Some functions seems to be deprecated, therefore some dsm() have been put in order to warn if called: this code has been commited and if it runs on production, it might cause WSOD because dsm() won't exist without devel module: maybe triggering a watchdog notice or such could be better?
See ctools.module lines 681 and 689.
| Comment | File | Size | Author |
|---|---|---|---|
| #5 | 1313382-dsm-remove-5.patch | 2.68 KB | andypost |
| #3 | 1313382-dsm-change2debug-3.patch | 2.53 KB | andypost |
Comments
Comment #1
Ashlar commentedThis bug report has not been active for six months. In an effort to clean-up the issue queue this item has been closed. If your modules are current and the report is still relevant please feel free to change the Status back to active.
Comment #2
andypostOnly a few places needs review
Comment #3
andypostComment #4
tim.plunkettWhy keep any of this? I'd say remove it completely.
Comment #5
andypostSuppose this commented out code is a thing of a past.
Suppose we need to ping someone from ctools team to carefully review, probably this clean-up could be deeper.
Comment #6
andypostcan we get this commited?
Comment #7
pounardCode that should not be called may throw logic error exception, to explicitely tell the code he did a bad mistake: finding errors right away forces the devs to fix them. I'm just sayin', I really don't know why you kept those functions in the first place, there's maybe a good reason.
Comment #8
merlinofchaos commentedIt was a bad idea of me to use dsm() to point out these errors, but pounard is right -- it's better for developers to get explicit errors.
I don't remember why those functions are still there at this point. Possibly worth doing a little code archeaology to see.
Comment #9
andypostso... let's change them to trigger_error()?
Comment #10
tim.plunkettThrow named exceptions.
Comment #11
pounardDepends on how is fatal the error, but if they are, exceptions are the way to go.
Comment #12
merlinofchaos commentedOkay, I went through the list. I removed the no longer needed context_cache functions. I replaced a dsm about $attributes['class'] with a correction to just make it an array. (Experience has shown that there's Always One More 'class' setting that was not updated to the array notation).
I left in the commented out dsm()s. In a way, they're a reminder of something I had intended but haven't gotten back to. Maybe someday I or someone will, and the commented code at least says something about it. Removing it guarantees nothing will ever happen there.
Committed and pushed.