechniać.
+
+Wersja lektury w opracowaniu merytorycznym i krytycznym (przypisy i motywy) dostępna jest na stronie %(url)s.
+-----
+
+
+
+%(text)s
+"""
 
 
 def strip(context, text):
@@ -37,17 +50,20 @@ def substitute_entities(context, text):
     return text
 
 
-def wrap_words(context, text):
+def wrap_words(context, text, wrapping):
     """XPath extension function automatically wrapping words in passed text"""
     if isinstance(text, list):
         text = ''.join(text)
+    if not wrapping:
+        return text
+    
     words = re.split(r'\s', text)
     
     line_length = 0
     lines = [[]]
     for word in words:
         line_length += len(word) + 1
-        if line_length > MAX_LINE_LENGTH:
+        if line_length > wrapping:
             # Max line length was exceeded. We create new line
             lines.append([])
             line_length = len(word)
@@ -62,28 +78,22 @@ ns['substitute_entities'] = substitute_entities
 ns['wrap_words'] = wrap_words
 
 
-def transform(input_filename, output_filename):
+def transform(input_filename, output_filename, **options):
     """Transforms file input_filename in XML to output_filename in TXT."""
     # Parse XSLT
     style_filename = os.path.join(os.path.dirname(__file__), 'book2txt.xslt')
     style = etree.parse(style_filename)
 
-    doc_file = cStringIO.StringIO()
-    expr = re.compile(r'/\s', re.MULTILINE | re.UNICODE);
-    
-    f = open(input_filename, 'r')
-    for line in f:
-        line = line.decode('utf-8')
-        line = expr.sub(u'<br/>\n', line)
-        doc_file.write(line.encode('utf-8'))
-    f.close()
+    if is_file:
+        document = WLDocument.from_file(input, True)
+    else:
+        document = WLDocument.from_string(input, True)
 
-    doc_file.seek(0)
+    result = document.transform(style, **options)
 
-    parser = etree.XMLParser(remove_blank_text=True)
-    doc = etree.parse(doc_file, parser)
-    
-    result = doc.xslt(style)
     output_file = codecs.open(output_filename, 'wb', encoding='utf-8')
-    output_file.write(unicode(result) % dcparser.parse(input_filename).url)
+    output_file.write(TEMPLATE % {
+        'url': dcparser.parse(input_filename).url,
+        'text': unicode(result),
+    })
 
diff --git a/librarian/text.pyc b/librarian/text.pyc
deleted file mode 100644
index 6c6eb91802d9dea061615b0f83adc4b5dc39c1c6..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 2993
zcmb_eUvC>l5T8B&N$kW;nxtvcLgj*lEY#RlUnmetX%irjy3xg<A!=kfzD>H#<<7a?
zYaHZAmC83LukgqlPdxH%_y~Lf_|2S?lp-V~>OXrkyK_6UJHOxT{`yCyZ0paz>eBEt
zj^|zc?JsBo{1ojIZJo)Aw$y3AM_b;hoTIIr$R6$IDJ;-dffT==qx~X<BeXRF{SpPU
z6kMiYj)HmGgViVn7ig<Q!32E|KV=FgMHnMpATv(EMS20B71CwW6$++kQDY3!G%W@*
zVtaw3T(4b0#O6<Ynl%N{70_sxXq~B}yr6EclUSXkIVcMY3-bK^`46cl`cQo-&w9N+
z)PI)Y8#jjPPbp9T&QSZ&tU>GY1o`3rKAx6)9nnCMQ}n{4cpB?hWIXP|l1>ywid6qS
zayeSszDAEdI>9f;<0el780Ppz7#_Y)kI=hbV=wkoG@EN3_oS-(PDj?nQFS+sx{ifz
z96i&CGtpkP6NXhCIpeggnyA|ASgV7o!%KApMpWYsa%D8en#j6N)YXpH5E|=5cp96?
z&7fzw*;;9R+g{(ew+>;WweqmJ?nW?>=w8_A>KeZ$9q`W{<aH5>7?CKB-35p?-LVKM
z#TVh$&_mMLk0CWSkQ>|hF5S^dV=alF>V0?IIEuqahY-``xZVpJq1oYo(n(AwYRGze
z$8P%7$#q*?8g1ven?X(9q?%SEG_GcqWQzmi9guAO2H`(s3#meCf-HxUFW?&~+auaT
zO7qm;=EN;+Z_r7B^9K+P3Y_jBPnSR7<f5-gPxBPNO=^%QGr}*!K%JSwijjfApg{95
z$YxN7U(=(Y$OsBX822z(!i)H=*BIP(3C+I(qm#P$pyNyzusE(BC7oW6L2?u)!Rzqw
z1!mklwgxR32E0(qvQ+MJ%i~NV%l~)K^smx@XnTec8^B(`kLbQphebqwieJ!ysG4!q
z=L86<9Zy%kImBP2-yAY*2{SM5=}7l`$(@@}IAW45^LYV2gI6MvAQCN=!}dbskZ;fd
zSen4|!DpN8)s^OQdv&?_Me9qXkqEn}qoBs3&pGhfnY=F>>&BplCem$8vge-22<uiB
z$~JFqp+aR&Y+HO^hil6<F+={8S5zh7zo=$Z#jB`EbxGMOMxQv>`>62`mYr4<>oa@&
zc{D;B)|mN1o7>1GXajrzl|uml&kN$E?4hJ#(Wc0>02Y94lnFCRFNo>lN@P&-VFO`I
z+HtzaMXx}v!f*n=SS?b2nFd8L>k~QwBQOW5zW8Ak9wwMA=x1y6fLW7bc5!G1=jt#*
z9R-7?*uAiCn!qR?;vi?3Snd)A)ye2-*r>#a7L%4}fFNh3Dq|sKa+k><Mt~iM7)~%m
z<EfJ|8jP`TH%EgqPwnPq`9gAYg=X1XPzGli8}=@mR?>;=Zk!z0>aGbjighn_Z5E@>
z0jt%`2dh;VSL4*ZKGNE3)olfn?8Nc@Er;6Px1qC)+s0!nTuUsoEOomJw^Ih{;!e`I
zQx;;oi_jB&5I@uHov;(_x20w?7ms!v16oJjI6$Fj)~-Smg2WkkdpLnI)-W6blW?%|
zPFrGcw|(?yWRO!AIR=K-U4rRW9*X6u;!m<A&(n<$dRVyFKG;}otuW(cWTUxq|KZZI
z5GMxfgnxlXcv@(RlO-cwf)h^$CH9QX-7vP<;bt{X5{?9Wx+a$anX#3m**I^Xyv-AL
zt+-Aj!<x^POMInCSYsl{k@oj=NvU`(QQ67ojzfKxRSDH{muy53cV$Ro;X);u+x5Bb
zNHp2KMV!M{x)N<u#L6&w#orsOCsNB!vH2m~NSUYHbv5CQtGB!fDJK(ZN?lQ9ui{Oq
zDfG=LDd=)NdgY3hbuzNfu3azB4PUuzog1FtxUxRp{<CXWjkm_X#?8CjFoOJBXk-h+
z{y~_*j~8)vTaXO;Qmy&Ea9-Mk2%irxQB45Z_HeBcxJTB%3)}O373u%pS1glj@gNS;
iP~YKQk-Z&P^7SUY$$U;_-%`GqE0s#q6RNDrxxWCiK#NEK

