Re: [SLUG] Convert AppleDouble encoded Macintosh fonts to truetype?
On 5/11/06, Erik de Castro Lopo [EMAIL PROTECTED] wrote: Erik de Castro Lopo wrote: Does anyone know how to convert AppleDouble encoded Macintosh fonts to truetype?Well, using the t1unmac program from the t1utils package I'vemanaged to convert them to Postscript Type 1. I still need to convert them to Truetype or OpenType. fontforge? -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Decent widescreen LCD display?
On 5/9/06, Michael Kedzierski [EMAIL PROTECTED] wrote: I have the same 24 panel and it's utterly brilliant, and good valuefor money. The brightness and the contrast are amazing. I highlyrecommend it.The panel itself may be a samsung one, or an LG - which is the same as in the apple cinema displays afaik (but with a better backlight). My vote is also for the Dell 24. Or better yet, two of them ;) % xdpyinfo | grep dimensions dimensions: 3840x1200 pixels (1060x331 millimeters) Hey thats more than a meter across :) -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] mod_perl question
At Thu, 19 Jan 2006 14:48:00 +1100, Dean Hamstead wrote: in using Crypt::RandPasswd (::word() specifically), it doesnt seem at all mod_perl safe. there definately seems to be something that loops of and consumes all the web servers cpu. From a quick glance through the Crypt::RandPasswd code, I don't see anything that leaps out as being mod_perl unsafe. To get an idea of where a mod_perl script is spinning, try the process described here: http://perl.apache.org/docs/1.0/guide/debug.html#Using_the_Perl_Trace If you can't work it out from that, post the resulting stacktrace here and we can all have a bash at it. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Timezone/Daylight savings query.
At Wed, 04 Jan 2006 10:02:26 +1030, Glen Turner wrote: Also note that there was a leap second this year. So your NTP-based time will be 1s wrong if you have a tzdata earlier than 2005k. From what I understand of things, Unix timezone data has nothing to do with leap seconds. NTP, on the other hand, does - so any NTP-based system will correctly adjust for leap seconds (effectively shifting the epoch by 1 second each time). -- - Gus.email.backlog -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Help Me - C codes
At Mon, 28 Nov 2005 05:53:24 +1100, Tess Snider wrote:
On 11/27/05, Crossfire [EMAIL PROTECTED] wrote:
This is a case of recursion.
What's totally crazy is that once you've been programming a while, and
really understand this recursion stuff well, you have to then learn to
stop using it. It's very sad, because recursion is good stuff, but
the trouble is that, in C, arbitrarily large recursions make your
stack the size of Godzilla.
I realise I'm a bit behind the times with continuing this thread, but
I'm surprised no-one mentioned that modern compilers are quite capable
of turning tail-recursion into in-place iteration.
As a silly example, just to make it obvious:
int recurse_forever(int n) {
if (++n % 1 == 0) printf(n is %d\n, n);
return recurse_forever(n); /* n will wrap occasionally */
}
int main(int argc, char **argv) {
recurse_forever(0);
return 0;
}
Compile and run it without optimisation and it grows in memory and
segfaults. Compile it with -O and it will happily run forever without
growing in memory. Compare the assembly output from each and the
change is obvious - the recursive call is changed to a jmp since none
of the original state is actually needed after the recursed call
returns.
Scheme interpreters/compilers, as another example, *require* this
ability since tail-recursion is the only method of doing loops in
scheme.
--
- Gus
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] latex question
At Tue, 25 Oct 2005 12:37:07 +1000, Taryn East wrote:
\begin{tabular}{l|r}
\makeleftpage \makerightpage \\
\end{tabular}
You could also use something explicit (and simpler?) like this:
\makeleftpage \hspace{3mm} \vrule \hspace{3mm} \makerightpage
--
- Gus
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Linux friendly flash mp3 players
At Thu, 20 Oct 2005 11:23:06 +1000, David Gillies wrote: Does anyone out there have any recommendations for Linux friendly flash mp3 players? A few of the guys at work have these: http://eng.iaudio.com/product/product_X5_feature.php MP3, OGG, FLAC, WMA, WAV player; FM radio; voice, radio or line-in recorder; MPEG4 (video) player (although its a pretty small screen); text/image viewer; USB *host* support (so it can pull photos, etc directly from cameras without an intermediate computer) This particular one is disk-based (20/30 GB) - they have flash versions that are smaller (in both size and storage). From the Linux side, it just looks like a mass storage device that you copy files onto (no ipod-style index rebuild required). (The web site specifically mentions Linux and MAC support) -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Your top-ten linux desktop apps
At Fri, 7 Oct 2005 19:33:29 +1000, Peter Chubb wrote: Aren't there any emacs users on this list? my top ten are: Xemacs (editing) Xemacs (mail reading/writing) Xemacs (web browsing) Xemacs (compiling, with make, distcc and ccache underneath) Xemacs (remote editing, with tramp) Xemacs (teminal window for other command line apps) Xemacs (games!) Xemacs (picture previewing) Xemacs (session multiplexing) Oh, and Xemacs's web browser sucks, so Mozilla-firefox for that... Amen! .. and IRC client, scientific calculator, dictionary client, address book, wiki, appointment reminder, daily planner, music player, LDAP gui, etc Oh, and it all works in either text or graphical interfaces and most of it runs just fine on win32 or mac OSX. -- - Gus M-x all-hail-xemacs -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Your top-ten linux desktop apps
At Sun, 9 Oct 2005 16:46:46 +1000, Alan L Tyree wrote: I've never tried Xemacs - are there any traps for young players when installing both? Not really. Default values for some options are different between the two, as are some elisp package versions and unusual keybindings (M-g is one that springs to mind). The way you enable/disable some features (like global font-locking) is a little different, so a complex .emacs probably won't work without modification. Faces (fonts, etc) have a totally different implementation, which is mostly hidden by the elisp functions. XEmacs defaults to using zmacs regions, where the region is only active until the next command then it goes away (C-x C-x will get it back though). The XEmacs `vc-mode' is a fork of a much older version and can't have new version control tools (like svn or tla) added in at run time. XEmacs comes with gnuserv/gnuclient so you can attach to running xemacs instances. I believe this is only available as an addon to GNU emacs, so it may change the way you use (x)emacs a bit. Apparently GNU emacs doesn't have the ability to know when one of its buffers is obscured - the XEmacs `pop-to-buffer' is significantly better, which has a subtle affect all over emacs. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] slow default route and website
At Sun, 09 Oct 2005 08:58:41 +1000, Ben Donohue wrote: just for the slug archives i found the problem here. in the apache config file with older versions of apache, for the virtual servers I would have per directory options of allow from all deny from none however the deny from none in apache 2.x seems to cause an exactly 15 second delay for every different part of the page. don't know why this is happening though... Afaik, there is no such thing as deny from none. Your 15 second delay is apache trying to resolve the hostname none so that it can deny accesses from it.. This will allow access to everyone: Order deny,allow Allow from all See http://httpd.apache.org/docs/2.0/mod/mod_access.html -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] C-Pointers and Perl ?
At Fri, 30 Sep 2005 07:58:22 +1000, Telford Tendys wrote: How about writing a network protocol stack. You get a packet and all you know about it is that here is a block of memory. You then have to figure out what sort of packet it is, how long it is and what structure to give it. C handles this very nicely with pointers to structures that can be cast into whatever you need. Since you dared me, here's some (untested) perl5 code that will parse a TCP header, including network byte order conversions and bitfields: my ($source, $dest, $seq, $ack_seq, $off, undef, undef, $urg, $ack, $psh, $rst, $syn, $fin, $window, $check, $urg_ptr) = unpack 'n2 N2 C B8 n3', $tcpdata; $off = 0x0F; (this is fun ;) -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] C-Pointers and Perl ?
At Thu, 29 Sep 2005 18:38:31 +1000, Oscar Plameras wrote:
Is their equivalent codes for ff in perl 6 ?
Sure, perl6 (just as in perl5) has coderefs. In fact, these can be
references to anonymous functions or dynamically created closures,
which certainly can't be done in C.
[EMAIL PROTECTED] oscarp]# cat f.c
int main(void)
{
int (*get_f())();
static int arr[] = { -1, 0, 1, 0, 2, +9, +3200, -3500 };
int *ptr, *pastend;
int (*fptr)();
pastend = arr + sizeof(arr)/sizeof(arr[0]);
for (ptr = arr; ptr pastend; ptr++)
if (fptr = get_f(ptr))
fptr(ptr);
return 0;
}
int (*get_f(ptr))()
int *ptr;
{
int is_neg(), is_pos();
static int (*cmds[])() = {
(int (*)())0,
is_neg,
is_pos
};
int index = 0;
if (*ptr 0)
index = 1;
if (*ptr 0)
index = 2;
return(cmds[index]);
}
is_neg (iptr)
int *iptr;
{
printf(%d is negative\n, *iptr);
}
is_pos (iptr)
int *iptr;
{
printf(%d is positive\n, *iptr);
}
Deliberately reproducing the structure of the C program:
#!/usr/bin/pugs
use v6;
sub get_f (Int $i) {
our @cmd is private //=
(undef, { say $^a is negative }, { say $^a is positive });
my $index = 0;
if( $i 0 ) { $index = 1 }
elsif ( $i 0 ) { $index = 2 }
return @cmd[$index];
}
my int @arr = (-1, 0, 1, 0, 2, 9, 3200, -3500);
for @arr {
my $fptr = get_f($_);
$fptr($_) if $fptr;
}
But you are choosing awkward examples around trivial numerical
operations. Basic numeric operations and branching are things C can
do quite easily (provided the numbers fit within C's types). Try
common coding needs that C can't handle easily, like string
manipulation or memory management during error recovery. Oscar, I
suggest you actually use a high level language for a while and then
try going back to C. Doing anything in C (or even java for that
matter) is just so much tedious typing.
