FS#35754 - update_pkgver() should properly fail when sed returns nonzero exit code

Attached to Project: Pacman
Opened by Zack Buhman (buhman) - Tuesday, 11 June 2013, 07:30 GMT
Last edited by Allan McRae (Allan) - Tuesday, 15 October 2013, 02:47 GMT
Task Type Bug Report
Category makepkg
Status Closed
Assigned To Allan McRae (Allan)
Architecture All
Severity Medium
Priority Normal
Reported Version git
Due in Version 4.2.0
Due Date Undecided
Percent Complete 100%
Votes 0
Private No

Details

Summary and Info:

update_pkgver() does not check to make sure @SEDINPLACE@ on line 813 actually succeeds. This causes a seemingly successful build, but lacking a rather important pkgver bump.

Admittedly a non-broken PKGBUILD would not cause this problem to begin with; but it would be quite useful for debugging.

Steps to Reproduce:

Attached is a sample PKGBUILD with a pkgver() that will pass check_pkgver(), but still result in an invalid pkgver and escape the sed on line 813 in scripts/makepkg.sh.in.

Suggested fix included.
   PKGBUILD (0.2 KiB)
   die-properly-on-invalid-newpk... (0.6 KiB)
This task depends upon

Closed by  Allan McRae (Allan)
Tuesday, 15 October 2013, 02:47 GMT
Reason for closing:  Fixed
Additional comments about closing:  git commit 968486ec
Comment by Zack Buhman (buhman) - Tuesday, 11 June 2013, 07:42 GMT
My mistake; I didn't realize check_pkgver happens *after* update_pkgver on line 1372, but the order of operations isn't exactly relevant in either case (sorry I placed implicit partial blame on check_pkgver).
Comment by Allan McRae (Allan) - Monday, 14 October 2013, 06:52 GMT
Sent to mailing list with the following minor change:

error "$(gettext "Failure updating %s from %s to %s")"
"pkgver" "$pkgver" "$newpkgver"

Loading...