This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH] localedata: Make IBM273 compatible with ISO-8859-1 [BZ #23290]
- From: Carlos O'Donell <carlos at redhat dot com>
- To: Florian Weimer <fweimer at redhat dot com>, libc-alpha at sourceware dot org
- Date: Thu, 14 Jun 2018 15:37:21 -0400
- Subject: Re: [PATCH] localedata: Make IBM273 compatible with ISO-8859-1 [BZ #23290]
- References: <20180614191115.62A754016AD37@oldenburg.str.redhat.com>
On 06/14/2018 03:11 PM, Florian Weimer wrote:
> 2018-06-14 Florian Weimer <fweimer@redhat.com>
>
> [BZ #23290]
> * localedata/charmaps/IBM273: Map codepoint 0xbc to U+00AF, so
> that the result stays within the ISO-8859-1 range.
> * iconvdata/ibm273.c (HAS_HOLES): Define as 0 because all 256
> characters are defined in IBM273.
OK with s/OVERLINE/MACRON/g in the charmap.
I reviewed:
ftp://ftp.software.ibm.com/software/globalization/gcoc/attachments/CP00273.pdf
just to see if macron/overline was expected and it was. I expect what you see
there, being slightly narrower than fullwidth, is a macron.
My review shows that the *intent* of IBM273 was to stay within Latin1, and
using "OVERLINE" <U203E> goes against that intent when MACRON is fine.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
> diff --git a/iconvdata/ibm273.c b/iconvdata/ibm273.c
> index c303cf306d..132c06f0eb 100644
> --- a/iconvdata/ibm273.c
> +++ b/iconvdata/ibm273.c
> @@ -23,6 +23,6 @@
> #define TABLES <ibm273.h>
>
> #define CHARSET_NAME "IBM273//"
> -#define HAS_HOLES 1 /* Not all 256 character are defined. */
> +#define HAS_HOLES 0
>
> #include <8bit-gap.c>
> diff --git a/localedata/charmaps/IBM273 b/localedata/charmaps/IBM273
> index c3f70e2a6f..233d0db57d 100644
> --- a/localedata/charmaps/IBM273
> +++ b/localedata/charmaps/IBM273
> @@ -194,7 +194,7 @@ CHARMAP
> <U00BE> /xb9 VULGAR FRACTION THREE QUARTERS
> <U00AC> /xba NOT SIGN
> <U007C> /xbb VERTICAL LINE
> -<U203E> /xbc OVERLINE
> +<U00AF> /xbc OVERLINE
This is no longer OVERLINE but MACRON, which is in theory
slightly narrower, but still fits the description of the
codepage.
> <U00A8> /xbd DIAERESIS
> <U00B4> /xbe ACUTE ACCENT
> <U00D7> /xbf MULTIPLICATION SIGN
>
c.