--
- Gus
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Your top-ten linux desktop apps
At Tue, 27 Sep 2005 14:12:54 +1000, Erik de Castro Lopo wrote:
let integer_array = [| 1 ; -2 ; 3 ; -4 ; 5 ; -6 ;
-7 ; 8 ; -9 ; 32727000 |] ;;
Array.mapi (fun i x
- Printf.printf integer_array[%d] = %d\n i x
) integer_array ;;
Hey, my first actual perl6 program:
#!/usr/bin/pugs
use v6;
my @integer_array = 1 -2 3 -4 5 -6 -7 8 -9 32727000;
for 0 .. @integer_array - 1 {
say integer_array[$_] = @integer_array[$_];
}
--
- Gus
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Your top-ten linux desktop apps
At Thu, 29 Sep 2005 17:17:21 +1000, Erik de Castro Lopo wrote:
for 1 .. @integer_array {
say integer_array[$_] = @integer_array[$_];
}
Yeah sorry. Did I mention it was my first ever perl6 program?
Try this version, note the iterator, the typed array (compile-time
checked/optimised) and the bigint.
#!/usr/bin/pugs
use v6;
my int @integer_array = (1, -2, 3, -4, 5, -6, -7, 8, -9,
32727000, 9876543210);
for @integer_array.kv - ($i, $value) {
say integer_array[$i] = $value;
}
--
- Gus
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: dynamic vs static type checking (was Re: [SLUG] Your top-ten linux desktop apps)
At Tue, 27 Sep 2005 12:00:09 +1000, Bruce Badger wrote: On 9/27/05, Erik de Castro Lopo [EMAIL PROTECTED] wrote: There are large classes of problems where running speed is an important issue. Static typing does make for faster run times and in cases where that moves your program from being too slow to being fast enough, that is not a premature optimisation. Modern VMs (e.g. many of the Smalltalk VMs) dynamically compile code, i.e. they JIT. [...] In fact, the very best of the JITing VMs can get performance that exceeds that attainable by static compilation - because there is more information available at run time to base the tuning decisions upon. If a program's use changes over its invocation, and the JIT continually shifts its optimisation targets, then I can see the potential benefit of this approach. I don't believe, however, that there are many programs that have this dynamic behaviour. You can gain the same runtime knowledge in a statically compiled C program by compiling with gcc's -ffprofile-arcs, running over some typical use cases (will write a bunch of .gcda files) and then recompiling with -fbranch-probabilities. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] FW: Script help
At Thu, 22 Sep 2005 17:01:24 +1000, Vino Fernando Crescini wrote:
If you want a list of all invalid usernames in smbpasswd that is
not in allusers.txt, this way might be easier:
cut -d: -f1 smbpasswd | while read name; do
grep -q ${name}\$ /tmp/list || echo $name
done
oops. that's allusers.txt instead of /tmp/list. the output of
this can then be fed to smbpasswd -x
Since apparently I've been getting some experience with large datasets
lately, this version scales better:
sort allusers.txt sortedusers.txt
sort /etc/samba/smbpasswd | join -v1 -t: -o1.1 - sortedusers.txt | \
xargs -n 1 smbpasswd -x
--
- Gus
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Asterisk Open Source PABX software
At Wed, 24 Aug 2005 15:16:04 +1000, Michael Kraus wrote: Has anyone had any experience with Asterisk (or any other) open source PABX software, and wouldn't mind commenting? My comments: Asterisk developers prioritises new features over stability and particularly clean code. There's a *whole* lot of cruft in the Asterisk code base - lots of features are hacked directly into the channel drivers (particularly the zaptel one), so don't expect new feature combinations to work without testing it. The Asterisk community is by and large made up of people who can't code and don't really know what they're doing. Because of this there is a huge amount of disinformation and rumours about certain features/bugs. Basically treat anything you read with scepticism and learn enough about things so you can work through them yourself. Asterisk is a nice toolkit with some good ideas (and a few bad ones). If you want to use it in real-life situations, I recommend reading the code to see what it actually does and reaching a point where you feel comfortable doing trivial changes to some of the hard-coded (typically American) defaults. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] ubuntu mod perl2
At Sat, 27 Aug 2005 13:30:42 +1000, Ashley Maher wrote: I loaded the mod perl2 package into Ubuntu. The registery scripts worked well. The handler modules test failed misserably. mod_perl/1.99_14 Perl/v5.8.4 PHP/4.3.10-10ubuntu4 configured -- resuming normal operations [...] [Sat Aug 27 12:39:46 2005] [error] [client 127.0.0.1] Can't locate Apache2/RequestRec.pm in @INC (@INC contains: /usr/lib/perl5/Apache2 /var/www /etc/perl /usr/local/lib/perl/5.8.4 /usr/local/share/perl/5.8.4 /usr/lib/perl5 /usr/share/perl5 /usr/lib/perl/5.8 /usr/share/perl/5.8 /usr/local/lib/site_perl . /etc/apache2/ /etc/apache2/lib/perl) at /var/www/MyApache2/Rocks.pm line 8.\nBEGIN failed--compilation aborted at /var/www/MyApache2/Rocks.pm line 8.\nCompilation failed in require at (eval 4) line 3.\n mod_perl2 changed the names of the critical libraries in one of the late prereleases (in order to avoid compatibility problems with mod_perl1) You are using a pre-API-change mod_perl2 with code written from post-API-change documentation. Basically the pre-API-change stuff was to use Apache2 somewhere and then use Apache::RequestRec, etc (ie without the 2 on the prefix). I don't have a box handy with pre-API-change mod_perl2 handy, but you should be able to work things out by looking at what files you actually have below /usr/lib/perl5/Apache2/. Be aware that any code you right to this API will not work against the released version of mod_perl2 (it was released too late into the sarge release process to make it into that release unfortunately). In most cases the changes are a simple search+replace however. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Patents and OSS Development
At Tue, 09 Aug 2005 22:26:47 +1000, James Purser wrote: Does anybody know of, or are involved in projects that have been hampered by software patents. zsync is a kind of reverse-rsync implementation, which makes the whole rsync public-server thing actually possible since the server no longer needs to recalculate checksums. Potentially very useful for things like Debian mirror servers where small changes to package index files need to be updated regularly. AFAIK, the idea is basically unused due to patent fears (probably the same ones that killed the rsync-over-HTTP rproxy effort). (you should probably confirm the details before committing to anything though..) -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Postfix, IMAP, Lotus Notes, Evolution OfflineIMAP
At Sun, 7 Aug 2005 19:28:57 +1000, steven wrote: 2. With the Lotus Notes client running in IMAP mode I can create emails offline. When I sync the client with the server the server will send out any unsent emails. Copies of sent mails are thus filed in the main server sent folder or as directed by the user. [...] is there an OSS client that would allow me to do this? My MUA, wanderlust can do this. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] LDAP config help sought (long post)
At Thu, 14 Jul 2005 12:28:56 +1000, Howard Lowndes wrote: I'm trying to get somewhere with setting up an LDAP database. The problem I am encountering is that all the examples that I can find assume the the top level has a dn: of the form dc=example,dc=com My problem is that I want a different top level, and I want my slapd.conf file to look like [o=myhosting example] The DN is supposed to be *globally* unique. Hence everyone using some sort of DNS name as the root of the data stored in their LDAP server. (The older convention was o=Company Name,c=AU - but DNS names are better in an internet age) I really think you should rethink your base DN. Note that the DN has very little to do with the actual data stored in that entry (or below that point in the tree) - it just has to be unique. Thus far my layout works just fine, but when I come to add dcObjects I start to run into problems. An LDIF of: dn: dc=example,dc=com,o=myhosting objectClass: dcObject dc: example returns the following error: # ldapadd -x -D 'cn=manager,o=myhosting' -W -f myhosting.ldif adding new entry dc=example,dc=com,o=myhosting ldap_add: Object class violation (65) additional info: no structural object class provided Thats because your objectclasses (only dcObject) are only auxiliary classes that augment other structural objectclasses. You need at least one structural objectclass for each LDAP entry. You might want to consider using objectClass domain for these entries (RFC1274/cosine). If I then expand this LDIF file to: dn: dc=example,dc=com,o=myhosting objectClass: dcObject dc: example objectClass: organizationalUnit ou: My Hosting I now get this error: # ldapadd -x -D 'cn=manager,o=myhosting' -W -f myhosting.ldif adding new entry dc=example,dc=com,o=myhosting ldap_add: No such object (32) matched DN: o=myhosting The question at this point is: What am I doing wrong here? I think that error means you have to create dc=com,o=myhosting first. Now, let us consider setting up my database more like the examples. The questions here are: 1: Can I have more than 1 database bdb entry, say 1 for each of several disparate domains? Yes, if you want. There isn't much reason to actually store the separate domains in separate files however. I wouldn't do it unless I had some technical reason to do so (large databases that want different indexing or different backend implementations, for example). 2. If I can have more than 1 database bdb entry, can they all co-exist in the same directory path /var/lib/ldap or do they need separate directory paths /var/lib/ldap/firstdomain, /var/lib/seconddomain, etc? It looks like it has to be separate directories. 3. Do the rootdn's have to match each dc= for its suffix or can it be quite different, and can I have a common rootdn for all domains? You may specify a rootdn for each database. According to slapd.conf(5), if the rootdn is outside the suffix (namingContext), then the rootpw directive won't work. The rootdn in OpenLDAP is like an LDAP superuser account. The most secure way to use openldap is to never use the superuser account. You can use openldap without specifying rootdn at all and enjoy finer grained access controls. 4. I understand that multiple database bdb entries are permissible and apparently multiple suffix entries are permissible. Are multiple/multiples permissible? yes (assuming I understand your question). 5. If multiple suffixes are permissible under any database bdb entry, then how is the following considered: suffix dc=example,dc=com suffix dc=sitea suffix dc=siteb Is the third line a subset of the second or of the first? Neither is a subset of another in this example. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Looking for lazy way out
At Wed, 22 Jun 2005 16:33:53 +1000, Simon Bryan wrote:
I have uploaded a web based CD to our Moodle setup, but all the links
are broken. In true sloppy MS style most of the filenames are in
uppercase whereas the html files refer to them in lower case. The
underlying webserver is Apache, (thought mod-speling might help from
some reading but it is already installed). I am looking for the lazy way
out of going through and editing all the links in the html files or
renaming all the disc files, is there a solution? NB there are hundreds
of files and corresponding links :-(
Totally untested (other than checking it compiles).
