minor fix

[librarian.git] / librarian / pdf.py

diff --git a/librarian/pdf.py b/librarian/pdf.py
index a868a2b..bb1ba64 100644 (file)
--- a/librarian/pdf.py
+++ b/librarian/pdf.py
@@ -19,14 +19,13 @@ from Texml.processor import process
 from lxml import etree
 from lxml.etree import XMLSyntaxError, XSLTApplyError
 
+from librarian.dcparser import Person
 from librarian.parser import WLDocument
-from librarian import ParseError
+from librarian import ParseError, DCNS
 from librarian import functions
 
 
-
 functions.reg_substitute_entities()
-functions.reg_person_name()
 functions.reg_strip()
 functions.reg_starts_white()
 functions.reg_ends_white()
@@ -45,27 +44,23 @@ def insert_tags(doc, split_re, tagname):
     <a><b>A<d/>B<d/>C</b>X<d/>Y<d/>Z</a>
     """
 
-    for elem in doc.iter():
-        try:
-            if elem.text:
-                chunks = split_re.split(elem.text)
-                while len(chunks) > 1:
-                    ins = etree.Element(tagname)
-                    ins.tail = chunks.pop()
-                    elem.insert(0, ins)
-                elem.text = chunks.pop(0)
-            if elem.tail:
-                chunks = split_re.split(elem.tail)
-                parent = elem.getparent()
-                ins_index = parent.index(elem) + 1
-                while len(chunks) > 1:
-                    ins = etree.Element(tagname)
-                    ins.tail = chunks.pop()
-                    parent.insert(ins_index, ins)
-                elem.tail = chunks.pop(0)
-        except TypeError, e:
-            # element with no children, like comment
-            pass
+    for elem in doc.iter(tag=etree.Element):
+        if elem.text:
+            chunks = split_re.split(elem.text)
+            while len(chunks) > 1:
+                ins = etree.Element(tagname)
+                ins.tail = chunks.pop()
+                elem.insert(0, ins)
+            elem.text = chunks.pop(0)
+        if elem.tail:
+            chunks = split_re.split(elem.tail)
+            parent = elem.getparent()
+            ins_index = parent.index(elem) + 1
+            while len(chunks) > 1:
+                ins = etree.Element(tagname)
+                ins.tail = chunks.pop()
+                parent.insert(ins_index, ins)
+            elem.tail = chunks.pop(0)
 
 
 def substitute_hyphens(doc):
@@ -84,7 +79,6 @@ def move_motifs_inside(doc):
     """ moves motifs to be into block elements """
     for master in doc.xpath('//powiesc|//opowiadanie|//liryka_l|//liryka_lp|//dramat_wierszowany_l|//dramat_wierszowany_lp|//dramat_wspolczesny'):
         for motif in master.xpath('motyw'):
-            print motif.text
             for sib in motif.itersiblings():
                 if sib.tag not in ('sekcja_swiatlo', 'sekcja_asterysk', 'separator_linia', 'begin', 'end', 'motyw', 'extra', 'uwaga'):
                     # motif shouldn't have a tail - it would be untagged text
@@ -126,6 +120,16 @@ def hack_motifs(doc):
                 break
 
 
+def parse_creator(doc):
+    """ find all dc:creator tags and add dc:creator_parsed with forenames first """
+    for creator in doc.findall('//'+DCNS('creator')):
+        p = Person.from_text(creator.text)
+        creator_parsed = deepcopy(creator)
+        creator_parsed.tag = DCNS('creator_parsed')
+        creator_parsed.text = ' '.join(p.first_names + (p.last_name,))
+        creator.getparent().insert(0, creator_parsed)
+
+
 def get_resource(path):
     return os.path.join(os.path.dirname(__file__), path)
 
@@ -185,6 +189,7 @@ def transform(provider, slug=None, file_path=None, output_file=None, output_dir=
         # hack the tree
         move_motifs_inside(document.edoc)
         hack_motifs(document.edoc)
+        parse_creator(document.edoc)
         substitute_hyphens(document.edoc)
         fix_hanging(document.edoc)