Closed benzvan closed 11 months ago
ping @Dridi , @hermunn
I would probably have gone with something simpler, replacing the offending word with another word:
method .blacklist --> .mark method .blacklist_count --> .mark_count
No opinion on the matter besides being happy to see $Alias
in action.
@Dridi I'm open to something simpler but also wanted something descriptive. I could go either way or use sick
in place of mark
. The complication for me is one is a action taken on an object (verb) and the other is a value (parameter).
Hmm...how about
method .blacklist --> .mark_for(duration) method .blacklist_count --> .count_marked() or .mark_count()
I like
method .blacklist --> .mark_for(duration)
method .blacklist_count --> .count_marked()
Let's see if anyone complains, if not, we can merge next week.
In the meantime, could you add a vtc
to test the new functions? You can just duplicate an existing one (so that the initial API is still tested), and scrub the old wording from all the original files.
actually, I just checked and Varnish Enterprise already looked at that, and are using denylist()
and denylist_count()
, we should align with that instead
I think they made the wrong decision :-D but I'm in favor of interoperability.
(PR updated as s/black/deny/g
and a new test added to duplicate the basic testing of test 01 but with the aliases)
Going for compatibility was the obvious choice, LGTM.
Removes archaic references to
blacklist
and replaces with more descriptive terms. Includes aliases, examples, and log messages.method .blacklist --> .denylist method .blacklist_count --> .denylist_count
TODO: saintmode.c still uses old code and terms.