- May 03, 2008
-
-
Gary King authored
Added around methods for perform loading and compiling cl-source-files. These add a simple-restart to retry the operation.
-
- Feb 15, 2008
-
-
Nikodemus Siivola authored
gah.
-
- Feb 02, 2008
-
-
Nikodemus Siivola authored
* Merged from SBCL CVS. * Original patch by Michael Weber. * Trailing whitespace cleanup.
-
Nikodemus Siivola authored
-
Nikodemus Siivola authored
* Detabify. * Fix indentation. * Wrap long docstrings.
-
- Dec 14, 2007
-
-
Gary King authored
Applied patch from Luis Oliveira to prevent unnecessary unreachable code warnings under SBCL when using the :pathname argument in defsystem.
-
- Sep 27, 2007
-
-
Gary King authored
(thanks to Peter Van Eynde).
-
- Jul 27, 2007
-
-
Gary King authored
-
- Jun 13, 2007
-
-
Gary King authored
Removes inline component methods more carefully (and, I believe, correctly!). Helps keep LispWorks happy.
-
- Mar 21, 2007
-
-
Gary King authored
system name and a partial pathname and combines them to return a new pathname _relative_ to the system. This is exported and lightly documented. this patch also updates preference-file-for-system/operation in an effort to prevent it from being confused by *default-pathname-defaults* that include a specific name and type.
-
- Feb 14, 2007
-
-
Gary King authored
rather than (package-name *package*); this keeps case sensitive Lisps happy all the time and keeps case insensitive lisps happily insensitive.
-
- Feb 06, 2007
-
-
Gary King authored
better with *load-truename* and *default-pathname-defaults*; this makes it easier to work with defsystems in the REPL.
-
- Jan 23, 2007
-
-
Gary King authored
-
- Jan 21, 2007
-
-
Gary King authored
-
- Nov 07, 2006
-
- Aug 21, 2006
-
-
Christophe Rhodes authored
types. It seems regrettably common for system authors, despite the documentation, to place their system in CL-USER, where the user can quite legitimately add symbols. Adjust the lookup code in CLASS-FOR-TYPE to ignore symbols not naming subclasses of ASDF:COMPONENT, and also to try the symbol itself first if it's not a keyword. Also adjust the documentation slightly to make this slightly clearer. Add test files, and make the clisp test-harness work for me.
-
- Jul 06, 2006
-
-
Gary King authored
them so that preferences are found in ~/.asdf/<name-of-system>.lisp by default and are loaded on either a load-op or a load-source-op. Refactored load-op and load-source-op to both be subclasses of basic-load-op to facilitate this. Added test case in test directory. Still need to add to documentation of ASDF and will once I figure out where such a beast truely lives.
-
- Jun 16, 2006
-
-
Gary King authored
-
- May 30, 2006
-
-
Gary King authored
Added &allow-other-keys to operate. Added test/test-force to test that forcing a system works (at least in a simple case).
-
- May 14, 2006
-
- Apr 20, 2006
-
-
Christophe Rhodes authored
-
Christophe Rhodes authored
is still subject to filesystem races (say, between probe-file and file-write-date) but they're a lot less likely. Attempt to do the "right" thing if an input file has gone missing.
-
Christophe Rhodes authored
-
- Mar 21, 2006
-
-
Christophe Rhodes authored
* don't create a package with gensym. Instead try creating packages with different names until one succeeds; * delete the scratch package when we're finished with it * if the user has requested a pathname for a component, don't try to construct the default one; it probably won't work.
-
- Feb 01, 2006
-
-
Christophe Rhodes authored
on Windows. No changes outside that specific section.
-
- Dec 16, 2005
-
-
Nikodemus Siivola authored
-
- Dec 07, 2005
- Sep 26, 2005
-
-
Peter Van Eynde authored
-
- May 25, 2005
-
-
Peter Van Eynde authored
defsystem files. Now works again with McClim. Closes: #310640 * Corrected dependencies.
-
- Jul 19, 2004
-
-
Christophe Rhodes authored
-
- May 16, 2004
-
-
Daniel Barlow authored
* "ECL cannot externalize packages in compiled files. In other words, compiled files cannot have package objects as literal constants. Things like LOAD-TIME-VALUE, or saving the package name and afterwards looking for the package is all right. This only affects your uses of #.*package* somewhere in the code." * Also an appropriate run-shell-command implementation
-
- May 05, 2004
-
-
Edi Weitz authored
-
Christophe Rhodes authored
workaround, and the clisp developers have the aim of making asdf work "out of the box" for their 2.34 release -- and I'd hate for them to target the workarounded version rather than the one that's idiomatic.
-
Edi Weitz authored
-
- Dec 21, 2003
-
-
Daniel Barlow authored
-
- Dec 01, 2003
-
-
Daniel Barlow authored
-
- Oct 17, 2003
-
-
Christophe Rhodes authored
deprecation, because (a) I want to undeprecate it; (b) its effect is trivially reimplementable even if it vanishes; (c) it will shortly become the right thing to do in sbcl.
-
- Oct 09, 2003
-
-
Daniel Barlow authored
asdf is saved in a core. When the core is reloaded, the stream is not attached to anything interesting.
-
- Jul 17, 2003
-
-
Christophe Rhodes authored
* make the "warned while..." and "failed while..." arguments go the right way round * use pretty printing (and the FORMATTER trick to keep CLISP happy) Tidy up the RESTARTs * new names, exported (RETRY and ACCEPT) * descriptions of the restarts for the debugger * set COMPONENT-OPERATION-TIMES in the ACCEPT restart
-