Permalink
Commits on Mar 21, 2018
  1. Version 0.5.14.8

    erikd committed Mar 21, 2018
  2. Support GHC 8.4.1 (0.5 branch).

    23Skidoo authored and erikd committed Mar 20, 2018
  3. Update .gitignore.

    23Skidoo authored and erikd committed Mar 20, 2018
Commits on Sep 2, 2017
  1. correct spelling mistake

    EdwardBetts authored and erikd committed Sep 1, 2017
Commits on Mar 27, 2016
  1. Version 0.5.14.7

    erikd committed Mar 27, 2016
  2. Backport unicode and other string escape fixes to 0.5

    teh authored and erikd committed Mar 27, 2016
    Signed-off-by: Erik de Castro Lopo <erikd@mega-nerd.com>
    Closes: #50
Commits on Mar 18, 2016
  1. Version 0.5.14.6

    erikd committed Mar 18, 2016
  2. Lexer.x: Fix regex lexing bug

    erikd committed Mar 18, 2016
    The simplest example of what was failing to compile was `/[/]/` where
    the forward slash in the middle was being incorrectly detected as the
    end of the regex.
    
    Closes: #44
  3. Version 0.5.14.5

    erikd committed Mar 18, 2016
  4. Lexer.x: Fix parsing of escaped quotes

    erikd committed Mar 18, 2016
    The existing lexer was failing to accept strings with escaped quotes
    like `"\""` and `'\''`.
Commits on Mar 17, 2016
  1. Version 0.5.14.4

    erikd committed Mar 17, 2016
  2. language-javascript.cabal: Remove upper bound on HUnit

    erikd committed Mar 17, 2016
    The `< 1.3` upper bound was originally added because the first
    `HUnut 1.3` had problems.
    
    Closes: #46
Commits on Mar 16, 2016
  1. Version 0.5.14.3

    erikd committed Mar 16, 2016
  2. Parser: Fix parsing of single line do/while statement

    erikd committed Mar 16, 2016
    New test included.
    
    Closes: #45
Commits on Aug 21, 2015
  1. Rename runtests.hs -> testsuite.hs

    erikd committed Aug 21, 2015
    Also update the cabal file to match so that the compiled test suite
    binary is `dist/build/testsuite/testsuite` and update travis file.
Commits on Aug 10, 2015
  1. Lexer: Improve handling of character literals in strings

    erikd committed Aug 10, 2015
    The reported issue was about '\x1b' (escape) embedded in a string
    but it turned out to be a problem with all characters with an ordinal
    value of < 0x20.
    
    Closes: #42
Commits on Aug 5, 2015
  1. travis: Rrefresh

    erikd committed Aug 5, 2015
    * Use cabal-install-1.22.
    * Switch to ghc 7.10.2.
    * Run tests directly instead of via cabal.
Commits on Aug 4, 2015
  1. runtests.hs: Add testcase IfElse3

    erikd committed Aug 4, 2015
    This test passes on the master branch but was failing on the
    new-ast branch.
Commits on Aug 3, 2015
Commits on Aug 2, 2015
  1. Version 0.5.14.2

    erikd committed Aug 2, 2015
  2. Updating alex dependency for ghc 7.10 to get the Applicative instances

    erdeszt authored and erikd committed Aug 2, 2015
    Signed-off-by: Erik de Castro Lopo <erikd@mega-nerd.com>
Commits on Jul 29, 2015
  1. Version 0.5.14.1

    erikd committed Jul 29, 2015
  2. fix haddock markup in Language/JavaScript/Parser/Token.hs

    trofi authored and erikd committed Jul 29, 2015
    haddock failed as:
    > src/Language/JavaScript/Parser/Token.hs:16:5:
    >     parse error on input ‘-- * The tokens’
    
    Signed-off-by: Sergei Trofimovich <siarheit@google.com>
    Signed-off-by: Erik de Castro Lopo <erikd@mega-nerd.com>
    
    Closes: #38
    Closes: #39
Commits on Jul 28, 2015
  1. Version 0.5.14.0

    erikd committed Jul 28, 2015
  2. Parser: Fix another break/continue/return issue

    erikd committed Jul 28, 2015
    After the previous lexer/parser fix, another problem showed up
    with these keywords enclosed in curly braces. The solution was
    to update the AutoSemi parser rule to accept any of the following:
    
    - A real semicolon
    - An AutoSemiToken inserted by the lexer
    - Nothing (which fixes this new case)
    
    Tests added as well.
  3. Lexer/Parser: Fix break/continue/return parsing

    erikd committed Jul 28, 2015
    There were a number of issues around parsing these keywords when
    when the code doesn't contain semi-colons. For example, this valid
    code fragment would fail to parse:
    
         function f (x) {
             if (x) return
             x.foo()
         }
    
    The solution was to modify the lexer so that it inserts an AutoSemiToken
    after break/continue/return if and only if the white space immediately
    following those keywords contain a newline.
    
    Closes: #30
    Closes: #36
  4. camelCase

    erikd committed Jul 28, 2015