pimppa-users Mailing List for PIMPPA

SourceForge Headquarters 225 Broadway Suite 1600 San Diego, CA 92101 +1 (858) 454-5900

good morning Pimppa

http://huso.se/everywhere.php?clean=vpg12dyghm236k

mr holzquack

As a further note, I did find that there _are_ other tools out there - =
though not many at all and I had to look deep...  Your note telling me =
to take a look at the competitors prompted me to make a few googles on =
key words like "linux news leech", "linux news reader", "usenet file =
grabber", etc...

I found at least one potential candidate in a list of two: BNR - Binary =
News Reaper (www.bnr2.org) - seems to be a more mature application than =
PIMPPA; KLibido - KDE version looks very similar to XNews in Windows (I =
haven't seen this one in operation and it's not what I'm looking for =
from what I can tell by the homepage and screenshots)

Laz

-----Original Message-----
From: LASZLO ACS
Sent: Tue 5/30/2006 10:07 AM
To: Igor Wronsky
Cc: pim...@li...
Subject: RE: [pimppa-users] I must be dense...
=20
Thanks, Igor, for your reply.

Since writing, I did figure out the source of my confusion.  It seemed =
that since I had tried using the Ubuntu "Synaptic" GUI to give me joy, =
and when that didn't happen from the initial 1st attempt in running =
Bowser by popping up an error message, I dug a little into the web and =
found your site to discover the "README" document.  The references your =
README document made to SETUP confused me since I didn't find that file =
in my system.  Getting MySQL up and running was a piece of cake; after =
that, I assumed clicking on and running Bowser would detect whether =
MySQL was running and take care of whatever was necessary to get the =
user up and running with the software.  No joy...  I still got the same =
error message as before.  I returned to the README file and perused it =
with a fine-toothed comb (ya know how hard this is with nappy and =
tangled substrates???)...  I searched for the files on my machine, =
eventually locating some of the scripts in a directory called "pimppa" =
(amazingly familiar).  There I found "create_database.sql" - hmmm...  =
viewing the file in my fav editor, I  got some insight to what was next. =
 I ran the script, it failed but gave me a useful error message [One of =
the things I absolutely abhor about the monopoly, M$, is the complete =
lack of information provided to the end-user with useless error messages =
and no intutive way to view the system to debug stuff...] letting me =
know I hadn't created the initial "god" MySQL user.  Re-running the =
script after remedying this - it WORKED!  I could log into MySQL or use =
the GUI, "Query Browser", to see the new "pimppa" database and the new =
tables created.=20

Now, I'm thinking to myself, I should be able to run Bowser.  And sure =
enough, it loads without error.  I go into its "preferences" and make =
some 'intelligent' choices  for configuring NNTP, etc.  I setup a =
newsgroup where I know there are a very small number of files just to =
test whether this will download anything.  I fire it off, up pops a =
terminal-like window letting me know what's happening - but it doesn't =
show anything happening after finding the group (long story - turns out =
there is a problem with Ubuntu and the GUI window - there's a lot of =
seg-faults and other issues using Bowser on Ubuntu [I'm running Dapper =
Beta, bleeding edge, so I kinda expect some problems...]).  After quite =
some time, I go back into non-GUI mode and see if I can figure out =
what's happening.  Looking through the files in the "pimppa" directory =
'cat'ing every file that I can, I discover the 'pleech' command and try =
'pleech --help' to discover this is a tool to invoke the downloading of =
files.  I run 'pleech -v' with no other arguments assuming (correctly, =
in this case, I've learned later) this will do the same thing Bowser did =
- sure enough, I got the same messages and data I saw in the =
Bowser-spawned output window.  Only this time, I saw that during the =
initial parsing there was an area being updated with progress numbers =
that didn't display in the Bowser-spawn.  I also noted behavior and =
determined there was a set of phases that was significantly different =
than the tool I had been familiar with in the Windows world.  [Side note =
- there is a tool (SBNews, have used this product for over 6 years) I =
can fire from the NT command-line AT scheduler that I really liked in =
the past but since I've now completely eliminated Winders products from =
most of my home net, I was looking for something similar and hopefully =
better in the open-source world.  This win tool had quite a usable GUI =
for configuration setting and could be used for leeching, but that =
wasn't it's prime purpose - it was a tool to be fired from the =
commandline and just run until it sucked down all the latest files (all =
of the new ones since the last time the tool was run) and quit.  I lived =
with all the duplicates and garbage files, wasting tons of time making =
tons of scripts and stuff to combat elimination of all this chaff.  but, =
all in all, it worked.  I had to turn it off from time to time due to =
running out of disk space before having time to go through the files =
determining what I wanted to keep and what to trash, what was good, what =
was incomplete, what was junk - and then burning the "good stuff" off to =
CD...  Ah, those were the days - I now have hundreds of CDs, hundreds of =
Gigs of useless, outdated stuff that was great state-of-art stuff back =
then...  But, I digress...]

