Apache Allura / Chat

Apache Allura / Chat is hosted on FreeNode IRC channel #allura

Log for 2013-04-24

  • 11:46:09
    confused

    hi

  • 11:46:51
    confused

    need hel p in fixing an error "userstats not available"

  • 11:49:39
    confused

    hi tim

  • 12:50:27
    confused

    error "userstats not available" and Attribute Error saying 'NoneType' object has no attribute '_id' while registering an user in the forge after installing allura.Can I get help in fixing this?

  • 13:51:21
    allura-ci

    Project allura build #2,060: FAILURE in 4 min 25 sec: http://engr.geek.net/hudson/job/allura/2060/

  • 13:53:25
    brondsem

    cory_fu: need to move the other WebError requirement from requirements-common.txt to requirements.txt so it's not duplicated on the SF build

  • 13:59:18
    cory_fu

    Ah, ok

  • 14:28:32
    allura-ci

    Yippie, build fixed!

  • 14:28:32
    allura-ci

    Project allura build #2,061: FIXED in 24 min: http://engr.geek.net/hudson/job/allura/2061/

  • 17:16:49
    brondsem

    i've made some followup changes to a ticket I was reviewing. Anyone want to review my 3 commits on https://git-wip-us.apache.org/repos/asf?p=incubator-allura.git;a=shortlog;h=refs/heads/db/2835 ?

  • 17:18:23
    tvansteenburgh

    looking

  • 17:21:08
    tvansteenburgh

    what's the diff b/t jinja Markup and escape

  • 17:21:31
    brondsem

    escape escapes it

  • 17:21:40
    brondsem

    Markup flags the string as safe

  • 17:21:46
    brondsem

    completely opposite :)

  • 17:22:38
    brondsem

    well not completely. escape will return a Markup object after its done escaping

  • 17:25:25
    tvansteenburgh

    +1

  • 17:26:11
    brondsem

    thx

  • 18:31:49
    brondsem

    Redie: does the rat report take a long time to run?

  • 18:32:01
    brondsem

    i'm trying it out after a big merge, to see if any files are missing license headers

  • 18:32:18
    brondsem

    seems to be taking a while

  • 18:35:54
    allura-ci

    Project allura build #2,062: UNSTABLE in 21 min: http://engr.geek.net/hudson/job/allura/2062/

  • 18:36:51
    brondsem

    i caused that, will fix

  • 18:54:03
    Redie

    brondsem: 1st run takes some time

  • 18:54:12
    Redie

    at least on my machine

  • 18:54:43
    Redie

    idk much about java but i think it caches some bytecode after this

  • 18:55:23
    Redie

    on another note :)

  • 18:55:49
    brondsem

    2nd run is slow too :P

  • 18:55:52
    brondsem

    oh well

  • 18:56:42
    Redie

    we have unportable test (probably more) in allura.tests.test_helpers.test_find_executable

  • 18:57:23
    Redie

    and i don't see any way to fix it in portable way, cause due to the very nature of it it has to make assumption about the absolute path

  • 18:58:50
    brondsem

    hm, could pick something more common than bash, like 'cp'

  • 18:59:05
    brondsem

    that only helps somewhat though

  • 18:59:39
    Redie

    yeah, actually the path is the problem, several distros moved all binaries to /usr/bin

  • 19:00:29
    allura-ci

    Yippie, build fixed!

  • 19:00:30
    allura-ci

    Project allura build #2,063: FIXED in 20 min: http://engr.geek.net/hudson/job/allura/2063/

  • 19:01:05
    brondsem

    ah

  • 19:01:38
    Redie

    okay, scrap that

  • 19:01:39
    brondsem

    if we assume at least some basic stuff, maybe could execute which bash and compare the result to the function?

  • 19:01:55
    Redie

    yeah, or just check if the output is executable

  • 19:02:14
    Redie

    sorry, haven't sleep much :P

  • 19:02:29
    brondsem

    you know what? i don't think find_executable is used anywhere :)

  • 19:03:34
    Redie

    hm, looks like it

  • 19:04:01
    brondsem

    looks like it's an old old helper from a totally different way of handling scm

  • 19:04:08
    Redie

    so i'll take the liberty of removing it

  • 19:04:40
    brondsem

    +1

  • 20:15:46
    brondsem

    https://sourceforge.net/p/allura/pastebin/51783d599095473a52f13635/ look reasonable for some licensing header updates?

  • 20:21:10
    tvansteenburgh

    +1

  • 20:22:27
    brondsem

    we should add "rat" (license header checker tool) to our test suite

  • 20:22:37
    tvansteenburgh

    +1

  • 20:29:53
    tvansteenburgh

    brondsem: i just asked you an asynchronous question

  • 20:30:14
    brondsem

    kk :)

  • 20:31:02
    brondsem

    i'll reply …. sometime ...

  • 20:31:46
    Redie

    brondsem: i thought about it, but i had doubts about adding huge dependency on java

  • 20:32:37
    brondsem

    Redie: i'm thinking it wouldn't be part of the main test suite for everyone but Jenkins could be set up to run it

  • 20:33:06
    brondsem

    fyi, I started making our build on the public Apache server

  • 20:33:15
    Redie

    ah, cool

  • 20:33:17
    Redie

    x2

  • 20:33:47
    brondsem

    2 major problems though: build results disappear pretty quickly, and the ubuntu hosts are way overloaded, takes on avg 12 hours to get a build through the queue

  • 20:33:52
    brondsem

    so i'll have to ask around about that

  • 20:44:37
    brondsem

    tvansteenburgh: i won't bother with another email but https://sourceforge.net/p/forge/site-support/2702/ is the use-case for a repo starting with a number

  • 20:46:30
    tvansteenburgh

    okay...

  • 20:47:17
    tvansteenburgh

    so allow . and _ and allow starting number, but only for repos?

  • 20:47:57
    brondsem

    yeah

  • 20:48:44
    tvansteenburgh

    mmmkay

  • 20:49:00
    tvansteenburgh

    i still think you should send the email ;)

  • 20:49:23
    tvansteenburgh

    otherwise the ML will think you're rude

  • 20:49:40
    tvansteenburgh

    oh, you did send one

  • 20:49:54
    brondsem

    :D

  • 21:02:39
    Redie

    i'm out for today, bye

  • 21:20:18
    allura-ci

    Project allura build #2,066: FAILURE in 38 sec: http://engr.geek.net/hudson/job/allura/2066/

  • 21:22:52
    brondsem

    cory_fu: do we need the new gitpython built?

  • 21:23:50
    brondsem

    fwiw I haven't merged 6078 changes since the 20130415 build of gitpython hadn't been done until today

  • 21:27:36
    brondsem

    found ops ticket 53588 going to reopen since it's not on our internal mirror yet

  • 21:30:34
    cory_fu

    Hrm. The new build I just ticketed is done, so I merged #6000

  • 21:30:38
    cory_fu
  • 21:31:06
    cory_fu

    Where is the 20130415 build from?

  • 21:31:15
    cory_fu

    Rather, what repo / branch / tag is it on?

  • 21:31:25
    cory_fu

    We need to get a standard location for our GitPython hacks

  • 21:32:45
    cory_fu

    Oh, ok

  • 21:32:49
    cory_fu

    So it's contained in the new one

  • 21:33:07
    brondsem

    yeah, but we should think about tracking these better

  • 21:33:41
    brondsem

    noting them in the requirements file is ok for now I guess

  • 21:40:13
    cory_fu

    What do you mean, tracking them better?

  • 21:41:24
    cory_fu

    You have a suggestion?

  • 21:44:15
    brondsem

    a gitpython fork within the /p/allura project?

  • 21:44:40
    cory_fu

    I was thinking something like that

  • 21:45:20
    cory_fu

    Would really encourage us to work on a "fork from github / send github merge-request" feature (if the latter is possible)

  • 21:45:54
    brondsem

    :)

  • 21:54:41
    allura-ci

    Project allura build #2,067: STILL FAILING in 39 sec: http://engr.geek.net/hudson/job/allura/2067/

Get latest updates about Open Source Projects, Conferences and News.

Sign up for the SourceForge newsletter:





No, thanks