| [00:00:44] |
<tux21b> |
things i would like to see in python is reloadable and revisioned modules, better optimazations and no gil :) |
| [00:05:56] |
* |
luks has quit IRC |
| [00:05:59] |
* |
luks has joined #pocoo |
| [00:08:11] |
* |
jinks1 has joined #pocoo |
| [00:10:13] |
* |
Kaelten has quit IRC |
| [00:10:38] |
* |
highwaychile has quit IRC |
| [00:17:11] |
<mitsuhiko> |
yeah. revisioned modules. that would be it |
| [00:19:19] |
* |
jinks has quit IRC |
| [00:21:41] |
<xorAxAx> |
mitsuhiko: how would that work? |
| [00:22:08] |
<mitsuhiko> |
xorAxAx: what. revisioned modules? |
| [00:22:16] |
<mitsuhiko> |
*versioned modules |
| [00:22:36] |
<tux21b> |
xorAxAx, do you know erlang? :) |
| [00:22:37] |
<xorAxAx> |
yes |
| [00:22:47] |
<xorAxAx> |
erlang has completly different binding semantics |
| [00:22:55] |
<mitsuhiko> |
xorAxAx: from jinja[>=2] import Template |
| [00:23:07] |
<mitsuhiko> |
requires of course that sys.modules goes baba |
| [00:23:18] |
<xorAxAx> |
thats something else than tux21b meant |
| [00:23:27] |
<tux21b> |
ah, that would be a good idea too, but i was speaking of code changes |
| [00:23:35] |
<xorAxAx> |
and its surely funny if you have components A and B both using C but in different versions |
| [00:23:41] |
<xorAxAx> |
so you have a split brain globals issue |
| [00:23:49] |
<xorAxAx> |
but globals are bad anyway |
| [00:24:13] |
<mitsuhiko> |
xorAxAx: yes. but i'm more interested in different versions of the same library in the same process |
| [00:24:30] |
<mitsuhiko> |
reason: sqlalchemy for example replaces the api every month :) |
| [00:24:34] |
<xorAxAx> |
yes, with the aforementioned issue |
| [00:24:41] |
<tux21b> |
i like the idea too, would be rally useful |
| [00:24:45] |
<xorAxAx> |
yeah, sqlalchemy is using a lot of globals |
| [00:25:14] |
* |
mitsuhiko likes the assembly loading system of .net |
| [00:25:23] |
<mitsuhiko> |
unfortunately that won't work in python |
| [00:25:33] |
<xorAxAx> |
imagine an object of an old sqlalchemy version passing from component a to b |
| [00:25:34] |
* |
luks has quit IRC |
| [00:25:36] |
* |
luks has joined #pocoo |
| [00:26:06] |
<mitsuhiko> |
xorAxAx: why would it do that? |
| [00:26:18] |
<xorAxAx> |
well, depends on your app |
| [00:26:29] |
<xorAxAx> |
whether it separates both domains correctly |
| [00:26:37] |
<xorAxAx> |
and whether they are clearly separated |
| [00:26:47] |
<mitsuhiko> |
i have django and sa running in the same process |
| [00:26:55] |
<mitsuhiko> |
i want sa1 and sa2 running there too :) |
| [00:26:58] |
<mitsuhiko> |
separated from each other |
| [00:27:17] |
* |
yashh has joined #pocoo |
| [00:27:21] |
<mitsuhiko> |
because for example lodgeit wants sa 0.4 and my textpress blog wants 0.5 and i want them to be "one website" :) |
| [00:27:24] |
<yashh> |
hello all |
| [00:27:38] |
<xorAxAx> |
mitsuhiko: yeah, if its separate apps ... |
| [00:27:49] |
<xorAxAx> |
mitsuhiko: but then you usually do http proxying |
| [00:27:57] |
<tux21b> |
or we would like to rewrite the django parts to sa 0.5 while the forum still uses 0.4 |
| [00:28:06] |
<mitsuhiko> |
sure. but it's a stupid limitation of python because sys.modules is a singleton |
| [00:28:28] |
<mitsuhiko> |
you can somewhat work around that issue in ruby |
| [00:28:46] |
<mitsuhiko> |
but it's not solved there either |
| [00:29:05] |
<mitsuhiko> |
because... ruby loves monkey business :) |
| [00:29:07] |
<mitsuhiko> |
hi yashh |
| [00:29:15] |
<yashh> |
hi yashh |
| [00:29:18] |
<yashh> |
oops |
| [00:29:21] |
<yashh> |
hi mitsuhiko |
| [00:29:50] |
<xorAxAx> |
mitsuhiko: one could imagine some kind of import context |
| [00:29:58] |
<xorAxAx> |
mitsuhiko: that replaces sys.modules with different sets |
| [00:30:24] |
<mitsuhiko> |
xorAxAx: that's basically what i do in zine for plugins loaded from different application instances |
| [00:30:36] |
<mitsuhiko> |
but that only works because the code limits plugins in what they are doing |
| [00:30:48] |
<mitsuhiko> |
eg, they are not allowed to monkey patch or access other versions of itself |
