Bug #253002 - Add a new variable.invalidchar check for metadata variables that

contain characters that are not part of the ASCII character set.

svn path=/main/trunk/; revision=12365

1 files changed, 14 insertions, 0 deletions

diff --git a/bin/repoman b/bin/repoman
index 441c7628..b2ae0a81 100755
--- a/bin/repoman
+++ b/bin/repoman
@@ -298,6 +298,7 @@ qahelp={
 	"ebuild.syntax":"Error generating cache entry for ebuild; typically caused by ebuild syntax error or digest verification failure",
 	"ebuild.output":"A simple sourcing of the ebuild produces output; this breaks ebuild policy.",
 	"ebuild.nesteddie":"Placing 'die' inside ( ) prints an error, but doesn't stop the ebuild.",
+	"variable.invalidchar":"A variable contains an invalid character that is not part of the ASCII character set",
 	"variable.readonly":"Assigning a readonly variable",
 	"LIVEVCS.stable":"This ebuild is a live checkout from a VCS but has stable keywords.",
 	"IUSE.invalid":"This ebuild has a variable in IUSE that is not in the use.desc or use.local.desc file",
@@ -358,6 +359,8 @@ qawarnings = set((
 "LIVEVCS.stable"
 ))
 
+non_ascii_re = re.compile(r'[^\x00-\x7f]')
+
 missingvars=["KEYWORDS","LICENSE","DESCRIPTION","HOMEPAGE","SLOT"]
 allvars = set(x for x in portage.auxdbkeys if not x.startswith("UNUSED_"))
 allvars.discard("CDEPEND")
@@ -1100,6 +1103,17 @@ for x in scanlist:
 		inherited = pkg.inherited
 		live_ebuild = live_eclasses.intersection(inherited)
 
+		for k, v in myaux.iteritems():
+			if not isinstance(v, basestring):
+				continue
+			m = non_ascii_re.search(v)
+			if m is not None:
+				stats["variable.invalidchar"] += 1
+				fails["variable.invalidchar"].append(
+					("%s: %s variable contains non-ASCII " + \
+					"character at position %s") % \
+					(relative_path, k, m.start() + 1))
+
 		if not src_uri_error:
 			# Check that URIs don't reference a server from thirdpartymirrors.
 			for uri in portage.flatten(portage.dep.use_reduce(