[Slackbuilds-users] intro, plus a query on packaging script suite

[Grant]

On Thu, 03 Aug 2006 01:26:55 -0500, Robby Workman <rworkman at slackbuilds.org> wrote:

...
>Welcome to the lists; if you get on IRC, come see us in 
>#slackbuilds on freenode.

I'm too old for IRC, cannot think / type fast enough ;)  Gave up 
trying IRC years ago...  
>
>In reference to junkview, I have to admit that it's a nice 
>project.

Thanks -- it has two known users, I'm one of them.

>  Honestly, I'm not sure if it's do-able in a way that 
>we'll be able to accept on the site.  It's not officially written 
>down anywhere, but the prevailing mentality among us was this: 
>users should be able to download all needed sources separately so 
>that they have the ability to do the install on a non-internet 
>connected box.  In this case, that may not be a valid concern, as 
>  the type of software pretty much assumes connectivity, but I'm 
>leery still - I mean, if you allow it once, then...  :/

The problem is that I have limited bandwidth, so I don't host 
the database files.  The tar exclusion lists (data/tar-X-*) allow 
creation of tarball with/without source database files.

>However, if the databases and other needed stuff are small, then 
>we can host them in the app's directory here, and the update 
>scripts can be included as part of the sources.  That would 
>(hopefully) allow for only one source tarball download (that of 
>the main junkview source), with the databases and other stuff, 
>including the SlackBuild script and README files hosted at SBo. 

Yep, I've not made a SlackBuild script (yet), and the install does 
scatter symlink files in quite a few locations, and I'm unsure 
where to put some files.  My thinking is that since this is not part 
of slackware, it belongs under /usr/local.  The project has no 
dependencies, no compile step.  It suits me here to symlink to the 
system so I can easily produce the tarball from the running project. 

That's where review will help, once I do the SB.

tarball with database source:
-rw-r--r-- 1 grant wheel  542372 2006-08-03 17:29 junkview+data.tar.bz2
-rw-r--r-- 1 grant wheel  573723 2006-08-03 17:29 junkview+data.tar.gz
without database:
-rw-r--r-- 1 grant wheel   17628 2006-08-03 17:31 junkview.tar.gz

I haven't looked at the database side for some time, a SlackBuild 
version could have the processed database files instead of the 
source data files?

-rw-r--r-- 1 grant wheel  541840 2006-08-03 18:02 junkview+database.tar.bz2

Ahh, not much of a size difference, no point -- the update script 
calls a data-downloader and a data-pre-processor script, so SlackBuild 
install script only needs to call the data-pre-processor script?

>If you decide to give it a try going this route, make sure you 
>have a good README file in the doc directory (and reference it in 
>the README that's hosted on our server) which explains the update 
>process.

cd /path/to/junkview/
./update.sh
su -c "/etc/rc.d/rc.junkview restart" # if using the ip2c-server
ccfind 123 # test the ip2c-server

I suppose the above could all be one script...
>
>Did all of that make any sense??  ;-)

Almost, I'll have to go through the SlackBuild process then see what 
reaction it stirs ;)

Grant.
-- 
<http://bugsplatter.mine.nu/>