From: "Joseph S. Myers" <joseph@codesourcery.com>
To: Shujing Zhao <pearly.zhao@oracle.com>
Cc: GCC Patches <gcc-patches@gcc.gnu.org>,
Paolo Carlini <paolo.carlini@oracle.com>
Subject: Re: [PATCH PR/42686] Align the help text output
Date: Wed, 17 Mar 2010 12:50:00 -0000 [thread overview]
Message-ID: <Pine.LNX.4.64.1003171201450.2148@digraph.polyomino.org.uk> (raw)
In-Reply-To: <4BA07D89.3040406@oracle.com>
On Wed, 17 Mar 2010, Shujing Zhao wrote:
> > > + if (item_width == item_col_width
> > > + || item_width == 0)
> > > + printf (" %-*.*s", col_width, item_width, item);
> > > + /* Handle the item that includes wide character. */
> > > + else
> >
> > I think the code would be a lot clearer with fewer conditionals. Why do you
> > need to handle specially the case where item_width == item_col_width? Can't
> > you just use the logic below unconditionally?
> The conditional item_width == item_col_width is used to handle the ITEM
> without wide characters. As your suggestion, I simplify the logic as
But you should be able to treat all ITEMs as if they have wide characters;
there should be no need to do something special for the case of no wide
characters.
> > > + if (len != gcc_gettext_width (help))
> > > + len = get_aligned_len (help, room, remaining);
> >
> > Again, I'd like to avoid conditionals if possible.
> After the logic is moved to intl.c, the conditional is still kept to tell
> calling get_str_aligned_len or get_wcs_aligned_len.
>
> + if (gcc_gettext_width (msgstr) == remaining)
> + return get_str_aligned_len (msgstr, room, remaining);
> + else
> + return get_wcs_aligned_len (msgstr, room, remaining);
>
> Is it ok or any suggestion?
The check still isn't appropriate. The width might equal the number of
bytes even when wide characters are present. I think the logic should be:
if NLS enabled and the relevant wide character functions available, use
get_wcs_aligned_len, otherwise use get_str_aligned_len. This means only
one of those functions is used in any compiler configuration, so you can
just make them two different implementations of get_aligned_len depending
on the configuration, rather than having three function names involved.
--
Joseph S. Myers
joseph@codesourcery.com
next prev parent reply other threads:[~2010-03-17 12:05 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-02-25 10:32 Shujing Zhao
2010-02-25 16:34 ` Joseph S. Myers
2010-03-10 10:26 ` Shujing Zhao
2010-03-10 17:19 ` Joseph S. Myers
2010-03-12 11:34 ` Shujing Zhao
2010-03-12 17:09 ` Joseph S. Myers
2010-03-15 9:28 ` Shujing Zhao
2010-03-15 10:58 ` Shujing Zhao
2010-03-15 11:42 ` Joseph S. Myers
2010-03-17 7:57 ` Shujing Zhao
2010-03-17 12:50 ` Joseph S. Myers [this message]
2010-03-18 11:48 ` Shujing Zhao
2010-03-18 17:54 ` Joseph S. Myers
2010-03-19 11:43 ` Shujing Zhao
2010-03-19 13:44 ` Joseph S. Myers
2010-03-22 9:17 ` Shujing Zhao
2010-03-31 7:03 ` Shujing Zhao
2010-04-06 16:56 ` Joseph S. Myers
2010-04-13 11:32 ` Shujing Zhao
2010-04-15 21:54 ` Joseph S. Myers
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=Pine.LNX.4.64.1003171201450.2148@digraph.polyomino.org.uk \
--to=joseph@codesourcery.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=paolo.carlini@oracle.com \
--cc=pearly.zhao@oracle.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).