Lucene - Core
  1. Lucene - Core
  2. LUCENE-4612

ant nightly-smoke leaves a dirty checkout

    Details

    • Type: Bug Bug
    • Status: Closed
    • Priority: Major Major
    • Resolution: Fixed
    • Affects Version/s: None
    • Fix Version/s: 4.2, 5.0
    • Component/s: None
    • Labels:
      None
    • Lucene Fields:
      New

      Description

      ? dev-tools/scripts/_pycache_

      Can we not leave this around?

        Activity

        Hide
        Uwe Schindler added a comment -

        Closed after release.

        Show
        Uwe Schindler added a comment - Closed after release.
        Hide
        Commit Tag Bot added a comment -

        [branch_4x commit] Michael McCandless
        http://svn.apache.org/viewvc?view=revision&revision=1437009

        LUCENE-4612: tell python not to write bytecode to the filesystem

        Show
        Commit Tag Bot added a comment - [branch_4x commit] Michael McCandless http://svn.apache.org/viewvc?view=revision&revision=1437009 LUCENE-4612 : tell python not to write bytecode to the filesystem
        Hide
        Commit Tag Bot added a comment -

        [trunk commit] Michael McCandless
        http://svn.apache.org/viewvc?view=revision&revision=1437007

        LUCENE-4612: tell python not to write bytecode to the filesystem

        Show
        Commit Tag Bot added a comment - [trunk commit] Michael McCandless http://svn.apache.org/viewvc?view=revision&revision=1437007 LUCENE-4612 : tell python not to write bytecode to the filesystem
        Hide
        Michael McCandless added a comment -

        -B seems to work ... I'll pass this to all places where we invoke Python.

        Show
        Michael McCandless added a comment - -B seems to work ... I'll pass this to all places where we invoke Python.
        Hide
        Michael McCandless added a comment -

        I think we can just pass -B when we run Python? http://stackoverflow.com/questions/154443/how-to-avoid-pyc-files

        Show
        Michael McCandless added a comment - I think we can just pass -B when we run Python? http://stackoverflow.com/questions/154443/how-to-avoid-pyc-files
        Hide
        Robert Muir added a comment -

        python doesn't care what the CWD is when it decides to make this cache. it puts it where the script is.

        so we should either disable the bytecode cache or copy the scripts to build/

        Show
        Robert Muir added a comment - python doesn't care what the CWD is when it decides to make this cache. it puts it where the script is. so we should either disable the bytecode cache or copy the scripts to build/
        Hide
        Robert Muir added a comment -

        fakeReleaseTmp wont work, it doesnt yet exist at the time the script runs. if we make it, the script complains.
        trying another place in build/ ....

        Show
        Robert Muir added a comment - fakeReleaseTmp wont work, it doesnt yet exist at the time the script runs. if we make it, the script complains. trying another place in build/ ....
        Hide
        Robert Muir added a comment -

        i don't like adding svn:ignores for things like this (in this case, really, its just like having an output directory full of java classes).

        for now i'm testing just running the py script with a CWD in the build directory:

        Index: build.xml
        ===================================================================
        --- build.xml	(revision 1419557)
        +++ build.xml	(working copy)
        @@ -276,7 +276,7 @@
              <fileset dir="solr/build/docs/changes"/>
            </copy>
            <makeurl file="${fakeRelease}" validate="false" property="fakeRelease.uri"/>
        -   <exec executable="${python32.exe}" failonerror="true">
        +   <exec executable="${python32.exe}" failonerror="true" dir="${fakeReleaseTmp}">
              <arg value="-u"/>
              <arg file="dev-tools/scripts/smokeTestRelease.py"/>
              <arg value="${fakeRelease.uri}"/>
        

        Unfortunately it takes 45 minutes to know if it works

        if this works i would recommend we just add an ant task for manual smoking too just to keep everything clean.

        Show
        Robert Muir added a comment - i don't like adding svn:ignores for things like this (in this case, really, its just like having an output directory full of java classes). for now i'm testing just running the py script with a CWD in the build directory: Index: build.xml =================================================================== --- build.xml (revision 1419557) +++ build.xml (working copy) @@ -276,7 +276,7 @@ <fileset dir="solr/build/docs/changes"/> </copy> <makeurl file="${fakeRelease}" validate="false" property="fakeRelease.uri"/> - <exec executable="${python32.exe}" failonerror="true"> + <exec executable="${python32.exe}" failonerror="true" dir="${fakeReleaseTmp}"> <arg value="-u"/> <arg file="dev-tools/scripts/smokeTestRelease.py"/> <arg value="${fakeRelease.uri}"/> Unfortunately it takes 45 minutes to know if it works if this works i would recommend we just add an ant task for manual smoking too just to keep everything clean.
        Hide
        Hoss Man added a comment -

        my limited python understanding is that you can't tell python to use a different dir for this, so if you want it in build you'd have to use the build dir as the working dir when running the script? (or maybe copy the script there?)

        or we could just python not to cache the bytecode...

        http://docs.python.org/2/using/cmdline.html#cmdoption-B
        http://docs.python.org/2/using/cmdline.html#envvar-PYTHONDONTWRITEBYTECODE

        ...but it still seems like maybe we should svn:ignore that dir to handle the case where people run those scripts manually

        Show
        Hoss Man added a comment - my limited python understanding is that you can't tell python to use a different dir for this, so if you want it in build you'd have to use the build dir as the working dir when running the script? (or maybe copy the script there?) or we could just python not to cache the bytecode... http://docs.python.org/2/using/cmdline.html#cmdoption-B http://docs.python.org/2/using/cmdline.html#envvar-PYTHONDONTWRITEBYTECODE ...but it still seems like maybe we should svn:ignore that dir to handle the case where people run those scripts manually
        Hide
        Robert Muir added a comment -

        I'm not happy with its location being in the source tree and not in a build/ directory.

        ideally we would put this cache in the build/ directory. this way its actually cleaned by 'ant clean' and so on.

        Show
        Robert Muir added a comment - I'm not happy with its location being in the source tree and not in a build/ directory. ideally we would put this cache in the build/ directory. this way its actually cleaned by 'ant clean' and so on.
        Hide
        Hoss Man added a comment -

        i've been meaning to ask this since the 4.0 vote ... should we add that dir to svn:ignore?

        Show
        Hoss Man added a comment - i've been meaning to ask this since the 4.0 vote ... should we add that dir to svn:ignore?

          People

          • Assignee:
            Michael McCandless
            Reporter:
            Robert Muir
          • Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

            Dates

            • Created:
              Updated:
              Resolved:

              Development