Package search page
Hi, imho the package search page should be made more fault tolerant wrt/ search terms. Example: While searching a package x, if one enters x or x , as can happen when you type too fast, or when you have marked for copy and paste too lazily, there is an ugly error message generated, saying something about only one word being possible. Trimming leading and trailing white space seems appropriate. Thank you! Best, --Toni++
stattrans.pl
Hi!
I object to the last changes to the stattrans.pl script :) I am going
to submit the following diff for the file:
===
RCS file: /cvs/webwml/webwml/stattrans.pl,v
retrieving revision 1.29
diff -u -r1.29 stattrans.pl
--- stattrans.pl2001/09/12 19:14:15 1.29
+++ stattrans.pl2001/09/13 14:32:42
@@ -262,9 +262,9 @@
$wml{$lang} = $translated{$lang};
$translated{$lang} = $translated{$lang} - $outdated{$lang};
-$percent_a{$lang} = int ($wml{$lang}/$nfiles * 100);
-$percent_t{$lang} = int ($translated{$lang}/$nfiles * 100);
-$percent_o{$lang} = int ($outdated{$lang}/$nfiles * 100);
+$percent_a{$lang} = int ($wml{$lang}/$nfiles * 100 + .5);
+$percent_t{$lang} = int ($translated{$lang}/$nfiles * 100 + .5);
+$percent_o{$lang} = 100 - $percent_t{$lang};
$percent_u{$lang} = 100 - $percent_a{$lang};
if (open (HTML, $config{'htmldir'}/$l.html)) {
===
There is no need at all to calculate percent_o, it's just the inverse
to percent_t, like percent_u is for percent_a. If you don't believe
please think of it. $outdated == $nfiles - $translated, and therefore
it yields the same value. So if we don't calculate percent_u we should
stop calculating percent_o too, if it's for performance-reasons.
Also, if we are going to int() the value of the percentage we should
round it properly, therefore the .5 - which I earlier sugguested. If
the int is removed the .5 should be removed in this place too, but as
long as it's int() the .5 should stay there.
Thanks,
Alfie
--
You never learn anything | /\ ,'~~.
by doing it right. | / chaos \ alfie.ist.org |o ?~\
-- unknown | \inside!/ [EMAIL PROTECTED] /_ ~\
| \_/ \__,~ \
Re: stattrans.pl
On Thu, Sep 13, 2001 at 04:37:43PM +0200, Gerfried Fuchs wrote:
I object to the last changes to the stattrans.pl script :) I am going
to submit the following diff for the file:
===
RCS file: /cvs/webwml/webwml/stattrans.pl,v
retrieving revision 1.29
diff -u -r1.29 stattrans.pl
--- stattrans.pl2001/09/12 19:14:15 1.29
+++ stattrans.pl2001/09/13 14:32:42
@@ -262,9 +262,9 @@
$wml{$lang} = $translated{$lang};
$translated{$lang} = $translated{$lang} - $outdated{$lang};
-$percent_a{$lang} = int ($wml{$lang}/$nfiles * 100);
-$percent_t{$lang} = int ($translated{$lang}/$nfiles * 100);
-$percent_o{$lang} = int ($outdated{$lang}/$nfiles * 100);
+$percent_a{$lang} = int ($wml{$lang}/$nfiles * 100 + .5);
+$percent_t{$lang} = int ($translated{$lang}/$nfiles * 100 + .5);
+$percent_o{$lang} = 100 - $percent_t{$lang};
$percent_u{$lang} = 100 - $percent_a{$lang};
if (open (HTML, $config{'htmldir'}/$l.html)) {
===
There is no need at all to calculate percent_o, it's just the inverse
to percent_t, like percent_u is for percent_a. If you don't believe
please think of it. $outdated == $nfiles - $translated, and therefore
it yields the same value. So if we don't calculate percent_u we should
stop calculating percent_o too, if it's for performance-reasons.
This is true only if you consider all of the untranslated files to
be outdated translations. I originally made this exact mistake,
but corrected it after committing it. Please see the CVS log.
