| [00:01:23] |
<pacopablo> |
why the time to index increases with each file |
| [00:01:53] |
<pacopablo> |
at the rate it's going, it's going to take a whole day to index one file by the time I get to the end |
| [00:01:54] |
<alect> |
your're not resetting the timer? :) |
| [00:02:21] |
* |
pacopablo slaps himself in the face |
| [00:04:01] |
<pacopablo> |
hmm, still experiencing the speed issue if the db isn't nuked |
| [00:04:17] |
<pacopablo> |
perhaps an artifact of killing the upgrade() process? |
| [00:04:38] |
<alect> |
it's a speedup or slowdown? |
| [00:05:05] |
<pacopablo> |
if I kill my update() process, and then rerun it, it slows down considerably |
| [00:05:10] |
<alect> |
it only reinserts words that have changed since the last update |
| [00:05:14] |
<pacopablo> |
going from ~10 seconds to over a minute |
| [00:05:19] |
<alect> |
so it has to fetch the words, then reinsert |
| [00:05:27] |
<alect> |
that could be the slowdown... :\ |
| [00:06:21] |
<pacopablo> |
hmm, there appears to be very little relation between the size of the file and the amount of time it takes to add it |
| [00:06:58] |
<alect> |
after an aborted update? |
| [00:07:08] |
<pacopablo> |
on a fresh update |
| [00:07:33] |
<pacopablo> |
well, huge slowdown after aborted update, no relation between size on fresh update |
| [00:08:09] |
<alect> |
that indicates the update is dominated by i/o access |
| [00:08:16] |
<alect> |
or gdbm mojo |
| [00:08:30] |
<alect> |
rather than processing of the document contents |
| [00:08:34] |
<alect> |
try bdb? |
| [00:08:46] |
<pacopablo> |
gdbm mojo perhaps, not i/o access |
| [00:09:06] |
<pacopablo> |
disk's aren't hammering |
| [00:10:01] |
<pacopablo> |
wonder how anydbm picks the db backend to use |
| [00:13:53] |
<pacopablo> |
hmm, will have to recompile python with support for bdb |
| [00:14:24] |
* |
hpnadig has joined #trac |
| [00:16:39] |
<alect> |
doh :( |
| [00:18:38] |
<alect> |
don't worry about it paco |
| [00:18:42] |
<alect> |
i'll do some more testing |
| [00:18:56] |
<pacopablo> |
bah, it's a challenge now :) |
| [00:19:04] |
<pacopablo> |
besides, I'll let it run while I do dishes |
| [00:19:05] |
<pacopablo> |
bbbl |
| [00:19:11] |
<pacopablo> |
or maybe not ;) |
| [00:19:25] |
<alect> |
hehe |
| [00:19:48] |
* |
Guardian has quit IRC |
| [00:24:23] |
<alect> |
hmmm |
| [00:24:27] |
<alect> |
uncool |
| [00:38:05] |
* |
klando has joined #trac |
| [00:38:31] |
* |
kop_ has joined #trac |
| [00:43:24] |
* |
lowkey has joined #trac |
| [01:12:22] |
* |
tsb has joined #trac |
| [01:34:35] |
* |
whitelynx|deskto has quit IRC |
| [01:43:15] |
<pacopablo> |
hmm, bdb backend starts out worse than the gdbm database |
| [01:47:37] |
* |
greg_72 has quit IRC |
| [01:50:24] |
<pacopablo> |
hmm, not super great time on my pentium m 2 GHz |
| [01:50:50] |
<pacopablo> |
and it still seems that the more files one has, the longer each file takes |
| [01:51:11] |
<pacopablo> |