Install and load mod_perl, then put this file in
/usr/share/perl5/Apache/IgnoreCase.pm (or wherever other Apache/foo.pm
modules are on your distro), then put something like this in your
httpd.conf (a Location block would be fine too):
Directory /path/to/my/web/based/cd
PerlFixupHandler Apache::IgnoreCase
/Directory
(How do you people use Apache without mod_perl?)
package Apache::IgnoreCase;
use Apache::Constants ':common';
use strict;
use warnings;
sub find_path {
my $path = $_[0];
my $new = '';
COMPONENT:
for my $component (split '/', $path) {
next if $component eq '';
my $tmp = $new/$component;
if (-e $tmp) {
$new = $tmp;
} else {
my ($dir, $entry);
opendir $dir, $new or return undef;
while (defined($entry = readdir $dir)) {
if (lc($entry) eq lc($component)) {
$new .= /$entry;
next COMPONENT;
}
}
return undef; # similar entry not found
}
}
$new = '/' if $new eq ''; # corner case
return $new;
}
sub handler {
my $r = $_[0];
my $fn = $r-filename;
unless (-e $fn) {
my $new = find_path($fn);
$r-filename($new) if $new;
}
return OK;
}
1;
--
- Gus
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] DirecPC under Linux (Telstra Sattelite)
At Wed, 22 Jun 2005 01:16:25 +0930, Ryan Verner wrote: Anybody had any experiences setting up 1-way DirecPC (Telstra Bigpond) satellite with an uplink through an NT1+II USB ISDN modem on Linux? err, yes. Its an extremely common setup for one of the boxes from my company. The DirecPC USB modem is a Hughes Network Systems (HNS) Sattelite Device, model ISU-R1. I seem to remember some commercial software a few years ago to achieve this, but I can't seem to find anything (at least, even remotely recent). As I recall it, there used to be some limited linux software from hughes themselves and even more limited free drivers elsewhere. To the point where we decided to develop our own drivers from scratch (almost entirely written by Herbert Xu). Now that the DW6000s are everywhere (ethernet based, no drivers needed) and the drivers are no longer a significant competitive advantage for us, Ursys could probably think about releasing the code. The main problem is that since the 4.2 release of the satellite system, a significant amount of user-level code is also required for various tunnels, key management, multicast support, commissioning, some proprietary PEP protocol used by most TCP connections, etc, etc. Unfortunately, this user-level code is tightly tied to our other pieces of software and wouldn't be particularly useful once cut out. It would be a significant undertaking for someone to turn it all back into working code again. Given the near-end-of-life of the DW3000/DW4000 range, I don't think anyone would (or should) expend that sort of effort now. If I were you, I'd seriously consider just buying a router from us (http://www.urnet.com.au/) - they're a little expensive for private use, but all the hard work is done and if you want to do VPNs over satellite then we have some features that would definately interest you. Failing that, your only real choices are: - get a two-way service and DW6000 from telstra - but you won't be able to do one-way stuff with it without layering some tunnel of your own on top, or policy routing certain traffic flows to the ISDN. - run the windows drivers and try to get a windows box to be your gateway/firewall. I think I even heard of someone who got that to work once. - beat on the sourceforge drivers and see if they support enough features to be useful for you. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Re: Execution via email ?
At Tue, 21 Jun 2005 10:23:03 +1000, Matthew Palmer wrote: For security, you can GPG-sign command messages, and then the script can just verify the signature before executing anything. ~% apt-cache show grunt Package: grunt Priority: optional Section: utils Installed-Size: 36 Maintainer: John Goerzen [EMAIL PROTECTED] Architecture: all Version: 0.5.0-4 Depends: python (= 2.3), python-gnupginterface Suggests: uucp, procmail | mailfilter Filename: pool/main/g/grunt/grunt_0.5.0-4_all.deb Size: 8784 MD5sum: a7383a25df3d75ec7f1eb065651c8323 Description: Secure remote execution via UUCP or e-mail using GPG GRUNT is a tool to let you execute commands remotely, offline. It will also let you copy files to a remote machine. . To do that, it will use GPG to digitally sign data that you send to your account on a remote machine. If the signature is valid and authorized, the remote will carry out the requested action or save the included file. Thus, it is similar to ssh, but for offline use. . Commands run remotely can have their stdin supplied, and when executed, the stdin will be sent to them -- just like uucp can. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Telephone recording?
At Fri, 27 May 2005 13:04:21 +1000, Jeff Waugh wrote: What's the most sensible and reliable way to record a phone conversation, assuming a standard phone, using Free Software (and probably a bit of hardware)? You can get phone audio into a computer by using one of those old voice-modems with the right AT commands (often not full-duplex, but that isn't a problem here), or a telephony card with an FXO port. Whether something like asterisk will make the software side easier or not, I'll leave as an exercise for the reader -- but even using asterisk for this step wouldn't require anyone to use VoIP, assuming you had a PSTN double-adaptor thingy at your end. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] recursive tree log grep ?
At Tue, 24 May 2005 09:36:56 +0800 (WST), jam wrote: I love vi, but do not use the vi-command-edit option of bash. My mate who does asked me how to do this with the standard (emacs) shell edit functions: /someword # look for a history event starting 'someword' up # previous history event starting 'someword' cr # execute THAT command The bash/readline function you're looking for is history-search-backward, which is not bound to any keys by default. I'd bind it to M-p (if I was using bash) to make it similar to the old tcsh behaviour (and its better than the default M-p function - non-incremental-reverse-search-history) Put this in ~/.inputrc to have it work for all readline-using programs: \M-p: history-search-backward or run this from your .bashrc if you only want to use it in bash: bind '\M-p: history-search-backward' For reference, zsh users would put this in .zshrc to get the same thing: bindkey -m # get a meta key bindkey '\M-p' history-beginning-search-backward -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Videos in presentation software
At Thu, 05 May 2005 12:30:01 +1000, James Gregory wrote: What other options do I have? I haven't tried it, but apparently advi is a DVI viewer (ie: TeX output) designed for presentations that can embed other X11 programs. Also note that PDF can embed movies which acroread might display. I doubt the xpdf-derivatives will cope with that though. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] running X11 app through Apache
At Mon, 2 May 2005 15:57:46 +1000, Julio Cesar Ody wrote: I tried to run a few GUI (X11) applications to run through my webserver, but no success so far. They simply don't run, and no error appear in any of my logs. Does anybody know how to do it? Heh, its a crazy idea but strangely enough not a new one. A lot of X11R6.3 was devoted to these sorts of ideas. See http://www.broadwayinfo.com/ and have a poke around http://cvs.freedesktop.org/xorg/xc/programs/xrx/ The x.org guys seem to have even updated the Xrx browser plugin to work with newer mozillas.. (Personally I think Xrx would be pretty cool for intranet settings. I'm sure daniels agrees ;) -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Wine setup
At Tue, 5 Apr 2005 13:09:18 +1000, Alan L Tyree wrote: I have a windows based CD that I would like to install and use with Wine. The setup.exe prog crashes. What's the best way to proceed? is there anyway to extract and install without using the setup.exe program? MSI is actually one of the most difficult programs for wine to deal with (it exercises all sorts of horrible win32 IPC mechanisms). *Latest* wine versions have the beginnings of support for MSI, including a custom msiexec program somewhere. Most (non-MSI) Windows installers are either self extracting zips or one of those .cab thingies. Native linux extractors for these archive formats exist and if you have some recognisable format I'd try them next. Failing that you can try to work out which DLL the crash is occurring in (bt in the winedebugger will give you a backtrace) and perhaps try switching to/from wine and real Windows DLLs. Various values for the WINEDEBUG environment variable may help here too (see manpage). Good luck ;) -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Postfix Virtual Domains and Timezones
At Wed, 23 Mar 2005 17:26:02 -0700 (MST), Dennis M. Gray wrote: I have set up a Postfix MTA with several virtual mail domains. So far in the doucmentation I have not found a way to have mail sent by Postfix to show a time that is different than that of the server, which is in Arizona, USA. I would like virtual domain users to have their mail sent with the correct time zone, i.e. +10 for Australia. The Date header is added by the original MUA -- the MTA just passes it along, as it does all the other headers. Its in a standard format and most MUAs will convert it to the local timezone when displaying. If it doesn't, its not something you want to fix in the server. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] reg exp q on eml
At Tue, 22 Mar 2005 08:32:07 +1100 (EST), Voytek Eymont wrote: em(ai)?l does it mean 'eml and/or ail' ? no, that means email or eml (ie, with or without the ai) -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] I wish to lowercase a character in a sed script
At Mon, 14 Mar 2005 17:57:08 +1100, Michael Lake wrote:
Suggested a perl script:
cat titles.html | perl -ne 'm/\.html([^,]{1,}),/; $name=lc($1); $_
=~ s/\.html/\.html#$name/; print $_;'
equivalent to:
perl -pe 's/\.html([^,]+),/.html#\L$1\E/' titles.html
(sorry, should have paid attention earlier :P )
--
- Gus
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] rdiff?
At Mon, 14 Mar 2005 11:44:26 +1100, Lyle Chapman wrote: Does anyone know of a gui for rdiff or something similar. Any suggestions are appreciated 'ediff' (in emacs) does side-by-side (or otherwise), coloured diffs - with interactive merging, etc. Does 3-way, from a patch, etc, etc. Its worth using, even if you use emacs for nothing else. (emacs followed by M-x ediff RET followed by ? if you get lost) -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Latex question: fi -- theta
At Fri, 11 Mar 2005 10:06:26 +1100, Nick Croft wrote:
Recently any occurrence of `fi' is rendered by a greek letter. So a word
like Office become OfØce. I'm not sure what to switch off or what kind of
package I'm using which does this.
Typical preamble:
\documentclass[12pt]{article}
\renewcommand{\familydefault}{put}
\usepackage{color}
\pagestyle{empty}
\begin{document}
I would guess that your font encoding has changed somehow. fi is a
very common ligature, and if your font encoding is wrong (or virtual
font out of whack) then you could end up with the wrong glyph
everywhere that ligature is used.
