Raku
source of file src/vm/js/nqp-runtime/unicode-data/names.js may be missing in nqp tarball
Hello
Debian has a strict policy to provide all sources for our packages. Long story short, a minified javascript file is not considered as source.
While creating nqp package for Debian, lintian issues the following error:
E: nqp source: source-is-missing src/vm/js/nqp-runtime/unicode-data/names.js line length is 32768 characters (>512)
To be able to package a version of nqp, I need:
- the list of files that are used to create the files found in
src/vm/js/nqp-runtime/unicode-data/(hopefully, they are available in the tarball) - instructions to build the content of
src/vm/js/nqp-runtime/unicode-data/from these files
All the best
The data in question would be Unicode Character Database. MoarVM has a similar situation, where we have source files generated from the UCD.
I don't think requiring MoarVM and NQP to bundle the Unicode Character Database in their tarballs is sensible, but fully agree the repository should include everything needed to reproduce the generated output given a copy of the UCD. I don't see such scripts in the NQP repo at first glance; probably @pmurias can help in that regard. In MoarVM we even provide a script (tools/download-UCD.p6) that takes care of obtaining the files; that's a developer convenience, but may be relevant in this context also.
On Debian, the UCD files are provided by uniciode-data package. So it would be possible to rebuild the content of src/vm/js/nqp-runtime/unicode-data/ without network access during the build.
Anyway, we don't plan to build js backend, so I'll remove the js files from nqp to be compliant with Debian policy.
I'll close this bug unless you want to keep it open until the unicode data files can be rebuilt,
Thanks for the help