diff --git a/scripts/book2html b/scripts/book2html
index a0229bb..02f2fa7 100755
--- a/scripts/book2html
+++ b/scripts/book2html
@@ -2,7 +2,7 @@
 import os
 import optparse
 
-from librarian import html
+from librarian import html, ParseError
 
 
 if __name__ == '__main__':
@@ -27,5 +27,25 @@ if __name__ == '__main__':
             print input_filename
         
         output_filename = os.path.splitext(input_filename)[0] + '.html'
-        html.transform(input_filename, output_filename)
+        try:
+            html.transform(input_filename, output_filename)
+        except ParseError, e:
+            print '%(file)s:%(name)s:%(message)s' % {
+                'file': input_filename,
+                'name': e.__class__.__name__,
+                'message': e.message
+            }
+        except IOError, e:
+            print '%(file)s:%(name)s:%(message)s' % {
+                'file': input_filename,
+                'name': e.__class__.__name__,
+                'message': e.strerror,
+            }
+        except BaseException, e:
+            print '%(file)s:%(etype)s:%(message)s' % {
+                'file': input_filename,
+                'etype': e.__class__.__name__,
+                'message': e.message,
+            }
+            raise e
 
diff --git a/scripts/book2txt b/scripts/book2txt
index 1ca4623..41a3978 100755
--- a/scripts/book2txt
+++ b/scripts/book2txt
@@ -3,6 +3,7 @@ import os
 import optparse
 
 from librarian import text
+from librarian import dcparser, ParseError
 
 
 if __name__ == '__main__':
@@ -14,9 +15,11 @@ if __name__ == '__main__':
 
     parser.add_option('-v', '--verbose', action='store_true', dest='verbose', default=False,
         help='print status messages to stdout')
-
+    parser.add_option('-w', '--wrap', action='store', type='int', dest='wrapping', default=0,
+        help='set line wrap column')
+    
     options, input_filenames = parser.parse_args()
-
+    
     if len(input_filenames) < 1:
         parser.print_help()
         exit(1)
@@ -27,5 +30,24 @@ if __name__ == '__main__':
             print input_filename
         
         output_filename = os.path.splitext(input_filename)[0] + '.txt'
-        text.transform(input_filename, output_filename)
-
+        try:
+            text.transform(input_filename, output_filename, wrapping=str(options.wrapping))
+        except ParseError, e:
+            print '%(file)s:%(name)s:%(message)s' % {
+                'file': input_filename,
+                'name': e.__class__.__name__,
+                'message': e.message
+            }
+        except IOError, e:
+            print '%(file)s:%(name)s:%(message)s' % {
+                'file': input_filename,
+                'name': e.__class__.__name__,
+                'message': e.strerror,
+            }
+        except BaseException, e:
+            print '%(file)s:%(etype)s:%(message)s' % {
+                'file': input_filename,
+                'etype': e.__class__.__name__,
+                'message': e.message,
+            }
+            raise e
diff --git a/setup.py b/setup.py
old mode 100644
new mode 100755
index 0988321..09bb42b
--- a/setup.py
+++ b/setup.py
@@ -1,19 +1,21 @@
+#!/usr/bin/env python
 # -*- coding: utf-8 -*-
 from distutils.core import setup
-
+from tests.utils import TestCommand
 