I've been rethinking this, though. It might be better to replace
it with the following:
$percent_o{$lang} = $percent_a{$lang} - $percent_t{$lang};
Also, if we are going to int() the value of the percentage we should
round it properly, therefore the .5 - which I earlier sugguested. If
the int is removed the .5 should be removed in this place too, but as
long as it's int() the .5 should stay there.
Heh, I'm a pessimist.
Matt
Re: stattrans.pl
On Thu, Sep 13, 2001, Matt Kraai wrote:
This is true only if you consider all of the untranslated files to
be outdated translations. I originally made this exact mistake,
but corrected it after committing it. Please see the CVS log.
I've been rethinking this, though. It might be better to replace
it with the following:
$percent_o{$lang} = $percent_a{$lang} - $percent_t{$lang};
*nirg* Thanks for fast response before any harm is done (I hope the
script is not running right now). I thought the up-to-date and outdated
values percentages are now calculated wrt/ the translated files and not
the full value? Wasn't that commited, too? I think that makes up a
much more valueable value than calculate them wrt/ all the files.
It was that I just saw it on the page that Matt did - Matt, didn't you
also submit that part? If not, why not? If yes, why doesn't it work?
:)
Heh, I'm a pessimist.
Alright, I'm usually more a realist, and rounding seems to be more to
the true values :) And, in the handling without rounding could have
been a mistake if all three values got rounded down, up-to-date and
outdated could make up one less in the sum than translated files (yes, I
could be quite pedantic with this, I know ;)
So long,
Alfie [sorry for the inconvenience]
--
You never learn anything | /\ ,'~~.
by doing it right. | / chaos \ alfie.ist.org |o ?~\
-- unknown | \inside!/ [EMAIL PROTECTED] /_ ~\
| \_/ \__,~ \
Re: stattrans.pl
On Thu, Sep 13, 2001 at 06:34:39PM +0200, Gerfried Fuchs wrote:
On Thu, Sep 13, 2001, Matt Kraai wrote:
This is true only if you consider all of the untranslated files to
be outdated translations. I originally made this exact mistake,
but corrected it after committing it. Please see the CVS log.
I've been rethinking this, though. It might be better to replace
it with the following:
$percent_o{$lang} = $percent_a{$lang} - $percent_t{$lang};
*nirg* Thanks for fast response before any harm is done (I hope the
script is not running right now). I thought the up-to-date and outdated
values percentages are now calculated wrt/ the translated files and not
the full value? Wasn't that commited, too? I think that makes up a
much more valueable value than calculate them wrt/ all the files.
It was that I just saw it on the page that Matt did - Matt, didn't you
also submit that part? If not, why not? If yes, why doesn't it work?
:)
I wasn't sure whether that part was OK, so I went ahead with my
other changes. If you'd like to fold that change in with your
solution below, I doubt anyone will have a problem.
Heh, I'm a pessimist.
Alright, I'm usually more a realist, and rounding seems to be more to
the true values :) And, in the handling without rounding could have
been a mistake if all three values got rounded down, up-to-date and
outdated could make up one less in the sum than translated files (yes, I
could be quite pedantic with this, I know ;)
Yeah, I realized this after committing. I think my proposed
change should take care of that problem, though. It sounds like
you have write permissions, so you're free to implement whatever
solution you prefer.
