Charles Oliver Nutter

Marshal data too short accessing mavengems

7

This seems to be a recent issue, possibly related to some unknown update to rubygems.org: ``` [INFO] --- maven-jar-plugin:3.0.0:jar (default-jar) @ jruby-stdlib --- [INFO] Building jar: /Users/headius/projects/jruby/lib/target/jruby-stdlib-9.2.13.0-SNAPSHOT.jar [INFO] [INFO] -----------------------<...

Improved representation of JRuby for JVM tools

2

This PR will encompass several improvements to JRuby's representation of code and objects, in order to make JVM tooling more useful. Among the ideas explored here: * Cleaner representation of...

This change detects when a literal array contains only fixnums or floats and uses indy to embed those values in a call site, greatly reducing the bytecode size for constructing...

This started out as a call site optimization to pass `$_` directly yo `Kernel#print` or `IO#print` rather than provide it through a heap frame, but the simpler change is to...

This is a modified version of a patch proposed in https://github.com/jruby/jruby/issues/7515#issuecomment-1368041882 that uses a third mechanism for getting the PID: the `pid` method added to ProcessImpl in 8 and made...

This change was applied to CRuby in 3.0. See https://bugs.ruby-lang.org/issues/12654 Fixes #7750

For #1088 I got into the quagmire of trying to get jzlib to receive compression level and compression strategy values from the GzipWriter constructors, and just ran into more errors...

ByteList.toString should not be used for encoded data

2

ByteList.toString seems like it should decode the bytes according to the ByteList's encoding, but instead it just decodes them using ISO-8859-1. This causes the resulting string to lose its encoded...

This previously decoded the bytelist always as an ISO-8859-1 string, which would obviously break for other encodings and any multibyte characters. This change uses the ByteList's Encoding's actual Charset to...

Fixes #8299