 setup(
     name='librarian',
-    version='1.1',
+    version='1.2.1',
     description='Converter from WolneLektury.pl XML-based language to XHTML, TXT and other formats',
     author='Marek Stępniowski',
     author_email='marek@stepniowski.com',
     url='http://redmine.nowoczesnapolska.org.pl/',
-    packages=['librarian', 'librarian.tests'],
-    package_dir={'librarian': 'librarian'},
+    packages=['librarian', 'tests'],
+    package_dir={'librarian': 'librarian', 'tests': 'tests'},
     package_data={
         'librarian': ['*.xslt'],
-        'librarian.tests': ['files/dcparser/*.xml', 'files/erroneous/*.xml'],
+        'tests': ['files/dcparser/*.xml', 'files/erroneous/*.xml'],
     },
     scripts=['scripts/book2html', 'scripts/book2txt', 'scripts/bookfragments', 'scripts/genslugs'],
+    cmdclass={'test': TestCommand},
 )
diff --git a/tests/__init__.py b/tests/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/tests/files/dcparser/andersen_brzydkie_kaczatko.out b/tests/files/dcparser/andersen_brzydkie_kaczatko.out
new file mode 100644
index 0000000..fda83eb
--- /dev/null
+++ b/tests/files/dcparser/andersen_brzydkie_kaczatko.out
@@ -0,0 +1,17 @@
+{
+    'publisher': u'Fundacja Nowoczesna Polska',
+    'about': u'http://wiki.wolnepodreczniki.pl/Lektury:Andersen/Brzydkie_kaczątko',
+    'source_name': u'Andersen, Hans Christian (1805-1875), Baśnie, Gebethner i Wolff, wyd. 7, Kraków, 1925',
+    'author': u'Andersen, Hans Christian',
+    'url': u'http://wolnelektury.pl/katalog/lektura/brzydkie-kaczatko',
+    'created_at': u'2007-08-14',
+    'title': u'Brzydkie kaczątko',
+    'kind': u'Epika',
+    'source_url': u'http://www.polona.pl/dlibra/doccontent2?id=3563&dirids=4',
+    'translators': [u'Niewiadomska, Cecylia'],
+    'released_to_public_domain_at': u'1925-01-01',
+    'epoch': u'Romantyzm',
+    'genre': u'Baśń',
+    'technical_editors': [u'Gałecki, Dariusz'],
+    'license_description': u'Domena publiczna - tłumacz Cecylia Niewiadomska zm. 1925',
+}
diff --git a/librarian/tests/files/dcparser/andersen_brzydkie_kaczatko.xml b/tests/files/dcparser/andersen_brzydkie_kaczatko.xml
similarity index 100%
rename from librarian/tests/files/dcparser/andersen_brzydkie_kaczatko.xml
rename to tests/files/dcparser/andersen_brzydkie_kaczatko.xml
diff --git a/tests/files/dcparser/biedrzycki_akslop.out b/tests/files/dcparser/biedrzycki_akslop.out
new file mode 100644
index 0000000..1111b9c
--- /dev/null
+++ b/tests/files/dcparser/biedrzycki_akslop.out
@@ -0,0 +1,17 @@
+{
+            'editors': [u'Sekuła, Aleksandra'],
+            'publisher': u'Fundacja Nowoczesna Polska',
+            'about': 'http://wiki.wolnepodreczniki.pl/Lektury:Biedrzycki/Akslop',
+            'source_name': u'Miłosz Biedrzycki, * ("Gwiazdka"), Fundacja "brulion", Kraków-Warszawa, 1993',
+            'author': u'Biedrzycki, Miłosz',
+            'url': u'http://wolnelektury.pl/katalog/lektura/akslop',
+            'created_at': u'2009-06-04',
+            'title': u'Akslop',
+            'kind': u'Liryka',
+            'source_url': u'http://free.art.pl/mlb/gwiazdka.html#t1',
+            'epoch': u'Współczesność',
+            'genre': u'Wiersz',
+            'technical_editors': [u'Sutkowska, Olga'],
+            'license': u'http://creativecommons.org/licenses/by-sa/3.0/',
+            'license_description': u'Creative Commons Uznanie Autorstwa - Na Tych Samych Warunkach 3.0.PL'
+}
diff --git a/librarian/tests/files/dcparser/biedrzycki_akslop.xml b/tests/files/dcparser/biedrzycki_akslop.xml
similarity index 100%
rename from librarian/tests/files/dcparser/biedrzycki_akslop.xml
rename to tests/files/dcparser/biedrzycki_akslop.xml
diff --git a/tests/files/dcparser/kochanowski_piesn7.out b/tests/files/dcparser/kochanowski_piesn7.out
new file mode 100644
index 0000000..4c1dc8f
--- /dev/null
+++ b/tests/files/dcparser/kochanowski_piesn7.out
@@ -0,0 +1,18 @@
+{
+            'publisher': u'Fundacja Nowoczesna Polska',
+            'about': u'http://wiki.wolnepodreczniki.pl/Lektury:Kochanowski/Pieśni/Pieśń_VII_(1)',
+            'source_name': u'Kochanowski, Jan (1530-1584), Dzieła polskie, tom 1, oprac. Julian Krzyżanowski, wyd. 8, Państwowy Instytut Wydawniczy, Warszawa, 1976',
+            'author': u'Kochanowski, Jan',
+            'url': u'http://wolnelektury.pl/katalog/lektura/piesni-ksiegi-pierwsze-piesn-vii-trudna-rada-w-tej-mierze-pr',
+            'created_at': u'2007-08-31',
+            'title': u'Pieśń VII (Trudna rada w tej mierze: przyjdzie się rozjechać...)',