So, I "saw" my initial attempt in config actually work, I saw articles =
flash by in the logging data and I saw that all articles were flagged as =
"NO MATCH" and ultimately saw that one file did download, the database =
got updated.  So, at this point I knew I had it working, but this tool =
was a "black box" for the moment and I knew I needed to dig further.  =
Also, I noted from initially visiting your 'Sourceforge' site, the =
version I'm running downloaded from some Ubuntu place was 0.5.7 and that =
'Sourceforge' site had 0.5.8 available.  After downloading the source =
from your site and extracting the contents, I was able to ascend the =
ranks of enlightenment!!!  I found the text file, 'SETUP' that gave me =
all the background and infrastructure information I needed to 'properly' =
configure the data contained in the pimppa tables and all the =
definitions and descriptions and options I need to use the software from =
the command-line.  It was the elusive SETUP file I needed all along!!! =20

I have now been up and running, happily leeching away while I sleep, as =
it were  :).  This tool IS what I was looking for and it did take some =
rocket science in getting it going - only because the GUI wasn't as =
useful as I assumed it to be and the initial error message confounding.  =
I feel amazed I was able to get something to work before I actually =
broke down, downloaded the latest source, and found the SETUP file.  As =
to whether there are better open-source tools out there - I haven't =
found any...  I suppose newsgroups are not in the realm of popularity as =
peer-to-peer stuff like Kazaa, Azureus and Torrents...

Thanks for the time and effort you took in creating this tool.  Perhaps, =
when I have a little time from my research and life's obligations, I'll =
assist in creating better documentation so the next person would have to =
go through the frustration I did to begin with (one of my many titles is =
"technical writer"...).

Cheers back at you,
Laz

-----Original Message-----
From: Igor Wronsky [mailto:iwr...@us...]
Sent: Sat 5/27/2006 8:48 AM
To: LASZLO ACS
Cc: pim...@li...
Subject: Re: [pimppa-users] I must be dense...
=20
On Wed, 24 May 2006, LASZLO ACS wrote:

> So, I guess this is one of those rocket-science tools where ya gotta =
be
> a whiz-bang genius geek or something. I do some searching (my soul and
> elsewhere), finally stumbling upon your PIMPPA home on the web.  I =
look
> at the site for a while, stumbling on a page that holds the "READ ME"
> document - GASP!  This looks really complicated...
> I guess I'll need to take a vacation to get this working...

As usually is the case with open source software, I made it for
my own use originally, and slick setup wasn't one of the most
important design criteria. :) For those who know their way
around Linux, its not too difficult. The text file SETUP included
in the original distribution archive (from pimppa.sourceforge.net,
for example) contains step-by-step instructions for the
installation starting from the compilation of the source code.

When it comes to distros like Ubuntu, I'm happy if its accessible
from them, but I feel its the role of the package maintainers
to make the installation procedure compatible with the general
way of doing things in the resp. distribution, or, to provide a
small piece of documentation how the package is actually
made to work there. I'm not currently personally maintaining
any binary packages for any distribution.

> I sure would appreciate a bit of assistance from someone having a =
sense
> of humor and patience while I continue to stumble through ...

One thing that comes to mind is that many drops of water have
passed under the bridges of the world since I originally coded
PIMPPA. It could be that there are sexier and easier to use
software packages available for news-harvesting today. I haven't
checked lately, I still run pimppa by CRON and I'm reasonably
happy with it. But it might be a good idea to take a look at the

competitors. I tried to provide some links on pimppa homepage
to other similar software, but I don't know if they're the
state of the art any more. :)

Cheers,
Igor

Thanks, Igor, for your reply.

Since writing, I did figure out the source of my confusion.  It seemed =
that since I had tried using the Ubuntu "Synaptic" GUI to give me joy, =
and when that didn't happen from the initial 1st attempt in running =
Bowser by popping up an error message, I dug a little into the web and =
found your site to discover the "README" document.  The references your =
README document made to SETUP confused me since I didn't find that file =
in my system.  Getting MySQL up and running was a piece of cake; after =
that, I assumed clicking on and running Bowser would detect whether =
MySQL was running and take care of whatever was necessary to get the =
user up and running with the software.  No joy...  I still got the same =
error message as before.  I returned to the README file and perused it =
with a fine-toothed comb (ya know how hard this is with nappy and =
tangled substrates???)...  I searched for the files on my machine, =
eventually locating some of the scripts in a directory called "pimppa" =
(amazingly familiar).  There I found "create_database.sql" - hmmm...  =
viewing the file in my fav editor, I  got some insight to what was next. =
 I ran the script, it failed but gave me a useful error message [One of =
the things I absolutely abhor about the monopoly, M$, is the complete =
lack of information provided to the end-user with useless error messages =
and no intutive way to view the system to debug stuff...] letting me =
know I hadn't created the initial "god" MySQL user.  Re-running the =
script after remedying this - it WORKED!  I could log into MySQL or use =
the GUI, "Query Browser", to see the new "pimppa" database and the new =
tables created.=20

