[1346] in BarnOwl Developers
Re: [Fwd: Owl patches]
daemon@ATHENA.MIT.EDU (James Kretchmar)
Thu Oct 29 18:15:38 2009
Resent-From: nelhage@mit.edu
Resent-To: barnowl-dev-mtg@charon.mit.edu
X-Original-To: nelhage@lunatique.mit.edu
In-Reply-To: <ab5ecb6c0903221022g737db974jf08a3f0b3d281c0d@mail.gmail.com>
Date: Tue, 24 Mar 2009 09:25:06 -0400
From: James Kretchmar <kretch@gmail.com>
To: =?ISO-8859-1?Q?Alejandro_R=2E_Sede=F1o?= <asedeno@mit.edu>
Cc: barnowl-dev@mit.edu
> Thanks. All these patches sound good, I'll get the crashes and
> leaks in for sure, and all of the miscellaneous ones that apply
> without too much effort.
FYI, I applied the patches on Sunday. Most of them applied cleanly
with no effort, four more applied with a trivial change or two. A few
had more extensive patch failures, so I'll look at them more closely
after getting out the security fix. Those were:
9c590d4400a3cc43b25cb33b4d9e06086f77ec90 -- CC behavior
d0961fe0bc68cec33bc23afbae4928bbd861d718 /
fba0f96c700b25bb5dbd1a6bef9aa520d76c0be6 -- format discuss messages
ad15610ead2550219224b67ce4c2d7da2819e4df -- More ANSI C
Also, the second patch listed as releated to the discuss formating
(4f15e8e9ceada0d4b2cc969ebf43b0a1fb3709ea) was a duplicate of an
earlier change number listed for ripping out the browser stuff.
> Ok, if someone can point me at office heads running the relevant
> releases that'd be great.
Anyone know which office heads are running the current releases?
kretch
2009/3/22 James Kretchmar <kretch@gmail.com>:
>> Which patches did you actually take?
>
> I took the barnowl_1.0.1-4.diff patch with the security fixes for
> zwrite.c and zcrypt.c.
>
> The select loop I integrated by hand. It replaces muxevents and is
> handling keyboard input, zephyrs, aims, the stderr redirector and the
> popexec input, just as you have it. The only difference is I left out
> the timer implementation for the time being.
>
>> There are some memory leak patches that would be worth incorporating
>> while you're updating, and a few simple things that I think would be
>> worth having.
>
> Thanks. All these patches sound good, I'll get the crashes and leaks
> in for sure, and all of the miscellaneous ones that apply without too
> much effort.
>
>> If you could get back to us on the licensing stuff we talked about
>> last summer, regarding relicensing as either MIT or GPL, that'd be
>> appreciated.
>
> Ok, I'll look into this. It's going to be a busy week but I should be
> able to say something concrete by next weekend.
>
>> As for build machines, I'm not sure which machines run what these
>> days, as I mostly use my laptop in the office. I'm hoping others on
>> barnowl-dev can better answer that question.
>
> Ok, if someone can point me at office heads running the relevant
> releases that'd be great.
>
> kretch
>
> On Sun, Mar 15, 2009 at 11:53 AM, "Alejandro R. Sede=F1o" <asedeno@mit.ed=
u> wrote:
>> [CC-ing barnowl-dev.]
>>
>> Which patches did you actually take?
>>
>> There are some memory leak patches that would be worth incorporating
>> while you're updating, and a few simple things that I think would be
>> worth having.