If you add \useencoding[T1] to your preamble, you'll end up using some
different font files that may bypass the problem. Either way, I'd
start filing a bug with your distro provider.
To test TeX's understanding of a font, you should be able to run tex
testfont and give the answer putr7t or something when asked which
font to test. At the * prompt type \sample\bye and you'll get a
dvi which shows the full font table (and a sample piece of text). If
this. In OT1 encoding (the LaTeX default) table, I believe the fi
ligature appears in position 014.
--
- Gus
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Weird login behaviour
At Thu, 24 Feb 2005 22:37:42 +1100, Oscar Plameras wrote: With my firewalls and other security critical servers, I require recompiling kernels by removing all UNUSED and REDUNDANT modules as part of the audit process so, when I got a problem such as the one illustrated above, I ONLY need to examine a few modules instead of TONS of them. Of course, nasty kernel code can be loaded and then the file containing that code deleted. (ie: modules on disk really have no relation to loaded modules). So this only really adds a level of inconvenience to hackers - assuming they wanted one of the standard modules (and it takes up less disk space if thats important too). -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Digital TV Tuner Cards On Linux?
At Sun, 13 Feb 2005 10:05:49 +1100, Oscar Plameras wrote: 'tv_grab_au' does not complete execution at all. Somewhere, along the execution process, it bombs out with an error that says 'something is missing'. I will post the exact message as soon as I have setup my Linux box. I'd guess that you needed to install some additional Perl library. The error message would have told you which one. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Digital TV Tuner Cards On Linux?
At Thu, 10 Feb 2005 12:21:23 +1100, Oscar Plameras wrote: Kevin Saenz wrote: There is NO 'tv_grab_au' from the de-facto http://www.xmltv.org server. I suspect because there is none that works consistently. ffs Oscar, the first hit on google is a perl script that I've been using fine for about 6 months now: http://www.onlinetractorparts.com.au/rohbags/xmltvau/ I don't get what part of mythtv even needs a howto. The docs on the mythtv site are excellent and using mdz's pre-built Debian packages makes it ridiculously simple: http://dijkstra.csh.rit.edu/~mdz/debian/dists/unstable/mythtv/ -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Digital TV Tuner Cards On Linux?
At Fri, 11 Feb 2005 02:37:41 +1100, Oscar Plameras wrote: Angus Lees wrote: This is because tv_grab_au (from http://www.onlinetractorparts.com.au/rohbags/xmltvau/) does not work successfully on my installation. I'll try this perl version again and the python version suggested in another post. I'm happy to help debug the problem with the script if you'd like, but I'm going to need more information than that. In what way doesn't it work? I presume you've tried running the script standalone (outside mythtv)? -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] diffstat, Ctrl-R, pushd/popd
At Wed, 9 Feb 2005 09:42:31 +1100, Ben Leslie wrote: 1/ Ctrl-R history searching When using the shell you pretty quickly work out that pressing up will search backwards through you history, however it tooks me ages to find out that you could search backs through the history by typing Ctrl-R and a search string. One of my favourite is M-. (thats alt or maybe windows key and full-stop) That will cycle through the last word on the last commands. (Of course zsh has all sorts of exciting keys, but I won't turn this into a zsh-specific mail..) -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] how to get shell script to supply password ?
At Wed, 12 Jan 2005 22:41:21 +1100, Robert Thorsby wrote: On 2005.01.12 22:33 Rod Butcher wrote: I need to get my shell script to login to something and then enter a password at a prompt.. i.e. unattended operation. I can't get the script to feed it the password, it always prompts me.. lets say userid=xx, password=yy xyz login xx password : at this point I want the shell script to feed it the password yy I've tried | , to no avail. Read the manpages. no dice. Try: http://expect.nist.gov/ To expand on Robert's answer a little: Almost all Password: prompts are done by talking directly to the programs controlling tty. The intention is precisely to bypass the normal shell pipeline mechanism and prompt the user, even though the input/output is doing something else. Now some programs (like gnupg) will also read the password from some arbitrary file descriptor when given a suitable command line option - you can use this to pass in the password from another program. In the general case, however, your program needs to setup a fake tty and run the program in that, so that it can catch the password prompt. The classic program for doing that is expect at the URL Robert gave you. These days, most other languages (eg perl, python) also have some sort of library for doing the same, usually called some variation on expect - so if you are more familiar with a particular language you might want to use its library. If you don't feel up to programming anything, the original expect has a learning mode where you can just do what you want (as a human) and it will watch and generate an expect script that will duplicate that - its pretty cool. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Postgresql case-insensitive mishmash brain twister.
At Thu, 6 Jan 2005 16:16:05 +1100 (EST), Stuart Guthrie wrote: So Debian seems to be initialising postgres dbs with the first of these collate sequences meaning that whatever 'C' is seems to return case-sensitive search results. Mandrake's standard RPM implements postgres with en_US collation sequences which means it sorts AaBbCcDd which is what I'd be hoping for. Run dpkg-reconfigure --priority=low postgresql. One of the questions is what encoding you want to use by default (although since you obviously *rely* on the encoding, then I'd suggest mentioning it explicitly in the CREATE DATABASE statement anyway). -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Font Antialiasing problems
At Tue, 28 Dec 2004 22:55:13 +1100, Indelible wrote: I don't have have gnome installed either, just WindowMaker. So I guess gnome-font-properties is not really an option. Ah. So edit /etc/fonts/fonts.conf (or ~/.fonts.conf) directly. This is assuming your fonts are rendered through fontconfig+Xft. If fonts.conf doesn't seem to change the fonts in question, you might have to describe which particular fonts/programs are affected. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] PDF generated from a server is missing fonts
At Thu, 30 Dec 2004 00:49:49 +1100, Michael Lake wrote: I have a server and I'm using pdflib to generate a PDF file and sending back to a user. At home it all works fine. But on the server when I create the PDF and view it ALL the text is missing. Even if I download the PDF file there are no fonts. I think I need some debian font package but what one. There are so many. Its a Debian system and it is not running XFree or any X. Any idea what font package to install so the libary can find a plain times-roman and a helvetica to embed? Ok. PDF (like Postscript) has 14 base fonts that are assumed to exist in any reader and there is no need to embed them. Courier, Helvetica and Times Roman are in those base 14. A very simple tool for finding out what fonts are referenced and/or embedded in a PDF is pdffonts that comes with xpdf (its in the xpdf-utils Debian package). For example, a simple pdftex-generated PDF I happened to have handy gives: /tmp% pdffonts intermezzo.pdf name type emb sub uni object ID --- --- --- - XSSGUM+CMR12 Type 1 yes yes no 6 0 LDZDIG+CMR10 Type 1 yes yes no 9 0 Note the emb column showing that both these type1 fonts are embedded in the PDF. (The random name prefix is a PDFTeX workaround for a bug on MSWindows Acroread) So what you should do is run pdffonts on your PDF file. Either you have forgotten to embed some fonts that need to be (ie: any fonts outside the base 14) and only some of your viewers have access to the font through other means (its installed elsewhere on your filesystem); or you are using the base fonts and your xpdf is unable to find its installed version of them (from the gsfonts package on Debian) and your PDF file should work fine elsewhere. From the error message you mention in another post, I'm suspecting the latter. Try fixing/removing your ~/.xpdfrc and see what happens. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] horizontal scrollbar terminal
At Thu, 30 Dec 2004 09:08:42 +1100, James A Coffey wrote: I have been using Linux for a number of years and the only gripe I have is that I can not find a terminal emulator that has horizontal scroll bars or I can't find away to configure my environment so that lines scroll horizontally and I can scroll through them using arrow keys for example. Why would I want them I hear you ask ? Simply because when you are tailing a log file ( eg syslog from cisco routers or worse snmp traps) having a line wrap makes viewing the logs very difficult. All I am after is something like xnmevent viewer from HP that doesn't wrap lines but alllows you to scroll horizontally without stopping the real time logs. Try xconsole for a simple light-weight solution. I'd use (x)emacs with `truncate-lines' set to true, but thats a bit of a sledgehammer if you aren't already using it for some other reason.. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Merry Christmas to you Down Under
At Fri, 24 Dec 2004 15:48:15 +0100, Michael Hayder wrote: Now evolution can use your webcal://foo.bar calendar. But I am not able to use webcal://foo.bar/mic.ics.. My problem is I could not find out how to enable webcal:// on my webserver or what program is necessary to use it. Can you send me examples from your webserver slug.org.au/events/event.ics uses a custom written perl script written by me (and with a few bugfixes by others ;) Its written using HTML::Embperl, but it should be obvious how to convert it to a simple CGI script. It generates the iCal entry from a row in a PostgreSQL database. Source is available from http://slug.org.au/source.html?path=/events/event.ics -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Font Antialiasing problems
At Wed, 22 Dec 2004 16:14:54 +1100, Indelible wrote: Yes, definitely the native res. Everything else is sharp as, it's just the fonts which give me trouble. Run gnome-font-properties(*) and play around with the anti-aliasing amounts and the various LCD RGBA orderings. You also get to tell gnome-font-properties all about your current resolution too, so make sure that matches what X already knows it is (oops I'm ranting again). (*) I think thats what its called - I don't have any gnome stuff installed on my current machine. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Samba printpdf utility (answer)
At Wed, 26 May 2004 14:40:51 +1000, Peter Rundle wrote: add 'printing = bsd' to the share, Ah yes, I saw someone mention that during my web searching. I had no luck with it since I think its only implemented in the *latest* version or so of samba. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] USB HID IR device - how generic can I go?