Now, I'm thinking to myself, I should be able to run Bowser.  And sure =
enough, it loads without error.  I go into its "preferences" and make =
some 'intelligent' choices  for configuring NNTP, etc.  I setup a =
newsgroup where I know there are a very small number of files just to =
test whether this will download anything.  I fire it off, up pops a =
terminal-like window letting me know what's happening - but it doesn't =
show anything happening after finding the group (long story - turns out =
there is a problem with Ubuntu and the GUI window - there's a lot of =
seg-faults and other issues using Bowser on Ubuntu [I'm running Dapper =
Beta, bleeding edge, so I kinda expect some problems...]).  After quite =
some time, I go back into non-GUI mode and see if I can figure out =
what's happening.  Looking through the files in the "pimppa" directory =
'cat'ing every file that I can, I discover the 'pleech' command and try =
'pleech --help' to discover this is a tool to invoke the downloading of =
files.  I run 'pleech -v' with no other arguments assuming (correctly, =
in this case, I've learned later) this will do the same thing Bowser did =
- sure enough, I got the same messages and data I saw in the =
Bowser-spawned output window.  Only this time, I saw that during the =
initial parsing there was an area being updated with progress numbers =
that didn't display in the Bowser-spawn.  I also noted behavior and =
determined there was a set of phases that was significantly different =
than the tool I had been familiar with in the Windows world.  [Side note =
- there is a tool (SBNews, have used this product for over 6 years) I =
can fire from the NT command-line AT scheduler that I really liked in =
the past but since I've now completely eliminated Winders products from =
most of my home net, I was looking for something similar and hopefully =
better in the open-source world.  This win tool had quite a usable GUI =
for configuration setting and could be used for leeching, but that =
wasn't it's prime purpose - it was a tool to be fired from the =
commandline and just run until it sucked down all the latest files (all =
of the new ones since the last time the tool was run) and quit.  I lived =
with all the duplicates and garbage files, wasting tons of time making =
tons of scripts and stuff to combat elimination of all this chaff.  but, =
all in all, it worked.  I had to turn it off from time to time due to =
running out of disk space before having time to go through the files =
determining what I wanted to keep and what to trash, what was good, what =
was incomplete, what was junk - and then burning the "good stuff" off to =
CD...  Ah, those were the days - I now have hundreds of CDs, hundreds of =
Gigs of useless, outdated stuff that was great state-of-art stuff back =
then...  But, I digress...]

So, I "saw" my initial attempt in config actually work, I saw articles =
flash by in the logging data and I saw that all articles were flagged as =
"NO MATCH" and ultimately saw that one file did download, the database =
got updated.  So, at this point I knew I had it working, but this tool =
was a "black box" for the moment and I knew I needed to dig further.  =
Also, I noted from initially visiting your 'Sourceforge' site, the =
version I'm running downloaded from some Ubuntu place was 0.5.7 and that =
'Sourceforge' site had 0.5.8 available.  After downloading the source =
from your site and extracting the contents, I was able to ascend the =
ranks of enlightenment!!!  I found the text file, 'SETUP' that gave me =
all the background and infrastructure information I needed to 'properly' =
configure the data contained in the pimppa tables and all the =
definitions and descriptions and options I need to use the software from =
the command-line.  It was the elusive SETUP file I needed all along!!! =20

I have now been up and running, happily leeching away while I sleep, as =
it were  :).  This tool IS what I was looking for and it did take some =
rocket science in getting it going - only because the GUI wasn't as =
useful as I assumed it to be and the initial error message confounding.  =
I feel amazed I was able to get something to work before I actually =
broke down, downloaded the latest source, and found the SETUP file.  As =
to whether there are better open-source tools out there - I haven't =
found any...  I suppose newsgroups are not in the realm of popularity as =
peer-to-peer stuff like Kazaa, Azureus and Torrents...

Thanks for the time and effort you took in creating this tool.  Perhaps, =
when I have a little time from my research and life's obligations, I'll =
assist in creating better documentation so the next person would have to =
go through the frustration I did to begin with (one of my many titles is =
"technical writer"...).

Cheers back at you,
Laz

-----Original Message-----
From: Igor Wronsky [mailto:iwr...@us...]
Sent: Sat 5/27/2006 8:48 AM
To: LASZLO ACS
Cc: pim...@li...
Subject: Re: [pimppa-users] I must be dense...
=20
On Wed, 24 May 2006, LASZLO ACS wrote:

> So, I guess this is one of those rocket-science tools where ya gotta =
be
> a whiz-bang genius geek or something. I do some searching (my soul and
> elsewhere), finally stumbling upon your PIMPPA home on the web.  I =
look
> at the site for a while, stumbling on a page that holds the "READ ME"
> document - GASP!  This looks really complicated...
> I guess I'll need to take a vacation to get this working...