| [00:30:55] |
<mitsuhiko> |
or interact with other applications at all |
| [00:31:11] |
* |
maze_ has joined #pocoo |
| [00:31:12] |
* |
maze has quit IRC |
| [00:38:30] |
<yashh> |
hey |
| [00:38:41] |
<yashh> |
whats the adv of SQLAlchemy |
| [00:38:54] |
<yashh> |
basically I am a Django guy |
| [00:40:14] |
<ronny> |
yashh: sqlalchemy actually is usefull when it comes to databases, while djangos orm basicaly uses them just for object persistence |
| [00:40:31] |
<yashh> |
yes |
| [00:40:42] |
<yashh> |
i heard its more good than Django ORM |
| [00:41:03] |
<ronny> |
well, djangos orm is like a nasty toy compared to sqlalchemy |
| [00:41:20] |
<ronny> |
also there is the sql expression layer |
| [00:42:06] |
<ronny> |
its basically the most powerfull python tool for database interaction |
| [00:42:39] |
<yashh> |
cool |
| [00:42:54] |
* |
maze_ is now known as maze |
| [00:43:01] |
<yashh> |
there was a small discussion on #django regarding Sql Alchemy |
| [00:43:12] |
<yashh> |
but to use it with django it needs a lot of customization |
| [00:45:45] |
* |
yashh has left #pocoo |
| [02:20:16] |
* |
heikkit has quit IRC |
| [02:22:10] |
* |
davidcramer has joined #pocoo |
| [02:32:09] |
* |
mtrichardson has quit IRC |
| [02:33:21] |
* |
alowry has quit IRC |
| [02:52:51] |
* |
kleinweby_ has joined #pocoo |
| [02:59:45] |
* |
alowry has joined #pocoo |
| [03:01:49] |
* |
kleinweby has quit IRC |
| [03:08:09] |
* |
saptah has quit IRC |
| [03:15:55] |
* |
maze_ has joined #pocoo |
| [03:20:43] |
* |
maze has quit IRC |
| [03:43:52] |
* |
Baumichel has quit IRC |
| [05:13:14] |
* |
saptah has joined #pocoo |
| [05:31:04] |
* |
rcscomp2 has quit IRC |
| [05:31:06] |
* |
aconbere has joined #pocoo |
| [05:36:37] |
* |
rcscomp has joined #pocoo |
| [05:39:33] |
<rcscomp> |
has anyone had problems with the werkzeug http server being really slow when being used form another machine? When I test a website built on my desktop with a browser on my laptop, the page takes about 12 seconds to load. When testing from my desktop its intantaneous. I have a gigabit connection between the two and don't have any other connection issues between the two computers. |
| [05:55:03] |
<prencher> |
rcscomp - i had a very odd stalling issue when my ip had no reverse dns.. maybe you're seeing similar |
| [05:55:56] |
<prencher> |
rcscomp - try to go to your hosts file, add something to the effect of <your ip> localhost, on the computer with the werkzeug instance running on |
| [05:56:08] |
<prencher> |
it's a stretch but it worked for me back when my isp didnt assign rdns to my ip |
| [05:56:33] |
* |
davidcramer has quit IRC |
| [05:56:45] |
<rcscomp> |
prencher: hmmmm...that might be the culprit since I do my own internal DNS. Do you know if I can turn off resolution on the HTTP server output? |
| [05:57:09] |
* |
davidcramer has joined #pocoo |
| [06:01:20] |
<rcscomp> |
prencher: bingo!!!! I did a test entry in my hosts file and that solved the delay. Thank you soooooooooooo much! |
| [06:04:52] |
<prencher> |
hehe, you're welcome |
| [06:06:46] |
<prencher> |
rcscomp - as far as turning off the rdns.. im going to write a patch for it soon.. mostly because it just pisses me off to have rdns as long as the shell in the dev logs |
| [06:08:32] |
<hads> |
I've run across the same issue actually, meant to look into patching it too. |
| [06:12:49] |
<prencher> |
remind me tomorrow |
| [06:12:56] |
<prencher> |
for now, bed time |
| [06:14:01] |
<rcscomp> |
prencher: thanks for your help. Good night. |
| [06:15:21] |
* |
aconbere has quit IRC |
| [06:27:03] |
* |
aconbere has joined #pocoo |
| [06:33:58] |
* |
KirkMcDonald has quit IRC |
| [06:34:02] |
* |
KirkMcDonald has joined #pocoo |
| [07:04:40] |
* |
rcscomp has quit IRC |
| [07:05:02] |
* |
rcscomp has joined #pocoo |
| [07:07:43] |
<rcscomp> |
anyone still around? I have a patch for the --no-dnslookups option but can't create a ticket on the Werkzeug Trac |
| [07:11:23] |
<rcscomp> |