+            'kind': u'Liryka',
+            'source_url': u'http://www.polona.pl/Content/1499',
+            'released_to_public_domain_at': u'1584-01-01',
+            'epoch': u'Renesans',
+            'genre': u'Pieśń',
+            'technical_editors': [u'Gałecki, Dariusz'],
+            'license_description': u'Domena publiczna - Jan Kochanowski zm. 1584 ',
+            'editors': [u'Sekuła, Aleksandra', u'Krzyżanowski, Julian', u'Otwinowska, Barbara'],
+}
+
diff --git a/librarian/tests/files/dcparser/kochanowski_piesn7.xml b/tests/files/dcparser/kochanowski_piesn7.xml
similarity index 99%
rename from librarian/tests/files/dcparser/kochanowski_piesn7.xml
rename to tests/files/dcparser/kochanowski_piesn7.xml
index 96be1ae..b4d8d2e 100644
--- a/librarian/tests/files/dcparser/kochanowski_piesn7.xml
+++ b/tests/files/dcparser/kochanowski_piesn7.xml
@@ -24,4 +24,4 @@
         <dc:audience xml:lang="pl">L</dc:audience>
         <dc:language xml:lang="pl">pol</dc:language>
     </rdf:Description>
-</rdf:RDF>
\ No newline at end of file
+</rdf:RDF>
diff --git a/tests/files/dcparser/mickiewicz_rybka.out b/tests/files/dcparser/mickiewicz_rybka.out
new file mode 100644
index 0000000..ff4bd98
--- /dev/null
+++ b/tests/files/dcparser/mickiewicz_rybka.out
@@ -0,0 +1,18 @@
+{
+            'editors': [u'Sekuła, Aleksandra', u'Kallenbach, Józef'],
+            'publisher': u'Fundacja Nowoczesna Polska',
+            'about': 'http://wiki.wolnepodreczniki.pl/Lektury:Mickiewicz/Ballady/Rybka',
+            'source_name': u'Mickiewicz, Adam (1798-1855), Poezje, tom 1 (Wiersze młodzieńcze - Ballady i romanse - Wiersze do r. 1824), Krakowska Spółdzielnia Wydawnicza, wyd. 2 zwiększone, Kraków, 1922',
+            'author': u'Mickiewicz, Adam',
+            'url': u'http://wolnelektury.pl/katalog/lektura/ballady-i-romanse-rybka',
+            'created_at': u'2007-09-06',
+            'title': u'Rybka',
+            'kind': u'Liryka',
+            'source_url': u'http://www.polona.pl/Content/2222',
+            'released_to_public_domain_at': u'1855-01-01',
+            'epoch': u'Romantyzm',
+            'genre': u'Ballada',
+            'technical_editors': [u'Sutkowska, Olga'],
+            'license_description': u'Domena publiczna - Adam Mickiewicz zm. 1855',
+}
+
diff --git a/librarian/tests/files/dcparser/mickiewicz_rybka.xml b/tests/files/dcparser/mickiewicz_rybka.xml
similarity index 100%
rename from librarian/tests/files/dcparser/mickiewicz_rybka.xml
rename to tests/files/dcparser/mickiewicz_rybka.xml
diff --git a/tests/files/dcparser/sofokles_antygona.out b/tests/files/dcparser/sofokles_antygona.out
new file mode 100644
index 0000000..0f2b4d0
--- /dev/null
+++ b/tests/files/dcparser/sofokles_antygona.out
@@ -0,0 +1,19 @@
+{
+            'editors': [u'Sekuła, Aleksandra'],
+            'publisher': u'Fundacja Nowoczesna Polska',
+            'about': 'http://wiki.wolnepodreczniki.pl/Lektury:Sofokles/Antygona',
+            'source_name': u'Sofokles (496-406 a.C.), Antygona, Zakład Narodowy im. Ossolińskich, wyd. 7, Lwów, 1939',
+            'author': u'Sofokles',
+            'url': u'http://wolnelektury.pl/katalog/lektura/antygona',
+            'created_at': u'2007-08-30',
+            'title': u'Antygona',
+            'kind': u'Dramat',
+            'source_url': u'http://www.polona.pl/Content/3768',
+            'translators': [u'Morawski, Kazimierz'],
+            'released_to_public_domain_at': u'1925-01-01',
+            'epoch': u'Starożytność',
+            'genre': u'Tragedia',
+            'technical_editors': [u'Gałecki, Dariusz'],
+            'license_description': u'Domena publiczna - tłumacz Kazimierz Morawski zm. 1925',
+}
+
diff --git a/librarian/tests/files/dcparser/sofokles_antygona.xml b/tests/files/dcparser/sofokles_antygona.xml
similarity index 100%
rename from librarian/tests/files/dcparser/sofokles_antygona.xml
rename to tests/files/dcparser/sofokles_antygona.xml
diff --git a/tests/files/dcserialize/andersen_brzydkie_kaczatko.out b/tests/files/dcserialize/andersen_brzydkie_kaczatko.out
new file mode 100644
index 0000000..e69de29
diff --git a/tests/files/dcserialize/andersen_brzydkie_kaczatko.xml b/tests/files/dcserialize/andersen_brzydkie_kaczatko.xml
new file mode 100644
index 0000000..d653a9b
--- /dev/null
+++ b/tests/files/dcserialize/andersen_brzydkie_kaczatko.xml
@@ -0,0 +1,24 @@
+<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:dc="http://purl.org/dc/elements/1.1/">
+    <rdf:Description rdf:about="http://wiki.wolnepodreczniki.pl/Lektury:Andersen/Brzydkie_kaczątko">
