Re: debian.org web pages
On Fri, Jun 02, 2000 at 02:05:47AM -0500, w trillich wrote: > > A search engine is coming but it will require some upgrading on one of > > the Debian hosts (VA I think) that the Debian admins quite sensibly do not > > want to do during a freeze. > > > > It's on my list of things to do, and it will get done. > > maybe a volunteer could help? :) Nope, unless you've suddenly become a debian-admin person. You may not understand the significance of that. debian-admin people run the computers we work on. For example, I cannot install a database program on one of our computers so the search data has a home to live, but a Debian admin can. They've basically said "It would be better to put a more modern version of the program on than an old one, let's wait until the next dist is released". So in a strange way the search pages are dependent on the release manager, so we can nicely blame Dark for everything :) > > What can help is things like the FAQoMAtic, but they need to be kept > > current. Have you seen PHPs manual with the user supplied comments? > > That is quite useful too. > > > um, php's manual? No, check out the online on on http://www.php.net/ Go to documentation, then look for annotated manual. It is very good idea of how to harness your users thinking caps. > i'd like to carry the banner "make the newbie's life easier > [in order to make the gurus' lives easier]". of course, i'm > not that far up on the totem pole myself... not just yet. > thus, suggestions (and direction) are welcome. Yes, I totally agree that Debian, in general, and the website in particular is not exactly newbie-friendly. There is a problem in that for people who has used Debian for years everything seems obvious. Part of the problem is either it is hard to find the person responsible for a page, or poor overworked JT is responsible. We're busy enough doing the day to days on the site. That does not mean suggestions to help should be ignored or rejected. It just means the suggester has to be patient sometimes. > Their is five errers in this sentance. there are errors sentence Hmm, we have a logical loop with the five, interesting. There are infinite errors perhaps, dunno. - Craig -- Craig Small VK2XLZ GnuPG:1C1B D893 1418 2AF4 45EE 95CB C76C E5AC 12CA DFA5 Eye-Net Consulting http://www.eye-net.com.au/<[EMAIL PROTECTED]> MIEEE <[EMAIL PROTECTED]> Debian developer <[EMAIL PROTECTED]>
/MailingLists/ should have an index
Hi, I noticed a few weeks ago that the /MailingLists/ directory on the web site has no index.html, so it displays just the directory listing. The subscription page and the support.wml page contain the general information that could be put in there, so I'd like to put that stuff in /MailingLists/index.wml, and move over HOWTO_start_list from /devel to that directory. Any objections? However, I'm not sure how to change the subscription page to point to index.wml for a general description of what our mailing lists do. Obviously the mklist script generates it, but does it have to be run from master.d.o or something? I'm not sure where it gets its configuration information from. -- Digital Electronic Being Intended for Assassination and Nullification
Re: the mirror run time
On Thu, Jun 01, 2000 at 10:33:24PM -0400, James A. Treacy wrote: > > That timing has been changed to 17:30 on purpose, to be 13:52 (dinstall > > start) + the time it took for dinstall to finish (while it was still on > > master) + a little bit of overhead, just in case. Therefore it would be > > better to change the script to remove that sleeping time, the problem it > > addresses has been dealt with already. > > > > Also, there is another `sleep 3600' there, but IIRC it doesn't take one hour > > for the primary mirrors to rsync the daily changes, usually, so (if I'm > > right) that one should be changed, too. > > > Just got rid of the first sleep and changed the second set of mirrors > to start in 1800s = 30m. Good! Although, you should have kept the parenthesis if you want the combined output of those first three signals to go to /dev/null, currently it doesn't do that (it writes a \n in /dev/null :) BTW /debian2/web/update.mirrors file is owned by UID 1167, which doesn't exist, better chown that to yourself or some working account, I don't think the suid will work with that. -- Digital Electronic Being Intended for Assassination and Nullification
/partners/
Hi, whoever is maintaining the /partners/ web page, Novare is now called Brainfood. Mindspring doesn't host any of our sites anymore, ftp.debian.org moved to Gigabell AG. Above.net could be mentioned, too, since they provided archive.debian.org. -- Digital Electronic Being Intended for Assassination and Nullification
fix to search_contents.pl
The following patches allow for different archs to be searched from, and
allows paging of the output(ie, page 1, page 2, page 20).
There is still the issue where the Contents files on master(where
cgi.debian.org is run) are not being updated from auric. For that, it might
require someone with more perms than me.
I do have a pulse coming into master, to update
/org/ftp.debian.org/ftp/indices/, because the BTS needs an uptodate
Maintainers file.
