Menu
▾
▴
htdig-dev
|
From: Neal R. <ne...@ri...> - 2003-07-07 16:25:00
|
On Sat, 5 Jul 2003, Lachlan Andrew wrote: > On Wed, 2 Jul 2003 04:24, Neal Richter wrote: > > > It is possible to create two different environments for the > > different DBs in the htdig classes that control them. > > True, but the "_weakcmpr" database is internally created by mp_cmpr.c > for any compressed database -- ht://Dig knows nothing about it. Of > course, we could *change* the API to pass in two database > environments (one for the database proper and one for _weakcmpr), but > that is far from fixing it at the API level. Ahh.. We could create a frunction in mp_cmpr.c which makes a copy of the current DBENV and use this copy for _weakcmpr. We would need to figure out which DBENV variable fixes the problem. There are around 25 elements to the DBENV structure.. a fair number of these a function pointers and it seems unlikely to be those. > Yes, it is exactly Loic's code (mp_cmpr.c) that I was proposing to > fix. In the past I hacked mp_alloc to try to work around the bug, > but I had planned to back those changes out once b5 was out and > there was time for a real fix. OK. If we confine ourselves to hacking mp_cmpr.c as much as possible that's cool. As long as we're on the same page here. (pun alert) > That sounds impressive. OK -- I'll stop tinkering with the database > code... Not sure you'll think its impressive when you see the < 20 line patch ;-) I'll post it to the group later today. Thanks. Neal Richter Knowledgebase Developer RightNow Technologies, Inc. Customer Service for Every Web Site Office: 406-522-1485 |
|
From: Lachlan A. <lh...@us...> - 2003-06-18 12:44:23
|
On Wed, 18 Jun 2003 05:32, J. op den Brouw wrote: > I'm still having trouble getting the Berkeley database ./configure > finding my select() function. I'm compiling a big post with > lots of output from various programs/files, but I may not > be able to produce a fix the planned release. That sounds good. It will be good to see the output. Don't worry too much about the release schedule -- as I just posted,=20 I've found another database bug :( > Add a (new) document containing the Unix/Windows flavors > htdig runs on + more information and hints on how to > compile and install htdig on those various flavors. > (HP-UX needs a huge list of parameters to get it > compiled anyway). The supported platforms are currently in htdoc/requirements.html. Is there really a need for a new document? The compile hints belong in htdoc/install.html. Could you post the compilation sequence for HP-UX? Cheers, Lachlan --=20 lh...@us... ht://Dig developer DownUnder (http://www.htdig.org) |
|
From: Geoff H. <ghu...@ws...> - 2003-06-19 05:30:19
|
On Wednesday, June 18, 2003, at 07:43 AM, Lachlan Andrew wrote: > The supported platforms are currently in htdoc/requirements.html. > Is there really a need for a new document? No, but they do need updating, even with your recent changes. For example, we can't really be sure we compile with gcc-2.7 anymore, right? (BTW, I wouldn't suggest putting an e-mail address in the page even as a comment. SPAMbots seem to harvest these too.) -Geoff |
|
From: Lachlan A. <lh...@us...> - 2003-06-19 14:13:37
|
Thanks for the tip about the address -- I was forgetting that these=20 pages are intended for the BBB (Big Bad WeB). We don't claim that we compile under gcc-2.7. The old systems are=20 commented out. (Just my fear of deleting...) Cheers, Lachlan On Thu, 19 Jun 2003 15:30, Geoff Hutchison wrote: > On Wednesday, June 18, 2003, at 07:43 AM, Lachlan Andrew wrote: > > The supported platforms are currently in htdoc/requirements.html. > > Is there really a need for a new document? > > No, but they do need updating, even with your recent changes. For > example, we can't really be sure we compile with gcc-2.7 anymore, > right? > > (BTW, I wouldn't suggest putting an e-mail address in the page even > as a comment. SPAMbots seem to harvest these too.) > > -Geoff --=20 lh...@us... ht://Dig developer DownUnder (http://www.htdig.org) |
|
From: J. op d. B. <ht...@op...> - 2003-06-19 14:47:59
|
Hmmm, i'm still compiling htdig with 2.8.1 since my admin will not update for now (too much work...) Any reason why compiling under 2.7 is discouraged? Lachlan Andrew wrote: > > Thanks for the tip about the address -- I was forgetting that these > pages are intended for the BBB (Big Bad WeB). > > We don't claim that we compile under gcc-2.7. The old systems are > commented out. (Just my fear of deleting...) --jesse |
|
From: Lachlan A. <lh...@us...> - 2003-06-20 06:31:20
|
Greetings Jesse, I don't think Geoff was saying we *shouldn't* use 2.7. It's just that we haven't actually *tested* it under 2.7. If (when :) you can confirm that it still works under 2.8, that's great! Cheers, Lachlan On Fri, 20 Jun 2003 00:45, J. op den Brouw wrote: > Hmmm, i'm still compiling htdig with 2.8.1 since my > admin will not update for now (too much work...) > > Any reason why compiling under 2.7 is discouraged? --=20 lh...@us... ht://Dig developer DownUnder (http://www.htdig.org) |
|
From: Geoff H. <ghu...@ws...> - 2003-06-23 16:46:01
|
> I don't think Geoff was saying we *shouldn't* use 2.7. > It's just that we haven't actually *tested* it under 2.7. > If (when :) you can confirm that it still works under 2.8, Right. Sorry for the confusion. If it works under gcc-2.7, so much the better! But IMHO, we should be pushing people towards newer compilers with newer releases. It's pretty clear that new GCC releases will stop compiling non-ISO C++, so we'll either need to do a lot of conditional compilation or clean up the code and get people to use newer compilers. -Geoff |