At Sat, 8 May 2004 10:38:33 +1000, Jeff Waugh wrote: So I got this USB IR port and remote bundled with my DVB card. I'd kinda like a generic IrDA port to use with my phone and iPaq, so I'm wondering if I can use this one. I think I read somewhere that IR remotes and IrDA use quite different hardware-level protocols so you can't turn one into the other. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] spam filters not working
At Fri, 7 May 2004 11:04:55 +1000, Mary Gardiner wrote: I train it on all my spam and non-spam, and I train it every week on mail received during that week. (With a cronjob, I just need to make sure false negatives and positives are moved into an appropriate folder.) I don't delete the existing token database ever. .. so with all that manual spam/ham classification/archiving, is there actually any point running an automatic spam filter anymore? From what I can see any spam filter that needs training is missing the point - but I've never actually run any of the Bayesian filters so its purely ignorant prejudice ;) -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Cross platform interpreter invocation
At Tue, 30 Mar 2004 13:47:12 +1000, Mary Gardiner wrote:
I'm accustomed to starting my various Python and Perl files with:
#!/usr/bin/env python
or
#!/usr/bin/env perl
However, you can't pass arguments to whatever you're invoking, thanks to
the limits of the #! interpretion (#!/usr/bin/env perl -w at the top
of a file causes a search for a binary named perl -w). What workaround
do people use for this problem in general? (I know -w is equivalent to
use warnings; so I know the Perl workaround)
#!perl -w
works on Linux, I'm not sure how it goes on other Unices..
For the perl case, you can use some wacky perl features (see
perlrun(1)). These all assume /bin/sh exists and use that to find
perl in $PATH:
#!/bin/sh
exec perl -x
#!perl -w
A variant on the standard MakeMaker-produced perl script:
#!/bin/sh
eval 'exec perl -w -S $0 ${1+$@}'
if 0; # not running under some shell
or even:
#!/bin/sh -- # -*- perl -*- -w
eval 'exec perl -S $0 ${1+$@}'
if 0; # not running under some shell
--
- Gus
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] NFS mount on Debian Woody not working
At Sat, 13 Mar 2004 19:55:26 +1100, Terry Collins wrote: Jeff Waugh wrote: Do you have nfs-common installed? Yes. That version has nfs-common nfs-kernel-server nfs-user-server running. All three were restarted with no difference and the machine was rebooted with no difference, aka same error. This is on the non-working *client* right? -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] rights of root from su
At Fri, 12 Mar 2004 14:35:31 +1100, mlh wrote: On Fri, Mar 12, 2004 at 02:30:12PM +1100, Geoffrey Cowling wrote: Xlib: connection to :0.0 refused by server Xlib: Client is not authorized to connect to Server ./setup: cannot connect to X server :0 so I have to logout as user and login as root on X11... Because of X authorisation (xauth). If you are the only using this box, do a 'xhost localhost' before su'ing. This will allow anyone on the local machine to connect. For the special case of root (since root can read everyone's files), you're better off doing as others have suggested and setting XAUTHORITY to the relevant user's .Xauthority file. If your gaining root process doesn't change $HOME (eg sudo without -H, su with -m or -p), then it should find the right .Xauthority file ok anyway. -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Press request: Legitimate uses of P2P
At Tue, 17 Feb 2004 16:03:30 +1100, Mary Gardiner wrote: If you are willing to comment on your use of P2P technology, please contact the SLUG committee, and we'll pass your details on. I used a web browser once and I believe there was no globally centralised server involved at any point. Does that count? (More seriously, all the good bits of the Internet are P2P networks.) -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Embedding fonts in pdf documents
At Wed, 4 Feb 2004 00:28:43 +1100, Tom Massey wrote: * Angus Lees [EMAIL PROTECTED] [2004-02-03 21:17]: At Mon, 12 Jan 2004 00:45:19 +1100, Tom Massey wrote: Is there a way to embed fonts in a pdf file created on Linux? ps2pdf can't embed a font it doesn't know about. You could tell ghostscript about your font (if you were using Debian, you'd register your font with Defoma) OK, something to look into. I haven't added any fonts to the system beyond the standard fonts in OOo, kword, abiword, etc. Do I need to register them? perhaps. Defoma isn't quite ready to be made a requirement for Debian packages just yet - so not all font-providing packages will do the correct registration. If you care, run dfontmgr(1) and drag+drop a font onto it. This won't help you here, however - see below. Which font in particular are you having trouble with? When I print to a ps file and use ps2pdf, uploading to lulu.com gives the error: You have not embedded certain fonts in your document, which may cause display or printing issues. Font Type Times-Bold is not embedded. Font Type Times-Roman is not embedded. Font Type Times-BoldItalic is not embedded. Font Type Times-Italic is not embedded. That's with OOo 1.1 Beta. 1.0.3 gives similar errors, though complains for each time the font changes so repeats fonts. Same for kword. aha. PostScript (and thus PDF) is allowed to assume the existence of the 14 base fonts -- Times is one of these. Its perfectly fine (a good idea even) for an otherwise enthusiastically font-embedding program to not embed these fonts (it makes the PDF file smaller). The problem is that not all Times are created equal and paranoid publishing companies (or people who run PitStop with options they don't understand) insist on embedding *all* fonts, including those in the base 14. There is probably a way to make OOo do this, but it might not be exposed in the GUI. I'd suggest asking on a OOo list. From reading the docs, I believe ps2pdf -dPDFSETTINGS=/prepress will do what you want (I haven't tried it though). To make pdfTeX do this, you need to run updmap --setoption pdftexDownloadBase14 true The pdffonts command that comes with xpdf is very good for finding out what fonts are in a PDF and if they're embedded - see the emb column. Personally, I always produce pdfs with pdfTeX (through ConTeXt, pdfLaTeX or pdfJadeTeX) and it embeds fonts quite nicely ;) Mmm, I suspect learning TeX would be useful. The document I'm experimenting with started out in Emacs. Only moved to OOoo as a quick means of making it look nice. It takes a bit of work to learn, but I'm confident you'll be impressed with the results. When you feel ready to start learning, there are several people here on the slug list who'd be eager to help. (Its not that hard - very similar to any other markup language, eg HTML) -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Embedding fonts in pdf documents
At Mon, 12 Jan 2004 00:45:19 +1100, Tom Massey wrote: I'm looking for a way to embed fonts in pdf documents so that I can be sure that they look exactly as intended wherever viewed. Using OOo 1.1 beta 2, or kword 1.1.1, I'm able to create pdf files, but the fonts aren't embedded, and although I've searched through all the menus I can't seem to find a way of doing it. Same for printing to a ps file and then using ps2pdf. Googling doesn't seem to give me any hints on this. Is there a way to embed fonts in a pdf file created on Linux? ps2pdf can't embed a font it doesn't know about. You could tell ghostscript about your font (if you were using Debian, you'd register your font with Defoma) -- but your best bet is to get OOo to do its job in the first place :P Unfortunately I have no experience with OOo or kword, but I find it strange that its hard to embed fonts. Which font in particular are you having trouble with? Personally, I always produce pdfs with pdfTeX (through ConTeXt, pdfLaTeX or pdfJadeTeX) and it embeds fonts quite nicely ;) -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] LaTex and the use of \symbol.
At Wed, 14 Jan 2004 15:34:07 +1100, Bill Bennett wrote:
1. I'm using the palatino package, ie., \usepackage{palatino}
2. I want to use the letter u with the tichy little hole above it
(apologies in advance to the Czechs, but it's not an umlaut and I
don't know what it's called). I looked up the Cork layout for the
symbol and was going to use that, ie., \symbol{xx}
try \r{u} (if thats the symbol you mean)
--
- Gus
--
SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Xinerama and fonts
At Wed, 21 Jan 2004 17:15:05 +1030, Jeff Waugh wrote: Matrox in both cases? Badness - Matrox drivers and Xinerama have this effect, which is really annoying, because the drivers and hardware otherwise rock. (my Matrox MilleniumII has never had any problems with Xinerama) -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] Open Source in Iraq.
At Thu, 18 Dec 2003 15:06:56 +1100, Jeff Waugh wrote: quote who=David Killen An interesting article I found about the possibilities of Open Source in Iraq Interestingly enough, we just kicked off a mailing list for GNOME in Iran. Hopefully we don't get a GNOME/KDE-Iran/Iraq war on our hands. Aren't Americans banned from exporting things to Iran or something? How does this affect GNOME? (or is this just a user list at the moment) (Luckily Debian hasn't got any developers in any of the 6(?) US-banned countries so hasn't had to deal with this issue yet) -- - Gus -- SLUG - Sydney Linux User's Group Mailing List - http://slug.org.au/ Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html
Re: [SLUG] ispell trick
At Tue, 02 Dec 2003 23:29:15 +1100, Ken Foskey wrote: I am working on some doco on programming languages in tex. There is a lot of words specific for that particular language and I don't want it polluting my personal list and I ouwld like the checks consistent for others to keep it easy. when using emacs/AUCTeX+ispell.el, ignoring words for this session adds the words to the end of the tex file in a % Local words comment block. I have no idea if detecting/using this list is an ispell feature, or ispell.el massages it somehow first. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] compiler warnings
At Tue, 25 Nov 2003 23:45:48 +1100, Ken Foskey wrote: I am getting this error: process.c:1018: warning: passing arg 1 of `putenv' discards qualifiers from pointer target type It turns out that I am sending a const string to putenv which is defined: ./stdlib.h:extern int putenv (char *__string) __THROW; This discards the constness of the string. Then I find this: http://sources.redhat.com/ml/libc-hacker/1999-08/msg8.html which implies that it was const ages ago. Should I raise this as a bug? quoting from putenv(3) (from manpages-dev.deb 1.60-3): NOTES The putenv() function is not required to be reentrant, and the one in libc4, libc5 and glibc2.0 is not, but the glibc2.1 version is. Description for libc4, libc5, glibc: If the argument string is of the form name, and does not contain an `=' character, then the variable name is removed from the environment. If putenv() has to allocate a new array environ, and the previous array was also allocated by putenv(), then it will be freed. In no case will the old storage asso- ciated to the environment variable itself be freed. The libc4 and libc5 and glibc 2.1.2 versions conform to SUSv2: the pointer string given to putenv() is used. In particular, this string becomes part of the environment; changing it later will change the environment. (Thus, it is an error is to call putenv() with an auto- matic variable as the argument, then return from the calling function while string is still part of the environment.) However, glibc 2.0-2.1.1 differs: a copy of the string is used. On the one hand this causes a memory leak, and on the other hand it violates SUSv2. This has been fixed in glibc2.1.2. The BSD4.4 version, like glibc 2.0, uses a copy. SUSv2 removes the `const' from the prototype, and so does glibc 2.1.3. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Anyone used Perl's Filter package !!