I'll post full results tomorrow (if they're done) |
| [01:51:21] |
<pacopablo> |
night all |
| [02:04:33] |
* |
whitelynx has joined #trac |
| [02:13:56] |
<alect> |
paco: send me your data |
| [02:16:02] |
<asmodai> |
alect, pacopablo: howdy folks |
| [02:19:13] |
<alect> |
hey asmodai |
| [02:22:14] |
<asmodai> |
alect: Sup dude? |
| [02:22:35] |
<alect> |
hax0ring |
| [02:25:20] |
<asmodai> |
haha |
| [02:25:22] |
<asmodai> |
The usual eh? |
| [02:35:08] |
* |
Guardian has joined #trac |
| [03:07:19] |
* |
kop_ is now known as kop-lunch |
| [03:12:19] |
<alect> |
hehe, yep :) |
| [03:12:21] |
<alect> |
how about you? |
| [03:15:58] |
* |
Synapse_ has joined #trac |
| [03:17:49] |
* |
Synapse has quit IRC |
| [03:18:00] |
* |
Synapse_ is now known as Synapse |
| [03:28:34] |
* |
whereami has joined #trac |
| [03:29:16] |
* |
whereami has left #trac |
| [03:51:45] |
* |
skar has left #trac |
| [03:52:06] |
* |
skar has joined #trac |
| [03:52:16] |
* |
skar has left #trac |
| [03:54:52] |
* |
kahn has joined #trac |
| [04:42:06] |
* |
pygi has joined #trac |
| [04:56:17] |
* |
pygi has quit IRC |
| [05:15:14] |
* |
kop-lunch is now known as kop_ |
| [05:20:41] |
* |
Semhirage has quit IRC |
| [05:21:12] |
* |
Semhirage has joined #Trac |
| [05:38:16] |
* |
stevegt has quit IRC |
| [05:59:06] |
* |
n0rus has joined #trac |
| [05:59:33] |
<n0rus> |
How do I protect the main page (Available Projects) with a password? Like with an .htaccess file |
| [06:01:04] |
<misc> |
with mod_python ? |
| [06:01:40] |
<n0rus> |
I'm running Trac without Apache |
| [06:02:53] |
* |
d0rt has joined #trac |
| [06:04:04] |
* |
n0rus has quit IRC |
| [06:14:46] |
* |
xjjk has quit IRC |
| [06:15:32] |
* |
LionsMane has joined #trac |
| [06:24:40] |
* |
idnar_ has joined #trac |
| [06:24:46] |
* |
idnar has quit IRC |
| [06:25:33] |
* |
agile has quit IRC |
| [06:26:12] |
* |
kahn has quit IRC |
| [06:33:28] |
* |
hpnadig_ has joined #trac |
| [06:35:12] |
* |
elehack|work has joined #trac |
| [06:41:10] |
* |
hpnadig has quit IRC |
| [07:17:02] |
* |
rjdave_ has joined #trac |
| [07:21:45] |
* |
rjdave has quit IRC |
| [07:23:02] |
* |
danbeck has joined #trac |
| [07:32:52] |
* |
agile has joined #trac |
| [07:36:10] |
* |
De_Mon has left #trac |
| [07:43:38] |
* |
chandlerc has quit IRC |
| [07:50:13] |
* |
kop_ has quit IRC |
| [07:50:31] |
* |
kop_ has joined #trac |
| [08:27:31] |
* |
m_g has joined #trac |
| [08:37:37] |
<pacopablo> |
morning |
| [08:39:32] |
<klando> |
coffee ? |
| [08:47:51] |
<m_g> |
alect, coderanger: one of you there? |
| [08:47:58] |
<coderanger> |
Indeed |
| [08:48:12] |
<coderanger> |
Though alect is probably asleep |
| [08:48:55] |
<m_g> |
i need to unset a custom field with the sqlite command tool because i can't do in the web interface, but i need to be sure that i don't fuck up anything. the custom field is named 'revision' and the ticket is '1813'. is UPDATE ticket_custom SET value='' WHERE ticket=1813 AND name='revision'; correct? |