As usually is the case with open source software, I made it for
my own use originally, and slick setup wasn't one of the most
important design criteria. :) For those who know their way
around Linux, its not too difficult. The text file SETUP included
in the original distribution archive (from pimppa.sourceforge.net,
for example) contains step-by-step instructions for the
installation starting from the compilation of the source code.

When it comes to distros like Ubuntu, I'm happy if its accessible
from them, but I feel its the role of the package maintainers
to make the installation procedure compatible with the general
way of doing things in the resp. distribution, or, to provide a
small piece of documentation how the package is actually
made to work there. I'm not currently personally maintaining
any binary packages for any distribution.

> I sure would appreciate a bit of assistance from someone having a =
sense
> of humor and patience while I continue to stumble through ...

One thing that comes to mind is that many drops of water have
passed under the bridges of the world since I originally coded
PIMPPA. It could be that there are sexier and easier to use
software packages available for news-harvesting today. I haven't
checked lately, I still run pimppa by CRON and I'm reasonably
happy with it. But it might be a good idea to take a look at the

competitors. I tried to provide some links on pimppa homepage
to other similar software, but I don't know if they're the
state of the art any more. :)

Cheers,
Igor

On Wed, 24 May 2006, LASZLO ACS wrote:

> So, I guess this is one of those rocket-science tools where ya gotta be
> a whiz-bang genius geek or something. I do some searching (my soul and
> elsewhere), finally stumbling upon your PIMPPA home on the web.  I look
> at the site for a while, stumbling on a page that holds the "READ ME"
> document - GASP!  This looks really complicated...
> I guess I'll need to take a vacation to get this working...

As usually is the case with open source software, I made it for
my own use originally, and slick setup wasn't one of the most
important design criteria. :) For those who know their way
around Linux, its not too difficult. The text file SETUP included
in the original distribution archive (from pimppa.sourceforge.net,
for example) contains step-by-step instructions for the
installation starting from the compilation of the source code.

When it comes to distros like Ubuntu, I'm happy if its accessible
from them, but I feel its the role of the package maintainers
to make the installation procedure compatible with the general
way of doing things in the resp. distribution, or, to provide a
small piece of documentation how the package is actually
made to work there. I'm not currently personally maintaining
any binary packages for any distribution.

> I sure would appreciate a bit of assistance from someone having a sense
> of humor and patience while I continue to stumble through ...

One thing that comes to mind is that many drops of water have
passed under the bridges of the world since I originally coded
PIMPPA. It could be that there are sexier and easier to use
software packages available for news-harvesting today. I haven't
checked lately, I still run pimppa by CRON and I'm reasonably
happy with it. But it might be a good idea to take a look at the
competitors. I tried to provide some links on pimppa homepage
to other similar software, but I don't know if they're the
state of the art any more. :)

Cheers,
Igor

I'm a raw recruit...
=20
I've got a newly installed Ubuntu workstation and did a search on the =
net for a way to automate newsgroup activity.  I learned about this =
PIMPPA thingy.  So, I fire up Synaptic and see that yup, there's an =
entry in the list for this software.  I click apply, it goes out in the =
world, installs everything (I expect), and sure enough, I now have =
"Bowser" up there (I read this is the GUI for this PIMPPA thingy).  =
Wanting to get going RIGHT NOW - I select the Bowser icon and I get an =
error message:
           "src_db mysql connect fail"
=20
Now, I know I ain't done nothing to install or set up MySQL.  I do =
recall seeing a ton of things in the dependency list, but not that...
=20
So, I guess this is one of those rocket-science tools where ya gotta be =
a whiz-bang genius geek or something.
=20
I do some searching (my soul and elsewhere), finally stumbling upon your =
PIMPPA home on the web.  I look at the site for a while, stumbling on a =
page that holds the "READ ME" document - GASP!  This looks really =
complicated...
=20
I guess I'll need to take a vacation to get this working...
=20
Could someone give me a step-by-step EASY HOW-TO guide to getting this =
working???  What's the order of operation - what do I need to get this =
running?  I've already figured out that I'll need to dive into getting =
MySQL running (and that's about as far as I know) ...
=20
I sure would appreciate a bit of assistance from someone having a sense =
of humor and patience while I continue to stumble through ...
=20
Zolo

On Wed, 11 May 2005, Neal Lafferty wrote:

> Hey all,
>      Just a quick question for everyone.  For adding a new server or
> a new group, there is pnewsrv and pnewgrp.  But I don't see how I
> would remove a group or server.
> How would I go about removing a group from the list, or removing a
> server?

You can use the GUI 'bowser', from there select 'preferences'.
If you can't use the GUI, then you'll have to hit mysql directly,

$ mysql
mysql> use pimppa;
mysql> delete from p_groups where g_name='somegroup';
mysql> delete from p_servers where s_name='someserver';

Thats it. And you can see the currently defined servers and groups
from mysql by e.g.

mysql> select s_name from p_servers;
mysql> select g_name from p_groups;

... I'm not saying this is the best interface but hopefully
those ops don't have to be used often. :)

Cheers,
Igor

Hey all,
     Just a quick question for everyone.  For adding a new server or  
a new group, there is pnewsrv and pnewgrp.  But I don't see how I  
would remove a group or server.
How would I go about removing a group from the list, or removing a  
server?

