Team Chat Logs

2007 1
Mo Tu We Th Fr Sa Su
      1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28        
<< <   > >>

February 21, 2007

[00:09:54] * m_g has joined #trac
[00:38:09] * cmlenz has joined #trac
[00:39:26] * cmlenz has quit IRC
[00:41:56] * cmlenz has joined #trac
[00:41:58] * klando has joined #trac
[00:42:06] <coderanger> cmlenz: You're alive!
[00:42:17] <cmlenz> yo :-P
[00:42:24] <coderanger> :)
[00:47:05] <m_g> ah good, the gurus are here :)
[00:47:47] <m_g> i have set the url_base in trac.ini, but it seems that trac ignores it in the notifications. the footer of the notification email shows an URL different from the url_base
[00:48:21] <m_g> does trac "officially" ignore url_base for the notifications ticket URL or do i have to set anything else for this to work?
[01:05:01] * pygi has left #trac
[01:06:38] * Sebastian has joined #trac
[01:06:59] <Sebastian> Anyone familiar with the RevTree plugin around?
[01:08:04] * pacopabl1 has joined #trac
[01:08:21] * grim has quit IRC
[01:09:04] * oblio_ has quit IRC
[01:09:04] * MrDebout has quit IRC
[01:09:04] * Fatal_2 has quit IRC
[01:09:04] * jae has quit IRC
[01:09:04] * tsb has quit IRC
[01:09:04] * synic has quit IRC
[01:09:04] * calculus has quit IRC
[01:09:04] * verm__ has quit IRC
[01:09:04] * Semhirage has quit IRC
[01:09:04] * s0undt3ch has quit IRC
[01:09:04] * gmoore has quit IRC
[01:09:04] * pacopablo has quit IRC
[01:09:08] * Semhirage has joined #trac
[01:09:08] * jae has joined #trac
[01:09:08] * tsb has joined #trac
[01:09:08] * MrDebout has joined #trac
[01:09:08] * s0undt3ch has joined #trac
[01:09:08] * gmoore has joined #trac
[01:09:08] * calculus has joined #trac
[01:09:08] * oblio_ has joined #trac
[01:09:08] * verm__ has joined #trac
[01:09:08] * synic has joined #trac
[01:09:08] * Fatal_2 has joined #trac
[01:15:04] * dilinger has joined #trac
[01:20:08] <m_g> coderanger, any ideas? :)
[01:20:24] <coderanger> m_g: what version is this?
[01:20:28] <m_g> 0.10.3
[01:22:25] <coderanger> Can you pastebin a copy of one of the emails
[01:23:59] <m_g> hm no, sorry, it contains private information, but basically the problem is that the line: "Ticket URL: ....." shows the "real url" while the project is being accessed through a subdomain and the url_base is set to that domain
[01:24:10] * nextime has joined #trac
[01:26:11] <coderanger> one sec
[01:27:51] <coderanger> You mean base_url, not url_base, right?
[01:28:31] <m_g> ah, yes
[01:29:11] <coderanger> python -c 'import trac.env; e = trac.env.Envionment('/path/to/env'); print e.base_url'
[01:29:52] <nextime> coderanger : is there somewhere an example of tracforge to see it in action? and more... how tickets are centralized in tracforge, there is something like a global unique is progression for tickets, or the same ticketid can be on multiple trac envs?
[01:30:15] <nextime> s/is progression/id progression/
[01:30:25] <m_g> coderanger, shows the correct url.. http://trac.domain.de/projectname/
[01:30:26] <coderanger> nextime: Nope, haven't felt like setting up a demo server
[01:30:40] <coderanger> nextime: Each ticket has a different ID in each Trac
[01:31:07] <coderanger> nextime: There is a hidden custom field that tracks which IDs map to which
[01:31:51] <nextime> coderanger : so, if there is a ticket #3 on the trac A, it can't be the same # on the trac B?