| [08:49:49] |
<coderanger> |
looks okay |
| [08:50:00] |
<coderanger> |
why can't you do it from the web? |
| [08:50:06] |
* |
BrianHV has quit IRC |
| [08:50:52] |
<m_g> |
because it's a hidden custom field and i don't want to enable it now. |
| [08:50:57] |
<m_g> |
the command worked, thanks := |
| [08:50:59] |
<m_g> |
:) |
| [08:51:00] |
<coderanger> |
ahh |
| [08:52:09] |
* |
klando has quit IRC |
| [08:52:43] |
<m_g> |
btw, were you able to help the guy who wrote to you about the screwed trac DB? |
| [08:53:57] |
<coderanger> |
which guy? |
| [08:54:43] |
<m_g> |
Hershel, it was a commercial help request |
| [08:54:47] |
<m_g> |
by email |
| [08:55:58] |
<coderanger> |
ahh, haven't responded to that yet |
| [08:56:19] |
<coderanger> |
been busy with school the last few days |
| [08:56:25] |
<m_g> |
ah, i see |
| [09:02:13] |
<pacopablo> |
gah, alect, your mailserver sucks ;) |
| [09:15:56] |
* |
siezer has left #trac |
| [09:23:24] |
* |
hpnadig_ has quit IRC |
| [09:38:48] |
* |
thm has joined #trac |
| [09:40:23] |
* |
BrianHV has joined #trac |
| [09:53:13] |
<unlord> |
hi |
| [09:53:23] |
<unlord> |
I'm using trac 0.10.1 and I had a request come up at work |
| [09:53:56] |
<unlord> |
we'd like to use keywords to classify our tickets, and would like to generate a report that lists all tickets sorted by keyword |
| [09:54:27] |
<pacopablo> |
whip out your SQL foo |
| [09:54:46] |
<pacopablo> |
if you have REPORT_ADMIN permissions, you can simply create a new report |
| [09:55:14] |
<unlord> |
I am TRAC_ADMIN and I have the 'Create new report' button |
| [09:55:28] |
<pacopablo> |
that's the ticket (no pun intended) |
| [09:55:32] |
<unlord> |
heh |
| [09:55:57] |
<unlord> |
our keywords are comma separated |
| [09:56:13] |
<unlord> |
what does this report editor expect |
| [09:56:16] |
<pacopablo> |
you could use the tags plugin to get most of what you want |
| [09:56:31] |
<unlord> |
hmm, thats a good idea |
| [09:56:50] |
* |
Guard][an has joined #trac |
| [09:58:38] |
* |
Guard][an has quit IRC |
| [09:58:57] |
<unlord> |
pacopablo: it says I can tag my tickets |
| [10:01:46] |
* |
lowkey has quit IRC |
| [10:01:59] |
* |
l0wkey has joined #trac |
| [10:02:17] |
* |
Guardian has quit IRC |
| [10:04:07] |
<unlord> |
hmm, I bet I could make an ebuild for this |
| [10:04:13] |
<coderanger> |
for tags? |
| [10:04:18] |
<unlord> |
yep |
| [10:04:22] |
<unlord> |
there is one for webadmin |
| [10:04:57] |
<coderanger> |
"easy_install http://trac-hacks.org/svn/tagsplugin/trunk" |
| [10:05:28] |
<unlord> |
yah, its dirt simple |
| [10:05:50] |
<unlord> |
but I don't believe in installing anything manually |
| [10:06:18] |
<coderanger> |
"emerge trac-tags" isn't much less manual ;-) |
| [10:06:51] |
<unlord> |
no, but I can then do emerge -C trac-tags and it removes it |
| [10:06:57] |
<coderanger> |
fair enough |
| [10:07:06] |
<unlord> |
or "emerge -DuN world" and it upgrades it |
| [10:07:10] |
<coderanger> |