At Mon, 27 Oct 2003 13:17:11 +1100 (EST), education wrote: I download the package Filter-1.26 from CPAN. I'm having problems understanding how to use this package to get Perl code to go through the filter before compilation. Also does this method clearly prevents anyone from seeing the source code of a Perl script ? Unless it uses some form of encryption, then all you've done is obscured the code - anyone who can use the same filter can read the source code. See Embperl for an example of a module which *does* do encrypted source code - you compile a private key into the embperl module and then use it to encrypt source code files (see crypto/README in Embperl source). Be aware that a determined person with access to the files and the ability to run them can always find out what it does. You'd have to have some peculiar needs to make this approach worth the effort (as opposed to simple restrictive file permissions, for example). -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Asking about NAT With Debian and tutorial about Firewall using iptables or ipchains
At Thu, 23 Oct 2003 12:30:09 +1000, Jeff Waugh wrote: quote who=pesoy misak I have 3 computer that one of them I want to become my gateway. I have read about the Masquarding HOWTOs but i seems too difficult to be understood. My question is there anyone that could teach me using NAT or is there any good tutorials that could teach me about the masquaring. If you want a quick start on Debian, just install the ipmasq module on your firewall/gateway. It will automagically set up NAT to do the right thing! where ipmasq module means ipmasq debian package, not ipmasq kernel module. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] K 2.6 help.
At Mon, 20 Oct 2003 23:07:40 +1000,Ken Foskey wrote: If anyone is on a very fresh version of K2.6 with extra patches can you please run this code and see if it crashes. It fails on all K2.6 up to Test6 release. I would be interested to hear of any success. Here is what I get: Getting from NULL Setting Jump Running Signal 11 caught After jump Setting to NULL Setting Jump Running Segmentation fault There should be no segfault and another signal caught. A C program is undefined after returning from a SIGSEGV handler. In particular that means its allowed to die in weird ways. (If this was intended for real use, you definately want to rethink your approach. Testing addresses for segfaults is always going to get you into trouble) -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] K 2.6 help.
Oh yeah - if you *really* want to do this, fork() first and test in the child. If the child exited with a segfault you know it was a bad idea. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Re: [activities] Codefest
At Tue, 14 Oct 2003 20:14:32 +1000, Erik de Castro Lopo wrote: I hereby offer to try and find/fix any bug in any piece of software that mets the following conditions: heh, that sounds like fun. -- - Gus (looking around for a meaty libc6/toolchain bug ;) -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Re.E;Tried to deque ...
At Wed, 08 Oct 2003 16:32:21 +1300, Adam Bogacki wrote:
This is in stable (!) and I've forgotten how to do it [apt] off CD-ROM.
% man -k 'apt.*cdrom'
apt-cdrom (8)- APT CDROM managment utility
% man apt-cdrom
NAME
apt-cdrom - APT CDROM managment utility
SYNOPSIS
apt-cdrom [ -hvrmfan ] [ -d=cdrom mount point ] [ -o=config string ] [
-c=file ] { add | ident }
DESCRIPTION
apt-cdrom is used to add a new CDROM to APTs list of available sources.
apt-cdrom takes care of determining the structure of the disc as well
as correcting for several possible mis-burns and verifying the index
files.
[...]
--
- Gus
--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Notice of motion at the next meeting.
Surely moderating lists based on From address is flawed, since that data can be so easily faked. I know I get plenty of spam claiming to be from [EMAIL PROTECTED].. If the intention is to have 0% spam make it through to the list, I don't see how this is going to achieve that. I do see that it will inconvenience newcomers, people with more than one email address and our volunteer list admin -- all people we should be *attracting*. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Emacs custom languages
At Wed, 17 Sep 2003 10:18:27 +1000, Andy Eager wrote: I have designed the language written an interpreter for it, but would like emacs to do some syntax highlighting for me. After you've written the rest of your major mode (which gives details like syntax definitions (what constitutes a word, etc), any relevant functions/keybindings, etc) then you need to give a list of keywords/regexps to highlight. On XEmacs, you need to put a `font-lock-defaults' property on the major mode symbol, on FSF Emacs you put an entry into `font-lock-defaults-alist'. Luckily its the same entry in both cases (unless you use exotic features). See the comments and docstrings in font-lock.el, particularly for `font-lock-defaults'. Start reading the lispref.info section on writing major modes and go from there - its not too hard if you're already familiar with elisp. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] ide-cd under Debian Woody
At Sat, 20 Sep 2003 10:22:47 +1000, Terry Collins wrote: Where do you get ide-cd module under Debian Woody. The search engine on http://packages.debian.org/ lists: lib/modules/2.4.16-386/kernel/drivers/ide/ide-cd.o base/kernel-image-2.4.16-386 lib/modules/2.4.16-586/kernel/drivers/ide/ide-cd.o base/kernel-image-2.4.16-586 lib/modules/2.4.16-586tsc/kernel/drivers/ide/ide-cd.o base/kernel-image-2.4.16-586tsc lib/modules/2.4.16-686-smp/kernel/drivers/ide/ide-cd.o base/kernel-image-2.4.16-686-smp lib/modules/2.4.16-686/kernel/drivers/ide/ide-cd.o base/kernel-image-2.4.16-686 lib/modules/2.4.16-k6/kernel/drivers/ide/ide-cd.o base/kernel-image-2.4.16-k6 lib/modules/2.4.16-k7/kernel/drivers/ide/ide-cd.o base/kernel-image-2.4.16-k7 lib/modules/2.4.18-386/kernel/drivers/ide/ide-cd.o base/kernel-image-2.4.18-386 lib/modules/2.4.18-586tsc/kernel/drivers/ide/ide-cd.o base/kernel-image-2.4.18-586tsc lib/modules/2.4.18-686-smp/kernel/drivers/ide/ide-cd.o base/kernel-image-2.4.18-686-smp lib/modules/2.4.18-686/kernel/drivers/ide/ide-cd.o base/kernel-image-2.4.18-686 lib/modules/2.4.18-k6/kernel/drivers/ide/ide-cd.o base/kernel-image-2.4.18-k6 lib/modules/2.4.18-k7/kernel/drivers/ide/ide-cd.o base/kernel-image-2.4.18-k7 -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Re: Font problem
At Mon, 15 Sep 2003 15:06:30 +1200, Adam Bogacki wrote: Checking the X font server's configuration file in /etc/X11/xfs/config I find I have no 'xfs' directory there. However, 'find' tells me I have Debian doesn't use/need xfs by default. I doubt you have a reason to use it either. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Font problem
At 15 Sep 2003 12:49:10 +1000, Peter Hardy wrote: On Mon, 2003-09-15 at 12:28, Adam Bogacki wrote: Of course, if they are installed, then you've got bigger problems that I'd rather not think about unless they're actually happening. :-) An easy way to see what fonts the XServer knows about is to run xfontsel (or xlsfonts for simple dump complete list to stdout tasks) Of course, client-side font rendering is entirely up to the client. GNOME and KDE now use fontconfig to find them; supposedly you can use fc-list from the command line, but I've found that mostly useless. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Installing a verion of xpdf later than my Debian stable version
At Wed, 10 Sep 2003 15:44:09 +1000, Michael Lake wrote: Debian stable contains xpdf-reader Version: 1.00-3 and that is what I have currently installed. I have just pulled down xpdf version 2.02 from foolabs.com and compiled it. It works fine but I have not yet done 'make install'. In this case xpdf is not needed for anything i.e. nothing depends on it so it's easy to remove. If there were things that needed a pdf viewer that would be more troublesome. Certainly I gather doing an install over the top of an existing package would confuse the packaging system. (In this case) I would just install xpdf from upstream source into /usr/local/ and remove the xpdf-reader package. If you found something that depended on xpdf-reader, you could just install it again and have both versions (the old .deb and the new /usr/local version) simultaneously installed - provided you put /usr/local/bin before /usr/bin in $PATH everything should just work. /usr/local exists for a reason, not everything has to be done through packages ;) -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Counting up in a shell script
At Fri, 29 Aug 2003 13:10:08 +1000, Jeff Waugh wrote: quote who=Douglas Stalker In a shell script, is the a better (i.e.: more elegant) way to get a variable to cycle from 1 to another value (such as 100) Currently I am using: $ I=1 ; while [ $I -le 10 ]; do echo $I ; I=`expr $I + 1`; done for i in $(seq 1 10); do echo $i done seq is good for short lists (such as 1 to 10), but no good for counting up to several thousand. Simple maths can be performed in the shell by using $(( ... )), ie: i=1 while [ $i -le 1000 ]; do echo $i i=$(( $i + 1 )) done -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Perl5.8 installed;Apt preconfigures, but fails to install.
At Fri, 29 Aug 2003 10:52:14 +1000, Jeff Waugh wrote:
You also want the list of installed software, which you should normally be
able to get from apt Hmm. Presumably you can get it from looking
directly at where apt stores its files.
dpkg -l
or for just the package names:
COLUMNS=150 dpkg -l | awk '{print $2}'
I would have used dpkg --get-selections and later piped that output
back into dpkg --set-selections.
Its not quite as robust as jdub's, since it will (pretty much) only
give you a list of the *installed* packages. So long as you're
starting from a minimal base system the second time round, it'll do
the right thing - and its nice and easy.
--
- Gus
--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Debian apt sources question
At Wed, 27 Aug 2003 01:35:09 +1000, Alex Sutcliffe wrote: I'm running woody and installed from a cd set. For some time I have been puzzled by the fact that my /apt/get/sources.list file lists the 3 cds I have as unstable. This has puzzled me for a while since I thought that cds were all stable. So I mounted one of the cds and looked at it. In the dists directory there are three links - stable, testing, unstable - to the a directory named woody which then has the directories contrib etc. So my question(s) is - can I just change the entries in my sources.list file to stable? I have already suffered a mini disaster with apt so want to be careful. no. apt needs some special fiddling to properly know about cds; apt-cdrom does the necessary stuff. Manual fiddling is likely to not work. If it works fine currently, there is no reason to want to change it. I seem to remember a bug in the original Debian 3.0r0 cds, where they specified Suite: unstable in the dists/woody/Release file (or was it the dists/woody/main/binary-i386/Release file?). It all looks ok on my Debian 3.0r1 cd here. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] neat tricks used for the purposes of evil
At Tue, 26 Aug 2003 23:40:40 +1000, Robert Collins wrote: I'm dubious about 'vastly more versatile' - that quite unsubstantiated. For example, you can't use random perl functions to control squid's behaviour. You can with apache+mod_perl, which in my book counts as vastly more versatility. -- - Gus (not defending any opinions expressed earlier in the thread) -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] LaTeX question, Linux command.
