Just Launched: You can now import projects and releases from Google Code onto SourceForge
We are excited to release new functionality to enable a 1-click import from Google Code onto the Allura platform on SourceForge. You can import tickets, wikis, source, releases, and more with a few simple steps. Read More
From: Erik Huelsmann <ehuels@gm...> - 2009-04-22 07:49:54
Last evening, I turned out to be on a fixing streak: when the evening
started, we had 41 failing ANSI tests. Around midnight, we had only 33
(!) in the interpreted case.
The failing tests in the interpreter also fail in the compiler.
Additionally, the compiler fails these tests:
The first 4 tests fail because the compiler - contrary to the
interpreter - doesn't support the use of non-constant initforms as
The remaining failures are mostly pretty-printer issues: the pretty
printer has an issue when *print-pretty* is bound to T, where it emits
"#=(A . #1#)" as "(A A . )", for example.
Finding this issue means we can reduce the number of failures by a
large number in 1 blow. Anyone with time on their hands: if you're
looking for something to fix, this is a nice one in terms of impact on
One thing that could help: the pretty printer is also used in SBCL.
Examining the changes in SBCL since the file was spawned off might
help to fix detected and latent issues in our version. Well, so much
for hinting. Anybody who starts working on this issue, please report
back before you do so: that'll help prevent double work.