what would be nice is something like g-cpan.pl for PyPI |
| [10:07:27] |
<unlord> |
isn't that what setuptools is? |
| [10:07:32] |
<coderanger> |
No |
| [10:07:47] |
* |
agile has quit IRC |
| [10:07:48] |
* |
thm has quit IRC |
| [10:07:50] |
<coderanger> |
g-cpan autogenerates ebuilds for any package listed on CPAN |
| [10:08:04] |
<coderanger> |
and then emerges it |
| [10:10:10] |
<unlord> |
hmm, so trac-tags wouldn't actually let me get a list of all tickets sorted by tags? |
| [10:10:16] |
<coderanger> |
sure it will |
| [10:10:39] |
<unlord> |
how do I do that? |
| [10:11:05] |
<unlord> |
I'm looking at http://www.trac-hacks.org/ fo how they use tags |
| [10:11:24] |
* |
em-dash has joined #trac |
| [10:12:43] |
<coderanger> |
Do you know what tags you will be using, or is it variable? |
| [10:12:59] |
<unlord> |
I'd like it to be ad-hoc |
| [10:13:12] |
<coderanger> |
That would require making a macro to do the listing |
| [10:13:16] |
<unlord> |
if I knew what they were and there was a 1-1 mapping to tickets, I'd just use one of the built in classifiers |
| [10:13:48] |
<unlord> |
hmm, how do macro's work? |
| [10:14:09] |
<coderanger> |
Take a look at the existing ListTags and ListTagged macros in the tags source |
| [10:14:13] |
<coderanger> |
basically just combine them |
| [10:14:24] |
<coderanger> |
So for each tag is calls ListTagged |
| [10:16:22] |
<unlord> |
but I only want to show tickets |
| [10:18:31] |
<coderanger> |
just use the tagspace "tickets" |
| [10:18:40] |
<unlord> |
oh yah |
| [10:18:42] |
<unlord> |
thats smart |
| [10:23:59] |
* |
pygi has joined #trac |
| [10:27:32] |
* |
pacopablo curses swapoff.org's mail system |
| [10:28:31] |
* |
em-dash has quit IRC |
| [10:28:35] |
<coderanger> |
pacopablo: Whats not working? |
| [10:28:49] |
* |
Ragnarok has joined #trac |
| [10:29:04] |
<pacopablo> |
stupid attachment size bloking |
| [10:29:06] |
<pacopablo> |
er, blocking |
| [10:29:19] |
<pacopablo> |
or some kind of av false-positive |
| [10:29:36] |
<pacopablo> |
Diagnostic code: smtp;550 This message contains a virus (BZip.ExceededFileSize). |
| [10:30:52] |
* |
omry has quit IRC |
| [10:30:58] |
<coderanger> |
Thats a normal defensive system in AVs. Prevents a DoS by forcing it to unarchive for a long time. |
| [10:31:21] |
<pacopablo> |
it's a 3 meg bzip |
| [10:31:27] |
<pacopablo> |
er, bzip'd tar |
| [10:31:27] |
* |
omry has joined #trac |
| [10:31:28] |
<coderanger> |
though it is usually set at a few hundred megs or something |
| [10:31:35] |
<coderanger> |
heh, now thats a little odd :P |
| [10:31:46] |
<pacopablo> |
it has 308 text files, all under 200k |
| [10:31:53] |
<coderanger> |
irc logs? |
| [10:31:57] |
<pacopablo> |
yup |
| [10:32:08] |
<pacopablo> |
so, I'm falling back to trac |
| [10:32:28] |
<coderanger> |
You have a login on trac-hacks no? |
| [10:32:40] |
<coderanger> |
Just SCP it there an let him download it himself :P |
| [10:32:48] |
<pacopablo> |
ahh, that's even better |
| [10:33:00] |
<pacopablo> |
I was going to just put it up on pacopablo.com |
| [10:33:17] |
<pacopablo> |
probably still will, since I'm halfway done |
| [10:33:39] |
<coderanger> |
heh |
| [10:35:45] |
* |
tpollari has joined #trac |
| [10:37:30] |
<Ragnarok> |
hi |
| [10:37:57] |
<Ragnarok> |
any form to manual install the webadmin plugin? |
| [10:38:12] |
<coderanger> |
form? |
| [10:38:30] |
<Ragnarok> |
form = idea |
| [10:38:32] |
<Ragnarok> |
sorry my english |
| [10:38:41] |
<coderanger> |
@wiki WebAdmin |
| [10:38:41] |
<evil_twin> |
coderanger: http://trac.edgewall.org/wiki/WebAdmin |
| [10:38:44] |
<Ragnarok> |
i download the code from svn |
| [10:38:46] |
<coderanger> |
@wiki TracPlugins |
| [10:38:46] |
<evil_twin> |
coderanger: http://trac.edgewall.org/wiki/TracPlugins |
| [10:39:14] |
<Ragnarok> |
when install i get this error: |
| [10:39:16] |
<Ragnarok> |
tom:/tmp/webadmin# python setup.py install |
| [10:39:17] |
<Ragnarok> |
Traceback (most recent call last): |
| [10:39:17] |
<Ragnarok> |
File "setup.py", line 25, in ? |
| [10:39:17] |
<Ragnarok> |
packages=find_packages(exclude=['ez_setup', '*.tests*']), |
| [10:39:17] |
<Ragnarok> |
TypeError: find_packages() got an unexpected keyword argument 'exclude' |
| [10:39:17] |
<Ragnarok> |
t |
| [10:39:28] |
* |
chandlerc has joined #trac |
| [10:40:43] |
<Ragnarok> |
the use of egg is required? or can i put the files in my project environment dirs? |
| [10:42:19] |
<s0undt3ch> |
what's the var I should check inside a cs template to see if current user is annonymous? |
| [10:44:02] |
* |
kop_ has quit IRC |
| [10:44:03] |
<Ragnarok> |
i think that is $USER |
| [10:45:30] |
* |
whitelynx|firest has joined #trac |
| [10:46:17] |
* |
tpollari has quit IRC |
| [10:46:40] |
* |
Ragnarok has left #trac |
| [10:46:47] |
<coderanger> |
s0undt3ch: trac.authname |
| [10:48:05] |
* |
pygi has quit IRC |
| [10:48:15] |
<s0undt3ch> |
ah ha |
| [10:48:16] |
<s0undt3ch> |
<?cs if:trac.authname == "anonymous" ?> |
| [10:48:22] |
<s0undt3ch> |
coderanger: oh, you said that :) |
| [10:55:48] |
* |
mrb_ has joined #trac |
| [10:58:09] |
* |
tpollari has joined #trac |
| [10:58:38] |
<mrb_> |
http://pastie.textmate.org/38594 <-- anyone have a hint for this error? |
| [11:00:16] |
* |
agile has joined #trac |
| [11:00:30] |
* |
coderanger_ has joined #trac |
| [11:02:08] |
<pacopablo> |
mrb_: is the monotone plugin known to work with 0.11dev? |
| [11:02:19] |
<mrb_> |
yeah, it has worked |
| [11:03:17] |
<mrb_> |
its home location runs 0.11dev too: http://tracmtn.1erlei.de/ |
| [11:07:18] |
* |
tolsen has joined #trac |
| [11:11:08] |
* |
elehack|work has quit IRC |
| [11:16:43] |
* |
Paulymer has joined #trac |
| [11:16:52] |
* |
hotte has joined #trac |
| [11:22:24] |
* |
l0wkey has quit IRC |
| [11:23:27] |
* |
whitelynx|firest has quit IRC |
| [11:23:51] |
* |
dilinger has quit IRC |
| [11:32:10] |
<pacopablo> |
mrb_: so what's changed? |
| [11:32:21] |
<mrb_> |
good question |
| [11:32:32] |
<mrb_> |
i'm tempted to say nothing |
| [11:33:02] |
<pacopablo> |
did you upgrade trac? |
| [11:33:08] |
<mrb_> |
nope |
| [11:33:45] |
<mrb_> |
i'm gonna think it ove thru dinner, back in a bit |
| [11:35:16] |
* |
mrb_ has quit IRC |
| [11:36:53] |
* |
mrb_ has joined #trac |
| [11:42:57] |
* |
idnar_ is now known as idnar |
| [11:56:04] |
* |
Guardian has joined #trac |
| [12:02:56] |
* |
dilinger has joined #trac |
| [12:05:22] |
* |
em-dash has joined #trac |
| [12:05:47] |
* |
Guard][an has joined #trac |
| [12:15:47] |
* |
rlo has joined #trac |
| [12:17:48] |
<rlo> |
hello trac |
| [12:17:53] |
<pacopablo> |
jello |
| [12:17:54] |
<coderanger_> |
alo |
| [12:18:13] |
<rlo> |
I have more trac inquiries:/ |
| [12:18:36] |
<pacopablo> |
coderanger_: no, no, it's 'rlo' :) |
| [12:20:54] |
<rlo> |
I need to allow a remotely hosted form to post to a trac instance. |
| [12:21:01] |
* |
thansen has left #trac |
| [12:21:09] |
<coderanger_> |
XmlRpcPlugin |
| [12:21:20] |
<rlo> |
I can't make the form on the other end and I don't know if I can have them handle cookies |
| [12:21:25] |
<rlo> |
ooooh |
| [12:21:33] |
<rlo> |
XMLRPCPlugin??? |
| [12:21:35] |
<rlo> |
where |
| [12:21:38] |
<coderanger_> |
Its on trac-hacks |
| [12:21:39] |
* |
rlo googles |
| [12:22:34] |
<rlo> |
yummy thanks |
| [12:22:58] |
<coderanger_> |
What is you goal? |
| [12:27:10] |
* |
Guardian has quit IRC |
| [12:28:44] |
* |
chandlerc has quit IRC |
| [12:31:25] |
<rlo> |
essentially to create a means of remotely posting new tickets. I know that trac uses a cooke with a new ticket form, so I would have to hack at trac or make a go between service |
| [12:31:43] |
<rlo> |
I think XMLRPCPlugin is probably perfect |
| [12:31:47] |
<rlo> |
thanks again |
| [12:31:58] |
<coderanger_> |
Yes, but beware of CSRF |
| [12:32:18] |
<coderanger_> |
Thats why the protection system was created in the first place |
| [12:33:14] |
* |
Blackhex has joined #trac |
| [12:37:05] |
* |
chandlerc has joined #trac |
| [12:37:11] |
<rlo> |
CSRF? |
| [12:37:30] |
<coderanger_> |
cross-site request forgery |
| [12:37:31] |
<misc> |
cross site request forgery ? |
| [12:38:09] |
<rlo> |
this will be on a closed Intranet. But elaborate anyways please |
| [12:38:18] |
* |
Paulymer has left #trac |
| [12:38:21] |
<coderanger_> |
Check google |
| [12:38:27] |
<coderanger_> |
its a type of site attack |
| [12:38:35] |
<rlo> |
THis I figured |
| [12:38:42] |
<coderanger_> |
and it doesn't matter that the app is on protected net |
| [12:38:43] |
<rlo> |
will look later |
| [12:38:52] |
<rlo> |
thanks for bringing it to my attention |
| [12:38:56] |
<coderanger_> |
it is still vulnerable if the users will have Internet access |
| [12:46:12] |
* |
stevegt has joined #trac |
| [12:55:59] |
<mrb_> |
pacopablo: any more thoughts on http://pastie.textmate.org/38594 ? |
| [13:16:48] |
<rlo> |
showstopper in xmlrppcplugin |
| [13:20:50] |
<coderanger_> |
hmm? |
| [13:21:03] |
* |
Blackhex has left #trac |
| [13:21:31] |
* |
tpollari has quit IRC |
| [13:22:01] |
* |
tpollari has joined #trac |
| [13:35:41] |
<pacopablo> |
mrb_: contact the monotone dev first, then email trac-dev |
| [13:36:41] |
<mrb_> |
pacopablo: will do, thanks |
| [13:41:45] |
<rlo> |
http://pastie.textmate.org/38654 |
| [13:48:15] |
<rlo> |
how do I make a plugin without ez_setup |
| [13:48:29] |
<rlo> |
I think the egg is doing it |
| [13:54:33] |
<pacopablo> |
rlo: do you have setuptools installed? |
| [13:54:47] |
<rlo> |
yes |
| [13:54:52] |
<rlo> |
it makes the egg |
| [13:54:56] |
<pacopablo> |
then you don't need ez_setup |
| [13:55:07] |
<pacopablo> |
that's just to install setuptools |
| [13:55:11] |
<rlo> |
but makes the error I pasted in pastie and posted earlier |
| [13:55:19] |
<pacopablo> |
and if you have setuptools, you have easy_install |
| [13:55:31] |
<rlo> |
http://pastie.textmate.org/38654 |
| [13:56:03] |
<pacopablo> |
hmm, that's not a bug with setuptools |
| [13:57:21] |
<pacopablo> |
that's an issue importing os |
| [13:57:36] |
<pacopablo> |
weird python installation? |
| [14:00:47] |
* |
hotte has quit IRC |
| [14:00:47] |
* |
Guardian has joined #trac |
| [14:01:01] |
<rlo> |
works fine without the egg |
| [14:01:01] |
<rlo> |
there are others with the same experience |
| [14:01:01] |
<rlo> |
os.py from posixpath |
| [14:01:03] |
<rlo> |
oops |
| [14:01:06] |
<rlo> |
http://trac.edgewall.org/ticket/4280 |
| [14:01:09] |
<rlo> |
want to install without setup tools if there is a way |
| [14:01:19] |
<pacopablo> |
python setup.py install |
| [14:02:25] |
<alect> |
hello |
| [14:05:51] |
<alect> |
those timings are odd paco |
| [14:06:30] |
* |
Guardian has quit IRC |
| [14:06:30] |
* |
Guard][an has quit IRC |
| [14:06:30] |
* |
mrb_ has quit IRC |
| [14:06:30] |
* |
agile has quit IRC |
| [14:06:30] |
* |
rlo has quit IRC |
| [14:06:30] |
* |
BrianHV has quit IRC |
| [14:06:39] |
* |
whitelynx|blackb has joined #trac |
| [14:06:39] |
* |
Guardian has joined #trac |
| [14:06:39] |
* |
rlo has joined #trac |
| [14:06:39] |
* |
Guard][an has joined #trac |
| [14:06:39] |
* |
mrb_ has joined #trac |
| [14:06:39] |
* |
agile has joined #trac |
| [14:06:39] |
* |
BrianHV has joined #trac |
| [14:09:04] |
* |
pygi has joined #trac |
| [14:09:33] |
* |
mario_ has joined #trac |
| [14:10:17] |
<alect> |
very strange indeed |
| [14:10:43] |
* |
mario_ has left #trac |
| [14:10:57] |
<alect> |
it does actually seem to be increasing in time for me too though |
| [14:11:12] |
<alect> |
but each file between 0.1 and 0.5 seconds |
| [14:13:00] |
<alect> |
hmmm |
| [14:13:02] |
<alect> |
and the index is 42M :) |
| [14:13:15] |
<alect> |
suboptimal |
| [14:13:40] |
<alect> |
i'll need to add a uri/word index i think |
| [14:15:51] |
<alect> |
Total time to index files: 119.464354992 |
| [14:16:37] |
<alect> |
you need to do a framework.close() at the ned btw |
| [14:17:30] |