[01:32:06] <coderanger> nextime: It could be, but chances are it won't
[01:32:23] <nextime> coderanger : right, thanks
[01:32:39] <coderanger> nextime: When a ticket is linked over to another env, it takes on the next sequential value
[01:33:11] <coderanger> Just like entering a new ticket normally
[01:33:17] <nextime> latest: how about "merging" already populated trac envs with existants tickets?
[01:33:28] <nextime> there are problems in do that?
[01:33:29] <coderanger> nextime: Not handled
[01:33:47] <coderanger> nextime: Now is there an easy way to do it manually
[01:33:51] <coderanger> er, nor
[01:34:26] <coderanger> At some point in the far future I may rig something up with the datamover plugin
[01:34:36] <coderanger> m_g: Dunno then
[01:34:44] <coderanger> m_g: It should work
[01:35:00] <m_g> coderanger, ok thanks :/ generally speaking Trac uses base_url for the ticket URL?
[01:35:02] <nextime> ok, thanks a lot.
[01:35:29] <coderanger> m_g: No, base_url is only used when generating absolute URLs (most of the time Trac uses relative links)
[01:35:49] <m_g> but this is an absolute url, isn't it?
[01:35:53] <m_g> the ticket URL in the footer
[01:35:58] <m_g> of a notification email
[01:36:00] <coderanger> m_g: You can make a macro to print out env.abs_href.ticket(1)
[01:36:05] <coderanger> m_g: Yes
[01:36:20] * klando has left #trac
[01:36:47] <coderanger> And the code clearly shows it using env.abs_href, which is req.abs_href, which is based on req.environ['trac.base_url'], which is set from env.base_url
[01:37:18] <m_g> oh.. then this is quite strange, isn't it?
[01:37:28] <coderanger> Yes
[01:37:46] <coderanger> Did you change any code at any point?
[01:37:52] <coderanger> Or do anything else funnky
[01:38:37] <m_g> no the base code is the same, i did only install the following plugins for this project: wiki rename, ticket delete, hier wiki, graphviz, ctxtnavadd, gantt
[01:40:56] * grim has joined #trac
[01:41:40] <coderanger> m_g: I would try removing the trac python modules and reinstalling them
[01:42:20] <m_g> rm -rf /usr/share/trac; python setup.py install?
[01:42:34] <coderanger> no
[01:42:51] <coderanger> /usr/lib/python2.?/site-packages/trac
[01:43:41] <m_g> ah, right, thanks! i will try this later
[01:43:43] * pratchett has joined #trac
[01:45:13] * oblio_ has quit IRC
[01:45:14] * MrDebout has quit IRC
[01:45:14] * synic has quit IRC
[01:45:14] * calculus has quit IRC
[01:45:14] * nextime has quit IRC
[01:45:14] * tsb has quit IRC
[01:45:14] * s0undt3ch has quit IRC
[01:45:14] * gmoore has quit IRC
[01:45:14] * verm__ has quit IRC
[01:45:14] * jae has quit IRC
[01:45:14] * Semhirage has quit IRC
[01:45:14] * Fatal_2 has quit IRC
[01:47:00] * grim has quit IRC
[01:47:29] * nextime has joined #trac
[01:47:29] * Semhirage has joined #trac
[01:47:29] * jae has joined #trac
[01:47:29] * tsb has joined #trac
[01:47:29] * MrDebout has joined #trac
[01:47:29] * s0undt3ch has joined #trac
[01:47:30] * gmoore has joined #trac
[01:47:30] * calculus has joined #trac
[01:47:30] * oblio_ has joined #trac
[01:47:30] * synic has joined #trac
[01:47:30] * Fatal_2 has joined #trac
[01:47:47] * grim has joined #trac
[01:49:06] * coderanger -> sleep
[01:49:19] <m_g> good night coderanger
[01:59:11] <Dwarfoo7> how can you edit trac.db to add permissions like TRAC_ADMIN ?
[01:59:25] <Dwarfoo7> good night coderanger
[02:04:29] <Dwarfoo7> yea actually I found out
[02:04:51] <Dwarfoo7> trac-admin [path/to/project] permission add [user] TRAC_ADMIN
[02:04:51] <wu-nix> i try to use basic auth and on wiki it says
[02:04:52] <wu-nix> to use Basic authentication, replace --auth with --basic-auth in the examples below, and omit the realm.