At Mon, 25 Aug 2003 17:41:14 +1000, Michael Lake wrote: Bill Bennett wrote: I was going to use a .jpg file in a figure in a LaTeX document, on the grounds that an.eps file would be too big. I don't have my cp of Goosens here to look up the graphics rule but it looks like you are wanting to use latex and generate postscript. An eps is not necessarily too big. Both convert and jpg2eps (which might even already be on your system as it comes with many teTeX distributions) just encapsulates the binary jpg and it wont be much bigger at all than the jpg. Then you wont need the graphics rule at all. .. also if you're aiming for postscript output, the jpg will have to be converted to postscript at some point in the process. (iirc, PDF can embed a jpg directly so thats a different story) -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Perl5.8 installed;Apt preconfigures, but fails to install.
Dude, you've hosed your system. At this point you can either: restore from backups (*ahem*) try to undo whatever you did (iirc you moved /usr and /lib somewhere else) try to duplicate exactly what Debian had put in those directories by hand (almost impossible) or save what data you can and reinstall. The choice is yours, but I know what I would have done by now. At Mon, 25 Aug 2003 22:52:32 +1200, Adam Bogacki wrote: In sum, I'm starting to contemplate reinstalling Debian, something I am prepared to go to lengths to avoid because I would lose a lot of hard-won material, because I'm living with an 80 yr old who is only starting to get over her computer-phobia, and because I also have other things to do. Where/what is this hard-won material? Just copy your important data off somewhere temporarily (maybe just another partition) and reinstall. The alternative is going to be *weeks* of hand hacking stuff. If you'd simply reinstalled you'd be done by now. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] LDAP server in Debian
At Sat, 23 Aug 2003 02:37:14 -0700 (PDT), Phillipus Gunawan wrote: My debian box is running LDAP server, slapd. It seems working fine because I can query the server from 127.0.0.1/localhost address. I dont have any 'ldap.conf' file in my box, but I do have the slapd.conf. Later on, I try to open/browse the LDAP database using LDAP browser from another computer (Softerra LDAP Browser 2.2) but no matter how many times I tried to, it always failed to connect. In what way does it fail to connect? Any useful error messages? What do the logs say on your ldap server? Have you restricted access using something like /etc/hosts.deny ? - Is this has anything to do with my missing 'ldap.conf'? No. ldap.conf configures the client-side tools (like ldapsearch(1)). - In debian, can I create it from a text file? Create what? ldap.conf *is* a text file. - Where I have to put it? /etc or /etc/ldap? Debian expects to find it as /etc/ldap/ldap.conf on my (testing/unstable) system and its in the libldap2 package. See http://packages.debian.org/ and search for packages containing ldap.conf for a full search of all packages from different Debian versions. Of the packges you currently have installed dpkg -S ldap.conf would do a similar search. If you have the relevant package installed (dpkg -l libldap2) and you still don't have an ldap.conf file, then you probably deleted it at some point. To get it back (as root): touch /etc/ldap/ldap.conf apt-get --reinstall install libldap2 choose yes, replace my version with the package version when prompted (ldap.conf is a conffile, and actions such as deleting are preserved across package upgrades) -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Problem using 'perl -MCPAN -e shell'
At Fri, 22 Aug 2003 21:15:35 +1200 (NZST), Andrew McNaughton wrote: That file contains the preferences you set up. You could edit it, or you could just back it up somewhere and delete it, in which case `perl -MCPAN -e shell` should ask you to provide the configuration details again. .. or type o conf init at the CPAN prompt. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] DocBook
At 12 Aug 2003 13:49:53 +1000, Alan L Tyree wrote: One of the other publishers is using DocBook as their primary system, so I thought I would have a look at changing from LaTeX. But, as I said, I need (or at least want) the good author support (outlining, citation, cross-referencing) that I get with emacs/auctex/reftex. psgml-mode will fold/unfold elements, thats about as close as I think you'll get to outlining. You could probably add speedbar support or something pretty easily though (I don't think it supports speedbar out of the box, but I haven't tried). It also has lots of ways of moving around the SGML source, so you might find that good enough for skipping around the document. Cross-referencing should work ok (even referencing particular pieces of other docbook files). Like everything else in docbook, the process is very markup-heavy compared to LaTeX. I haven't looked, but I'm guessing citation/bibtex-stuff is one of those its so easy with XML/SGML that no-one bothers publishing their solution things. Indeed, given a good bibtex-esque database, it should be quite simple to find all your citations and build another XML/SGML file containing the relevant bibliography. With docbook, be prepared to get familiar with DSSSL or XSLT+some other language. Unlike TeX, your source is data (not code) so will almost certainly need a more complicated build process surrounding it. (I alternated between preferring LaTeX or DocBook for years. I think they each have clear strengths/weaknesses and should be considered carefully on a case-by-case basis. I'm completely sold on ConTeXt at the moment though ;) -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] How to set standard settings for all users?
At Mon, 11 Aug 2003 22:53:41 +1000 (EST), Simon Bryan wrote: Now I am looking at how do I set other Mozilla prefs - or would it be easier in some other browser? I need to set the home page and the proxy settings in particular. In our windows setup we can do this with a combination of profiles and login scripts. I assume that Linux can do the same? On Debian at least (I don't think its a Debian-specific feature), you can edit /etc/mozilla/prefs.js to alter mozilla defaults for all users. Have a look at an existing ~/.mozilla/default/*/prefs.js or the system files in /usr/lib/mozilla/defaults/pref/*.js for some suitable lines to copy. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] basename files and paths with embedded spaces
At Thu, 7 Aug 2003 23:45:58 +1000, Matthew Hannigan wrote:
This behaviour has long been considered a bug
in the shell by many. But unfortunately it's
too late to change it for the standard shell.
zsh by default does the right thing, which of course irritates the
hell out of everyone who expects it to do the legacy thing.
For reference, to expand a zsh variable and re-break it up into words
(as normal bourne shell does), you have to do ${=variable_name}.
There are standard ways of making bourne shell scripts immune to
spaces in filenames, they basically revolve around using
$variable_name (with the quotes) everywhere and $@ (with the
quotes) if you want to pass all args through to some other command.
--
- Gus
--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] multiple recursive s'n'replace, what's a good tool ?
At Mon, 4 Aug 2003 23:18:19 , Voytek Eymont wrote: what's the recommendation for a m-r-s-r tool ? I'm looking to replace multiple string pairs in some web files. all I need is simple 'string a' for 'string b' swap, every instance, For an interactive tool, you can use emacs' dired mode to select files (or directory trees). 'Q' will run a regex search/replace over the selected files. (to get into dired mode, just File-open (or C-x C-f) a directory) -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] mixED CAse to lower.case ?
At Wed, 30 Jul 2003 21:02:29 , Voytek Eymont wrote: what can I use to recursively change file names/extension to all lower case ? I have some files and/directories like: I tried rename few times with little effect: [EMAIL PROTECTED] photos]# rename .JPG *.jpg [EMAIL PROTECTED] photos]# ls atomfactory1.JPG makitapage03-s.jpg makitapage10-s.jpg atomfactory2.JPG makitapage04-s.jpg makitapage11-s.jpg atomfactory3.JPG makitapage05-s.jpg makitapage12-450.jpg makitapage01-450.jpg makitapage06-s.jpg makitapage12-s.jpg makitapage01-n.jpgmakitapage07-s.jpg shopfront.jpg rename takes a perl expression that modifies $_. .JPG is not a particular useful perl expression. try: rename 's/\.JPG$/.jpg/' *.jpg i note the rename manpage has an example which translates uppercase filenames to lowercase. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] stripping CtrlM: CRLF vs LF vs CR dillema
At Tue, 29 Jul 2003 21:50:53 +1000, Mary wrote: On Tue, Jul 29, 2003, Voytek Eymont wrote: - what can I use to strip the CR ? The command dos2unix will do this. tr -d '\r' dosfile unixfile will also do the trick for simple cases. I usually just load it up in an editor (emacs or vi) and delete the \r's using the normal replace all strings matching functionality. vi: :%s/.$// - which is a cop-out, since I can never remember how to put a literal \r in emacs: M-x replace-string RET C-q RET RET RET - ie: replace literal RET (C-q quotes the next key) with the empty string -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] stripping CtrlM: CRLF vs LF vs CR dillema
At Tue, 29 Jul 2003 23:30:18 , Voytek Eymont wrote: [EMAIL PROTECTED] voytek]# tr -d '[\200-\377]' httpd.linux unixfile1 didn't work... no it wouldn't since \r is \015, which isn't between \200 and \377. (see man ascii) the above command would strip 8 bit characters (the top half of the charset - which has bit 8 set). \r and the other control characters are below \040, not in the top half of the charset. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] tiff - multi-page viewer solution
At Wed, 30 Jul 2003 08:32:03 +1000, Stuart Guthrie wrote: Receiving faxes via efax/jfax whatever. All come in as multi-page tif. Which apparently is a bit of a standard in 'document management software' circles. For what its worth, the best format for scanned (or faxed) documents sounds like its DjVu. I've never tried it myself, but there's a bunch of conversion tools/display programs/browser plugins for it packaged for Debian under the names djview and djvulibre. If you can install the appropriate windows software, this sounds like a very nice format speed+size-wise for these types of documents - particularly if you want to archive/serve them through HTTP. Alternatively, you could use the tiffsplit tool from libtiff and a bit of shell coding to split the file and attach multiple PNGs or something to the email. This may make archiving harder though. You could also use a similar shell script on the MUA end, by dropping a similar command line into /etc/mailcap or ~/.mailcap. I can't remember whether evolution honours mailcap or whether its like the rest of those sucky GNOME programs that have forgotten what makes UNIX great in the course of trying to make UNIX greater(*). (*) blatently unnecessary troll. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Finding all config files... (Debian question))
At Mon, 28 Jul 2003 10:36:38 +1000, Peter Chubb wrote:
I'm trying to set up a backup strategy. What I want to find
is all the configuration files for the various installed packages.
