Bug 147935 - [MAINTAINER] security/botan: update to 1.8.9
Summary: [MAINTAINER] security/botan: update to 1.8.9
Status: Closed FIXED
Alias: None
Product: Ports & Packages
Classification: Unclassified
Component: Individual Port(s) (show other bugs)
Version: Latest
Hardware: Any Any
: Normal Affects Only Me
Assignee: Sahil Tandon
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2010-06-17 11:10 UTC by Lapo Luchini
Modified: 2010-07-24 07:00 UTC (History)
1 user (show)

See Also:


Attachments
botan-1.8.9.patch (1.05 KB, patch)
2010-06-17 11:10 UTC, Lapo Luchini
no flags Details | Diff
botan-2.diff (1.33 KB, patch)
2010-06-30 14:11 UTC, Lapo Luchini
no flags Details | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Lapo Luchini 2010-06-17 11:10:01 UTC
- Update to 1.8.9

Generated with FreeBSD Port Tools 0.99
Comment 1 niels freebsd_committer freebsd_triage 2010-06-20 23:16:48 UTC
Responsible Changed
From-To: freebsd-ports-bugs->niels

Thanks! I'll take it
Comment 2 niels freebsd_committer freebsd_triage 2010-06-24 12:58:23 UTC
State Changed
From-To: open->feedback



Hi Lapo, 

The update builds OK but there are many compiler warnings. Can you fix these ? 
http://freebsd.heinen.ws/tb/logs/8.0-STABLE/botan-1.8.9.log 

Kind regards, 
Niels
Comment 3 Lapo Luchini 2010-06-30 14:11:13 UTC
I asked the author (Jack Lloyd) about that, and that's what he told me:

> It only shows up on 32 bit compilations. It's actually caused by a bug
> in GCC, fixed in 4.5.0 (PR 13358). It seems strange to me that this
> would be 'too many' warnings - it's effectively one warning.
>
> There is no actual functionality problem that this warning is pointing
> out; GCC is still doing exactly the right thing, just printing an
> obnoxious warning for every 64-bit constant it sees that is impossible
> to disable via a -Wno-obnoxious-warning flag. The workaround is to
> append ULL to each of the constants. I don't want to do that upstream
> because every other compiler accepts them without any suffix, but some
> Windows compilers will reject the suffixed version.

I wonder if it's the best approach to have a hundreds' line patch adding
ULLs to a DES cryptographic routine only to silence warnings that
produce no problems (except spamming the console, of course!).

IMHO it's probably better to avoid that, but I can put some clever
inline-sed code to add them, if you think that's really necessary.

PS: it would probably be nice to add LICENSE lines in the Makefile
though, as in the attached patch.
Comment 4 niels freebsd_committer freebsd_triage 2010-07-22 20:28:22 UTC
State Changed
From-To: feedback->open

Throw back in the pool. Apologies, I don't want to delay this longer (lack of time) 



Comment 5 niels freebsd_committer freebsd_triage 2010-07-22 20:28:22 UTC
Responsible Changed
From-To: niels->ports
Comment 6 niels freebsd_committer freebsd_triage 2010-07-22 20:30:01 UTC
Hi Lapo,
sorry for delaying this one. I lack the time to follow up and threw the 
PR back in the pool.

Niels

On 06/30/10 15:11, Lapo Luchini wrote:
> I asked the author (Jack Lloyd) about that, and that's what he told me:
>
>> It only shows up on 32 bit compilations. It's actually caused by a bug
>> in GCC, fixed in 4.5.0 (PR 13358). It seems strange to me that this
>> would be 'too many' warnings - it's effectively one warning.
>>
>> There is no actual functionality problem that this warning is pointing
>> out; GCC is still doing exactly the right thing, just printing an
>> obnoxious warning for every 64-bit constant it sees that is impossible
>> to disable via a -Wno-obnoxious-warning flag. The workaround is to
>> append ULL to each of the constants. I don't want to do that upstream
>> because every other compiler accepts them without any suffix, but some
>> Windows compilers will reject the suffixed version.
>
> I wonder if it's the best approach to have a hundreds' line patch adding
> ULLs to a DES cryptographic routine only to silence warnings that
> produce no problems (except spamming the console, of course!).
>
> IMHO it's probably better to avoid that, but I can put some clever
> inline-sed code to add them, if you think that's really necessary.
>
> PS: it would probably be nice to add LICENSE lines in the Makefile
> though, as in the attached patch.

-- 
Niels Heinen
FreeBSD committer | www.freebsd.org
PGP: 0x5FE39B80
Comment 7 Mark Linimon freebsd_committer freebsd_triage 2010-07-22 20:58:19 UTC
Responsible Changed
From-To: ports->freebsd-ports-bugs

Canonicalize assignment.
Comment 8 Sahil Tandon freebsd_committer freebsd_triage 2010-07-23 03:05:54 UTC
Responsible Changed
From-To: freebsd-ports-bugs->sahil

I'll take it.
Comment 9 dfilter service freebsd_committer freebsd_triage 2010-07-24 06:55:40 UTC
sahil       2010-07-24 05:55:26 UTC

  FreeBSD ports repository

  Modified files:
    security/botan       Makefile distinfo 
  Log:
  - Update to 1.8.9
  - Add LICENSE
  
  PR:             ports/147935
  Submitted by:   Lapo Luchini <lapo@lapo.it> (maintainer)
  
  Revision  Changes    Path
  1.53      +4 -2      ports/security/botan/Makefile
  1.33      +3 -3      ports/security/botan/distinfo
_______________________________________________
cvs-all@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/cvs-all
To unsubscribe, send any mail to "cvs-all-unsubscribe@freebsd.org"
Comment 10 Sahil Tandon freebsd_committer freebsd_triage 2010-07-24 06:57:27 UTC
State Changed
From-To: open->closed

Committed. Thanks!