DXR is a code search and navigation tool aimed at making sense of large projects. It supports full-text and regex searches as well as structural queries.

Untracked file

Line Code
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163
Release ???

    New Time columns for use in environments with multiple time zones.

    Spelling fixes 

    Add support for popup windows which work "on click"

    Add support for users who want to see tinderbox's notion of what was
    checked in not just Bonsai's query.

    Add support for historical Tinderbox images
        (flames, stars)

    Mozilla layout preferences are now standard 
      (non uniform row spacing, 
      truly empty cells, 
      legend on bottom of page).

    Allow users to backdate notices

    automate regeneration of HTML when user uses CGI interface

    Put the C in all build cells.

    VC code did the wrong thing when generating old pages and there was
    no TreeState data.

    Fix Spelling mistakes throughout code.	

    Add support for build machines to tell the server to add text/links 
    into build cells (TinderboxPrint).

    When builds are started too frequently (more than once every five minutes) 
    newest build is now kept and oldest build is thrown away.

    New Configure script contributed by Stephen Lamm (slamm@yahoo-inc.com)
    This includes a fix to allow Perl to be in different locations on the 
    machine where make is run and on the machine where the web server runs.

    Now it is possible to include the time column on the left and right 
    side of the status table.

    There is now a working Perforce module for those who use this product
    instead of CVS.

    There is a module to interface with the req ticketing system.

    New Summary format and number of errors displayed on build page by
    Adam Di Carlo <adam@onshore-devel.com>,

    Spelling and consistency fixes from:
        Adam Di Carlo <adam@onshore-devel.com>, 
        Orion Poplawski <OPoplawski@cqg.com>,
        John L. Villanlovos <john.l.villalovos@intel.com>,

    Change HTML generation so that the text browser 'links' 
    (http://artax.karlin.mff.cuni.cz/~mikulas/links/) can
    work with tinderbox.

    Allow the build script to use a single process and alternate builds 
    of different types.

    Add a sample /etc/rc.d/init.d file for the RedHat Build machines.

    Mail processing programs were not writing error files correctly.  
    This has been broken since I separated out the mail processing module.

    Add script to cull old builds from a directory of completed builds.

    Let the VC and BT headers be configurable.  Too many companies find
    this an important feature.

    Allow Tinderbox to run via suid wrappers.  This is needed for some
    times of mail delivery systems

    No longer exclude 'privileged id's from running Tinderbox. 
    Now we require tinderbox to run as the id which the user specifies 
    is the tinderbox id.

    Better errors for failures in Storable.pm

    BasicTxtHeader.pm had a bug which prevented it from working when using Storable

    Rework the BT modules to allow use when URL is not easily mapped from the bugid

    Begin work to allow tinderbox admin pages to open and close bonsai trees

    Fix code which allowed tinderbox to determine the state of bonsai trees

    Major bug in the untainting code, fixed. This was a security
    problem. I suspect people may have trouble with the fix but this was
    how the code was intended to work.  I do not understand how this got
    through my unit tests.

    Give a useful example of how use the new build script.

    Improve the spec files, split out build machine packages.


Release 0.9

    Add full Bonsai integration functionality.

    Allow users to define a display order for the buildnames on a per tree basis.
    This adds a new function into TreeData.pm check your local files.

    Fix computation of 'deadtime' for eazel.com's build setup,
    this will improve their estimate of when the next build will be
    complete.

    Develop portable client build scripts.

    Add RPM spec files.

    Relax check of root id for all redhat users.  
    uid nobody is 99 and by default redhat apache runs all cgi as nobody.

    VC_CVS.pm was not removing excess treestates from the Database correctly, 
      now fixed.

Release 0.8

    This release moved some files into new directories update with
      cvs update -d -P

    Remove the strict time checking on mail, it caused too many problems.

    Fix the Unmonitored Builds in admintree.cgi and in status.html page.

    Storable.pm has been tested as a Persistence implementation.

    Improve the prediction of when builds will end by introducing the
    notion of 'deadtime'.

    Separate out the BuildStatus table for easier configuration of the
    status without changing the build display library.

    Better comments in the ReadMe files

    Better security:

      - All scripts run in taint perl mode

      - Separation of HTML and DATA directories in 
        addition to the CGI directory

      - Additional checks of ownership for the data directories

      - Check the network address of all administrative commands

    Improved Makefile/Configure support

    New permissions help prevent accidental edit of files in the
    build or install directory

    Inter-positioning of libraries, for local configuration,
      fully supported

    Test arguments added to Makefile

    Test compiles of the perl are no longer required