package w3m
tag 291735 patch
thanks
On Sat, Jan 22, 2005 at 09:59:27PM +0100, Samuel Thibault wrote:
> Package: w3m
> Version: 0.5.1-3
> Severity: normal
>
>
> Hi,
>
> Say I have a test.html.utf-8 page on some web server:
> <body>
> test
> </body>
>
> The web server properly announces that it is an utf-8 encoded page:
> Content-Type: text/html; charset=utf-8
>
> But w3m simplify this into US-ASCII, because the page indeed doesn't
> contain anything than can't be coded in plain ascii:
>
> [snipped]
>
> The problem comes if I put a form in my page. Since the page is
> announced as utf-8-encoded, w3m should default to using utf-8 to code
> the values. But since w3m simplifies charset into US-ASCII, it will
> default to that to code the values (and won't know how to code accents &
> co).
>
> W3m should *not* simplify charset.
>
Hi,
here is a patch so that w3m does not simplify the charset with
autodetect of charsets on. Another possibility is to set the option
"Automatic charset detect when loading" to OFF.
Regards,
--
Karsten Schölzel | Email: [EMAIL PROTECTED]
Väderleden 9 4:98 | Jabber: [EMAIL PROTECTED]
97633 Luleå | VoIP: sip:[EMAIL PROTECTED]
Sweden | sip:[EMAIL PROTECTED]
| Tel: +4918015855857712
| Mobile: +46706725974
Use the hint instead of US_ASCII in wc_auto_detect.
Fixes Debian bug #291735: w3m shouldn't "simplify" page's charset
---
commit 5ab3cec76b0514cc1cb333889ba34de5f82800c7
tree 249e2fcf17a83a378caeb7f829afe4e592723ccf
parent a3449ff39ec4a3cda629873f4e2fc37b026a9327
author Karsten Schoelzel <[EMAIL PROTECTED](none)> Sun, 06 Nov 2005 00:25:35
+0100
committer Karsten Schoelzel <[EMAIL PROTECTED](none)> Sun, 06 Nov 2005 00:25:35
+0100
libwc/detect.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/libwc/detect.c b/libwc/detect.c
--- a/libwc/detect.c
+++ b/libwc/detect.c
@@ -99,7 +99,7 @@ wc_auto_detect(char *is, size_t len, wc_
for (; p < ep && ! WC_DETECT_MAP[*p]; p++)
;
if (p == ep)
- return WC_CES_US_ASCII;
+ return hint;
switch (hint) {
case WC_CES_ISO_2022_JP:
