slyrus
Fix typo in resolve method
Currently,
CL-USER(21): (abcl-asdf:resolve "com.xeiam.xchange:xchange-core")
#<THREAD "interpreter" {4E3DB38D}>: Debugger invoked on condition of type PROGRAM-ERROR
Wrong number of arguments for #<RESOLVE-DEPENDENCIES {164FBB94}>; at least 2 expected.
fails, and this succeeds instead:
CL-USER(23): (abcl-asdf:resolve "com.xeiam.xchange/xchange-core")
WARNING: Using LATEST for unspecified version.
...
Interesting. I've been (erroneously?) using slash there. I have a hard time finding a specification for maven artifact (?) designators (or whatever these strings are called). Can you point me to some documentation that describes the rules/conventions for these things? thanks!
I was working off the contrib's own documentation:
https://github.com/adlai/abcl/blob/patch-1/contrib/abcl-asdf/README.markdown#example-2
This page from Maven's documentation uses colon syntax:
https://maven.apache.org/plugins/maven-dependency-plugin/examples/filtering-the-dependency-tree.html
It's not clear to me if the correct thing to do is to fix the code or the documentation. I was expecting some sort of standard maven URI description. It is true that the syntax for filter patterns is as you describe but these guys, e.g., https://ops4j1.jira.com/wiki/display/paxurl/Mvn+Protocol use a syntax with slashes. I guess I don't really care but since there doesn't seem to really be a standard here I'm not super inclined to change the format unless there's a good reason to do so.
OK, modified the commit to leave behavior as-is, and updated the documentation.
I'm still not sure what the right thing to do is here and am open to suggestions. The maven resolution code could definitely use some help but it sounds like making sure the code and documentation (and hopefully the people who use both) are on the same page regarding the spec for the maven artifact URI-like things would be a good thing.
ABCL-ASDF:RESOLVE is deliberately generous, attempting to parse either format (i.e. accepting both "group-id/artifact-id" and "group-id:artifact-id"). But the NAMESTRING of the parsed object provides the canonical CL:TRUENAME of the "mvn" schema.
I wonder whether "Close with comment" is appropriate; obviously, the cleanest click.