details and link to patch: http://groups.google.com/group/pocoo-libs/browse_thread/thread/54846669425914f1 |
| [07:11:40] |
* |
rcscomp has left #pocoo |
| [07:15:52] |
* |
rcscomp has joined #pocoo |
| [07:16:03] |
<hads> |
rcscomp: Good stuff |
| [07:19:11] |
<rcscomp> |
hads: thanks. Not sure if the name works, seems kind of long. Maybe --no-rdns would be better. But it works for now. |
| [07:35:33] |
* |
alowry has quit IRC |
| [07:45:30] |
* |
marchon has joined #pocoo |
| [07:45:52] |
<marchon> |
mitsuhiko - are you here? |
| [07:49:07] |
* |
marchon has quit IRC |
| [08:02:58] |
* |
rcscomp has left #pocoo |
| [08:10:58] |
* |
sashav has joined #pocoo |
| [08:13:07] |
<birkenfeld> |
moin! |
| [08:24:16] |
* |
davidcramer has quit IRC |
| [09:19:10] |
* |
davidcramer has joined #pocoo |
| [09:31:42] |
<asmodai> |
birkenfeld: dude! |
| [09:32:48] |
<asmodai> |
http://www.openplans.org/projects/topp-engineering/blog/2008/09/24/pyinstall-a-new-hope/ |
| [09:36:09] |
<birkenfeld> |
let's hope the next chapter isn't "The Eby Strikes Back" |
| [09:37:51] |
<asmodai> |
but did you see Ian's mail |
| [09:39:14] |
<birkenfeld> |
where to? |
| [09:40:46] |
<birkenfeld> |
mitsuhiko: ping |
| [09:40:53] |
<asmodai> |
the TIP list |
| [09:41:18] |
<asmodai> |
part of said email: |
| [09:41:20] |
<asmodai> |
Single-file executable for you to try: |
| [09:41:20] |
<asmodai> |
https://svn.openplans.org/svn/pyinstall/trunk/pyinstall.py |
| [09:41:20] |
<asmodai> |
|
| [09:41:20] |
<asmodai> |
I think it's good. |
| [09:49:09] |
<mitsuhiko> |
birkenfeld: pong |
| [09:49:43] |
<birkenfeld> |
mitsuhiko: keyword search doesn't seem to work |
| [09:50:54] |
<mitsuhiko> |
birkenfeld: hmm. i tried with "sphinx.ext.refcount" yesterday and some other modules, and i got output |
| [09:50:59] |
<mitsuhiko> |
which keyword doesn't work? |
| [09:51:15] |
<birkenfeld> |
e.g. "os.path" in the python docs |
| [09:51:50] |
<mitsuhiko> |
birkenfeld: can you do "Search._index.keywords['os.path']" in firebug? |
| [09:51:55] |
<mitsuhiko> |
does that return an item? |
| [09:52:16] |
<birkenfeld> |
yes |
| [09:53:45] |
<mitsuhiko> |
hmm |
| [09:53:51] |
<birkenfeld> |
also there's something wrong with incremental index building, but I can't exactly say if it's new with that commit |
| [09:54:01] |
<mitsuhiko> |
that could be |
| [09:54:06] |
<mitsuhiko> |
i changed the indexer format |
| [09:54:08] |
<mitsuhiko> |
looking into that |
| [09:54:20] |
* |
luks has quit IRC |
| [09:54:47] |
* |
luks has joined #pocoo |
| [09:56:22] |
<mitsuhiko> |
birkenfeld: hwere is the search index loaded? |
| [09:56:45] |
<birkenfeld> |
load_indexer |
| [09:57:12] |
<mitsuhiko> |
hmm |
| [09:57:27] |
<mitsuhiko> |
ah. got it |
| [09:59:46] |
* |
aconbere has quit IRC |
| [10:01:44] |
<mitsuhiko> |
birkenfeld: if i raise the environment version, will it still try to load the index? |
| [10:02:22] |
<birkenfeld> |
it seems so, yes |
| [10:02:42] |
<birkenfeld> |
but that can be changed |
| [10:05:03] |
<asmodai> |
birkenfeld: finished BH btw |
| [10:05:08] |
<birkenfeld> |
asmodai: nice |
| [10:11:11] |
<mitsuhiko> |
birkenfeld: sphinx eats 100% of the cpu here and doesn't progress |
| [10:11:15] |
<mitsuhiko> |
(when building the python docs) |
| [10:11:20] |
<mitsuhiko> |
reference/datamodel |
| [10:12:02] |
<birkenfeld> |
mitsuhiko: didn't happen here just now |
| [10:12:12] |
<mitsuhiko> |
hmm |
| [10:12:21] |
* |
mitsuhiko reruns and straces |
| [10:13:22] |
<mitsuhiko> |
birkenfeld: heisenbug |
| [10:13:29] |
<mitsuhiko> |
but it's weird |
| [10:13:58] |
* |
davidcramer has quit IRC |
| [10:18:30] |
<birkenfeld> |
mitsuhiko: how do I do "ab" in "abc" in JS? |
| [10:19:01] |
<mitsuhiko> |
"abc".indexOf("ab") > -1 afair |
| [10:22:48] |
* |
jpcw has joined #pocoo |
| [10:23:25] |
<mitsuhiko> |
birkenfeld: does the search in the python docs work at all for you? |
| [10:23:29] |
<mitsuhiko> |
it looks up my browser |
| [10:23:37] |
<asmodai> |
hehehe |
| [10:23:42] |
* |
asmodai points and laughs |
| [10:24:00] |
<birkenfeld> |
mitsuhiko: yes |
| [10:25:13] |
<mitsuhiko> |