+        <dc:creator xml:lang="pl">Andersen, Hans Christian</dc:creator>
+        <dc:title xml:lang="pl">Brzydkie kaczątko</dc:title>
+        <dc:contributor.translator xml:lang="pl">Niewiadomska, Cecylia</dc:contributor.translator>
+        <dc:contributor.technical_editor xml:lang="pl">Gałecki, Dariusz</dc:contributor.technical_editor>
+        <dc:publisher xml:lang="pl">Fundacja Nowoczesna Polska</dc:publisher>
+        <dc:subject.period xml:lang="pl">Romantyzm</dc:subject.period>
+        <dc:subject.type xml:lang="pl">Epika</dc:subject.type>
+        <dc:subject.genre xml:lang="pl">Baśń</dc:subject.genre>
+        <dc:description xml:lang="pl">Publikacja zrealizowana w ramach projektu Wolne Lektury (http://wolnelektury.pl). Reprodukcja cyfrowa wykonana przez Bibliotekę Narodową z egzemplarza pochodzącego ze zbiorów BN.</dc:description>
+        <dc:identifier.url xml:lang="pl">http://wolnelektury.pl/katalog/lektura/brzydkie-kaczatko</dc:identifier.url>
+        <dc:source.URL xml:lang="pl">http://www.polona.pl/dlibra/doccontent2?id=3563&amp;dirids=4</dc:source.URL>
+        <dc:source xml:lang="pl">Andersen, Hans Christian (1805-1875), Baśnie, Gebethner i Wolff, wyd. 7, Kraków, 1925</dc:source>
+        <dc:rights xml:lang="pl">Domena publiczna - tłumacz Cecylia Niewiadomska zm. 1925</dc:rights>
+        <dc:date.pd xml:lang="pl">1925</dc:date.pd>
+        <dc:format xml:lang="pl">xml</dc:format>
+        <dc:type xml:lang="pl">text</dc:type>
+        <dc:type xml:lang="en">text</dc:type>
+        <dc:date xml:lang="pl">2007-08-14</dc:date>
+        <dc:audience xml:lang="pl">SP1</dc:audience>
+        <dc:language xml:lang="pl">pol</dc:language>
+    </rdf:Description>
+</rdf:RDF>
\ No newline at end of file
diff --git a/tests/files/dcserialize/biedrzycki_akslop.out b/tests/files/dcserialize/biedrzycki_akslop.out
new file mode 100644
index 0000000..e69de29
diff --git a/tests/files/dcserialize/biedrzycki_akslop.xml b/tests/files/dcserialize/biedrzycki_akslop.xml
new file mode 100644
index 0000000..da0cd9f
--- /dev/null
+++ b/tests/files/dcserialize/biedrzycki_akslop.xml
@@ -0,0 +1,25 @@
+<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+xmlns:dc="http://purl.org/dc/elements/1.1/">
+    <rdf:Description rdf:about="http://wiki.wolnepodreczniki.pl/Lektury:Biedrzycki/Akslop">
+        <dc:creator xml:lang="pl">Biedrzycki, Miłosz</dc:creator>
+        <dc:title xml:lang="pl">Akslop</dc:title>
+        <dc:contributor.editor xml:lang="pl">Sekuła, Aleksandra</dc:contributor.editor>
+        <dc:contributor.technical_editor xml:lang="pl">Sutkowska, Olga</dc:contributor.technical_editor>
+        <dc:publisher xml:lang="pl">Fundacja Nowoczesna Polska</dc:publisher>
+        <dc:subject.period xml:lang="pl">Współczesność</dc:subject.period>
+        <dc:subject.type xml:lang="pl">Liryka</dc:subject.type>
+        <dc:subject.genre xml:lang="pl">Wiersz</dc:subject.genre>
+        <dc:description xml:lang="pl">Publikacja zrealizowana w ramach projektu Wolne Lektury (http://wolnelektury.pl).</dc:description>
+        <dc:identifier.url xml:lang="pl">http://wolnelektury.pl/katalog/lektura/akslop</dc:identifier.url>
+        <dc:source.URL xml:lang="pl">http://free.art.pl/mlb/gwiazdka.html#t1</dc:source.URL>
+        <dc:source xml:lang="pl">Miłosz Biedrzycki, * ("Gwiazdka"), Fundacja "brulion", Kraków-Warszawa, 1993</dc:source>
+        <dc:rights xml:lang="pl">Creative Commons Uznanie Autorstwa - Na Tych Samych Warunkach 3.0.PL</dc:rights>
+        <dc:rights.license>http://creativecommons.org/licenses/by-sa/3.0/</dc:rights.license>
+        <dc:format xml:lang="pl">xml</dc:format>
+        <dc:type xml:lang="pl">text</dc:type>
+        <dc:type xml:lang="en">text</dc:type>
+        <dc:date xml:lang="pl">2009-06-04</dc:date>
+        <dc:audience xml:lang="pl">L</dc:audience>
+        <dc:language xml:lang="pl">pol</dc:language>
+    </rdf:Description>
+</rdf:RDF>
\ No newline at end of file
diff --git a/tests/files/dcserialize/kochanowski_piesn7.out b/tests/files/dcserialize/kochanowski_piesn7.out
new file mode 100644
index 0000000..e69de29
diff --git a/tests/files/dcserialize/kochanowski_piesn7.xml b/tests/files/dcserialize/kochanowski_piesn7.xml
new file mode 100644
index 0000000..b4d8d2e
--- /dev/null
+++ b/tests/files/dcserialize/kochanowski_piesn7.xml
@@ -0,0 +1,27 @@
+<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:dc="http://purl.org/dc/elements/1.1/">