Matt
Bug#112175: www.debian.org: Please link to s390 buildd logs
Package: www.debian.org Version: N/A; reported 2001-09-13 Severity: normal http://www.debian.org/devel/buildds currently states: The build daemons for alpha, hppa, i386, ia64, mips, mipsel, and s390 do not have web interfaces. This is not (any longer) true in the case of s390: http://source.rfc822.org/pub/mirror/s390-ibm-linux/buildbot/s390-buildd/ -- - mdz
Bug#112175: marked as done (www.debian.org: Please link to s390 buildd logs)
Your message dated Thu, 13 Sep 2001 14:07:30 -0600 with message-id [EMAIL PROTECTED] and subject line Bug#112175: www.debian.org: Please link to s390 buildd logs has caused the attached Bug report to be marked as done. This means that you claim that the problem has been dealt with. If this is not the case it is now your responsibility to reopen the Bug report if necessary, and/or fix the problem forthwith. (NB: If you are a system administrator and have no idea what I am talking about this indicates a serious mail system misconfiguration somewhere. Please contact me immediately.) Darren Benham (administrator, Debian Bugs database) -- Received: (at submit) by bugs.debian.org; 13 Sep 2001 18:57:13 + From [EMAIL PROTECTED] Thu Sep 13 13:57:13 2001 Return-path: [EMAIL PROTECTED] Received: from smtp02.mrf.mail.rcn.net [207.172.4.61] by master.debian.org with esmtp (Exim 3.12 1 (Debian)) id 15hbfl-0008Bx-00; Thu, 13 Sep 2001 13:57:13 -0500 Received: from 146-115-127-203.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com ([146.115.127.203] helo=mizar.alcor.net) by smtp02.mrf.mail.rcn.net with esmtp (Exim 3.33 #10) id 15hbfk-0006uw-00 for [EMAIL PROTECTED]; Thu, 13 Sep 2001 14:57:12 -0400 Received: from mdz by mizar.alcor.net with local (Exim 3.32 #1 (Debian)) id 15hbfj-0005nu-00 for [EMAIL PROTECTED]; Thu, 13 Sep 2001 14:57:11 -0400 Date: Thu, 13 Sep 2001 14:57:11 -0400 From: Matt Zimmerman [EMAIL PROTECTED] To: Debian Bug Tracking System [EMAIL PROTECTED] Subject: www.debian.org: Please link to s390 buildd logs Message-ID: [EMAIL PROTECTED] Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.3.20i X-Reportbug-Version: 1.28 Sender: Matt Zimmerman [EMAIL PROTECTED] Delivered-To: [EMAIL PROTECTED] Package: www.debian.org Version: N/A; reported 2001-09-13 Severity: normal http://www.debian.org/devel/buildds currently states: The build daemons for alpha, hppa, i386, ia64, mips, mipsel, and s390 do not have web interfaces. This is not (any longer) true in the case of s390: http://source.rfc822.org/pub/mirror/s390-ibm-linux/buildbot/s390-buildd/ -- - mdz --- Received: (at 112175-done) by bugs.debian.org; 13 Sep 2001 20:07:31 + From [EMAIL PROTECTED] Thu Sep 13 15:07:31 2001 Return-path: [EMAIL PROTECTED] Received: from opensource.lineo.com [204.246.147.130] by master.debian.org with esmtp (Exim 3.12 1 (Debian)) id 15hcln-0006CE-00; Thu, 13 Sep 2001 15:07:31 -0500 Received: by opensource.lineo.com (Postfix, from userid 1011) id 247611935E3; Thu, 13 Sep 2001 14:07:30 -0600 (MDT) Date: Thu, 13 Sep 2001 14:07:30 -0600 From: Matt Kraai [EMAIL PROTECTED] To: [EMAIL PROTECTED] Subject: Re: Bug#112175: www.debian.org: Please link to s390 buildd logs Message-ID: [EMAIL PROTECTED] References: [EMAIL PROTECTED] Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: [EMAIL PROTECTED]; from [EMAIL PROTECTED] on Thu, Sep 13, 2001 at 02:57:11PM -0400 Delivered-To: [EMAIL PROTECTED] On Thu, Sep 13, 2001 at 02:57:11PM -0400, Matt Zimmerman wrote: The build daemons for alpha, hppa, i386, ia64, mips, mipsel, and s390 do not have web interfaces. This is not (any longer) true in the case of s390: http://source.rfc822.org/pub/mirror/s390-ibm-linux/buildbot/s390-buildd/ Thanks for the information. Please let us know of any other unlisted buildd web interfaces that you discover. Matt