what the hack? looks like it's built against a terrible old version of sphinx |
| [10:25:28] |
<birkenfeld> |
docs.python.org? |
| [10:25:47] |
<mitsuhiko> |
no, locally |
| [10:25:52] |
<mitsuhiko> |
i have a searchindex.json |
| [10:26:01] |
<mitsuhiko> |
but i just built it :) |
| [10:26:06] |
<birkenfeld> |
oh |
| [10:26:07] |
<birkenfeld> |
:) |
| [10:36:56] |
<mitsuhiko> |
birkenfeld: got a searchindex.json again |
| [10:37:02] |
<mitsuhiko> |
do the python docs have a special builder? |
| [10:37:15] |
<birkenfeld> |
nope |
| [10:37:22] |
<mitsuhiko> |
argh. tools/sphinx |
| [10:37:24] |
<birkenfeld> |
is your tools/sphinx |
| [10:37:28] |
<birkenfeld> |
yeah :) |
| [10:37:32] |
<birkenfeld> |
it's a symlink here |
| [10:37:48] |
<birkenfeld> |
mitsuhiko: how much did you change in searchtools.js? |
| [10:38:10] |
<mitsuhiko> |
birkenfeld: that explains soooo much now |
| [10:38:24] |
<mitsuhiko> |
oh boy. that explains a lot |
| [10:38:54] |
<mitsuhiko> |
birkenfeld: locally or on the server? |
| [10:38:56] |
<mitsuhiko> |
in general not that much |
| [10:38:58] |
<birkenfeld> |
locally |
| [10:39:02] |
<mitsuhiko> |
a bit. why? |
| [10:39:07] |
<birkenfeld> |
ok, because I'm editing it right now |
| [10:39:10] |
<mitsuhiko> |
ic an check it in |
| [10:39:10] |
<mitsuhiko> |
sec |
| [10:39:34] |
<mitsuhiko> |
checked in |
| [10:40:36] |
<CIA-49> |
python-doctools: armin.ronacher * r66573 /doctools/trunk/sphinx/ (static/searchtools.js search.py builder.py): Some tiny search fixes. |
| [10:40:44] |
<birkenfeld> |
ok |
| [10:40:56] |
<birkenfeld> |
mitsuhiko: I'm encoding the "module" etc. now |
| [10:41:12] |
<mitsuhiko> |
encoding -> giving it a number? |
| [10:41:16] |
<birkenfeld> |
yes |
| [10:41:21] |
<mitsuhiko> |
actually, that information is unused at the moment i think |
| [10:41:26] |
<birkenfeld> |
no more |
| [10:41:35] |
<mitsuhiko> |
ah |
| [10:46:52] |
* |
davidcramer has joined #pocoo |
| [10:47:22] |
* |
davidcramer has joined #pocoo |
| [10:47:30] |
<CIA-49> |
python-doctools: georg.brandl * r66574 /doctools/trunk/sphinx/ (static/searchtools.js search.py): |
| [10:47:31] |
<CIA-49> |
python-doctools: * Search for partial keyword matches and be case insensitive. |
| [10:47:31] |
<CIA-49> |
python-doctools: * Show keyword results before regular ones. |
| [10:47:31] |
<CIA-49> |
python-doctools: * Show full name, type of keyword and title of containing doc. |
| [10:47:39] |
<davidcramer> |
how do you all typically handle large siege attempts? |
| [10:47:52] |
<davidcramer> |
i have awesome response times, im just not that familiar with using siege to pusht he limits |
| [10:48:27] |
<davidcramer> |
e.g. can only pull 25-30 req/s but the response time is 0.01 |
| [10:49:07] |
<davidcramer> |
(i am sieging from the same machine as well) |
| [10:54:23] |
<CIA-49> |
python-doctools: georg.brandl * r66575 /doctools/trunk/sphinx/ (static/searchtools.js search.py): Compress the index more, by not storing the label name for keywords. |
| [11:07:23] |
<CIA-49> |
python-doctools: georg.brandl * r66576 /doctools/trunk/sphinx/search.py: Add stopword list. |
| [11:24:13] |
* |
davidcramer has quit IRC |
| [11:32:45] |
<CIA-49> |
python-doctools: georg.brandl * r66578 /doctools/trunk/sphinx/ (static/searchtools.js search.py): More compression: group objects by prefix. |
| [11:33:25] |
* |
delroth has quit IRC |
| [11:33:46] |
* |
delroth has joined #pocoo |
| [11:34:23] |
<CIA-49> |
python-doctools: georg.brandl * r66579 /doctools/trunk/CHANGES: Add changelog entry about search change. |
| [11:44:06] |
<birkenfeld> |
mitsuhiko: ping |
| [11:44:12] |
<mitsuhiko> |
birkenfeld: pong |
| [11:44:30] |
<birkenfeld> |
mitsuhiko: isn't it somehow possible in JS to not quote strings? |
| [11:44:35] |
<birkenfeld> |
when used as dict keys |
| [11:48:55] |
<mitsuhiko> |
{foo: "bar"} |
| [11:49:00] |
<mitsuhiko> |
unless foo is a keyword or contains spaces |
| [11:49:06] |
<mitsuhiko> |
however. that's not valid json, just valid javascript |
| [11:49:17] |
<birkenfeld> |
it only seems to work for assignments? |
