Re: The Inbox: MethodMassage-dtl.73.mcz

Eliot Miranda <[email protected]>
Newsgroups gmane.comp.lang.smalltalk.squeak.general
Message-ID <[email protected]>
Hi Dave,

On May 23, 2026, at 3:48 PM, [email protected] wrote:



A question regarding BytecodeCoverageTests>>testCoverageOfTestBasicBlockLabelsForMethod - is this a test that we actually expect to work? By my reading, it is attempting to assert that a method that has been modified by ByteCodeCoverer>>coverMethod: will be clever enough to fix itself if someone tries to evaluate the modified method. It would be nice if that worked, but this a realistic scenario that needs to be implemented? We already have ByteCodeCoverer>>uncoverMethod: which does the expected thing. I am asking because testCoverageOfTestBasicBlockLabelsForMethod is not easy to understand, and if it is testing something that we do not really expect to work. then maybe we should get rid of the test?

I’m ok with that. This part of MethodMassage comes from a line oriented debugger I had to implement for a Cadence system built above Newspeak/Newsqueak and is not something I expect we’ll ever want to revisit. Testing uncoverMethod: is much more relevant.

Dave

On 2026-05-23 22:24, [email protected] wrote:

David T. Lewis uploaded a new version of MethodMassage to project The Inbox:
http://source.squeak.org/inbox/MethodMassage-dtl.73.mcz

==================== Summary ====================

Name: MethodMassage-dtl.73
Author: dtl
Time: 23 May 2026, 6:24:40.21969 pm
UUID: 3148219f-bb82-4666-9fd7-5c6d6e727ef1
Ancestors: MethodMassage-dtl.72

Make BytecodeCoverageTests>>expectedFailures do the right thing for EncoderForV3PlusClosures
The #expectedFailures for BytecodeCoverageTests appears first in MethodMassage-eem.58 but was mistakenly implemented on the class side and later moved to instance side in MethodMassage-ct.69, which now results in an expected failure for #testCoverageOfTestBasicBlockLabelsForMethod for the test running with EncoderForSistaV1, and a not expected failure for EncoderForV3PlusClosures. The test fails with either encoder, so just mark it as an expected failure.

=============== Diff against MethodMassage-dtl.72 ===============

Item was changed:
----- Method: BytecodeCoverageTests>>expectedFailures (in category 'failures') -----
expectedFailures
"Alas this class's single test only passes if there are methods in the MethodMassage package that are compiled without full blocks."
+
- self thisPackage actualMethodsDo:
- [:m|
- m isQuick ifFalse:
- [m encoderClass supportsFullBlocks ifFalse:
- [^#()]]].
^#(testCoverageOfTestBasicBlockLabelsForMethod)!

Squeak-dev mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Squeak-dev mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Squeak-dev mailing list -- [email protected]
To unsubscribe send an email to [email protected]
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.