Skip to content
Commits on May 10, 2016
  1. Add and configure rubocop.

    - Create .rubocop_todo.yml with the command `rubocop --auto-gen-config`.
    - Add .rubocop.yml that inherits from .rubocop_todo.yml.
    committed May 9, 2016
  2. Update Gemfile.lock.

    committed May 9, 2016
  3. Add '.bundle' to .gitignore

    The '.bundle' directory is specific to the local environment and should
    be ignored.
    
    By default, bundler should install gems system-wide. However, as of
    bundler version 1.12, a .bundle directory is created in the project
    root on my system for reasons currently unknown.
    committed May 9, 2016
Commits on Feb 9, 2015
  1. Bump version to 0.5.1

    committed Feb 8, 2015
  2. Update depenency versioning

    committed Feb 8, 2015
  3. Bump version to 0.5.0

    committed Feb 8, 2015
  4. Update copyright date

    committed Feb 8, 2015
  5. Remove caveats from Readme

    These are outdated and not particularly informative.
    committed Feb 8, 2015
  6. Print unicode to stdout with both Python 2 and 3

    When attempting to print a unicode character, Python 2 requires use of
    `#encode('utf8')`, otherwise it erros with the "'ascii' codec can't
    decode byte" error message, discussed here:
    
    https://docs.python.org/2/howto/unicode.html#the-unicode-type
    
    Meanwhile, if `#encode()` is used on a string in Python 3,
    sys.stdout.write errors with the message "must be str, not bytes".
    
    The best solution seems to be to check against the python version
    and use a utf8 writer for sys.stdout in Python 2, as noted here:
    
    http://stackoverflow.com/a/24104423
    
    Other resources consulted include these:
    
    https://docs.python.org/3/howto/unicode.html
    http://www.pythoncentral.io/encoding-and-decoding-strings-in-python-3-x/
    committed Feb 8, 2015
Commits on Feb 8, 2015
  1. Use sys.stdout.write for rst2latex

    This is preferrable to print() since it doesn't do any additional
    formatting. This is also what is used for rst2html, so this change
    brings them to the same functionality.
    committed Feb 8, 2015
  2. Reverse argument order in conversion asserts

    The first argument should be the expected value
    committed Feb 8, 2015
  3. Split unicode tests to one each of latex and html

    Since both conditions need testing, add a different test for each one.
    
    As of this commit, these two tests both fail since unicode input causes
    empty strings to be returned. Also, the output statements in the python
    executables are different and if this is deliberate, then a note needs
    to be added.
    committed Feb 8, 2015
  4. Use multiline method chaining and assert_equals in tests

    Multilining everything makes it easier to navigate and read, and it
    avoids having arguments far off to the side after long strings.
    
    Backslash / newline / period format for multiline chaining is used
    because it's the method that communicates the most about the fact that
    the chain is multilined. With this approach, it's immediately obvious
    that it's a chain no matter what line you look at.
    committed Feb 8, 2015
  5. Remove trailing whitespace

    committed Feb 8, 2015
  6. Merge pull request #3 from hsoft/master

    Add Python 3 support
    committed Feb 8, 2015
  7. Bump version to 0.4.0

    committed Feb 8, 2015
  8. Remove VERSION file

    The version already exists within the gemspec
    committed Feb 8, 2015
  9. Update test latex file

    committed Feb 8, 2015
  10. Remove jeweler dependency

    This made gem building easier in the past, but hasn't been needed for a
    while.
    committed Feb 8, 2015
Commits on Jun 2, 2014
  1. @hsoft

    Add Python 3 support

    Under Python 3, `sys.stdout.write()` doesn't accept bytes so we need to
    give it a unicode string.
    
    `transform()` returned its result encoded in the part's original
    encoding, which didn't make much sense so I've removed that part to keep
    the result in a unicode string.
    
    I've tested the result with both Python 2 and 3 on a utf8 source file
    containing non-ascii characters. Works fine.
    
    Results are a bit fuzzier with source encoding that aren't utf-8, but
    that problem was already present before this change.
    hsoft committed Jun 2, 2014
Commits on Nov 2, 2012
  1. Bumping version to 0.3.0

    committed Nov 1, 2012
  2. Updating test strings and files for changes in 81d5b5f

    Newlines at ends of output strings
    committed Nov 1, 2012
  3. Merge remote branch 'frankmichler/master'

    * frankmichler/master:
      make it working on Windows by using IO.popen instead of popen3
      Instead of printing the output we write it to stdout directly to avoid some encoding issues. This will probably make RbST work as long as you use UTF-8.
      encodes output correctly
    committed Nov 1, 2012
Commits on Nov 1, 2012
  1. Bumping version to 0.2.0

    committed Nov 1, 2012
Something went wrong with that request. Please try again.