| [11:51:33] |
<mitsuhiko> |
birkenfeld: assignments? |
| [11:51:41] |
<mitsuhiko> |
what do you mean? |
| [11:52:00] |
<birkenfeld> |
mitsuhiko: at least in firebug, when I eval {a:1} it doesn't work, but a={a:1} works |
| [11:52:11] |
<mitsuhiko> |
({a: 1}) works |
| [11:52:15] |
<mitsuhiko> |
{} alone is not a valid statement |
| [11:52:29] |
<birkenfeld> |
ah |
| [11:52:31] |
<birkenfeld> |
interesting :) |
| [11:52:40] |
<birkenfeld> |
and somehow understandable |
| [11:52:48] |
<mitsuhiko> |
one of the better design decisions of javascript |
| [11:52:51] |
<birkenfeld> |
in a language with braces... |
| [11:59:23] |
<mitsuhiko> |
birkenfeld: i'm thinking about adding jinja2 support to sphinx. should it prefer jinja2 over jinja or provide a separate template bridge to set in the config? |
| [11:59:36] |
* |
davidcramer has joined #pocoo |
| [12:00:12] |
<birkenfeld> |
mitsuhiko: depends on the template compatibility |
| [12:00:15] |
<birkenfeld> |
mitsuhiko: urgs |
| [12:00:29] |
<mitsuhiko> |
urgs? |
| [12:00:30] |
<birkenfeld> |
({"true": 1, true: 2}) is {true: 2} |
| [12:00:40] |
<mitsuhiko> |
you can't use variables in object literals |
| [12:00:47] |
<birkenfeld> |
isn't true a keyword? |
| [12:01:04] |
<mitsuhiko> |
birkenfeld: welcome to the wonderful world of "microsoft does it wrong, let's make our code compatible" |
| [12:01:19] |
<mitsuhiko> |
birkenfeld: class is a keyword in javascript too, but you can use it as identifier |
| [12:01:20] |
<birkenfeld> |
grr |
| [12:01:48] |
<mitsuhiko> |
is especially great if you load your code into rhino or some standard compliant interpreter that gives a syntax error |
| [12:02:03] |
<birkenfeld> |
ok, I'm not surprised anymore that "1" and 1 are also treated the same |
| [12:02:12] |
<mitsuhiko> |
birkenfeld: http://www.javascripter.net/faq/reserved.htm |
| [12:02:28] |
<mitsuhiko> |
birkenfeld: sure. that's however intentional because there are no non string keys allowed |
| [12:02:48] |
<mitsuhiko> |
that's actually how arrays are implemented |
| [12:02:50] |
<mitsuhiko> |
very, very crazy |
| [12:03:14] |
<birkenfeld> |
reminds me of PHP |
| [12:03:18] |
<mitsuhiko> |
no, it's worse |
| [12:03:23] |
<birkenfeld> |
haha |
| [12:03:41] |
<mitsuhiko> |
birkenfeld: the worst are older jscript versions that implemented objects as linked lists |
| [12:03:58] |
<mitsuhiko> |
nice for arrays, but god. that sucks for everything else :) |
| [12:04:53] |
<birkenfeld> |
ahh, now I know why incremental building doesn't work for docs.python.org |
| [12:04:58] |
<birkenfeld> |
the probably don't have simplejson |
| [12:08:41] |
* |
maze_ is now known as maze |
| [12:15:42] |
<davidcramer> |
http://code.google.com/p/django-command-extensions/wiki/RunServerPlus |
| [12:15:49] |
<davidcramer> |
ok the werkzeug utils are shiny :) |
| [12:15:50] |
<davidcramer> |
props on those |
| [12:19:17] |
* |
grumpy_ has joined #pocoo |
| [12:46:16] |
* |
rodfersou[away] is now known as rodfersou |
| [12:51:25] |
<asmodai> |
nice to see jquery.com dropping the drawing from the homepage |
| [13:06:21] |
* |
grumpy_ has quit IRC |
| [13:11:57] |
<davidcramer> |
jquery site looks nice |
| [13:12:16] |
<davidcramer> |
i still prefer mootools :) |
| [13:12:58] |
<davidcramer> |
mmm even has nice plugin hosting now |
| [13:13:34] |
* |
grumpy_ has joined #pocoo |
| [13:16:05] |
<birkenfeld> |
mitsuhiko: okay. simplejson is soon history |
| [13:16:56] |
<asmodai> |
no! |
| [13:17:04] |
<asmodai> |
It will get speed ups in 2.6.1! |
| [13:17:57] |
<CIA-49> |
python-doctools: georg.brandl * r66581 /doctools/trunk/sphinx/static/jquery.js: Update jQuery. |
| [13:21:16] |
<birkenfeld> |
hah, another 50k saved |
| [13:22:20] |
<mitsuhiko> |
heh |
| [13:23:05] |
<birkenfeld> |
mitsuhiko: okay, one more thing |
| [13:23:10] |
<mitsuhiko> |
yep? |
| [13:23:55] |
<birkenfeld> |
if the list of filenumbers contains only one, it needn't be a list |
| [13:24:33] |
<birkenfeld> |
but after that, I can't imagine any more ways to make it smaller |
| [13:24:47] |
<mitsuhiko> |