Thanks

N

On Wed, 14 Jan 2004, Karsten Prey wrote:

> Hello,
> despite the quietness on this mailing list, can somebody please help.
> I have just set up pimppa with the example newsgroup of alt.binaries
> pictures.cartoons. Everything looks fine so far only pleech doesn't
> download anything.

To me, this looks like "suck" that pimppa uses for news
leeching and your news server do not like each other. You
perhaps might be able to affect this by reading "man suck"
and changing the values P_SUCKOPTS_HEADER and P_SUCKOPTS_BODY
in src/pimppa.h and recompiling. I'm not certain from
your debug output whether the actual failure is in
header download or body download phase or both. You might for
example try to remove the '-n' option from P_SUCKOPTS_HEADER
and recompile. Or maybe adding '-Z' to either might help. Note
that I'm only guessing here, I haven't seen this previously
and I don't know what the real problem is. If you can
download messages using "suck" (it can be used separately too)
with your newsserver, then its probably the way pimppa uses it
that is wrong and I really should fix it, if not, then the
problem is suck-specific and should be reported to the
author of suck.

Hope this helps any.

Igor

I had that same problem before and it was something I had setup wrong. 
The computer I have it setup on is dead right now so I can't be much help.

Karsten Prey wrote:
> Hello,
> despite the quietness on this mailing list, can somebody please help.
> I have just set up pimppa with the example newsgroup of alt.binaries 
> pictures.cartoons. Everything looks fine so far only pleech doesn't 
> download anything.
> Thanks a lot
> Karsten Prey

Hello,
despite the quietness on this mailing list, can somebody please help.
I have just set up pimppa with the example newsgroup of alt.binaries 
pictures.cartoons. Everything looks fine so far only pleech doesn't 
download anything.
Thanks a lot
Karsten Prey

Messages (extract) are as follows:

Server:~$ pleech -v
[PIMPPA] OPT_VERBOSE enabled
[PIMPPA] Date: Wed Jan 14 17:08:20 2004
[PIMPPA] <= Headers of alt.binaries.pictures.cartoons
Attempting to connect to news1.surfino.com
Using Port 119
Official host name: treonet.news-service.com
Alias news1.surfino.com
Address: 195.114.230.35
Address: 195.114.230.36
Connected to treonet.news-service.com
200 nf2 NNTPSwitch-0.09-CVS-20031229, 60158 groups available, posting 
allowed, slot 534, connections 1
alt.binaries.pictures.cartoons - 767 articles 124419-125185
Elapsed Time = 0 mins 0.84 seconds
732 Articles to download
Deduping Elapsed Time = 0 mins 0.00 seconds
Deduped, 732 items remaining, 0 dupes removed.
/var/lib/news/history: No such file or directory
Total articles to download: 732
***Unexpected response to command, head 124420

423 Bad article number

***Unexpected response to command, head 124421

423 Bad article number

***Unexpected response to command, head 124422

423 Bad article number

***Unexpected response to command, head 124424

423 Bad article number

< counting numbers and a lot of error messages as follows >
.....

***Error - no such article - article <_IWKb.775180$Fm2.752523@attbi_s04>

***Error - no such article - article <Fb7Lb.343$ke...@ne...>

***Error - no such article - article <kaXKb.758842$HS4.5994359@attbi_s01>

***Error - no such article - article 
<UhhMb.5565$zj7...@ne...>

***Error - no such article - article 
<ZELFb.2781212$be....@ne...>

***Error - no such article - article 
<alt.binaries.pictures.cartoons@236.252.124.184>

***Error - no such article - article 
<dis...@he...>

***Error - no such article -
34374 Bytes received in 2 mins 32.58 secs, BPS = 225.3
Closed connection to news1.surfino.com
Cleaning up after myself
[PIMPPA] Decoding to /tmp/pleech.7917/decode_tmp
[PIMPPA] Launching uudeview at 1074096756
[PIMPPA] uudeview took 0 sec
[PIMPPA] Processing decoded files...
Ok, 0 files and 0 bytes gained, 0 files and 0 bytes were deleted.
[PIMPPA] Processing took 0 secs

Finished: Wed Jan 14 17:12:36 2004

Gained Files: 0

On Sat, 15 Nov 2003, Suran wrote:

> I just can't find out how to get pimppa to ignore images in a
> binary-newsgroup where artists post their work based on the sender as
> there are some artists that I just don't like the style of. Matching on
> the filename or the subject does not work here.