+    <rdf:Description rdf:about="http://wiki.wolnepodreczniki.pl/Lektury:Kochanowski/Pieśni/Pieśń_VII_(1)">
+        <dc:creator xml:lang="pl">Kochanowski, Jan</dc:creator>
+        <dc:title xml:lang="pl">Pieśń VII (Trudna rada w tej mierze: przyjdzie się rozjechać...)</dc:title>
+        <dc:relation.isPartOf xml:lang="pl">http://www.wolnelektury.pl/lektura/piesni-ksiegi-pierwsze</dc:relation.isPartOf>
+        <dc:contributor.editor xml:lang="pl">Sekuła, Aleksandra</dc:contributor.editor>
+        <dc:contributor.editor xml:lang="pl">Krzyżanowski, Julian</dc:contributor.editor>
+        <dc:contributor.editor xml:lang="pl">Otwinowska, Barbara</dc:contributor.editor>
+        <dc:contributor.technical_editor xml:lang="pl">Gałecki, Dariusz</dc:contributor.technical_editor>
+        <dc:publisher xml:lang="pl">Fundacja Nowoczesna Polska</dc:publisher>
+        <dc:subject.period xml:lang="pl">Renesans</dc:subject.period>
+        <dc:subject.type xml:lang="pl">Liryka</dc:subject.type>
+        <dc:subject.genre xml:lang="pl">Pieśń</dc:subject.genre>
+        <dc:description xml:lang="pl">Publikacja zrealizowana w ramach projektu Wolne Lektury (http://wolnelektury.pl). Reprodukcja cyfrowa wykonana przez Bibliotekę Narodową z egzemplarza pochodzącego ze zbiorów BN.</dc:description>
+        <dc:identifier.url xml:lang="pl">http://wolnelektury.pl/katalog/lektura/piesni-ksiegi-pierwsze-piesn-vii-trudna-rada-w-tej-mierze-pr</dc:identifier.url>
+        <dc:source.URL xml:lang="pl">http://www.polona.pl/Content/1499</dc:source.URL>
+        <dc:source xml:lang="pl">Kochanowski, Jan (1530-1584), Dzieła polskie, tom 1, oprac. Julian Krzyżanowski, wyd. 8, Państwowy Instytut Wydawniczy, Warszawa, 1976</dc:source>
+        <dc:rights xml:lang="pl">Domena publiczna - Jan Kochanowski zm. 1584 </dc:rights>
+        <dc:date.pd xml:lang="pl">1584</dc:date.pd>
+        <dc:format xml:lang="pl">xml</dc:format>
+        <dc:type xml:lang="pl">text</dc:type>
+        <dc:type xml:lang="en">text</dc:type>
+        <dc:date xml:lang="pl">2007-08-31</dc:date>
+        <dc:audience xml:lang="pl">L</dc:audience>
+        <dc:language xml:lang="pl">pol</dc:language>
+    </rdf:Description>
+</rdf:RDF>
diff --git a/tests/files/dcserialize/mickiewicz_rybka.out b/tests/files/dcserialize/mickiewicz_rybka.out
new file mode 100644
index 0000000..e69de29
diff --git a/tests/files/dcserialize/mickiewicz_rybka.xml b/tests/files/dcserialize/mickiewicz_rybka.xml
new file mode 100644
index 0000000..0796a5b
--- /dev/null
+++ b/tests/files/dcserialize/mickiewicz_rybka.xml
@@ -0,0 +1,28 @@
+<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:dc="http://purl.org/dc/elements/1.1/">
+    <rdf:Description rdf:about="http://wiki.wolnepodreczniki.pl/Lektury:Mickiewicz/Ballady/Rybka">
+        <dc:creator xml:lang="pl">Mickiewicz, Adam</dc:creator>
+        <dc:title xml:lang="pl">Rybka</dc:title>
+        <dc:relation.isPartOf xml:lang="pl">http://www.wolnelektury.pl/lektura/ballady-i-romanse</dc:relation.isPartOf>
+        <dc:contributor.editor xml:lang="pl">Sekuła, Aleksandra</dc:contributor.editor>
+        <dc:contributor.editor xml:lang="pl">Kallenbach, Józef</dc:contributor.editor>
+        <dc:contributor.technical_editor xml:lang="pl">Sutkowska, Olga</dc:contributor.technical_editor>
+        <dc:publisher xml:lang="pl">Fundacja Nowoczesna Polska</dc:publisher>
+        <dc:subject.period xml:lang="pl">Romantyzm</dc:subject.period>
+        <dc:subject.type xml:lang="pl">Liryka</dc:subject.type>
+        <dc:subject.genre xml:lang="pl">Ballada</dc:subject.genre>
+        <dc:description xml:lang="pl">Publikacja zrealizowana w ramach projektu Wolne Lektury (http://wolnelektury.pl). Reprodukcja cyfrowa wykonana przez Bibliotekę Narodową z egzemplarza pochodzącego ze zbiorów BN.</dc:description>
+        <dc:identifier.url xml:lang="pl">http://wolnelektury.pl/katalog/lektura/ballady-i-romanse-rybka</dc:identifier.url>
+        <dc:source.URL xml:lang="pl">http://www.polona.pl/Content/2222</dc:source.URL>
+        <dc:source xml:lang="pl">Mickiewicz, Adam (1798-1855), Poezje, tom 1 (Wiersze młodzieńcze - Ballady i romanse - Wiersze do r. 1824), Krakowska Spółdzielnia Wydawnicza, wyd. 2 zwiększone, Kraków, 1922</dc:source>
+        <dc:rights xml:lang="pl">Domena publiczna - Adam Mickiewicz zm. 1855</dc:rights>
+        <dc:date.pd xml:lang="pl">1855</dc:date.pd>
+        <dc:format xml:lang="pl">xml</dc:format>
+        <dc:type xml:lang="pl">text</dc:type>
+        <dc:type xml:lang="en">text</dc:type>