birkenfeld: implement gzip for javascript :_D |
| [13:24:58] |
<birkenfeld> |
hehe |
| [13:31:25] |
<CIA-49> |
python-doctools: georg.brandl * r66582 /doctools/trunk/sphinx/ (util/_json.py builder.py util/json.py): |
| [13:31:26] |
<CIA-49> |
python-doctools: * Always use our own JS dumper instead of simplejson. |
| [13:31:26] |
<CIA-49> |
python-doctools: * Compress JS further by omitting quotes for dict keys where possible. |
| [13:31:43] |
<birkenfeld> |
mitsuhiko: how do I test if a value is int or list? |
| [13:41:45] |
* |
grumpy_ has quit IRC |
| [13:42:00] |
<mitsuhiko> |
birkenfeld: if (typeof obj === 'number') |
| [13:42:25] |
<birkenfeld> |
mitsuhiko: is if (obj.length == undefined) bad? ;) |
| [13:45:07] |
* |
davidcramer_ has joined #pocoo |
| [13:45:30] |
* |
davidcramer has quit IRC |
| [13:45:34] |
* |
davidcramer_ has left #pocoo |
| [13:46:42] |
<birkenfeld> |
'nother 20k |
| [13:48:05] |
<asmodai> |
hahaha |
| [13:48:09] |
<asmodai> |
silly |
| [13:48:36] |
<mitsuhiko> |
birkenfeld: implemented for the json dumper only or all searchindex formats? |
| [13:49:12] |
<birkenfeld> |
mitsuhiko: in search.py, so it will be in all formats |
| [13:49:20] |
<birkenfeld> |
makes no sense otherwise |
| [13:50:22] |
<CIA-49> |
python-doctools: georg.brandl * r66583 /doctools/trunk/ (sphinx/templates/layout.html CHANGES doc/templating.rst): Rename rellinks to linktags. |
| [13:51:10] |
<mitsuhiko> |
birkenfeld: can't say that i like the idea :) but currently the search only works for js anyways |
| [13:52:07] |
<CIA-49> |
python-doctools: georg.brandl * r66584 /doctools/trunk/sphinx/ (static/searchtools.js search.py util/json.py): A fix in the JS dumper and further compression by omitting redundant braces. |
| [13:53:34] |
* |
rafael has joined #pocoo |
| [14:03:01] |
* |
davidcramer has joined #pocoo |
| [14:03:42] |
<CIA-49> |
python-doctools: georg.brandl * r66585 /doctools/trunk/sphinx/ (util/jsdump.py search.py util/json.py builder.py): |
| [14:03:43] |
<CIA-49> |
python-doctools: Rename util.json to util.jsdump because it doesn't generate valid JSON anymore. |
| [14:03:43] |
<CIA-49> |
python-doctools: The JSON html builder still needs simplejson to work. |
| [14:03:45] |
<birkenfeld> |
oh yes, another optimization: convert [120,129,193,283] to "120129193283" :D |
| [14:04:45] |
<mitsuhiko> |
birkenfeld: short integers :D |
| [14:05:07] |
<birkenfeld> |
mitsuhiko: as a string of course :) |
| [14:05:12] |
<mitsuhiko> |
right. not escaping :) |
| [14:10:32] |
<rafael> |
hi |
| [14:10:50] |
<rafael> |
mitsuhiko: how to make default values in forms? |
| [14:16:01] |
<mitsuhiko> |
rafael: a) hg pull |
| [14:16:08] |
<mitsuhiko> |
i pushed the zeml stuff |
| [14:16:44] |
<CIA-49> |
Zine: mitsuhiko default * 499:44f12da69ccb /zine/ (30 files in 11 dirs): |
| [14:16:45] |
<CIA-49> |
Zine: Monster commit: fragment goes, zeml comes. database changes and totally |
| [14:16:45] |
<CIA-49> |
Zine: incompatible with current zine installations. If one runs a blog currently, |
| [14:16:45] |
<CIA-49> |
Zine: change the schema with "alter table" yourself and refresh the parser_data with a |
| [14:16:45] |
<CIA-49> |
Zine: little script. |
| [14:16:47] |
<CIA-49> |
Zine: mitsuhiko default * 498:bc438ac05cea /zine/utils/forms.py: Improved field descriptor for forms. |
| [14:17:31] |
<mitsuhiko> |
b) what do you mean by default |
| [14:17:34] |
<mitsuhiko> |
what kind of default |
| [14:17:46] |
<rafael> |
nice |
| [14:17:56] |
<rafael> |
b) i mean i saved an option eg for pages options |
| [14:18:11] |
<rafael> |
but then the fields should have the config value by default |
| [14:18:25] |
<mitsuhiko> |
that's what the initial parameter in the form constructor is for |
| [14:18:32] |
<mitsuhiko> |
f = Form({'your_field': 'your_value'}) |
| [14:18:59] |
<rafael> |
ah, alright :) |
| [14:19:51] |
<mitsuhiko> |
rafael: make sure to merge, i changed a few things |
| [14:21:30] |
<rafael> |
mitsuhiko: and how to label correctly? currently i use <label> in the template |
| [14:21:45] |
<mitsuhiko> |
