[imp] improved i18n to scan all files

bzr revid: nicolas.vanhoren@openerp.com-20110817112828-tghtfms3zm3r0044

1 files changed, 36 insertions, 17 deletions

diff --git a/npybabel.py b/npybabel.py
index 3d9deaec676..0a3c75025c9 100755
--- a/npybabel.py
+++ b/npybabel.py
@@ -5,15 +5,30 @@ import sys
 from pkg_resources import load_entry_point
 import re
 import json
+import xml.etree.ElementTree as elt
 
 if __name__ == '__main__':
     sys.exit(
         load_entry_point('Babel==0.9.6', 'console_scripts', 'pybabel')()
     )
     
-QWEB_EXPR = re.compile(r"""(?:\< *t\-tr *\>(.*?)\< *\/t\-tr *\>)|(?:\_t *\( *((?:"(?:[^"\\]|\\.)*")|(?:'(?:[^'\\]|\\.)*')) *\))""")
-XML_GROUP = 1
-JS_GROUP = 2
+XMLJS_EXPR = re.compile(r"""(?:\_t *\( *((?:"(?:[^"\\]|\\.)*")|(?:'(?:[^'\\]|\\.)*')) *\))""")
+
+def extract_xmljs(fileobj, keywords, comment_tags, options):
+    content = fileobj.read()
+    found = XMLJS_EXPR.finditer(content)
+    result = []
+    index = 0
+    line_nbr = 0
+    for f in found:
+        mes = f.group(1)
+        mes = json.loads(mes)
+        while index < f.start():
+            if content[index] == "\n":
+                line_nbr += 1
+            index += 1
+        result.append((line_nbr, None, mes, ""))
+    return result
 
 def extract_qweb(fileobj, keywords, comment_tags, options):
     """Extract messages from XXX files.
@@ -28,19 +43,23 @@ def extract_qweb(fileobj, keywords, comment_tags, options):
              tuples
     :rtype: ``iterator``
     """
-    content = fileobj.read()
-    found = QWEB_EXPR.finditer(content)
     result = []
-    index = 0
-    line_nbr = 0
-    for f in found:
-        group = XML_GROUP if f.group(XML_GROUP) else JS_GROUP
-        mes = f.group(group)
-        if group == JS_GROUP:
-            mes = json.loads(mes)
-        while index < f.start():
-            if content[index] == "\n":
-                line_nbr += 1
-            index += 1
-        result.append((line_nbr, None, mes, ""))
+    def handle_text(str):
+        str = (str or "").strip()
+        if not str:
+            return
+        result.append((0, None, str, ""))
+    
+    def iter_elements(current_element):
+        for el in current_element:
+            if "t-js" not in el.attrib and \
+                    not ("t-jquery" in el.attrib and "t-operation" not in el.attrib) and \
+                    not ("t-trad" in el.attrib and el.attrib["t-trad"].strip() == "off"):
+                handle_text(el.text)
+                iter_elements(el)
+            handle_text(el.tail)
+    
+    tree = elt.parse(fileobj)
+    iter_elements(tree.getroot())
+
     return result