summaryrefslogtreecommitdiff
path: root/web/prefix.py
diff options
context:
space:
mode:
authorGabriel Schneider <[email protected]>2026-11-17 13:11:46 -0300
committerGabriel Schneider <[email protected]>2026-11-17 13:13:37 -0300
commit2fb9cbad0c88dbc3f2db86e2490fca40bc6b47bf (patch)
tree564d315cd8ee60fd743ca66782a1251b1bf6037d /web/prefix.py
parent8c32753594ce0d6852336206b86cbdd88a8878b1 (diff)
downloadcloud9-2fb9cbad0c88dbc3f2db86e2490fca40bc6b47bf.tar.gz
cloud9-2fb9cbad0c88dbc3f2db86e2490fca40bc6b47bf.zip
README: a package that uses cloud9, as test/consumer doesHEADmain
The README and 9proc's told a consumer to pass .target and .optimize to b.dependency("cloud9", ...), and to swap the URL pin for a path while working on both packages. Now: the two modules take no options, and the whole build.zig of a program on cloud9 is written out (eleven lines, the form test/consumer uses); nothing else of cloud9's build runs in it. The sibling checkout is Zig's --fork=../cloud9, with the manifest untouched. Checked outside the tree: that build.zig, in a package pinning cloud9 by URL and hash (pardes's pin, 7f418a6b, from a copy of its zig-pkg entry), builds and runs; with --fork=<this checkout> it builds against this one ("fork ... matched 1 cloud9 packages", and no jj or date at configure); and against this tree as a fetched package (a tarball of its tracked files, so the manifest's paths filter applies) it builds too. Co-Authored-By: Claude Opus 5.5 <[email protected]> Claude-Session: https://claude.ai/code/session_016Y9MNBECMWcJD5pdS4SR6T
Diffstat (limited to 'web/prefix.py')
0 files changed, 0 insertions, 0 deletions