+        <dc:date xml:lang="pl">2007-09-06</dc:date>
+        <dc:audience xml:lang="pl">SP2</dc:audience>
+        <dc:audience xml:lang="pl">G</dc:audience>
+        <dc:audience xml:lang="pl">L</dc:audience>
+        <dc:language xml:lang="pl">pol</dc:language>
+    </rdf:Description>
+</rdf:RDF>
\ No newline at end of file
diff --git a/tests/files/dcserialize/sofokles_antygona.out b/tests/files/dcserialize/sofokles_antygona.out
new file mode 100644
index 0000000..e69de29
diff --git a/tests/files/dcserialize/sofokles_antygona.xml b/tests/files/dcserialize/sofokles_antygona.xml
new file mode 100644
index 0000000..4acb2d4
--- /dev/null
+++ b/tests/files/dcserialize/sofokles_antygona.xml
@@ -0,0 +1,25 @@
+<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:dc="http://purl.org/dc/elements/1.1/">
+    <rdf:Description rdf:about="http://wiki.wolnepodreczniki.pl/Lektury:Sofokles/Antygona">
+        <dc:creator xml:lang="pl">Sofokles</dc:creator>
+        <dc:title xml:lang="pl">Antygona</dc:title>
+        <dc:contributor.editor xml:lang="pl">Sekuła, Aleksandra</dc:contributor.editor>
+        <dc:contributor.translator xml:lang="pl">Morawski, Kazimierz</dc:contributor.translator>
+        <dc:contributor.technical_editor xml:lang="pl">Gałecki, Dariusz</dc:contributor.technical_editor>
+        <dc:publisher xml:lang="pl">Fundacja Nowoczesna Polska</dc:publisher>
+        <dc:subject.period xml:lang="pl">Starożytność</dc:subject.period>
+        <dc:subject.type xml:lang="pl">Dramat</dc:subject.type>
+        <dc:subject.genre xml:lang="pl">Tragedia</dc:subject.genre>
+        <dc:description xml:lang="pl">Publikacja zrealizowana w ramach projektu Wolne Lektury (http://wolnelektury.pl). Reprodukcja cyfrowa wykonana przez Bibliotekę Narodową z egzemplarza pochodzącego ze zbiorów BN.</dc:description>
+        <dc:identifier.url xml:lang="pl">http://wolnelektury.pl/katalog/lektura/antygona</dc:identifier.url>
+        <dc:source.URL xml:lang="pl">http://www.polona.pl/Content/3768</dc:source.URL>
+        <dc:source xml:lang="pl">Sofokles (496-406 a.C.), Antygona, Zakład Narodowy im. Ossolińskich, wyd. 7, Lwów, 1939</dc:source>
+        <dc:rights xml:lang="pl">Domena publiczna - tłumacz Kazimierz Morawski zm. 1925</dc:rights>
+        <dc:date.pd xml:lang="pl">1925</dc:date.pd>
+        <dc:format xml:lang="pl">xml</dc:format>
+        <dc:type xml:lang="pl">text</dc:type>
+        <dc:type xml:lang="en">text</dc:type>
+        <dc:date xml:lang="pl">2007-08-30</dc:date>
+        <dc:audience xml:lang="pl">G</dc:audience>
+        <dc:language xml:lang="pl">pol</dc:language>
+    </rdf:Description>
+</rdf:RDF>
\ No newline at end of file
diff --git a/librarian/tests/files/erroneous/asnyk_miedzy_nami.html b/tests/files/erroneous/asnyk_miedzy_nami.html
similarity index 100%
rename from librarian/tests/files/erroneous/asnyk_miedzy_nami.html
rename to tests/files/erroneous/asnyk_miedzy_nami.html
diff --git a/librarian/tests/files/erroneous/asnyk_miedzy_nami.xml b/tests/files/erroneous/asnyk_miedzy_nami.xml
similarity index 100%
rename from librarian/tests/files/erroneous/asnyk_miedzy_nami.xml
rename to tests/files/erroneous/asnyk_miedzy_nami.xml
diff --git a/tests/files/text/asnyk_miedzy_nami.txt b/tests/files/text/asnyk_miedzy_nami.txt
new file mode 100644
index 0000000..e69de29
diff --git a/tests/files/text/asnyk_miedzy_nami.xml b/tests/files/text/asnyk_miedzy_nami.xml
new file mode 100644
index 0000000..5716a28
--- /dev/null
+++ b/tests/files/text/asnyk_miedzy_nami.xml
@@ -0,0 +1,25 @@
+<?xml version='1.0' encoding='utf-8'?>
+<utwor><liryka_lp>
+    <autor_utworu>Adam Asnyk</autor_utworu>
+    <nazwa_utworu>Między nami nic nie było</nazwa_utworu>
+
+    <strofa>Między nami nic nie było!/
+    Żadnych zwierzeń, wyznań żadnych!/
+    Nic nas z sobą nie łączyło ---/
+    Prócz wiosennych marzeń zdradnych;</strofa>
+
+    <strofa>Prócz tych woni, barw i blasków,/
+    Unoszących się w przestrzeni;/
+    Prócz szumiących śpiewem lasków/
+    I tej świeżej łąk zieleni;</strofa>
+
+    <strofa>Prócz tych kaskad i potoków,/
+    Zraszających każdy parów,/
+    Prócz girlandy tęcz, obłoków,/
+    Prócz natury słodkich czarów;</strofa>
+
+    <strofa>Prócz tych wspólnych, jasnych zdrojów,/
+    Z których serce zachwyt piło;/
+    Prócz pierwiosnków i powojów,---/
+    Między nami nic nie było!</strofa>
+</liryka_lp></utwor>
diff --git a/tests/test_dcparser.py b/tests/test_dcparser.py
new file mode 100755
index 0000000..62e664c
--- /dev/null
+++ b/tests/test_dcparser.py
