Timestamps are in GMT/BST.
[0:40] * robbyoconnor (~wakawaka@guifications/user/r0bby) has joined #jython
[0:45] * robbyoconnor (~wakawaka@guifications/user/r0bby) Quit (Remote host closed the connection)
[0:46] * robbyoconnor (~wakawaka@guifications/user/r0bby) has joined #jython
[1:10] * lheuer1 (~Adium@blfd-4db0f4e2.pool.mediaWays.net) has joined #jython
[1:11] * lheuer (~Adium@unaffiliated/lheuer) Quit (Ping timeout: 246 seconds)
[1:14] * clajo04_ (~clajo04_@pool-74-108-95-112.nycmny.fios.verizon.net) has joined #jython
[2:01] * mister_m_ (~matthew@98.156.58.98) has joined #jython
[2:01] <mister_m_> does jython have the GIL problem?
[2:02] <agronholm> no
[2:02] <mister_m_> :)
[2:04] <mister_m_> I have an application I am working on in regular python that uses a library called pyglet - I am unsure if it uses C extensions - but as it is a sort of wrapper over opengl, I am thinking it does
[2:04] <mister_m_> How can I see if something will be compatibile with Jython
[2:05] <agronholm> try to install it on jython
[2:05] <agronholm> although that's not conclusive evidence either
[2:09] <mister_m_> fair point, I will give it a try
[2:10] * mister_m_ (~matthew@98.156.58.98) Quit (Quit: Lost terminal)
[2:14] * vayer (~vayer@173-164-129-225-SFBA.hfc.comcastbusiness.net) Quit (Quit: Leaving)
[2:35] * clajo04_ (~clajo04_@pool-74-108-95-112.nycmny.fios.verizon.net) Quit (Quit: clajo04_)
[2:46] * clajo04_ (~clajo04_@pool-74-108-95-112.nycmny.fios.verizon.net) has joined #jython
[5:54] * lheuer1 is now known as lheuer
[9:07] * wmeissner (~wmeissner@60-241-132-27.static.tpgi.com.au) has joined #jython
[9:09] * seletz (~seletz@business-178-015-118-087.static.arcor-ip.net) has joined #jython
[9:52] * aquarius1973 (~aquarius1@egress-frr.sf.nat.aol.com) has joined #jython
[10:05] * juneau001 (~juneau@50-45-109-120.dklb.il.frontiernet.net) Quit (Quit: juneau001)
[11:09] * juneau001 (~juneau@fess-116326.dhcp.fnal.gov) has joined #jython
[12:17] * wainersm (~wainersm@189.61.217.121) has joined #jython
[12:38] * wainersm (~wainersm@189.61.217.121) Quit (Quit: Ex-Chat)
[12:44] * wmeissner (~wmeissner@60-241-132-27.static.tpgi.com.au) Quit (Quit: wmeissner)
[13:29] * ctismer is now known as ctismer_afk
[13:40] * ctismer_afk is now known as ctismer
[13:52] * Arfrever (~Arfrever@apache/committer/Arfrever) has joined #jython
[14:27] * wainersm (~wainersm@189.61.217.121) has joined #jython
[14:40] * wainersm (~wainersm@189.61.217.121) Quit (Ping timeout: 248 seconds)
[14:47] * wainersm (~wainersm@189.61.217.121) has joined #jython
[16:13] * seletz (~seletz@business-178-015-118-087.static.arcor-ip.net) Quit (Quit: seletz)
[16:46] * jedir0x (~jedir0x@cpe-72-134-20-41.socal.res.rr.com) has joined #jython
[16:46] <jedir0x> I'm considering Python, Flask, Jinja2 for a web app that i will be deploying to amazon web services. We currently have some java apps running in Beanstalk (auto scaling stuff). How do you guys feel about Jython and deploying python apps in a war file to a tomcat servlet container? I'd like to take advantage of Beanstalk with this python app as well.
[16:47] <agronholm> I don't feel confident about deploying web apps on jython
[16:47] <agronholm> I do have one CPython app deployed on Heroku though
[16:48] <jedir0x> what makes you not confident about deploying web apps with Jython?
[16:49] <agronholm> well for one, that jython is so much slower than CPython
[16:49] <agronholm> then Jython has holes in its stdlib implementation
[16:49] <agronholm> sometimes in strange places
[16:49] <agronholm> like socket.gethostbyname_ex missing
[16:49] <agronholm> and then, what's the benefit really?
[16:49] <agronholm> if you're not using java libs
[16:50] <jedir0x> i see
[16:51] <agronholm> for web apps, not even the absence of GIL is a factor
[16:52] <agronholm> since the apps are mostly I/O bound anyway
[16:52] <jedir0x> i see
[17:03] <jedir0x> So this app wouldn't be talking to any databases directly
[17:03] <jedir0x> it'd be using REST services provided by our java platform
[17:04] <agronholm> it'd be a web client?
[17:04] <agronholm> http client I mean
[17:05] * lheuer (~Adium@blfd-4db0f4e2.pool.mediaWays.net) Quit (Quit: Leaving.)
[17:05] <jedir0x> not sure what you're asking... but the java platform provides REST apis that the python webapp will use.
[17:05] <jedir0x> so the python wont be communicating directly with any databases or other resources
[17:06] <agronholm> so how will the webapp access these REST apis if not by means of http?
[17:13] * Wolfzz (~Wolfzz@host-94-101-1-9.igua.fi) has joined #jython
[17:16] * wainersm (~wainersm@189.61.217.121) Quit (Ping timeout: 260 seconds)
[17:31] * vayer (~vayer@173-164-129-225-SFBA.hfc.comcastbusiness.net) has joined #jython
[17:40] <agronholm> juneau001: no 2.5.3rc1 release news on jython.org?
