Better handling of multipart DAISY.

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

diff --git a/src/librarian/document.py b/src/librarian/document.py
index c0efd3a..1c8f223 100644 (file)
--- a/src/librarian/document.py
+++ b/src/librarian/document.py
@@ -1,20 +1,22 @@
 import gettext
 import os
 import re
+from urllib.request import urlopen
 from lxml import etree
-from .builders import builders
 from .parser import parser
-from . import dcparser
+from . import dcparser, DCNS
 from .functions import lang_code_3to2
 
 
 class WLDocument:
-    def __init__(self, tree=None, filename=None):
-        if filename is not None:
-            tree = etree.parse(filename, parser=parser)
+    def __init__(self, filename=None, url=None):
+        source = filename or urlopen(url)
+        tree = etree.parse(source, parser=parser)
         self.tree = tree
         tree.getroot().document = self
-        self.base_meta = dcparser.BookInfo({}, {}, validate_required=False)
+        self.base_meta = dcparser.BookInfo({}, {
+            DCNS('language'): ["pol"],
+        }, validate_required=False)
 
     @property
     def meta(self):
@@ -23,8 +25,8 @@ class WLDocument:
         return self.tree.getroot().meta
         return master.meta
 
-    def build(self, builder_id, **kwargs):
-        return builders[builder_id]().build(self, **kwargs)
+    def build(self, builder, **kwargs):
+        return builder().build(self, **kwargs)
 
     def _compat_assign_ordered_ids(self):
         """