Bug 39664

Summary:	Lots of error messages while building bonobo (Malformed UTF-8 character)
Product:	Gentoo Linux	Reporter:	Rui Malheiro <rmalheiro>
Component:	[OLD] GNOME	Assignee:	Gentoo Linux Gnome Desktop Team <gnome>
Status:	RESOLVED WORKSFORME
Severity:	normal
Priority:	High
Version:	unspecified
Hardware:	All
OS:	Linux
Whiteboard:
Package list:		Runtime testing required:	---
Attachments:	Hardcode LC_CTYPE=C so users with UTF-8 charset don't get error messages during build.

Description Rui Malheiro 2004-01-28 07:08:07 UTC

While emerging gnome-base/bonobo I got thousands of lines like these:

Malformed UTF-8 character (byte 0xfe) in pattern match (m//) at ../../xml-i18n-merge line 364, <PO_FILE> line 23699.
Malformed UTF-8 character (byte 0xfe) in pattern match (m//) at ../../xml-i18n-merge line 364, <PO_FILE> line 23699.
Malformed UTF-8 character (unexpected non-continuation byte 0x22, immediately after start byte 0xed) in pattern match (m//) at ../../xml-i18n-merge line 364, <PO_FILE> line 23699.
Malformed UTF-8 character (unexpected non-continuation byte 0x22, immediately after start byte 0xed) in pattern match (m//) at ../../xml-i18n-merge line 364, <PO_FILE> line 23699.

The emerge completes successefuly(?) inspite of the errors. Is this just cosmetic?

Reproducible: Always
Steps to Reproduce:
1. export LC_CTYPE=pt_PT.UTF-8
2. emerge bonobo

Actual Results:  
 

Expected Results:  
 

I'm using an UTF-8 locale (pt_PT.UTF-8).

Comment 1 Rui Malheiro 2004-01-28 07:14:55 UTC

Created attachment 24533 [details, diff]
Hardcode LC_CTYPE=C so users with UTF-8 charset don't get error messages during build.

After searching bugs.gentoo.org I found one similar bug (#22058) reported
against apache. Can we use the same approach here?

Comment 2 foser (RETIRED) gentoo-dev

2004-02-29 09:46:38 UTC

we could to stay on the safe side.

 afaik some gnome devs are on UTF-8 locale as well, can this problem be confirmed ?

Comment 3 foser (RETIRED) gentoo-dev

2004-08-18 09:03:07 UTC

no conformation, no further reports.

for now i'm gonna close this as an incident, since we're not officially utf-8 powered anyway