|
From: Gabriele B. <g.b...@co...> - 2003-06-24 11:06:06
|
Hi friends! > But IMHO, we should be pushing people towards newer compilers with > newer releases. It's pretty clear that new GCC releases will stop > compiling non-ISO C++, so we'll either need to do a lot of conditional > compilation or clean up the code and get people to use newer compilers. I agree with Geoff. Also, if you could be interested, thanks to my friend Marco, ht://Check since last sunday has got automatic checks for C++ standard library and conditional compilation which for fstream, iomanip, etc. and namespaces. If you are interested in this, I volounteer myself to port the code to this, maybe asking my friend Marco (mn...@sf...) to join me. I don't know though if at this time it is better to wait for 3.2.0b5 to be out. Let me know Ciao -Gabriele -- Gabriele Bartolini - Web Programmer Comune di Prato - Prato - Tuscany - Italy g.b...@co... | http://www.comune.prato.it > find bin/laden -name osama -exec rm {} ; |
|
From: Lachlan A. <lh...@us...> - 2003-06-24 12:51:38
|
On Tue, 24 Jun 2003 21:06, Gabriele Bartolini wrote: > > do a lot of conditional compilation or clean up the code and get > > people to use newer compilers. > > I agree with Geoff. I don't know though if at this time it is > better to wait for 3.2.0b5 to be out. Hear, hear! :) Lachlan --=20 lh...@us... ht://Dig developer DownUnder (http://www.htdig.org) |
|
From: Gabriele B. <bar...@in...> - 2003-07-04 17:06:28
|
At 22.23 24/06/2003 +1000, Lachlan Andrew wrote:
> > I agree with Geoff. I don't know though if at this time it is
> > better to wait for 3.2.0b5 to be out.
>
>Hear, hear! :)
Hi there,
I gave a look at autotools stuff (autoconf, autoheader, autoscan,
automake ...) and I must confess that now life is sooooooo much *easier*
than a couple of years ago.
First (thanks to my friend Marco's work) I could port configure scripts
to a clearer situation: I remember when I first introduced it to ht://Check
it was more a cut+paste operation, sniffing code from basically Loic's work
on ht://Dig.
Now, especially with autoscan and autoheader, the situation is much better.
Having said this, I have a few proposals to do: during this week-end I
will bring my laptop to my countryhouse (where I don't have the Internet so
I may not answer you) and try and hack the code in order to include at
least these 2 things:
- check for C++ standard library
- check for C++ namespaces
Essentially these two m4 macros (taken from the configure archive) will
define 2 preprocessor variables in config.h (we must get rid of acconfig.h
according to new autoconf's directives because it is deprecated) called
HAVE_STD and HAVE_NAMESPACES and I'll modify the code whenever iostream.h
(and similar like 'iomanip', etc.) is called.
Sounds good?
However, I really hope I can contribute this patch early next week.
Ciao and thanks
-Gabriele
--
Gabriele Bartolini: Web Programmer, ht://Dig & IWA/HWG Member, ht://Check
maintainer
Current Location: Prato, Tuscany, Italy
bar...@in... | http://www.prato.linux.it/~gbartolini | ICQ#129221447
> "Leave every hope, ye who enter!", Dante Alighieri, Divine Comedy, The
Inferno
|
|
From: Lachlan A. <lh...@us...> - 2003-07-05 08:38:23
|
Greetings Gabriele, Is there a reason to do this before 3.2.0b5 is released? If you are=20 going to recreate configure, remember that it has been manually=20 hacked to work with OS X, ready for the 'imminent' release a month or=20 two ago. Perhaps I should face reality about the release, but I'm still=20 hopeful(ish). Lachlan On Sat, 5 Jul 2003 03:06, Gabriele Bartolini wrote: > I have a few proposals to do: > > - check for C++ standard library > - check for C++ namespaces --=20 lh...@us... ht://Dig developer DownUnder (http://www.htdig.org) |
|
From: Gabriele B. <bar...@in...> - 2003-07-06 21:36:03
|
Ciao Lachlan! >Is there a reason to do this before 3.2.0b5 is released? No ... :-P > If you are going to recreate configure, remember that it has been > manually >hacked to work with OS X, ready for the 'imminent' release a month or >two ago. Well ... however that's gonna be something we'll have to face anyway. Unfortunately I can't do much on OS X systems as I don't have any computer with it; also on SF compile farm the Darwin shell has been removed! >Perhaps I should face reality about the release, but I'm still >hopeful(ish). I understand what you mean but I'd rather send a patch to the group. Also many things have changed as far as autotools are concerned and many MACROs we use or other techniques have been declared 'deprecated'. I refer in particular to: - acconfig.h which should be substituted by calls to AC_DEFINE(QUOTED) - the htconfig.h.in must be changed to htconfig.h and the default config.h file should be created by autoheader There are other small changes that should be applied but I don't wanna talk about them now. Only when I have a patch ready maybe. For your curiosity, I just changed the configure.in file on my laptop and I was able to run: - autoscan - aclocal - autoconf - autoheader - automake and ... finally ... autoreconf without getting any single warning. I hope I can provide a patch very soon (I just came back and I did not have any Internet connection and I just need to cross my fingers now for CVS update). :-) Ciao ciao -Gabriele -- Gabriele Bartolini: Web Programmer, ht://Dig & IWA/HWG Member, ht://Check maintainer Current Location: Prato, Tuscany, Italy bar...@in... | http://www.prato.linux.it/~gbartolini | ICQ#129221447 > "Leave every hope, ye who enter!", Dante Alighieri, Divine Comedy, The Inferno |
Thanks for helping keep SourceForge clean.
X