[02:05:10] <Dwarfoo7> what is basic auth?
[02:05:21] <wu-nix> But if i try to start without realm i get
[02:05:21] <wu-nix> tracd: error: Incorrect number of parameters for --basic-auth
[02:05:35] <Dwarfoo7> ah
[02:05:45] <Dwarfoo7> thought you were replying to my question
[02:15:10] * nextime has left #trac
[02:15:24] * cerw has joined #trac
[02:27:17] * goldeagle has joined #trac
[03:05:04] * Kronoss has joined #trac
[03:05:16] <Kronoss> hi
[03:06:19] <Kronoss> im trying to use a setuided python script to modify a trac database using sqlite backend
[03:07:13] <Kronoss> but... sqlite_backend.py uses os.access to validate if I cannot use the database
[03:07:42] <Kronoss> and os.access seems to watch only uid and not euid
[03:07:57] <Kronoss> so, i cannot access to database
[03:09:10] <Kronoss> what can i do?
[03:43:49] * lisppaste5 has quit IRC
[03:59:20] * rvsjoen has quit IRC
[04:28:03] * Sebastian has quit IRC
[04:31:14] * Miek has joined #trac
[04:45:19] * corp has joined #trac
[04:45:29] <corp> anyone got trac running on 64bit centos/fedora?
[04:48:12] * LeeH has left #trac
[04:49:05] * d0rt has joined #trac
[04:53:12] * Miek has quit IRC
[04:55:57] * Miek has joined #trac
[04:56:44] * corp has quit IRC
[04:59:39] * Kronoss has quit IRC
[05:08:18] * jtoy has joined #trac
[05:11:21] * jtoy has quit IRC
[05:24:05] * jnorthrup_pb17 has joined #trac
[05:25:05] <jnorthrup_pb17> i am running trac 11 svn, and for some reason 'search' has become ignorant of the search modules. is there a conf entry i should know about to turn on various search modules?
[05:27:17] * lisppaste5 has joined #trac
[05:33:44] * cboos has joined #trac
[05:33:58] <cboos> hi
[05:34:15] <cboos> jnorthrup_pb17: just read the IRC logs and saw your question...
[05:34:47] <cboos> the ISearchProvider interface should be imported from another location (the Search.py file has been restructured in a module)
[05:36:12] <cboos> see http://trac-hacks.org/browser/pydocplugin/0.10/tracpydoc/tracpydoc.py?rev=2010#L10 for the required change
[05:38:23] <jnorthrup_pb17> cboos: got it. i seem to be having more pronounced issues by nuking python site-packages and reinstalling fresh.
[05:39:12] <jnorthrup_pb17> since, of course, the bugfix on the matter says the installer has fixed this isssue, and apparently apre-existing trac wont get the benefit of the bugfix in my observation
[05:41:10] * _dmax has joined #trac
[05:47:36] * tkp has joined #trac
[05:56:11] * dmax has quit IRC
[05:56:14] * _dmax is now known as dmax
[06:07:32] * boorad_ has joined #trac
[06:07:51] * jtoy has joined #trac
[06:09:40] * Blackhex has joined #trac
[06:11:36] * Blackhex has left #trac
[06:12:54] * corp has joined #trac
[06:22:37] * LeeH has joined #trac
[06:23:19] * boorad has quit IRC
[06:53:25] * danbeck has joined #trac
[07:02:42] * chandlerc has quit IRC
[07:08:15] * corp has left #trac
[07:09:29] * elehack|work has joined #trac
[07:11:31] * retracile has joined #trac
[07:40:40] * Blackhex has joined #trac
[07:41:15] <Blackhex> hello, may I have a school question about english grammar?