That is not explicitly supported. :( But you could probably hack
the same result by using suckkillfile. Find out the place where
pimppa put the file "suckkillfile". Then make a new file
with lines like (confirm that this usage is correct from "man suck")

HEADER: From:.*junkartist.*
HEADER: From:.*stewman.*

etc. and replace the original file with the file you made.
Pimppa should now skip the messages listed in that file.

One problem with this approach is that if you wish different
killfiles for different groups, you should have multiple suckkillfiles
and copy the correct one over before using pleech... I might
change this in the future to be more user friendly.

If you bother to try it, let me know if it works or not. ;)

Igor

For those that may still be reading these lists,

PIMPPA v0.5.7 has been released after a while
of inactivity. But hey, if you can get PIMPPA
running, it seems to run fairly well, atleast
for me. And no mindblowing patch submissions
or suggestions have arrived on my doorstep. ;)

Here's the ChangeLog:

 v0.5.7
 ------
 - More options for suck can now be appended in "src/pimppa.h".
 - Removed "ydec" dependency, as current uudeview supports yyencode
 - Added a table defining the existing contexts + some
   bowser combo boxes.
 - Added keyword hunting behaviour (accept only messages
   that contain specified keyword pattern in the subject line),
   pleech switch '-k'. See README.
 - Tweaked autoconf stuff
 - src/news.c didn't create directories properly if the
   last path component wasn't '/'. Fixed.
 - (Group in "~/.my.cnf" is now by default [pimppa], not [client])
   This change was not good!!! Backtracked.
 - Tried to fix mysql_init() related crashes seen on
   some MySQL-dists [but it appeared to be a bug in mysql v3.23.49].
 - Tried to fix build for FreeBSD
 - When parsing filename from the subject: line, the
   parsing is now stopped at " -character (yenc stuff
   is often in "file.jpg" style format on the subject line)
 CHANGE:
   bash> mysql -u root -p < sql/056to057.sql

Thats all, folks. Happy leechings...

Igor

It seems that for a long time, there has been a bug
in Suck that can cause file corruption together with
PIMPPA. This problem has recently been more pronounced
because of increasing amounts of yEnc postings that
this bug especially likes to destroy. I'm sorry that
I didn't notice this earlier (and no-one reported). :(

The fix for Suck will probably appear in its next
version. Meanwhile, if you wish to get more good
files out of your PIMPPA system, download suck 4.3.1
sources and use the included patch as

$ cd suck-4.3.1; patch -p0 <suck_431_dotbug.patch

and recompile/reinstall. Note that the bug - if not
fixed - can also break uuencoded content.

 - Igor

Though pimppa has been working reasonably well around here,
recently I had trouble with yEncoded files. Either the nntp 
servers have been mangling the stuff, nonstandard encoders 
have been used or the standard has drifted from the current 
linux decoders. It always appears as something on the lines of

/tmp/yydecode: ../messages.txt: evilporn.jpg:1: Part longer than 
expected

... and in the end (especially as it was a 200mb file to
be processed and decoded) 

/tmp/yydecode: ../messages.txt: yucky.jpg:1: Too many open files

This behaviour was identical to both ydec (which pimppa
is currently using) and yydecode. In addition, I tried
something called just 'ydecode', but that never processed
more than one file from the archive and is thus not
straighforwardly applicable to pimppa.

Comments on this, anyone?

Anyway, the point of this message is twofold: 1) be careful
with those yenc bastards 2) if you have trouble, find out 
if some decoder can handle the file. If you can find a decoder 
that works better than the current, please report, and I'll 
change the default.

Igor

How exactly would one create a rule/filter in PIMPPA to only grab jpegs that 
are 200kb or larger ?

Garf

Occasionally when I am running pimppa my box locks up.
 It appears to happen in the leech stage.  Has anyone
else experienced this, and if so, what corrective
measures have you taken?  I am running Red Hat 7.2 on
a 700 MHZ intel machine.

Thanks,

Jason

__________________________________________________
Do You Yahoo!?
Yahoo! Movies - coverage of the 74th Academy Awards®
http://movies.yahoo.com/

On Thu, 21 Mar 2002, John Doe wrote:

> Now... the next problem is... the gui will not compile. I get an error during
> ./configure that tells me that GNOME is not installed and the GUI will not be
> compiled. I am running XIMIAN-GNOME so there must be a check that fails based
> on that. Any thoughts?

Argh. No. See "configure.in", after "dnl Gnome check" -line. It
should base the gnome detection on "gnome-config" which should
be executable and especially the output of "gnome-config gnomeui --cflags"
should print the necessary gnome compilation flags for your
system.

The problem can be fixed by replacing the current routine with
something better. Perhaps if you know a configure script (from 
some other program, perhaps?) that correctly detects 
XIMIAN-GNOME, send me that and I'll try to tweak this. Or does
the GNOME org provide some configure-script template
for all gnomes?

Meanwhile, you can use the system without the gui, all
functionality can be accessed from command line or mysql.