@@ -0,0 +1,56 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+import unittest
+
+from lxml import etree
+from utils import get_file_path
+from librarian import dcparser, html, ParseError
+from utils import AutoTestMetaclass
+
+class TestDCParser(unittest.TestCase):
+    __metaclass__ = AutoTestMetaclass
+
+    TEST_DIR = 'dcparser'
+
+    def run_auto_test(self, in_data, out_data):
+        info = dcparser.BookInfo.from_string(in_data).to_dict()
+        should_be = eval(out_data)
+        for key in should_be:
+            self.assertEqual( info[key], should_be[key] )
+
+class TestDCSerialize(unittest.TestCase):
+    __metaclass__ = AutoTestMetaclass
+
+    TEST_DIR = 'dcserialize'
+
+    def run_auto_test(self, in_data, out_data):
+        import lxml.etree
+        # first parse the input
+        info = dcparser.BookInfo.from_string(in_data)
+
+        # serialize
+        serialized = lxml.etree.tostring(info.to_etree(), encoding=unicode).encode('utf-8')
+
+        # then parse again
+        info_bis = dcparser.BookInfo.from_string(serialized)
+
+        # check if they are the same
+        for key in vars(info):
+            self.assertEqual( getattr(info, key), getattr(info_bis, key))
+
+        for key in vars(info_bis):
+            self.assertEqual( getattr(info, key), getattr(info_bis, key))
+
+class TestParserErrors(unittest.TestCase):
+    def test_error(self):
+        try:
+            html.transform(get_file_path('erroneous', 'asnyk_miedzy_nami.xml'),
+                           get_file_path('erroneous', 'asnyk_miedzy_nami.html'))
+            self.fail()
+        except ParseError:
+            pass
+            #self.assertEqual(e.position, (25, 13))    
+
+if __name__ == '__main__':
+    unittest.main()
diff --git a/tests/test_text.py b/tests/test_text.py
new file mode 100755
index 0000000..00fd787
--- /dev/null
+++ b/tests/test_text.py
@@ -0,0 +1,22 @@
+#!/usr/bin/env python
+# encoding: utf-8
+
+import unittest
+
+from utils import get_file_path
+from librarian import dcparser
+from librarian import text, NoDublinCore
+
+
+class TestXML(unittest.TestCase):
+    def test_no_dublincore(self):
+        try:
+            text.transform(get_file_path('text', 'asnyk_miedzy_nami.xml'),
+                           get_file_path('text', 'asnyk_miedzy_nami.txt'))
+            self.fail()
+        except NoDublinCore, e:
+            pass
+
+
+if __name__ == '__main__':
+    unittest.main()
diff --git a/tests/utils.py b/tests/utils.py
new file mode 100644
index 0000000..1870a07
--- /dev/null
+++ b/tests/utils.py
@@ -0,0 +1,62 @@
+from __future__ import with_statement
+
+import os
+from distutils.core import Command
+from unittest import TextTestRunner, TestLoader
+from glob import glob
+from os.path import dirname, join, realpath, splitext, basename, walk
+from os import listdir
+import codecs
+
+class AutoTestMetaclass(type):
+
+    def __new__(cls, name, bases, class_dict):        
+        test_dir = class_dict.pop('TEST_DIR')
+        path = realpath( join(dirname(__file__), 'files', test_dir) )
+
+        for file in listdir(path):
+            base, ext = splitext(file)
+            if ext != '.xml':
+                continue
+
+            class_dict['test_'+base] = cls.make_test_runner(base, \
+                    join(path, base +'.xml'), join(path, base + '.out') )
+
+        return type.__new__(cls, name, bases, class_dict)
+    
+    @staticmethod
+    def make_test_runner(name, inputf, outputf):
+        def runner(self):
+            with open(inputf, 'rb') as ifd:
+                with codecs.open(outputf, 'rb', encoding='utf-8') as ofd:
+                    self.run_auto_test(ifd.read(), ofd.read())            
+        return runner
+
+
+def get_file_path(dir_name, file_name):
+    return realpath(join(dirname(__file__), 'files', dir_name, file_name))
+
+class TestCommand(Command):
+    user_options = []
+
+    def initialize_options(self):
+        self._dir = os.getcwd()
+
+    def finalize_options(self):
+        pass
+
+    def run(self):
+        '''
+        Finds all the tests modules in tests/, and runs them.
+        '''
+        testfiles = []
+        for t in glob(join(self._dir, 'tests', '*.py')):
+            module_name = splitext(basename(t))[0]
+            if module_name.startswith('test'):
+                testfiles.append('.'.join(['tests', module_name])
+                )
+
+        tests = TestLoader().loadTestsFromNames(testfiles)
+        t = TextTestRunner(verbosity=2)
+        t.run(tests)
+
-- 
2.20.1