[07:44:05] * retracile has quit IRC
[07:44:27] * retracile has joined #trac
[07:52:56] * retracile has quit IRC
[07:54:16] <pacopabl1> Blackhex: you can always ask, and if no one wants to answer, they won't
[07:54:41] <Blackhex> so what is correct formating of lists in englis?
[07:54:56] <pacopabl1> what do you mean?
[07:55:17] <Blackhex> for example this is correct formating in my language...
[07:55:21] <Blackhex> This is a list:
[07:55:29] <Blackhex> - This is a first item,
[07:55:40] <Blackhex> - this is a second item,
[07:55:47] <Blackhex> - this is a last item.
[07:56:05] <pacopabl1> same in english
[07:56:17] <pacopabl1> though one can use numbers, or letters
[07:56:19] <Blackhex> even with big letters?
[07:56:31] <pacopabl1> basically, what trac provides ;)
[07:56:34] <Blackhex> and dot and strokesM
[07:56:50] <pacopabl1> one sec I'll whip up some examples
[07:56:54] <pacopabl1> lisppaste5: url
[07:56:55] <lisppaste5> To use the lisppaste bot, visit http://paste.lisp.org/new/trac and enter your paste.
[08:00:13] <lisppaste5> pacopablo pasted "list grammar" at http://paste.lisp.org/display/37161
[08:00:24] <pacopabl1> that is by no means exhaustive
[08:00:27] <pacopabl1> but you get the idea
[08:00:29] * chonarop has joined #trac
[08:00:45] <pacopabl1> basically, just about anything goes, as long as it looks fairly reasonable
[08:01:07] <pacopabl1> and if it's an ordered list, then the numbers/letters should have some type of sorting
[08:02:18] <Blackhex> type of bullets or number are not essential, the gramar rule tells about big/small letter and dot and strokes in my language
[08:02:38] <Blackhex> because whole list is considered as two sentences
[08:02:39] * pacopabl1 is now known as pacopablo
[08:03:15] <Blackhex> first is that ending with colon and second is the rest
[08:03:30] <Blackhex> Is that same?
[08:05:13] <pacopablo> um, I'm sure there are some official grammatical rules about that, but no one bothers to follow them?
[08:05:26] <pacopablo> basically, each item in a list is a sentence
[08:05:40] <pacopablo> except when it's not, and it's only a fragment
[08:05:46] <pacopablo> see, it gets complicated ;)
[08:06:32] <Blackhex> yes but I'll stick that evely item is sentecne opposite to what I'm used to
[08:06:41] <jnorthrup_pb17> i have trac dev 11, i have installed subversion 1.3, 1.4, on python 2.4, and manually installed pysvn as well as subversion's own swig bindings. i repeatedly get 'TracError: Unsupported version control system "svn". Check that the Python bindings for "svn" are correctly installed.'
[08:07:01] <jnorthrup_pb17> any suggestions would be greatly appreciated
[08:07:23] <pacopablo> jnorthrup_pb17: try to import the svn bindings in the python interpreter
[08:08:38] <jnorthrup_pb17> lets back up a tad, this worked before on blew away site-packages. which package with which configure --with-xxxx will get me to the finish line? this is not a repeatable process atm
[08:09:18] * goldeagle has quit IRC
[08:09:48] <jnorthrup_pb17> does python 2.4 install svn bindings if svn is present? does svn 1.3 install the swig bindings? i dont recall this being a stimbling point before
[08:10:40] <jnorthrup_pb17> and fwiw, pysvn package imports fine from commandline
[08:11:33] <Blackhex> pacopablo: btw thanks
[08:11:35] * LionsMane has joined #trac
[08:13:26] <Blackhex> bye for now
[08:13:30] * Blackhex has left #trac
[08:16:31] <cboos> jnorthrup_pb17: pysvn is not used by trac, only the swig bindings are
[08:16:39] <cboos> @wiki TracSubversion
[08:16:39] <evil_twin> cboos: http://trac.edgewall.org/wiki/TracSubversion
[08:16:50] <cboos> see the build and install sections in the above
[08:18:23] <jnorthrup_pb17> package svn imports fine
[08:19:06] * retracile has joined #trac
[08:19:28] <jnorthrup_pb17> wow what a clusterfuck
[08:19:31] <jnorthrup_pb17> thanks
[08:19:34] * jnorthrup_pb17 has quit IRC
[08:23:59] * cboos has quit IRC
[08:32:15] * chandlerc has joined #trac
[08:37:02] * chonarop has quit IRC
[09:00:18] * chandlerc has quit IRC
[09:04:51] * em-dash has joined #trac
[09:05:18] * dilinger has quit IRC
[09:14:05] * em-dash has quit IRC
[09:18:58] * pygi has joined #trac
[09:19:18] * kenichi has joined #trac
[09:22:26] * pombreda has quit IRC
[09:27:35] <kenichi> helo, i'm running 0.10.3 on gentoo with mysql. i originally started with 0.10.2, and had to run a special ALTER to get the schema to work with trac at all. now everything seems to work, except certain ticket reports. when i paste the SQL from the report editor (replacing $USER with valid username), it works. when trac tries to run it, i get a syntax error. can anyone help me?