>>>>
  Did you use the RPM or really compile from source? If you did compile
yourself, then all I can think of is that the gcc3 and/or =

  the libs coming with RH7.2 are broken. gcc coming with RH7.2 =

is old, as 3.0.4 =

>>>>>
OK... solved my previous problem... I uninstalled the .rpm and compiled f=
rom
the tarball... the rpm doesn't work on redhat with gcc 3.0.4, which is wh=
at I
have.

Now... the next problem is... the gui will not compile. I get an error du=
ring
=2E/configure that tells me that GNOME is not installed and the GUI will =
not be
compiled. I am running XIMIAN-GNOME so there must be a check that fails b=
ased
on that. Any thoughts?

--Ishtek

On Tue, 19 Mar 2002, John Doe wrote:

> I have installed the latest version of Pimppa on a RH 7.2 system. Everything
> compiled and installed correctly but when I try and run bowser or any of the
> command line tools I get an error:
> bowser: /lib/libc.so.6 version:'GCC_3.0' not found (required by bowser)
> I have libc.so.6 in my /lib directory so it appears that the requirements are
> met but I'm not sure what is causing the issue.

Did you use the RPM or really compile from source? If you did
compile yourself, then all I can think of is that the gcc3 and/or 
the libs coming with RH7.2 are broken. gcc coming with RH7.2 
is old, as 3.0.4 and 3.1.0 are available at least. Check

http://www.rpmfind.net/

and try some never gcc and the associated libs. I used
gcc3.0.3 myself which might cause .rpm -related troubles.

Anyone else with this problem? Or anybody successfully using 
v0.5.6, for that matter? ;D

 - I. 

I have installed the latest version of Pimppa on a RH 7.2 system. Everyth=
ing
compiled and installed correctly but when I try and run bowser or any of =
the
command line tools I get an error:

bowser: /lib/libc.so.6 version:'GCC_3.0' not found (required by bowser)

I have libc.so.6 in my /lib directory so it appears that the requirements=
 are
met but I'm not sure what is causing the issue.

--Ishtek

Hi

I've been jumping through hoops trying
to get pleech in 0.5.5a working to no avail.

I even thought my database might be corrupted
so I started from scratch.

I keep getting mysql errors when tryng to run it.

I edited common.c to include debug information comparing 0.5.4 and 0.5.5a.

Here is the output from 0.5.4

FROM p_groups, p_areas, p_servers
WHERE g_dest=3Darea_id AND g_server=3Ds_id
AND s_name like '%%'

AND NOT (g_flags & 1)
ORDER BY g_name

Here is the output from 0.5.5a

FROM p_groups, p_areas, p_servers
WHERE g_dest=3Darea_id AND g_server=3Ds_id
AND s_name like '%%'
=C2
AND NOT (g_flags & 1)
ORDER BY g_name

It appears that in this part of news.h ...

p_query(&dst_db, "SELECT g_name, g_last, area_path, area_id, s_name, s_id=
,
s_user, s_pass, area_name
FROM p_groups, p_areas, p_servers
WHERE g_dest=3Darea_id AND g_server=3Ds_id
AND s_name like '%s'
%s
AND NOT (g_flags & %ld)

... some crap is being assigned to the second %s
and fsking up the mysql string.

It seems that the problem lies somewhere in common.c
but I'll be darned if I can find it.

One other thing...

this line in sql/uninstall_db

mysql -u root -p <uninstall.sql

should be

mysql -u root -p <uninstall_db.sql

Boz

---
4.12.2001 V.I.P. Chat Live s
Terry Gilliamem - MONTY PYTHON=20
http://chat.volny.cz

Okay, you insatiable leechers. I have a new pimppa
release for you. The most important feature is a
change on the security side, making the system more
usable in a multiuser system (e.g. enabling you
to prevent your MOTHER from accessing your pimppa db). 

Here's the complete changelog.

----cut---
 - Enhanced security, especially important for multiuser systems. 
   Accessing pimppa DB now has a password, but you don't have to type
   it all the time, as it is automatically read from a file.
 CHANGE: Edit ~/.my.cnf to contain 

   ---cut---
   [client]
   user=pimppa
   password=your_new_pw
   ---cut---

   and make sure other people can't read the file, e.g.
   $ chmod 0600 ~/.my.cnf

   Change mysql password for the user:

   $ mysql -u root -p --execute="USE mysql; UPDATE user \
     SET password=PASSWORD('your_new_pw') WHERE user='pimppa';"

   If you already have some mysql user other than that, you 
   probably know what you're doing; just give that user priviledges 
   to the pimppa database.
 - More intelligent (& conservative) free space check; pimppa
   doesn't now download more messages than the tmpdir has 
   space for tmpfile and decoding.
 - Filenames from article subjects in format like "C:\bull\bull10.jpg" 
   will now be correctly extracted without the path.
 - Entering context in pnewarea was broken, fixed
 - Changed Bowser add_newsgroup default server id to 1
 - Changed the Bowser selection mode to one where you can select 
   multiple files by pressing down ctrl, and select blocks of 
   files by pressing down shift. Thanks to Boz Boz for idea.
 - Added "-n" to suck header leech phase. Seems to speed things up.
 - Viewing scripts now read the file viewer from p_misc 
   key CFG_VIEWER and default to gqview if not found. (Thanks to I.O.)
----cut----

The files are available from 

http://pimppa.sf.net/

as usual. Keep 'em bugreports etc coming.