[17:53] <juneau001> agronholm: not yet, i have to get to that
[17:54] <agronholm> also if you're making pydev releases news, 2.6.0 was released quite a while ago
[18:19] <juneau001> thanks
[18:22] * juneau001 (~juneau@fess-116326.dhcp.fnal.gov) Quit (Quit: juneau001)
[18:32] * lheuer (~Adium@unaffiliated/lheuer) has joined #jython
[18:40] <tcrawley> is the .pyc format the same between Jython and CPython? or is a pyc JVM bytecode under Jython?
[18:40] <agronholm> jython does not currently use .pyc files
[18:41] <agronholm> instead it produces $py.class files that are JVM bytecode
[18:41] <tcrawley> agronholm: great, thanks. my .pyc files must have shown up when I ran under CPython
[18:43] * enebo (~enebo@71-215-194-220.mpls.qwest.net) has joined #jython
[20:02] * jedir0x (~jedir0x@cpe-72-134-20-41.socal.res.rr.com) Quit (Ping timeout: 276 seconds)
[20:05] * vayer (~vayer@173-164-129-225-SFBA.hfc.comcastbusiness.net) Quit (Quit: Leaving)
[21:00] * sera_ (~sera@gentoo/developer/sera) has joined #jython
[21:03] * sera (~sera@gentoo/developer/sera) Quit (Ping timeout: 260 seconds)
[21:17] * also (u1186@gateway/web/irccloud.com/x-gyzsebtackoxwrtw) has joined #jython
[21:18] <also> anyone have experience running jython and jruby in the same jvm?
[21:19] <also> or, why "java.lang.IncompatibleClassChangeError: Found class com.kenai.jffi.InvocationBuffer, but interface was expected"
[21:20] <agronholm> possibly incompatible versions of jffi
[21:21] <also> that's what i thought.
[21:21] <agronholm> what version of jython btw
[21:21] <agronholm> I recall the jffi was recently updated
[21:22] <agronholm> but I don't know if that was in 2.5.3 or 2.7
[21:22] <also> 2.7a2
[21:22] <agronholm> if may have been after a2
[21:22] <agronholm> check the version history
[21:22] <agronholm> from hg.python.org that is
[21:22] <also> ok, i'll try the latest
[21:23] <also> i also tried using jarjar to rename the class in the jython jar
[21:23] <agronholm> won't work
[21:23] <also> why's that?
[21:24] <agronholm> well because either jython or jruby expects the class to be an interface, and the other expects it to be a class
[21:24] <agronholm> either one will break
[21:25] <agronholm> for jruby and jython to live happily in the same jvm, they either need to be using the same jffi implementation, or you need to load both in separate class loaders with separate class paths
[21:26] <agronholm> how THAT can be accomplished, I don't know
[21:26] <agronholm> but in theory, it should work like that
[21:27] <also> I was thinking renaming all the com.kenai classes to org.python.kenai and rebuilding jython would also work
[21:27] <also> so jython and jruby are using completely separate classes
[21:27] <agronholm> then you'd also need to rebuild jffi
[21:28] <also> you think jarjar wouldn't work?
[21:28] <agronholm> frankly, using separate class loaders would probably be easier
[21:28] <agronholm> think about it -- jffi certainly refers to its own classes too
[21:30] <also> oh, i guess the native stuff makes it tricky
[21:30] * JStoker (jstoker@unaffiliated/jstoker) Quit (Excess Flood)
[21:30] <agronholm> even without the native stuff
[21:30] <agronholm> will jarjar also change the class name references inside the class files?
[21:31] <also> yup
[21:31] <agronholm> I see
[21:31] <agronholm> then it might work
[21:31] <also> but i think there'd be references from the natively compiled libs
[21:31] <agronholm> the separate class loaders are probably your best choice
[21:33] <also> seems like you're probably right
[21:33] <also> i was hoping it would be easier to call between them
[21:34] <also> it works great until the jffi stuff comes into play
[21:35] * JStoker (jstoker@unaffiliated/jstoker) has joined #jython
[21:36] <pjenvey> we actually jarjar almost every jar, but I believe we don't w/ jffi because its native lib stubs are basically hardcoded
[21:36] <pjenvey> and then jarjaring the .class files ends up breaking that
[21:36] <pjenvey> you could probably manually workaround that somehow but it would take more than just jarjar'ing (last time i looked at this, anyway)
[21:37] <agronholm> pjenvey: did the jffi get updated yet?
[21:37] <agronholm> in jython
[21:37] <pjenvey> i think Arfrever had a patch for it
[21:37] <pjenvey> don't recall seeing it committed
[21:37] <agronholm> I see
[21:38] <pjenvey> https://bitbucket.org/jython/jython/pull-request/2/updated-java-native-runtime-jars-to-latest
[21:38] <pjenvey> that is against default (2.7)
[21:39] <also> i'll give that a try
[21:39] <also> thanks!
[21:41] <pjenvey> let us know how it turns out, then one of us can commit it
[22:21] * also (u1186@gateway/web/irccloud.com/x-gyzsebtackoxwrtw) has left #jython
[22:35] * enebo (~enebo@71-215-194-220.mpls.qwest.net) Quit (Quit: enebo)
[23:06] * Wolfzz (~Wolfzz@host-94-101-1-9.igua.fi) Quit (Quit: L?hd?ss?)
[23:33] * juneau001 (~juneau@50-45-109-120.dklb.il.frontiernet.net) has joined #jython
[23:34] <fwierzbicki_> I haven't yet updated the jffi jars
[23:34] <fwierzbicki_> it's finally getting close to the top of my list though
These logs were automatically created by JythonLogBot_ on irc.freenode.net using a slightly modified version of the Java IRC LogBot (github).