[09:28:53] <kenichi> the report in question is "my tickets" and the reported error is snipped, "check the manual that corresponds to your MySQL server version for the right syntax to use near ' 'knakamur', ')'\n ORDER BY (status = 'assigned') DESC, p.value, milestone, t.ty' at line 9" (knakamur is my username)
[09:31:15] <coderanger> pastebin the whole report
[09:33:31] <kenichi> ok, it's the default "my tickets" report. one sec...
[09:34:46] <coderanger> You might want to make sure that it is the most recent copy of that report
[09:37:24] * pygi has quit IRC
[09:37:44] <kenichi> http://pastebin.com/885963
[09:41:53] <coderanger> That should be t.user
[09:42:05] <coderanger> Or use the most recent SQL for it
[09:42:19] <coderanger> http://trac.edgewall.org/browser/tags/trac-0.10.3/trac/db_default.py#L288
[09:43:24] * em-dash has joined #trac
[09:43:45] <coderanger> Hmm, the 0.10.3 version has it as just "owner" too, if it fails try "t.owner"
[09:44:23] * cmlenz_ has joined #trac
[09:46:47] * cboos has joined #trac
[09:47:10] <cboos> cmlenz? hi, are you online?
[09:47:14] <kenichi> coderanger: thx, i assume you're taking about the AND of WHERE, i tried both the "new" SQL, and appending 't.' to the owner clause of WHERE. no luck. still get syntax error from MySQL server.
[09:49:32] <cboos> maybe cmlenz_ is online then ... ;)
[09:49:57] <kenichi> coderanger: the "old" SQL would work, when i paste it into the cmdline, and replace $USER with a proper username. i feel like this is some problem with the regex or w/e it is that replaces '$USER'. is there a debug flag i can set to see the actualy statements it's trying to execute?
[09:50:23] <coderanger> kenichi: It should show it when the log is set to debug
[09:50:47] * cmlenz_ has quit IRC
[09:50:53] * cboos thinks it's not even worth trying to reply when coderange is around, he's way too fast ;)
[09:50:55] <coderanger> cboos: Guess not :P
[09:51:19] * cmlenz_ has joined #trac
[09:51:46] * cmlenz has quit IRC
[09:52:08] <cboos> cmlenz_, hey, I'm trying to catch you ;-)
[09:53:53] <kenichi> thx, sigh... log_type was = none... dur...
[09:56:14] <kenichi> aha, so i see this: AND owner = 'concat(', %s, ')'
[09:56:22] * _shawn has quit IRC
[09:59:10] <kenichi> with ([u'username']) after the statement. according to the syntax error message, it seems like it's asking for "...owner=''username'') ORDER..."
[10:01:52] * dilinger has joined #trac
[10:09:02] <retracile> cboos: You used "2 << 30" for #4794; I would have expected to see "1 << 31"... I'm curious about the choice.
[10:10:27] <retracile> luuuuuunch.....
[10:14:00] * kenichi has left #trac
[10:19:47] <cmlenz_> cboos: what's up?