rafael: check out the new changepasswordform |
| [14:21:49] |
<mitsuhiko> |
it's not part of the form system |
| [14:22:10] |
<rafael> |
k |
| [14:23:00] |
<CIA-49> |
python-doctools: georg.brandl * r66586 /doctools/trunk/ (CHANGES sphinx/util/jsdump.py): Group changelog entries and fix docstring for jsdump. |
| [14:24:14] |
<rafael> |
mitsuhiko: http://paste.pocoo.org/show/86093/ after installation |
| [14:26:34] |
* |
highwaychile has joined #pocoo |
| [14:32:16] |
<asmodai> |
mmm, I am starting to appreciate itsalltext more |
| [14:42:48] |
* |
\3TATUK has joined #pocoo |
| [14:43:09] |
* |
\3TATUK has left #pocoo |
| [15:12:47] |
* |
Kaelten has joined #pocoo |
| [15:16:57] |
<asmodai> |
Kaelten: dude! |
| [15:17:36] |
<Kaelten> |
asmodai: dude! |
| [15:17:42] |
<Kaelten> |
like, where's the car? |
| [15:19:55] |
<asmodai> |
Kaelten: :) |
| [15:19:59] |
<asmodai> |
Kaelten: Nod32 fixed |
| [15:20:05] |
<Kaelten> |
Nod32? |
| [15:20:11] |
<asmodai> |
Kaelten: curse-client.exe starts normally again and can update |
| [15:20:17] |
<Kaelten> |
ah, cool |
| [15:20:18] |
<asmodai> |
the anti-virus |
| [15:20:29] |
<Kaelten> |
ya I think it was some kinda issue with him trying to compress the binary |
| [15:20:35] |
<asmodai> |
possibly! |
| [15:21:13] |
* |
pete__c has quit IRC |
| [15:21:13] |
* |
delroth has quit IRC |
| [15:21:13] |
* |
[diecast] has quit IRC |
| [15:21:13] |
* |
dev has quit IRC |
| [15:22:45] |
* |
delroth has joined #pocoo |
| [15:22:46] |
* |
[diecast] has joined #pocoo |
| [15:22:46] |
* |
pete__c has joined #pocoo |
| [15:22:46] |
* |
dev has joined #pocoo |
| [15:26:08] |
* |
rodfersou is now known as rodfersou[away] |
| [15:28:44] |
* |
davidcramer_ has joined #pocoo |
| [15:30:03] |
* |
davidcramer has quit IRC |
| [15:32:06] |
* |
davidcramer_ has quit IRC |
| [15:33:04] |
* |
davidcramer has joined #pocoo |
| [16:04:45] |
* |
rafael__ has joined #pocoo |
| [16:06:30] |
* |
rafael has quit IRC |
| [16:14:23] |
* |
maix has joined #pocoo |
| [16:20:17] |
* |
sashav has quit IRC |
| [16:38:12] |
* |
tux21b has quit IRC |
| [16:39:06] |
* |
tux21b has joined #pocoo |
| [16:40:41] |
<asmodai> |
well ain't that bizarre |
| [16:40:46] |
<asmodai> |
US army training for US domestic missions |
| [16:40:57] |
<asmodai> |
I wonder what happened to the National Guard. |
| [16:41:00] |
* |
_paneb has quit IRC |
| [16:50:47] |
* |
Kaelten has quit IRC |
| [17:00:31] |
<birkenfeld> |
asmodai: you mean the National Guards? |
| [17:00:49] |
<birkenfeld> |
afaik each state has its own |
| [17:05:14] |
* |
Kaelten has joined #pocoo |
| [17:09:22] |
<CIA-49> |
python-doctools: georg.brandl * r66587 /doctools/trunk/sphinx/locale/pl/LC_MESSAGES/ (sphinx.mo sphinx.js sphinx.po): Update to Polish locale. |
| [17:24:52] |
* |
|[diecast]| has joined #pocoo |
| [17:25:57] |
* |
[diecast] has quit IRC |
| [17:26:19] |
* |
alowry has joined #pocoo |
| [17:27:35] |
* |
|[diecast]| has quit IRC |
| [17:32:45] |
<CIA-49> |
python-doctools: georg.brandl * r66588 /doctools/branches/0.4.x/sphinx/texinputs/sphinx.sty: Fix footer marking. |
| [17:32:46] |
<CIA-49> |
python-doctools: georg.brandl * r66589 /doctools/trunk/sphinx/texinputs/ (Makefile sphinx.sty): #17: fix Makefile syntax. |
| [17:32:46] |
<CIA-49> |
python-doctools: georg.brandl * r66590 /doctools/converter/converter/latexparser.py: Fix url parsing type. |
| [17:36:55] |
* |
mtrichardson has joined #pocoo |
| [17:38:08] |
<CIA-49> |
python-doctools: georg.brandl * r66591 /doctools/trunk/ (CHANGES doc/markup/inline.rst sphinx/environment.py): #16: allow referring to figures without explicit text. |
| [17:44:09] |
* |
alowry has quit IRC |
| [17:49:24] |
<CIA-49> |
python-doctools: georg.brandl * r66592 /doctools/trunk/ (CHANGES sphinx/quickstart.py): Don't overwrite existing sphinx projects from quickstart. |
| [17:56:00] |
<CIA-49> |
python-doctools: georg.brandl * r66593 /doctools/trunk/doc/intro.rst: Converter is now fairly usable for non-Python core projects. |
| [18:00:22] |
* |
[diecast] has joined #pocoo |
| [18:07:29] |
