On Apr 23, 2025, at 09:50, Christoph Berg <myon@debian.org> wrote:
> Remembering which path the .control file was found in and from there
> open the extension "directory" doesn't sound too hard. Why does it
> have to be more complicated?
This was my question, as well. Do you have a WIP patch to share, Matheus?
> Also, re-running a search path discovery for the directory is probably
> just wrong, if there are different extension versions in the "control"
> search path and the "extensions" search path, it might lead to weird
> version skew problems.
I assumed we would just have one or the other GUCs, not both.
> The number of extensions using that feature is limited, though, so it
> wouldn't be a huge problem:
FWIW it’s a a simple patch to make semver work, and probably also for the others. It’s just the reverse of this
change[1]:
```patch
--- a/Makefile
+++ b/Makefile
@@ -5,7 +5,6 @@ EXTVERSION = $(shell grep -m 1 '[[:space:]]\{8\}"version":' META.json | \
DISTVERSION = $(shell grep -m 1 '[[:space:]]\{3\}"version":' META.json | \
sed -e 's/[[:space:]]*"version":[[:space:]]*"\([^"]*\)",\{0,1\}/\1/')
-MODULEDIR = semver
DATA = $(wildcard sql/*.sql)
DOCS = $(wildcard doc/*.mmd)
TESTS = $(wildcard test/sql/*.sql)
--- a/semver.control
+++ b/semver.control
@@ -1,7 +1,5 @@
# semver extension
comment = 'Semantic version data type'
default_version = '0.32.1'
-
-directory = 'semver'
module_pathname = '$libdir/semver'
relocatable = true
```
I think I’ll write a blog post this week recommending people not use these directives, and also to remove `$lib/` from
`module_pathname`.
Best,
David
[1]: https://github.com/theory/pg-semver/commit/88b3abd