[10:19:52] <cboos> hi
[10:19:58] <cboos> about the removal of Undefined...
[10:20:09] <cboos> I've done the change, just wondered if you'd like me to commit it
[10:20:23] <cmlenz_> the changes to trac, you mean?
[10:20:26] <cboos> or if that would conflict with what you've already done so far
[10:20:28] <cboos> yes
[10:20:32] <cboos> in ticket_view.html
[10:20:48] <cboos> explicit calll to isinstance(version, Undefined)
[10:21:03] <cmlenz_> well, we need more changes than that
[10:21:14] <cboos> sure
[10:21:22] <cmlenz_> I think I'd like to do those changes myself to get an idea of whether this is a good idea at all
[10:21:40] <cboos> ok
[10:22:38] * tpollari has joined #trac
[10:22:41] <cboos> but that explicit use of Undefined was easily done in another way, so if you don't have a pending change there, I thnk it's better if I commit it...
[10:24:16] <cboos> retracile: well, no idea why I used 2 << 30 instead of 1 << 31 ...
[10:25:32] <cboos> what about 16 << 27, even nicer pattern?
[10:25:39] * _shawn has joined #trac
[10:35:55] * goldeagle has joined #trac
[10:40:42] * MrRagga has joined #trac
[10:45:02] <cmlenz_> cboos: sure
[10:45:13] <cboos> oh, ok
[10:47:11] * bobbin has joined #trac
[10:48:51] <cboos> et voila! (http://trac.edgewall.org/changeset/4812)
[10:53:49] <cmlenz_> yeah, that's better
[10:56:13] * AnyutKa has joined #Trac
[10:57:18] * AnyutKa has left #Trac
[11:05:08] * coderanger_ has joined #trac
[11:05:54] * LeeH has quit IRC
[11:06:41] * cmlenz_ is now known as cmlenz
[11:06:41] * cmlenz is now known as cmlenz_
[11:07:02] * cmlenz_ is now known as cmlenz
[11:07:50] * _shawn has quit IRC
[11:14:46] * _shawn has joined #trac
[11:14:53] * cboos has quit IRC
[11:15:21] * LeeH has joined #trac
[11:18:25] * iLux has joined #trac
[11:18:31] * iLux has left #trac
[11:24:28] * coderanger_ has quit IRC
[11:25:13] * coderanger_ has joined #trac
[11:28:36] * Blackhex has joined #trac
[11:33:27] * goldeagle has quit IRC
[11:40:05] * hotte has joined #trac
[11:42:08] * coderanger_ has quit IRC
[11:45:03] * tpollari has quit IRC
[11:51:41] * The_Tick has joined #trac
[11:57:28] * converter has joined #trac
[12:00:00] * retracile has quit IRC
[12:04:59] * cmlenz has quit IRC
[12:07:00] * elehack|work has quit IRC
[12:07:05] <The_Tick> coderanger: you around?
[12:13:11] * retracile has joined #trac
[12:25:04] * bobbin has quit IRC
[12:25:11] * bobbin has joined #trac
[12:25:57] * coderanger_ has joined #trac
[12:38:59] * cmlenz has joined #trac
[12:43:35] * bobbin wonders why coderanger is angry ;)
[12:43:47] * bobbin has left #trac
[12:45:47] * Blackhex has left #trac
[12:48:30] * matt_good has joined #trac
[12:51:00] <mitsuhiko> matt_good: got a second?
[12:51:23] <matt_good> yeah, what's up?
[12:51:42] <mitsuhiko> matt_good: i think the issue with the fucked up wiki formatting system still exists although it's better now
[12:52:03] <mitsuhiko> however, someone found my tekisuto draft and mailed me
[12:52:15] <mitsuhiko> and that someone apparently is the dokuwiki lexer developer
[12:52:31] <matt_good> which wiki formatting issue?
