| Lists: | Postg토토 핫SQL : Postg토토 |
|---|
| From: | Claudio Natoli <claudio(dot)natoli(at)memetrics(dot)com> |
|---|---|
| To: | 'Andreas Pflug' <pgadmin(at)pse-consulting(dot)de>, Mark Kirkwood <markir(at)coretech(dot)co(dot)nz> |
| Cc: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, pgsql-hackers-win32(at)postgresql(dot)org |
| Subject: | Re: Win32 testing needed |
| Date: | 2004-08-06 08:24:35 |
| Message-ID: | A02DEC4D1073D611BAE8525405FCCE2B55F539@harris.memetrics.local |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Lists: | pgsql-hackers-win32 |
> I know this message. It does not appear for me if I do a clean checkout,
> and make from the pgsql top directory. If I make from src/backend (to
> skip the lengthy locale makes on my slow machine), that said undefined
> reference comes up, and won't go even if I make from the top dir unless
> dirmod.c is touched.
I can confirm that I've also experienced this previously, in the mode
described. However, with the introduction of rmtree, it looks like it is now
broken even from the top directory.
> So it's better if src/port sources don't use backend's functions.
Agree. Or rather, better if src/port doesn't use backend DATA.
Cheers,
Claudio
---
Certain disclaimers and policies apply to all email sent from Memetrics.
For the full text of these disclaimers and policies see
<a
href="http://www.memetrics.com/emailpolicy.html">http://www.memetrics.com/em
ailpolicy.html</a>
| From: | Andreas Pflug <pgadmin(at)pse-consulting(dot)de> |
|---|---|
| To: | Claudio Natoli <claudio(dot)natoli(at)memetrics(dot)com> |
| Cc: | Mark Kirkwood <markir(at)coretech(dot)co(dot)nz>, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, pgsql-hackers-win32(at)postgresql(dot)org |
| Subject: | Re: Win32 testing needed |
| Date: | 2004-08-06 10:56:44 |
| Message-ID: | 411363EC.3010909@pse-consulting.de |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Lists: | Postg토토 핫SQL : Postg토토 |
Claudio Natoli wrote:
>
>
>>So it's better if src/port sources don't use backend's functions.
>
>
> Agree. Or rather, better if src/port doesn't use backend DATA.
Not easy to avoid, many functions are really macros using backend data.
I wonder why rmtree uses xmalloc at all; instead of creating an array of
dir names and then iterating the array this could be done in one step.
There may be more than one DIR* open, right?
Regards,
Andreas
| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Andreas Pflug <pgadmin(at)pse-consulting(dot)de> |
| Cc: | Claudio Natoli <claudio(dot)natoli(at)memetrics(dot)com>, Mark Kirkwood <markir(at)coretech(dot)co(dot)nz>, pgsql-hackers-win32(at)postgresql(dot)org |
| Subject: | Re: Win32 testing needed |
| Date: | 2004-08-06 14:37:59 |
| Message-ID: | 11680.1091803079@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Lists: | pgsql-hackers-win32 |
Andreas Pflug <pgadmin(at)pse-consulting(dot)de> writes:
> Claudio Natoli wrote:
>> Agree. Or rather, better if src/port doesn't use backend DATA.
> Not easy to avoid, many functions are really macros using backend data.
Perhaps we should insist that nothing in src/port can include postgres.h.
If it needs something that's not in c.h, it doesn't belong in src/port.
These are the affected files:
$ grep postgres.h *.c
copydir.c:#include "postgres.h"
dirmod.c:#include "postgres.h"
exec.c:#include "postgres.h"
pipe.c:#include "postgres.h"
regards, tom lane
| From: | Andreas Pflug <pgadmin(at)pse-consulting(dot)de> |
|---|---|
| To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
| Cc: | Claudio Natoli <claudio(dot)natoli(at)memetrics(dot)com>, Mark Kirkwood <markir(at)coretech(dot)co(dot)nz>, pgsql-hackers-win32(at)postgresql(dot)org |
| Subject: | Re: Win32 testing needed |
| Date: | 2004-08-06 17:29:09 |
| Message-ID: | 4113BFE5.5080601@pse-consulting.de |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Lists: | pgsql-hackers-win32 |
Tom Lane wrote:
> Perhaps we should insist that nothing in src/port can include postgres.h.
> If it needs something that's not in c.h, it doesn't belong in src/port.
>
This is just consequent; files using postgres.h should be under
src/backend/port, not src/port which might be used by frontend tools too.
Regards,
Andreas