  - Igor

Thank you for your comments.

I see what is happening with the other
problem I mentioned. Every once in a while
I'll end up with files listed in the
database which are not in the incoming
area they are supposed to be in.

Looking through my logs I see what is happening.

For some reason, (uudeview not decoding them?),
some files will not be moved from the
/decode_tmp/ area.

p_cp(): Unable to open /tmp/pleech.4133/decode_tmp/test.jpg
p_mv() error from /tmp/pleech.4133/decode_tmp/test.jpg to /incoming_directory/test.jpg

But the filename is still entered into the
database as being moved to the incoming directory.

Trying to move them from the incoming area
gives an error, but the in the database they are
listed as being moved. So the filename exist in the
database as being in the area it should have been moved to,
but the file does not exist in the actual directory.

I'm still trying to work my way around through your code :-)

In the meantime, what I have done is add
pmarkoff
pclean -o

to p_maint

Thanks again

Boz

---
4.12.2001 V.I.P. Chat Live s
Terry Gilliamem - MONTY PYTHON 
http://chat.volny.cz

On Tue, 11 Dec 2001, boz boz wrote:

> Then I realized that in crontab, at least in my crontab, the full path
> to an executable must be used in a command or else it won't find it.
> Crontab is a limited environment and doesn't know what the paths to
> executables are unless you tell it.

In the beginning of the crontab, just add

PATH=/bin:/usr/bin:/usr/local/bin:/usr/whatever:/blah/blah
                   ^^^^^^^^^^^^^^
etc ... ;) There's probably some path defined already. Just
add the usr/local/bin to it... Though it could be that
the path is not exported to the programs ran from crontab
in your system. Dunno.

> Since ptest couldn't read anything from the tmpfile it assumed the
> file tested bad and marked it as such.

No. That's not it. The tmpfile is not used at all if no
'ok string' is defined - then just the return value of
the command is used (some test programs do not follow
this return value mechanism, and we have to read result from 
the tmpfile). But in this case a return value '0' from 
jpeginfo means that the file was correct, and if the command
was not found, like in this case, shell returns something
nonzero, which usually in the unix world means an error,
and ptest thinks similarly. It could be adjusted to
check for 'command not found' too but I haven't done
that... 

Ok. End of lecture. ;)

>
> > when p_maint is run from crontab, jpeginfo
> will
> > mark all new files as corrupt. This probably
> has more
> > to do with jpeginfo, than pimppa.
>
> If you're using settings in
> 'example_types.sql',
> you need jpeginfo v1.5a or later. It should be
> the
> one by Timo Kokkonen. If you have that, I
> don't
> know whats wrong then.
>

Hey, I figured this bug out.

sql/example_types.sql contains this line for example

INSERT INTO p_types VALUES ('JPG','jpeginfo -c',0,'','jpegoptim','jpg');

I found that ptest was calling jpeginfo by the command line

"jpeginfo -c /pathtofile > tmpfile" 2>dev/null

It was after I removed the 2>/dev/null that I saw the error
"command not found"

Then I realized that in crontab, at least in my crontab, the full path to an executable must be used in a command or else it won't find it. Crontab is a limited environment
and doesn't know what the paths to executables are unless you tell it.

Since ptest couldn't read anything from the tmpfile it assumed the file tested bad and marked it as such.

Changing sql/example_types.sql to the following made ptest run ok from crontab

INSERT INTO p_types VALUES ('JPG','/usr/local/bin/jpeginfo -c',0,'','/usr/local/bin/jpegoptim','jpg');

Thanks Again

Boz

---
V.I.P. Chat - Miro Zbirka
11.12.2001 od 14:00 na
http://chat.volny.cz

2000	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug (1)	Sep (3)	Oct	Nov	Dec
2001	Jan	Feb	Mar (2)	Apr	May (15)	Jun	Jul (1)	Aug	Sep	Oct	Nov (2)	Dec (7)
2002	Jan	Feb	Mar (6)	Apr	May	Jun	Jul (1)	Aug	Sep	Oct	Nov	Dec (1)
2003	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct (1)	Nov (1)	Dec
2004	Jan (3)	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec
2005	Jan	Feb	Mar	Apr	May (2)	Jun	Jul	Aug	Sep	Oct	Nov	Dec
2006	Jan	Feb	Mar	Apr	May (4)	Jun	Jul	Aug	Sep	Oct	Nov	Dec
2016	Jan	Feb (1)	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec

pimppa-users Mailing List for PIMPPA

pimppa-users — General discussion list for users of pimppa.