Some are obvious (/etc/apache/http.d.conf, /etc/passwd) Others are not
(/usr/lib/python2.2/site-packages/MoinMoin/config.py).
Is there a way of extracting from the dpkg information a list of all
changed configuration files? Dpkg must know this info...
Those files tagged as conffiles are listed in
/var/lib/dpkg/info/*.conffiles, but there doesn't seem to be any way
to know if one has changed (without unpacking a .deb and diffing).
The debconf database is also a good thing to back up, its
/var/cache/debconf/{config,passwords}.dat (by default).
--
- Gus
--
SLUG - Sydney Linux User's Group - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] anyone tried 2.6.0-test-beta?
At Mon, 28 Jul 2003 19:44:13 +1000, Steven Kowalik wrote: At 6:50 pm, Monday, July 28 2003, Kevin Saenz mumbled: Just wondering if anyone has started playing with 2.6.0 kernel? What's it like? I have. It's ... different. It uses modprobe.conf, rather than modules.conf, PCMCIA requires tweaking, and X doesn't work. And I've panic'd it twice. Oh well. I was going to say I hadn't noticed any difference - but I just tried getting X to work and I can't find the right module to make /dev/psaux work :( ppp_deflate seems to be broken (I can't receive UDP packets?), but adding nodeflate to pppd/options works fine. otherwise I haven't noticed any problems (running it right now) :) -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] C - exec fn call env vars
At Mon, 28 Jul 2003 19:15:07 +1000, Andy Eager wrote: Is there any way of setting an environment variable in the shell script so that it modifies the environment of the *parent* process? nope. I fear not - but is there any way of returning something from a shell other than through an exit status? stdout (or some other file descriptor), some file somewhere. all the other types of IPC too, but they're a bit hard from a shell script (without a helper (non-shell) command). -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] anyone tried 2.6.0-test-beta?
At Mon, 28 Jul 2003 23:05:35 +1000, Andrew Bennetts wrote: On Mon, Jul 28, 2003 at 09:03:05PM +, Angus Lees wrote: I was going to say I hadn't noticed any difference - but I just tried getting X to work and I can't find the right module to make /dev/psaux work :( I'm guessing that maybe the new HID stuff puts the mouse device at /dev/input/mouse0 regradless of physical source/protocol? i tried loading psmouse, which seems to detect a synaptics touchpad on my laptop (i presume this is correct). this seems to be enough for X to open /dev/input/mice successfully - although scratching the pad doesn't make the pointer move at all. loading mousedev claims to have loaded a PS/2 mouse device common for all mice, but this doesn't seem to be sufficient for X to load /dev/psaux (no such device). The Debian ALSA (user-space) packages don't yet cope with modprobe.conf transparently either (/me pokes stevenk) -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] lpnr to cups printer configuration
At Sat, 26 Jul 2003 12:19:41 +1000, David Kempe wrote: or there is a way to get cups to write out /etc/printcap from memory, the woody version of cups writes a /etc/printcap.cups you should be able to symlink /etc/printcap to that file (if you actually need /etc/printcap. you should be using native cups support where possible.) -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] Re: Multi-page TIFF Viewer/Printer
At Sat, 26 Jul 2003 21:20:09 , Voytek Eymont wrote: I'd say, for higher quality, there is no subsitute for TIFF, as far as bit mapped images go. tiff is a container format, so what format you really have inside a .tiff is really what determines the image's characteristics. in pretty much all cases you can strip off the tiff header and use the contents in its native format (bmp, jpeg, g3, etc) - there is of course no different in image quality. even if you were converting between lossless image formats (pretty much anything except jpeg), you shouldn't see any difference (they're lossless after all). -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] PDF fonts - thanks
At Sat, 26 Jul 2003 17:49:00 -0700 (PDT), Mark A. Bell wrote: Being a beginner at LaTeX, I produced my document using LyX (on top of Debian Unstable). I just exported the file as a PDF. The 'default' font that Lyx uses is apparently a Type 3 font. Pdffonts reveals: nametype emb sub uni object ID --- --- --- --- - [none] Type 3 no no no 9 0 weird. not only was there no font name, but it wasn't embedded either (emb). Is there any way of getting a log or something from lyx that might show what it does exactly when exporting as PDF? alternatively, try exporting as LaTeX source and running latex over it yourself. If that still produces the above, send me the LaTeX source and I should be able to tell you what's happening. (If it produces something that *does* correctly embed the fonts, then you've found a way to work around the problem) I also wrote a quick test file using Emacs without specifying any fonts or packages. I ran LaTeX and dvipdf, and tried out the file in Acrobat. It came out perfectly with Times Type 1 fonts. Pdffonts shows: nametype emb sub uni object ID --- --- --- --- - VCWIIB+CMR17Type 1C yes yes no 10 0 XACQQL+CMR12~f Type 1C yes yes no 16 0 GGLQBJ+CMR12Type 1C yes yes no 13 0 I presume that these are the 'Computer Modern' fonts mentioned in the TeX FAQ [http://www.tex.ac.uk/cgi-bin/texfaq2html?label=fuzzy-type3] and that they are infact the default for my Debian TeTeX installation. yes, thats Computer Modern (Roman) in a few different sizes. the random font name prefix on these and your Times-Roman example is a workaround for another acrobat bug, btw. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] LaTeX, fonts, Acrobat, and printing
At Fri, 25 Jul 2003 09:52:40 +1000, James Gray wrote: On Fri, 25 Jul 2003 09:37 am, James Gray wrote: If the fonts you're using in Linux (LyX/LaTeX) are Type1 or PostScript, you can copy them to the C:\Program Files\Adobe\Acrobat x.x\Resource\Font directory. Not sure how you go about editing Acrobat's Adobefnt.lst file automagically, but it's plain text and pretty easy to figure out. I also created a PDF using Font only available on my Linux machine and viewed it in windows before copying all my Linux Type1's over - font substitution was awful! AFTER copying the fonts, it displayed perfectly :) it would of course be much better to embed the fonts (as TeX tools usually do) - that way no modification is necessary on other machines. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] LaTeX, fonts, Acrobat, and printing
At Fri, 25 Jul 2003 09:45:04 +1000, Michael Lake wrote: Mark A. Bell wrote: The PDFs that I produce look beautiful in GhostView at home, but when I open them in Acrobat for Windows, the fonts look hideous! I don't much care because the printed copy looks great - better than I could produce with Word, with so much less frustration. Its most likely caused by having type3 fonts (bitmapped ones) in your document. They can be quite good in resolution even 1200 dpi and Acrobat 5.0 and below will rebder then shithouse. I concur. Why does this happen? Is it because the (expensive) laser-printer at work has the right fonts programmed into it, but regular Windows desktops don't? The fact that it prints OK but renders poorly on screen is the classic sign that you have type3 fonts in your document which acroshit renders badly. Open Acroread and under File/Document Properties it will list your fonts. I bet they are type3. You can also use the pdffonts tool from xpdf on Linux. Most likely the fonts you were using were derived from METAFONT fonts (.mf), converted to .pk and then embedded as type3 (a bitmap font format). If you can find/use a type1 version of these fonts (.pfa or .pfb) or even a truetype version, then you can embed them directly in the .pdf. These formats are rendered correctly by acroread. The progress output from pdfTeX or dvipdfm should also give you an indication about which formats are being used. (The default Debian tetex install should prefer vector fonts over bitmap fonts, and should correctly embed them. Try installing the tetex-extra package if you haven't already) Should I worry that if I try to print on a different printer, I might get an incompatible font? Should I somehow install more fonts on my Linux box? I'm running Debian Unstable. You shouldnt need more fonts. I am still myself working out how to convert my setup to use Type1 fonts and thats prob the best way to go as then Acroreader on Windows will render it OK. Fonts in TeX are hard I think. Installing new fonts under TeX is indeed hard. Its pretty much entirely automatable though - and I'm trying to do that with a tetex defoma script currently.. Once this is working, you should be able to register your type1/truetyp fonts with defoma and it will just work :) (a Debian/tetex -specific solution I'm afraid) Mark, I'm happy to give better explanations/solutions if wanted. I'll need to know exactly what TeX tools you're using to generate your .pdf though (and sample input/output from them). -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] latex gnuplot missing colour
for simple graphs, you might also want to look into metapost. a little harder than gnuplot, but you have complete control over the output (its a drawing language, rather than a plotting tool). -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] command Line mailer that can send an attachment
At Wed, 09 Jul 2003 10:13:34 +1000, Terry Collins wrote: As per subject, need a command line mailer that can send a file as an attachment. for MIME/binary attachment stuff, you really want the scripts/tools that come from metamail. includes mimencode, which is good for doing the gruntwork, mailto which is an interactive composer along the lines of mail(1), and mimeit which is a good wrapper for shell scripts to just send this binary file to this user with this mime type. -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
Re: [SLUG] request tracker -- worth it?
At Mon, 7 Jul 2003 22:45:14 +1000, mlh Matthew Hannigan wrote: How about gnats? Anyone work with that? I used/set it up once about 4 years ago. I probably prefer gnats to RT, but gnats is *heavily* email based so probably only suitable for power users (ie: people who are comfortable with text). We use RT at work and it works fine. The extent to which you can customise it (if you know perl) is a big plus - we've done all sorts of rude things to it (makes upgrading to RT3 harder though ;) (re interface: I think the dense interface is a plus too, since the people who use it use it every day. We don't have casual guest users like an open-source project might have) -- - Gus -- SLUG - Sydney Linux User's Group - http://slug.org.au/ More Info: http://lists.slug.org.au/listinfo/slug
