Switch branches/tags
Nothing to show
Commits on Feb 12, 2016
  1. Forgot to remove garbage.

    atsushieno committed Feb 12, 2016
  2. Merge commit 'c7700bf'

    atsushieno committed Feb 12, 2016
Commits on Oct 7, 2015
Commits on Aug 26, 2015
Commits on Aug 3, 2015
  1. Detecting @deprecated was too hacky.

    atsushieno committed Aug 3, 2015
    It was just checking "caution" section. It covered @deprecated but in few
    times it had false positives (negatives?) and other kind of "caution"s
    were treated as @deprecated.
    Now it also checks text content (still hacky, but seems to work).
Commits on Jul 9, 2015
  1. Update M to preview2.

    atsushieno committed Jul 9, 2015
Commits on May 28, 2015
  1. Add M preview.

    atsushieno committed May 28, 2015
Commits on Mar 12, 2015
Commits on Nov 26, 2014
  1. Merge pull request #3 from xamarin/bug22155

    jonpryor committed Nov 26, 2014
    [jar2ml] GoogleApiClientBuilder.AddApi(Api, Object) throws Java.Lang.NoSuchMethodError Exception
  2. [jar2ml] GoogleApiClientBuilder.AddApi(Api, Object) throws Java.Lang.…

    dellis1972 committed Nov 26, 2014
    …NoSuchMethodError Exception
    Jar2Xml was not emitting generic type contraints in a
    format that would work with generator. It was including
    nested types with a $ rather than a .
    This commit fixes the output of those types to use
    the correct seperator.
Commits on Oct 20, 2014
  1. android-L -> android-21.

    atsushieno committed Oct 20, 2014
Commits on Jul 31, 2014
Commits on Jul 30, 2014
  1. Merge pull request #2 from xamarin/javacfix

    jonpryor committed Jul 30, 2014
    Updated the Makefile to use the -source and -target parameters
  2. Updated the Makefile to use the -source and -target parameters to ensure

    dellis1972 committed Jul 30, 2014
    we generate .class files with a known version of java
Commits on Jul 29, 2014
Commits on Jun 26, 2014
  1. Add API Level 20 and L.

    atsushieno committed Jun 26, 2014
Commits on Apr 15, 2014
  1. Catch any kind of error that is caused by ClassLoader.

    atsushieno committed Apr 15, 2014
    This change is nasty. Well, it has been already nasty, but worse. The thing is
    Oracle ClassLoader could throw unexpectable variety of Errors, not even
    exceptions. And we cannot really predict what they could be.
    So we catch everything.
Commits on Jan 10, 2014
  1. regarding bug #15885, explicitly show Java stack trace before J2XA006…

    atsushieno committed Jan 10, 2014
    … error so that people can find how it is impossible to retrieve entire Java methods.
Commits on Nov 12, 2013
  1. Add deploy and undeploy targets.

    jonpryor committed Nov 12, 2013
    Previously, the install and uninstall targets would install things
    onto a target device, instead of the traditional semantic of
    installing the SDK into a target directory.
    We're changing that so that install & uninstall follow the traditional
    semantic, so introduce new deploy and undeploy targets for the
    previous install/uninstall semantics.
Commits on Oct 31, 2013
  1. add API Level 19.

    atsushieno committed Oct 31, 2013
Commits on Aug 29, 2013
Commits on Jul 24, 2013
  1. add API Level 18.

    atsushieno committed Jul 24, 2013
Commits on Jul 19, 2013
Commits on Jun 13, 2013
Commits on May 1, 2013
Commits on Mar 6, 2013
  1. Revert "avoid "java.lang.Void" output."

    atsushieno committed Mar 6, 2013
    This reverts commit 963c647.
Commits on Feb 28, 2013
  1. Oracle is buggy and throws NullPointerException for getGenericTypeNam…

    atsushieno committed Feb 28, 2013
    …e() too, so recover from that too.
    This should fix bug #10744.
Commits on Jan 31, 2013
Commits on Jan 23, 2013
  1. Add install, uninstall targets.

    jonpryor committed Jan 23, 2013
    This is to allow a toplevel `make uninstall install` to work.
Commits on Jan 8, 2013
Commits on Dec 5, 2012
  1. Workaround for Java bug that returns null in ParameterizedType.getAct…

    atsushieno committed Dec 5, 2012
    …ualTypeArguments() return array.
Commits on Nov 28, 2012
  1. From this version, any failed attempt to get generic type retrieval f…

    atsushieno committed Nov 28, 2012
    …alls back to non-generic one.
    This is to fix issues like bug #8619.
    Basically this is proguard bug. It should not expose any obfuscated type names
    in its output, not just something that the library code invokes.
    Though there are likely too many libraries that will be blocked in this way,
    we likely had better live with some workaround. Hence this "fix".