diff mbox series

[2/5] locale: Fix signed char bug in lr_getc

Message ID 619cade7e73dc33184bf4247b739d54cd9d7d8b3.1652994079.git.fweimer@redhat.com
State Committed
Commit 19d494445981a09503e4a0175732745c39dd7c21
Headers
Series Assume UTF-8 encoding for localedef input files |

Checks

Context Check Description
dj/TryBot-apply_patch success Patch applied to master at the time it was sent

Commit Message

Florian Weimer May 19, 2022, 9:06 p.m. UTC 
  The array lr->buf contains characters, which can be signed.  A 0xff
byte in the input could be incorrectly reported as EOF.  More
importantly, get_string in linereader.c converts a signed input byte
to a Unicode code point using ADDWC ((uint32_t) ch), under the
assumption that this decodes the ISO-8859-1 input encoding.  If char
is signed, this does not give the correct result.  This means that
ISO-8859-1 input files for localedef are not actually supported,
contrary to the comment in get_string.  This is a happy accident because
we can therefore change the file encoding to UTF-8 without impacting
backwards compatibility.

While at it, remove the \32 check for MS-DOS end-of-file character (^Z).
---
 locale/programs/linereader.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Comments

Carlos O'Donell July 4, 2022, 7:54 p.m. UTC | #1 
On 5/19/22 17:06, Florian Weimer via Libc-alpha wrote:
> The array lr->buf contains characters, which can be signed.  A 0xff
> byte in the input could be incorrectly reported as EOF.  More
> importantly, get_string in linereader.c converts a signed input byte
> to a Unicode code point using ADDWC ((uint32_t) ch), under the
> assumption that this decodes the ISO-8859-1 input encoding.  If char
> is signed, this does not give the correct result.  This means that
> ISO-8859-1 input files for localedef are not actually supported,
> contrary to the comment in get_string.  This is a happy accident because
> we can therefore change the file encoding to UTF-8 without impacting
> backwards compatibility.

LGTM.

Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Tested-by: Carlos O'Donell <carlos@redhat.com>

> 
> While at it, remove the \32 check for MS-DOS end-of-file character (^Z).

OK. We don't need this, files should have the correct EOF.

> ---
>  locale/programs/linereader.h | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/locale/programs/linereader.h b/locale/programs/linereader.h
> index 0fb10ec833..653a71d2d1 100644
> --- a/locale/programs/linereader.h
> +++ b/locale/programs/linereader.h
> @@ -134,7 +134,7 @@ lr_getc (struct linereader *lr)
>  	return EOF;
>      }
>  
> -  return lr->buf[lr->idx] == '\32' ? EOF : lr->buf[lr->idx++];
> +  return lr->buf[lr->idx++] & 0xff;

OK. Agreed, this should not be sign extended. It's a byte in the buffer not EOF.
With the original MS-DOS checking it might have been *needed* to return -1.

>  }
>  
>
diff mbox series

Patch

diff --git a/locale/programs/linereader.h b/locale/programs/linereader.h
index 0fb10ec833..653a71d2d1 100644
--- a/locale/programs/linereader.h
+++ b/locale/programs/linereader.h
@@ -134,7 +134,7 @@  lr_getc (struct linereader *lr)
 	return EOF;
     }
 
-  return lr->buf[lr->idx] == '\32' ? EOF : lr->buf[lr->idx++];
+  return lr->buf[lr->idx++] & 0xff;
 }