Closed Bug 551474 Opened 14 years ago Closed 13 years ago

Allow specification of reftest manifest in testing/testsuite-targets.mk

Categories

(Firefox Build System :: General, defect)

defect
Not set
normal

Tracking

(Not tracked)

RESOLVED DUPLICATE of bug 612677

People

(Reporter: bc, Assigned: bc)

Details

Attachments

(1 file, 1 obsolete file)

Attached patch patch v1 (obsolete) — Splinter Review
Right now it is not possible to specify the manifest file for any of the reftest based tests in testing/testsuite-targets.mk since TEST_PATH is hijacked by mochitest and xpcshell tests.

If we used REFTEST_PATH instead of TEST_PATH for the reftest style tests, we could do it.

This would be good on 1.9.2 too.
Attachment #431643 - Flags: review?(ted.mielczarek)
Using TEST_PATH works fine for me:
make -C ../opt-mozilla-central/ reftest TEST_PATH=layout/reftests/reftest-sanity/reftest.list

The Mochitest targets don't touch TEST_PATH, they just use the value if set.
Attached patch patch v2Splinter Review
patch v2 without the var name change.

I see the error of my ways. I had originally tried

export TEST_PATH=...
make -C ...

which had used the originally defined manifest in the testsuite-targets.mk. I just ass-umed it was because of the TEST_PATH conditional at the top completely missing that it was setting TEST_PATH_ARG and not TEST_PATH and never tried

make -C ... TEST_PATH=...

My problem is I didn't understand 

<quote>
Target-specific variables have the same priority as any other makefile variable. Variables provided on the command-line (and in the environment if the `-e' option is in force) will take precedence. Specifying the override directive will allow the target-specific variable value to be preferred. 
</quote>

I think this is probably either an invalid or wontfix issue unless other people hit the environment variable vs. command line variable issue like I did.

Ted, I'll leave that up to you.
Attachment #431643 - Attachment is obsolete: true
Attachment #431721 - Flags: review?(ted.mielczarek)
Attachment #431643 - Flags: review?(ted.mielczarek)
Comment on attachment 431721 [details] [diff] [review]
patch v2

No, this looks totally reasonable. The precedence of make variables is definitely unintuitive.
Attachment #431721 - Flags: review?(ted.mielczarek) → review+
Assignee: nobody → bclary
Status: NEW → RESOLVED
Closed: 13 years ago
Resolution: --- → DUPLICATE
Product: Core → Firefox Build System
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: