Permalink
Commits on Apr 6, 2018
  1. rename a variable

    nico committed Apr 6, 2018
  2. Merge pull request #1415 from nico/readident

    nico committed Apr 6, 2018
    Improve location of error messages around identifiers.
  3. Improve location of error messages around identifiers.

    nico committed Apr 6, 2018
    Lexer::ReadIdent() now sets last_token_ before returning, like
    Lexer::ReadEvalString() does. So all "expected identifiers" and things
    that call ReadIdent (pool parser, rule parser, let parser, code parsing
    the rule name after a : in a build line) now point the "^ near here" at
    what was there instead of the previous last_token
    
    According to manifest_parser_perftest, this is perf-neutral.
  4. Merge pull request #1409 from atetubou/appveyor

    nico committed Apr 6, 2018
    Add appveyor.yml for build & test on windows
  5. Merge pull request #1410 from ryandesign/configure.py-CXXFLAGS

    nico committed Apr 6, 2018
    Honor CXXFLAGS in configure.py
  6. Add appveyor.yml

    Takuto Ikuta
    Takuto Ikuta committed Apr 6, 2018
Commits on Apr 5, 2018
  1. Merge pull request #1407 from nico/vim

    nico committed Apr 5, 2018
    update ninja.vim version after bff884d
  2. Merge pull request #1406 from nico/dupeerr

    nico committed Apr 5, 2018
    make `-w dupbuild` default to `err`
  3. make `-w dupbuild` default to `err`

    nico committed Apr 5, 2018
    You can still opt out of this by passing `-w dupbuild=warn`.
    But if you're getting this diagnostic, your build files are incorrect
    and you should ideally just fix them.
    
    This is step 3 on ninja-build#931
    I sent an RfC to ninja-build a few months ago; nobody objected.
  4. Merge pull request #1348 from ngg/winunicode

    nico committed Apr 5, 2018
    Fix building on Windows in UNICODE mode
  5. Merge pull request #1223 from asankah/master

    nico committed Apr 5, 2018
    [compdb] Expand response files inline based on a switch.
  6. Merge pull request #1405 from nico/showoldest

    nico committed Apr 5, 2018
    Fix confusing smart console output from concurrent builds
  7. Fix confusing smart console output from concurrent builds

    nico committed Apr 5, 2018
    Developers tend to blame the last printed line when a build takes too
    long. Unfortunately, when building concurrently, the last printed line
    may have actually finished a long time ago. Under the current system,
    ninja does not update the status line to reflect what jobs are still
    running. This change makes ninja always print the oldest still running job
    instead. In other words, the likely build bottlenecks.
    
    Patch from David Zarzycki, originally uploaded at #1320.
  8. Merge pull request #1294 from bradking/plan-track-scheduling

    nico committed Apr 5, 2018
    Track in Plan whether wanted edges have been scheduled
  9. Merge pull request #1404 from nico/c++17

    nico committed Apr 5, 2018
    make ninja build with -std=c++17
  10. make ninja build with -std=c++17

    nico committed Apr 5, 2018
    Ninja is supposed to be able to build as C++98 so it can run on old
    systems, but it should also be possible to optionally build it with
    newer dialects.
  11. Merge pull request #1361 from ppluciennik/ppluciennik/flush_log

    nico committed Apr 5, 2018
    Flush changes into .ninja_log right away.
  12. Merge pull request #1365 from moroten/fix-disk-interface-tests-64-bit…

    nico committed Apr 5, 2018
    …-timestamp
    
    Fix disk_interface_test.cc on Windows for 64-bit timestamp
  13. [compdb] Expand response files inline based on a switch.

    asankah committed Jan 9, 2017
    References to response files in a clang compile_commands.json file can
    be tricky to deal with when tooling expects all the command flags to be
    present in the 'command' field.
    
    This change introduces a '-x' option to '-t compdb' that will expand
    @rspfile style response file invocations inline.
    
    E.g.
    
    ```sh
    $ ninja -t compdb cc
    [
      {
        "directory": "/src/foo",
        "command": "cc -foo -bar @foo.obj.rsp",
        "file": "foo.cc"
      }
    ]
    
    $ ninja -t compdb -x cc
    [
      {
        "directory": "/src/foo",
        "command": "cc -foo -bar foo.cc",
        "file": "foo.cc"
      }
    ]
    ```
  14. Merge pull request #1366 from fjhenigman/clean

    nico committed Apr 5, 2018
    Canonicalize targets of clean command.
  15. Canonicalize targets of clean command.

    fjhenigman committed Dec 18, 2017
    This corrects an inconsistency where build targets were canonicalized
    but clean targets were not.  For example you could build ./foo but not
    clean ./foo.
  16. Merge pull request #1386 from jmgao/exit

    nico committed Apr 5, 2018
    Don't clean up after ourselves when exiting.
  17. Merge pull request #1356 from RedBeard0531/vim_syntax

    nico committed Apr 5, 2018
    Improve vim syntax definition
  18. Merge pull request #1392 from alekseyshl/master

    nico committed Apr 5, 2018
    Fix build on Solaris.
  19. Merge pull request #1373 from BillyDonahue/help_k_zero

    nico committed Apr 5, 2018
    Update Usage to show "-k 0" behavior
  20. Merge pull request #1381 from sgraham/output-in-compdb

    nico committed Apr 5, 2018
    Add 'output' field to compdb output
  21. Merge pull request #1388 from karkhaz/kk-add-pool-to-build

    nico committed Apr 5, 2018
    Support pool for Writer.build() in ninja_syntax.py
Commits on Mar 5, 2018
  1. Fix build on Solaris.

    alekseyshl committed Mar 5, 2018
    Solaris also does not define struct stat's st_mtimensec field.
Commits on Feb 20, 2018
  1. Support pool for Writer.build() in ninja_syntax.py

    karkhaz committed Feb 20, 2018
    ninja_syntax.py now supports a pool being specified for individual
    builds, as well as rules.
Commits on Feb 19, 2018
  1. Don't clean up after ourselves when exiting.

    jmgao committed Feb 19, 2018
    Destruction of NinjaMain can be an expensive operation when dealing
    with stupidly large ninjafiles. exit directly instead of returning out
    of real_main to avoid doing so.
Commits on Feb 7, 2018
Commits on Jan 12, 2018
  1. Update Usage to show "-k 0" behavior

    BillyDonahue committed Jan 12, 2018
    For "-k N", N==0 is interpreted as infinite. It's useful but not documented in the help, unfortunately.