Implement layman-global.txt to repositories.xml translator

3 files changed, 96 insertions, 32 deletions

diff --git a/layman-parser.py b/layman-parser.py
index d20b20a..5a9d295 100755
--- a/layman-parser.py
+++ b/layman-parser.py
@@ -1,42 +1,91 @@
 #!/usr/bin/env python
 
 import xml.etree.ElementTree as ET
+import codecs
 
-def indent(elem, level=0):
-    i = "\n" + level*"  "
-    if len(elem):
-        if not elem.text or not elem.text.strip():
-            elem.text = i + "  "
-        if not elem.tail or not elem.tail.strip():
-            elem.tail = i
-        for elem in elem:
-            indent(elem, level+1)
-        if not elem.tail or not elem.tail.strip():
-            elem.tail = i
+def to_ascii(o, current_encoding='utf-8'):
+    if not isinstance(o, basestring):
+        return o
+
+    if isinstance(o, unicode):
+        s = o
     else:
-        if level and (not elem.tail or not elem.tail.strip()):
-            elem.tail = i
+        s = unicode(o, current_encoding)
+    return codecs.encode(s, 'ascii', 'ignore')
 
-def comp_elems (a, b):
-    return cmp(a.get('name'), b.get('name'))
+def recurse_print(e, level=0):
+	t = e.text
+	if t:
+		if t == '' or t.isspace():
+			t = None
+		else:
+			t = t.strip('\n\r')
+	print '%s<%s>%s' % ('  '*level, e.tag, t and ('  .. %s' % to_ascii(t)) or '')
+	for k, v in sorted(e.attrib.items()):
+		print '%s| %s  = %s' % ('  '*(level+2), k, v)
+	for c in e.getchildren():
+		recurse_print(c, level=level + 1)
 
 a = ET.parse(open('layman-global.txt'))
+overlays = a.getroot()
 
-x = a.getroot()
-
-x.getchildren().sort(cmp=comp_elems)
-
-#for c in x.getchildren():
-#	f = ET.SubElement(c, 'feed')
-#	f.text = c.get('src')
-
-indent(x)
-a.write('file')
+for overlay in overlays:
+	# Transform 'overlay' tag
+	overlay.tag = 'repo'
+	
+	# Transform 'link' tag
+	link = overlay.find('link')
+	if link != None:
+		link.tag = 'homepage'
+		
+	# Move 'description' tag
+	description = overlay.find('description')
+	overlay.remove(description)
+	overlay.insert(0, description)
 
-C = [t.get('name') for t in x.getchildren()]
-for n in set(C):
-    C.remove(n)
+	# Transform 'contact' attribute
+	maintainer = ET.Element('maintainer')
+	overlay.append(maintainer)
+	email = ET.Element('email')
+	email.text = overlay.attrib['contact']
+	del overlay.attrib['contact']
+	maintainer.append(email)
+	
+	# Transform 'src' and 'type' attribute
+	source = ET.Element('source')
+	source.text = overlay.attrib['src']
+	del overlay.attrib['src']
+	source.attrib['type'] = overlay.attrib['type']
+	del overlay.attrib['type']
+	overlay.append(source)
+	
+	# Extend by quality label
+	overlay.attrib['quality'] = 'experimental' # TODO
+	
+	# Extend by maintainer project flag
+	maintainer.attrib['project'] = 'no' # TODO
+	
+	# Extend by maintainer name
+	name = ET.Element('name')
+	name.text = 'TODO' # TODO
+	maintainer.append(name)
+	
+	# Extend by feed URI
+	feed = ET.Element('feed')
+	overlay.append(feed)
+	feed.text = 'http://www.example.org/TODO/' # TODO
+	
+	# Fix issues on the fly
+	if email.text.find('@') == -1 or \
+			email.text.find(' ') != -1:
+		email.text = 'todo@example.org' # TODO
+	
+# Transform 'overlays' tag
+overlays.tag = 'repositories'
 
-print 'dupes:' + str(C)
+# Extend by format version
+overlays.attrib['version'] = '1.0'
 
 
+recurse_print(overlays)
+a.write('repositories.xml')
diff --git a/overlay-rockage.txt b/overlay-rockage.txt
index eb0dbe7..a6e6b7c 100644
--- a/overlay-rockage.txt
+++ b/overlay-rockage.txt
@@ -19,15 +19,15 @@ Dead = Dumping ground for dead packages
 			status="official"
 			quality="experimental">
 		<description>Opal and other junk</description>
+		<homepage>asdsads</homepage>
 		<maintainer project="no">
 			(<name>Robert Buchholz</name>)
 			<email>rbu@gentoo.org</email>
 		</owner>
-		<feed>sdsds</feed>
-		<feed>sdsds</feed>
-		<homepage>asdsads</homepage>
 		<source type="rsync">git://sadasdsdsd</source>
 		<source type="git">git://sadasdsdsd</source>
+		<feed>sdsds</feed>
+		<feed>sdsds</feed>
 	<repo>
 </repositories>
 
diff --git a/validate.sh b/validate.sh
index c0eae3a..2b259f5 100755
--- a/validate.sh
+++ b/validate.sh
@@ -1,4 +1,5 @@
 #!/usr/bin/env bash
+echo '==================== layman-global.txt ===================='
 echo '== DTD =='
 xmllint --noout --dtdvalid layman-global.dtd layman-global.xml \
 	&& echo VALID \
@@ -10,3 +11,17 @@ xmllint --noout --relaxng layman-global.rng layman-global.xml \
 	&& echo VALID \
 	|| echo BROKEN
 echo
+
+echo
+echo '==================== repositories.xml ===================='
+echo '== DTD =='
+xmllint --noout --dtdvalid repositories.dtd repositories.xml \
+	&& echo VALID \
+	|| echo BROKEN
+echo
+
+echo '== Relax NG =='
+xmllint --noout --relaxng repositories.rng repositories.xml \
+	&& echo VALID \
+	|| echo BROKEN
+echo