>>
>> Crashes:
>> 4c46dfdebc294ca24fef59924f01688b8ee07dee -- fixing a hang
>> 99dabeed7de6d19acb50f1e73aa0bfe5f9469c02 -- no segfault on missing subs
>> 559076cd86edc3a8317819814dd5877b8bc7c3cb -- zcrypt buffer size bug
>> e97c4a306ae2c9d2147d83da29fee59918198e70 -- fix doublefree
>>
>> Leaks:
>> c0a90c2c7ab97b9465c9873757faf312faa0021e -- memory leaks
>> 95caa16b2e9ba10770d87d0955198c14c2d0e16a -- memory leaks
>> 1716fed8c2650e46892cf6571555eac937266c6e -- memory leaks
>>
>> Misc:
>> 216c73421653e3ef0e957aa9a9804e208a77c80e -- getstyle command
>> 0c8ab5eadbb6ecc97a120c91b9a824b33538c764 -- binary search for msgid
>> e78397d6ac5da0de31a4e269c0ba7f3d691857a3 -- close a file-handle leak
>> 801b7ac63b962640debbcfd422cb9a60da5fea31 -- fix delay in jump from top
>> to bottom
>> 93e883d60051b80bf6d35391f9d76fd7dfd198e3 -- load subs in chunks
>> f6050ee9b0a171a5031f84409eb181062afacd18 \- tweak to 93e... above
>> d857b667a5a9b108b1a2a26b4a5513bef2b53f80 -- logging -- better extraction
>> of zsig
>> d524c838ac7c115628424b8ac171c3489622ea3a -- free() -> owlfree()
>> af9b92e3e7ccb7de276a94b5c7e5861b24e71eff -- escape AIM users in smartfil=
ters
>> 80e54a7631f091be8c9762adb9746bad38104738 -- escale {}()
>> 4099cf83702763fa8d1efc4f1582a605431bdb77 -- s/"'/./ in smartfilters
>>
>> 48609ce969ee2eb01da64338f4e477e0049acc22 \
>> 9c590d4400a3cc43b25cb33b4d9e06086f77ec90 -- CC behavior
>> d0961fe0bc68cec33bc23afbae4928bbd861d718 /
>>
>> 128171aaf7cefa91a4bb1eada93a19d8fd0c355c \
>> 50e29e35c64d64e223d378d350a7bc4f038d78f5 -- note 0-len message fields
>> 804ab8af8b6d00bcd7e2402df892db8fbd61a3ec /
>>
>> c0f9e3009bc03e80a44de64cd5f2b4033290236e -- curmsg on screen after resiz=
e
>> 8e5935d11c699a7ce5a3e6e9a47799564c696d6a -- rip out browser stuff
>> 4f15e8e9ceada0d4b2cc969ebf43b0a1fb3709ea /
>> 6e400cc71aa59e041dce677aadf50dc1f25228e2 -- no passwords in command hist=
ory
>>
>> a1bb1980e4bca23b8329cc0e7c0bd5027055ea0a -- format NOC messages
>> fba0f96c700b25bb5dbd1a6bef9aa520d76c0be6 -- format discuss messages
>> 4f15e8e9ceada0d4b2cc969ebf43b0a1fb3709ea /
>> ad15610ead2550219224b67ce4c2d7da2819e4df -- More ANSI C
>> 27d8d835dc6d58c08fae10e75aae306c49215143 -- expand ~ in loadsubs
>> 10d67d57cb29221f63a43a30643c697fc7b38911 -- expand ~ in source
>>
>> Also, anything that andersk and geofft looked at the the C code is
>> probably worth looking at closely. Some of it is already included above.
>>
>> Again, barnowl's trac is located at http://barnowl.mit.edu/, and that
>> has links to the current repo.
>>
>> If you could get back to us on the licensing stuff we talked about last
>> summer, regarding relicensing as either MIT or GPL, that'd be appreciate=
d.
>>
>> As for build machines, I'm not sure which machines run what these days,
>> as I mostly use my laptop in the office. I'm hoping others on
>> barnowl-dev can better answer that question. I don't think we have any
>> explicit build machines in the office; protocol is as you remember for
>> office heads.
>>
>> -Alejandro
>>
>> -------- Original Message --------
>> Subject: Owl patches
>> Date: Sun, 15 Mar 2009 00:31:50 -0400
>> From: James Kretchmar <kretch@gmail.com>
>> To: asedeno@mit.edu
>>
>> Hey Alejandro,
>>
>> I went ahead and incorporated the zwrite/zcrypt patch and also the
>> select loop this evening. Can you point me at the current SIPB build
>> machines or office heads for the relevant Athena platforms?
>>
>> Is the protocol these days still to ask before logging in under
>> someone on an office head, but not on a build machine?
>>
>> Thanks,
>> kretch
>>
>>
>