* |
stifal has joined #pocoo |
| [18:12:09] |
* |
alowry has joined #pocoo |
| [18:14:34] |
<CIA-49> |
python-doctools: georg.brandl * r66594 /doctools/trunk/ (4 files in 3 dirs): #18: put footnotes at the correct location in the LaTeX writer. |
| [18:15:16] |
<CIA-49> |
Pygments: thatch trunk * 681:38e669d11284 / (CHANGES pygments/lexers/agile.py): Support simple prompt from irb (Closes #363) |
| [18:15:18] |
<CIA-49> |
Pygments: thatch trunk * 682:6f317b427b60 /pygments/lexers/agile.py: Further fixes for irb lexer |
| [18:15:18] |
<CIA-49> |
Pygments: thatch trunk * 683:c3d87fc41a23 /pygments/lexers/agile.py: Fix RubyLexer heredoc issue as described in #336 |
| [18:15:19] |
<CIA-49> |
Pygments: thatch trunk * 684:72573bca3fd0 /tests/examplefiles/irb_heredoc: Regression test for Ruby heredoc fix in #366 |
| [18:15:21] |
<CIA-49> |
Pygments: gbrandl trunk * 685:b376b453559d / (3 files in 3 dirs): Merge with Tim. |
| [18:18:42] |
* |
[diecast] has quit IRC |
| [18:18:54] |
* |
[diecast] has joined #pocoo |
| [18:20:14] |
<CIA-49> |
python-doctools: georg.brandl * r66595 /doctools/trunk/doc/ (concepts.rst conf.py): Use palatino for the Sphinx docs. |
| [18:39:12] |
* |
rodfersou[away] is now known as rodfersou |
| [18:55:28] |
* |
rcscomp2 has joined #pocoo |
| [18:56:54] |
<rcscomp2> |
what happens with a WSGI application when a string rather than a list is returned? I.e. r = Response() |
| [18:56:55] |
<rcscomp2> |
r.response = 'hw' |
| [18:56:55] |
<rcscomp2> |
return r(env, start_response) |
| [18:57:48] |
<mitsuhiko> |
rcscomp2: it becomes terrible slow |
| [18:58:02] |
<mitsuhiko> |
because the server will write one char, flush the stream, write one char, flush the stream etc. |
| [19:03:39] |
<CIA-49> |
python-doctools: georg.brandl * r66602 /doctools/branches/0.4.x/sphinx/util/texescape.py: Escape backtick in texescape. |
| [19:04:10] |
<rcscomp2> |
mitsuhiko: interesting. I have been doing this by accident. Would it make sense to make BaseResponse.response a property and if its set as a string, turn it into a one-element list? |
| [19:04:34] |
<mitsuhiko> |
rcscomp2: Response("foo") is perfectly fine |
| [19:04:50] |
<mitsuhiko> |
werkzeug makes sure strings are yielded as one |
| [19:05:50] |
<rcscomp2> |
mitsuhiko: correct. But currently I am needing to create the response object before I know what the actual response output is going to be. So the response object gets set up, I do some processing, and then when I am done with the processing I use response.response to set the actual content. |
| [19:06:28] |
<rcscomp2> |
I can obviously make sure I use a list, but I thought it might be helpful and avoid assumption problems in the future for newbs like myself. |
| [19:08:40] |
<mitsuhiko> |
rcscomp2: set response.data |
| [19:08:50] |
<mitsuhiko> |
if it's a string |
| [19:09:31] |
<rcscomp2> |
mitsuhiko: oh, I see. I was using the wrong property. Thanks. |
| [19:10:40] |
<CIA-49> |
python-doctools: georg.brandl * r66604 /doctools/ (trunk/sphinx/util/texescape.py trunk): (log message trimmed) |
| [19:10:40] |
<CIA-49> |
python-doctools: Merged revisions 66588,66602 via svnmerge from |
| [19:10:40] |
<CIA-49> |
python-doctools: svn+ssh://pythondev@svn.python.org/doctools/branches/0.4.x |
| [19:10:40] |
<CIA-49> |
python-doctools: ........ |
| [19:10:40] |
<CIA-49> |
python-doctools: r66588 | georg.brandl | 2008-09-24 17:25:43 +0200 (Wed, 24 Sep 2008) | 2 lines |
| [19:10:43] |
<CIA-49> |
python-doctools: Fix footer marking. |
| [19:10:45] |
<CIA-49> |
python-doctools: ........ |
| [19:10:47] |
<CIA-49> |
python-doctools: georg.brandl * r66606 /doctools/trunk/ (CHANGES sphinx/environment.py sphinx/roles.py): Add cmember role (#3875). |
| [19:11:44] |
<asmodai> |
birkenfeld: yeah, but the army is not for domestic deployment |
| [19:11:55] |
<birkenfeld> |
yep |
| [19:13:06] |
* |
leche has joined #pocoo |
| [19:24:12] |
* |
leche has quit IRC |
| [19:29:14] |
* |
leche has joined #pocoo |
| [19:35:54] |
<rafael__> |
mitsuhiko: what do you say to the pasted error? |
| [19:36:09]< |