<alect> |
paco! talk to me man! |
| [14:17:49] |
* |
Guard][an has quit IRC |
| [14:18:06] |
* |
tpollari has quit IRC |
| [14:18:26] |
* |
whitelynx has quit IRC |
| [14:18:37] |
* |
tpollari has joined #trac |
| [14:21:53] |
* |
mitsuhiko is waiting for the next trac clone... |
| [14:22:20] |
<mitsuhiko> |
it's interesting how fast clones and copies go 1.0 Oo |
| [14:22:35] |
<alect> |
mitsuhiko: which ones are you referring to? |
| [14:22:43] |
<mitsuhiko> |
alect: at the moment retrospectiva |
| [14:22:48] |
<mitsuhiko> |
that collaboa fork |
| [14:22:49] |
<alect> |
paco: on the upside, the search is almost instantaneous :) |
| [14:23:48] |
<alect> |
mitsu: don't think i've seen that one |
| [14:25:01] |
<mitsuhiko> |
i noticed two weeks ago for the first time |
| [14:25:17] |
<mitsuhiko> |
(when i found out that someone continued maintaining collaboa) |
| [14:25:42] |
* |
tolsen has quit IRC |
| [14:26:05] |
<alect> |
doesn't seem too bad |
| [14:30:23] |
<alect> |
paco: hype: Total time to index files: 9.9451 |
| [14:32:07] |
<alect> |
pcao: xapian: Total time to index files: 11.1465 |
| [14:32:24] |
<coderanger_> |
Is that minutes or seconds? |
| [14:32:26] |
<alect> |
evidently the builtin indexer needs some love |
| [14:32:28] |
<alect> |
seconds |
| [14:32:40] |
<alect> |
search time is negligible for all of them |
| [14:32:48] |
<alect> |
100ths of a second |
| [14:33:00] |
<alect> |
the indexes are...large however |
| [14:33:11] |
<alect> |
xapian's is 30M |
| [14:35:10] |
<alect> |
paco: does file pyndexter.idx/builtin.db/* show them as bdb files? |
| [14:41:00] |
<pacopablo> |
alect: one sec |
| [14:42:15] |
* |
d0rt has quit IRC |
| [14:42:16] |
* |
kahn has joined #trac |
| [14:45:06] |
* |
danbeck has quit IRC |
| [14:48:00] |
<alect> |
ok |
| [14:48:04] |
<alect> |
off to work, bbiab |
| [14:51:45] |
<alect> |
actually i just forced gdbm and it was about the same speed |
| [14:52:12] |
* |
alect ejects |
| [14:52:34] |
<alect> |
paco: updated version: http://swapoff.org/files/pydtest.py |
| [14:52:45] |
<pacopablo> |
k |
| [14:53:01] |
<pacopablo> |
hey, do you have the patches that I submitted applied, also? |
| [15:03:01] |
* |
rlo has quit IRC |
| [15:06:58] |
<pacopablo> |
oh my goodness! |
| [15:07:05] |
<pacopablo> |
it's the self.db.keys() patch |
| [15:07:26] |
<pacopablo> |
removed that and it's down to 1 to 2 seconds a file |
| [15:07:35] |
<pacopablo> |
which is much more what I expected |
| [15:10:02] |
* |
agile has quit IRC |
| [15:14:58] |
<pacopablo> |
length of time to index still increases with each file, but it's manageable, around 4 secs per file now |
| [15:32:56] |
* |
m_g has quit IRC |
| [15:36:05] |
* |
coderanger_ has quit IRC |
| [15:36:39] |
<pacopablo> |
yummy! pyndexter works! |
| [15:36:52] |
<pacopablo> |
of course, inow need to figure out how to use ti properly ;) |
| [15:46:17] |
<alect> |
what the? |
| [15:46:28] |
<alect> |
paco: you're fired! |
| [15:47:37] |
<alect>< |
