RT 2.1.46 - We've been autoconfiscated

One big change here…autoconf support has landed.

The other big thing that will happen to the installation process before
3.0 is that all the database creation stuff will get broken out as a
seperate binary that can be run seperately from the build/install
process.

All this should make packaging suck much less.

Commentary or patches to the autoconf process are definitely
appreciated.

-j

»|« http://www.bestpractical.com/rt – Trouble Ticketing. Free.

Yikes. I didn’t mean to not credit the autoconf stuff.
Autoconfiscation would have taken much more time and been much more
painful if I didn’t have darren chamberlain’s autoconfiscation of RT 2.0
to base things on.

Best,
JesseOn Wed, Nov 13, 2002 at 12:40:31AM -0500, Jesse Vincent wrote:

One big change here…autoconf support has landed.

The other big thing that will happen to the installation process before
3.0 is that all the database creation stuff will get broken out as a
seperate binary that can be run seperately from the build/install
process.

All this should make packaging suck much less.

Commentary or patches to the autoconf process are definitely
appreciated.

-j


»|« http://www.bestpractical.com/rt – Trouble Ticketing. Free.

»|« http://www.bestpractical.com/rt – Trouble Ticketing. Free.

At 12:40 AM 13/11/2002 -0500, Jesse Vincent wrote:

One big change here…autoconf support has landed.

Minor issue in Makefile.in:

DB_PORT = `@DB_PORT@

(the open quote may be a problem).

Philip Warner | ___
Albatross Consulting Pty. Ltd. |----/ -
(A.B.N. 75 008 659 498) | /(@)
Tel: (+61) 0500 83 82 81 | _________
Fax: (+61) 03 5330 3172 | ___________ |
Http://www.rhyme.com.au | / |
| –
___–
PGP key available upon request, | /
and from pgp5.ai.mit.edu:11371 |/

Ugh. Autoconf.

:slight_smile:

Jesse Vincent writes:

Ugh. Autoconf.

Come on, m4 rules!

But seriously, I don’t know of any tools that do the job better.

:slight_smile:

(darren)

There are worse things in life than death. Have you ever spent an
evening with an insurance salesman?
– Woody Allen

Ugh. Autoconf.

Come on, m4 rules!

But seriously, I don’t know of any tools that do the job better.

:slight_smile:

If there was something that did the job better and people were
expecting, I’d jump on it. As it is, I discussed a number of alternative
plans with a bunch of very clueful folks…who shot holes in them.
Autoconf may not be fantastic, but it appears to be the best current
option.

-j

»|« http://www.bestpractical.com/rt – Trouble Ticketing. Free.

At 12:40 AM 13/11/2002 -0500, Jesse Vincent wrote:

One big change here…autoconf support has landed.

The other big thing that will happen to the installation process before
3.0 is that all the database creation stuff will get broken out as a
seperate binary that can be run seperately from the build/install
process.

All this should make packaging suck much less.

Commentary or patches to the autoconf process are definitely
appreciated.

As a heads-up, the autoconf stuff could use a little work. I’m putting
something together now…

I’m using autoconf 2.54 – I hope no-one has a problem with this?

Cheers,

  • Andrew

As a heads-up, the autoconf stuff could use a little work. I’m putting
something together now…

Cool. what are you doing? (I’m still working on stuff myself)

I’m using autoconf 2.54 – I hope no-one has a problem with this?

My boxes seem to end up with vendor shipped 2.52 and 2.53. How much pain
am I going to be in?

Cheers,

  • Andrew

»|« http://www.bestpractical.com/rt – Trouble Ticketing. Free.

At 02:19 PM 16/11/2002 -0500, Jesse Vincent wrote:

As a heads-up, the autoconf stuff could use a little work. I’m putting
something together now…

Cool. what are you doing? (I’m still working on stuff myself)

  1. Cleaning up the output stage (it’s wrong).
  2. Looking into dropping Apache-style ‘layout’ support in. Using AC_WITH_ARG
    for directories is a bit naughty. Autoconf doesn’t support easy addition
    of more --blahdir options, so maybe the apache approach is better.
  3. Rename of variables, to follow the autoconf style.

I’m using autoconf 2.54 – I hope no-one has a problem with this?

My boxes seem to end up with vendor shipped 2.52 and 2.53. How much pain
am I going to be in?

Shouldn’t be too much pain – I’m using the docs at Gnu.org as my
reference, and they’re based on 2.53. I’ve added AC_PREREQ(2.53) to
document this.

I’ll try and have something ready by the end of the day to send in.

Cheers,

  • Andrew

At 02:19 PM 16/11/2002 -0500, Jesse Vincent wrote:

As a heads-up, the autoconf stuff could use a little work. I’m putting
something together now…

Cool. what are you doing? (I’m still working on stuff myself)

  1. Cleaning up the output stage (it’s wrong).

Thanks. I mostly cargo-culted this stuff, so that will be very useful.

  1. Looking into dropping Apache-style ‘layout’ support in. Using AC_WITH_ARG
    for directories is a bit naughty. Autoconf doesn’t support easy addition
    of more --blahdir options, so maybe the apache approach is better.

I’ll be interested to see what you come up with.

  1. Rename of variables, to follow the autoconf style.

Cool. Wil lthis touch the tags in other files?

I’ll try and have something ready by the end of the day to send in.

Excellent. That oughta be just about when I’m next cracking at this
stuff.

Best,
	Jesse

Cheers,

  • Andrew

»|« http://www.bestpractical.com/rt – Trouble Ticketing. Free.

At 02:19 PM 16/11/2002 -0500, Jesse Vincent wrote:

As a heads-up, the autoconf stuff could use a little work. I’m
putting something together now…

Cool. what are you doing? (I’m still working on stuff myself)

  1. Cleaning up the output stage (it’s wrong).

Thanks. I mostly cargo-culted this stuff, so that will be very useful.

So did I – please don’t mock my autoconf’ing too much. :slight_smile:

  1. Rename of variables, to follow the autoconf style.

Cool. Wil lthis touch the tags in other files?

I kept the exact names of the RT variables from the Makefile, in the
hopes that it would be easier for people used to modifying the Makefile
directly.

(darren)

The Feynman Problem Solving Algorithm:
1) Write down the problem.
2) Think real hard.
3) Write down the answer.