[12:52:48] <mitsuhiko> matt_good: cboos afair mentioned that the wiki formatter code is just crap :)
[12:52:48] * d0rt has quit IRC
[12:53:02] <mitsuhiko> at least the moin developers say so and trac took the moin code :)
[12:53:22] <matt_good> well, it is a bit messy
[12:53:40] <matt_good> actually Trac uses none of Moin's code
[12:53:45] <matt_good> just a similar syntax
[12:54:01] <mitsuhiko> matt_good: the trac formatter is written with exactly the same design as the moin formatter
[12:54:11] <mitsuhiko> so it's not a 1:1 port, but the architecture is the same
[12:54:50] <matt_good> hrm, well I haven't looked at how Moin does it, but we shouldn't be using their code AFAIK
[12:55:26] <mitsuhiko> matt_good: well. the variables are named different
[12:55:33] <matt_good> Moin is GPLed so I think jonas would've said something if he used it when we switched the license
[12:56:31] <matt_good> so, what about tekisuto?
[12:56:50] <mitsuhiko> well. i tried to write a clean text lexing solution
[12:57:13] <mitsuhiko> clean as of in: plugins can extend the syntax easily without hacking or screwing things up
[12:58:35] * tuxipuxi_ has joined #trac
[12:58:57] <matt_good> mitsuhiko: that could be another good project for the sprint next week
[12:59:46] <mitsuhiko> good idea. i'm currently replaying to that mail. maybe i can discuss that with harry next weekend on #pocoo
[12:59:57] <mitsuhiko> a text lexing solution for python would be a great thing imho
[13:02:57] <cmlenz> how far along did you get with that?
[13:03:09] <mitsuhiko> cmlenz: who with what?
[13:03:21] <cmlenz> you with your text-lexing thing ;-)
[13:03:35] <mitsuhiko> cmlenz: i have it since one month and stopped working on it for about the same time
[13:03:43] <mitsuhiko> nobody really had a look at it and so i forgot about it
[13:03:48] <mitsuhiko> apparently someone found it now ;D
[13:04:42] * m_g has quit IRC
[13:07:09] <cmlenz> ok
[13:08:08] <cmlenz> yeah, the examples look kinda neat
[13:08:44] <matt_good> mitsuhiko: is there a home page for tekisuto, or just the code?
[13:09:01] <mitsuhiko> matt_good: just the code
[13:09:34] <cmlenz> I wonder whether you need tokens to be class instances, instead of just (name, data) tuples
[13:09:43] <cmlenz> you'll be instantiating tons of objects
[13:09:47] <mitsuhiko> yeah
[13:09:57] <mitsuhiko> and what's a bigger problem: it's working char for char for not handled data
[13:10:00] <mitsuhiko> which is quite slow
[13:10:19] <mitsuhiko> it would be faster if i would be able to create regular expressions and use .search()
[13:10:28] <cmlenz> right
[13:10:28] <mitsuhiko> and yield data between matches as unhandled data
[13:12:04] <cmlenz> well, the problems always overlapping pattern matches
[13:12:17] <cmlenz> going char by char may be the only thing that works correctly
[13:12:19] <mitsuhiko> i hope the brainstorming with him will show up some new ideas
[13:12:39] <mitsuhiko> cmlenz: a c implementation would be much faster then
[13:13:29] <matt_good> well, I think my flight's about to board
[13:13:59] <matt_good> I'll talk to you when I get to PyCon
[13:14:57] <retracile> matt_good: have a safe trip. :)
[13:15:09] * matt_good has quit IRC
[13:36:27] * synic has left #trac
[13:54:22] * _shawn has quit IRC
[13:54:30] * _shawn has joined #trac
[14:25:08] * maxb has joined #trac
[14:26:46] <pacopablo> matt_good: getting to pycon friday morning
[14:27:00] * Blackhex has joined #trac
[14:29:48] * coderanger_ has quit IRC
[14:40:51] * hotte has quit IRC
[14:44:11] * _shawn has quit IRC
[14:51:18] * cmlenz has quit IRC
[14:53:32] * _shawn has joined #trac
[14:59:05] * jtoy has quit IRC
[15:25:27]