Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with
or
.
Download ZIP
Branch: nom
Commits on Aug 4, 2015
  1. @lizmat
  2. @jnthn
  3. @jnthn

    Bring Scalar creation on in line with JVM.

    jnthn authored
    Which also makes it usefuller for GLR.
  4. @coke

    Merge pull request #486 from jonathanstowe/test-errors

    coke authored
    skip-rest has the wrong name in the error message
  5. @jonathanstowe
  6. @lizmat
  7. @lizmat

    Add indices() as a sub

    lizmat authored
  8. @lizmat
  9. @lizmat
Commits on Aug 3, 2015
  1. @niner
Commits on Aug 1, 2015
  1. @moritz

    Remove "Failure changes" from ROADMAP

    moritz authored
    nobody seems to quite know anymore what it is supposed to be,
    and our Failure looks pretty much up to spec
Commits on Jul 31, 2015
  1. @TimToady
  2. @jnthn

    Re-implement &?ROUTINE.

    jnthn authored
    This uses the same appraoch as &?BLOCK, meaning you only pay for the
    feature if you use it. This unbreaks MoarVM's ability to do a bunch of
    inlining, which has rather positive performance consequences.
  3. @jnthn

    Fix samewith.

    jnthn authored
    It did rely on &?ROUTINE really being in the lexpad, but further to
    that it failed to search hard enough for a routine, so you could not
    do samewith inside of a closure inside of the routine.
  4. @jnthn
  5. @jnthn

    Avoid repeated calls to .name.

    jnthn authored
Commits on Jul 30, 2015
  1. @jnthn
  2. @jnthn
  3. @jnthn
  4. @jnthn

    Bump NQP_REVISION.

    jnthn authored
    Gets a MoarVM with a race condition in frame validation fixed, along
    with a bug in extop/exception interaction fixed.
  5. @jnthn

    Test for concise error on sinking last statement.

    jnthn authored
    If the last statement in a file used to need sinking, and doing so
    threw an exception, we'd spit out an epic low-level backtrace rather
    than a nice high-level one. Fix that by making non-EVAL and non-REPL
    stuff we compile put the last statement in sink context in the code we
    generate rather than having MAIN do the sinking. Also means a trailing
    0; in a file now warns.
  6. @jnthn

    Fix private method call in role code-gen.

    jnthn authored
    We can't just assume $*PACKAGE holds the type we should locate the
    private method in. Rather, ::?CLASS must be used in this case.
  7. @jnthn
  8. @jnthn

    Always decont things used as type constraints.

    jnthn authored
    Fixes RT #125715.
  9. @jnthn

    Avoid boolifying code objects in compiler.

    jnthn authored
    Some of them react to boolification by trying to match against $_,
    when we only want to check if we have one. Fixes RT #125670.
  10. @jnthn

    Tighten up types on things that must be callable.

    jnthn authored
    This avoids whining (potentially at a distance) about missing .count,
    .arity, and so forth, and instead showing a more helpful error about
    how we can't bind the signature here.
Commits on Jul 29, 2015
  1. @lizmat

    Streamline (r)index some more

    lizmat authored
    - separate candidates with/without pos
    - in signature coercion to Str/Int
  2. @lizmat

    Merge pull request #485 from Leont/exit-status

    lizmat authored
    Added leont to CREDITS
  3. @Leont

    Added leont to CREDITS

    Leont authored
  4. @lizmat
  5. @FROGGS

    Merge pull request #484 from Leont/exit-status

    FROGGS authored
    Pass correct exitstatus and signal values from Proc::Async to Proc
  6. @Leont
  7. @lizmat

    Don't call BUILD in new

    lizmat authored
  8. @lizmat
  9. @hoelzro

    Replace send with print on IO::Socket::Async

    hoelzro authored
    For consistency with IO::Socket
Something went wrong with that request. Please try again.