=== cut
Index: english/distrib/packages.wml
===
RCS file: /cvs/webwml/webwml/english/distrib/packages.wml,v
retrieving revision 1.12
diff -u -r1.12 packages.wml
--- english/distrib/packages.wml2000/01/20 23:10:55 1.12
+++ english/distrib/packages.wml2000/06/04 03:23:21
@@ -67,8 +67,21 @@
keyword in any filename.
http://cgi.debian.org/cgi-bin/search_contents.pl";>
Search in:
-
stable
- unstable
+
+ stable
+ frozen
+ unstable
+
+
+
+ Alpha
+ Arm
+ Hurd(i386)
+ m68k
+ Intel
+ Powerpc
+ Sparc
+
case-insensitive
=== cut
--- /org/cgi.debian.org/cgi-bin/search_contents.pl Fri Jan 7 11:48:45 2000
+++ search_contents.pl Sat Jun 3 21:54:52 2000
@@ -5,15 +5,19 @@
require 5.001;
use strict;
use CGI;
+use POSIX;
MAIN:
{
my ($input, # The CGI data
- $keyword,$file,$version,@results,$case);
+
$keyword,$file,$version,@results,$case,$page,$arch,$start,$end,%line,$numpages,$index_line);
my ($command, $number);
- my $stable_x86 = '/debian2/debian/dists/stable/Contents-i386.gz';
- my $unstable_x86 = '/debian2/debian/dists/unstable/Contents-i386.gz';
+ my $ftp_base = '/org/ftp.debian.org/ftp';
+ my $def_arch = "i386";
+ my $header = "header goes here";
+ my $footer = "footer goes here";
+ my $results_per_page = 40;
$ENV{PATH} = "/bin:/usr/bin:/usr/bin/zgrep:/usr/sbin/zgrep";
# Read in all the variables set by the form
@@ -34,15 +38,18 @@
print "No version given for search.";
exit 0;
}
- $case = '';
- $case= $input->param('case');
-
- if ($version =~ /^stable$/) {
-$file = $stable_x86;
+ $arch = $input->param('arch');
+ if (! defined $arch) {
+ $arch = $def_arch;
+ print "No arch given for search, using default($def_arch).";
}
- else {
-$file = $unstable_x86;
+ $page = $input->param('page');
+ if (! defined $page) {
+ $page = 1;
}
+ $case = $input->param('case');
+
+ $file = "$ftp_base/dists/$version/Contents-$arch.gz";
$keyword =~ s,^/,,;
if ($keyword =~ /^([EMAIL PROTECTED]/.]+)$/) {
@@ -53,8 +60,6 @@
}
if (-f $file) {
- print "\n";
- print "FILE
PACKAGE\n";
if ($case =~ /^sensitive/) {
$command = "zgrep".$keyword." ".$file;
}
@@ -63,16 +68,33 @@
}
@results = qx( $command );
$number = 1;
+ $start = ($page - 1) * $results_per_page + 1;
+ $end = $page * $results_per_page + 1;
+
foreach (@results) {
$number++;
-print $_;
-if ($number > 40) {
- print "\nToo many results. Not printing any more.";
- exit 0;
-}
+if ($start <= $number && $number < $end) {
+ $line{$number - $start} = $_;
+ }
+ }
+ $numpages = floor($number / $results_per_page);
+ $index_line = "";
+ for(my $i = 1; $i <= $numpages; $i++) {
+ $index_line .= "$i";
+ if($i < $numpages) {
+ $index_line .= " | ";
+ }
+ }
+ print "$index_line\n";
+ print "\n";
+ print "FILE
PACKAGE\n";
+ for(my $i = 0; $i < $results_per_page; $i++) {
+ if(defined($line{$i})) {
+ print $line{$i} . "\n";
+ }
}
- # print @results;
print "\n";
+ print "$index_line\n";
}
else {
print "Error: No Contents file.\n";
=== cut
BEGIN GEEK CODE BLOCK
Version: 3.12
GCS d- s: a-- c+++ UL P+ L !E W+ M o+ K- W--- !O M- !V PS--
PE++ Y+ PGP++ t* 5++ X+ tv b+ D++ G e h*! !r z?
-END GEEK CODE BLOCK-
BEGIN PGP INFO
Adam Heath <[EMAIL PROTECTED]>Finger Print | KeyID
67 01 42 93 CA 37 FB 1E63 C9 80 1D 08 CF 84 0A | DE656B05 PGP
AD46 C888 F587 F8A3 A6DA 3261 8A2C 7DC2 8BD4 A489 | 8BD4A489 GPG
-END PGP INFO-
