bug#27841: 25.2; Patch for completion-table-with-predicate

[Ryan Thompson]

Ok, here is the revised patch. I added a test and reformatted the commit message in (hopefully) ChangeLog format. The test includes a few basic tests of "completion-table-with-predicate" including one which is fixed by this patch (noted in the test comments). 

I've dropped the second patch entirely, because I realized it was inappropriately optimizing for a rare case (i.e. PRED1 being nil).

Also, a few other related things: First, I noticed that while the test file name is minibuffer-tests.el, the lisp header and provide statement both contain "completion-tests.el" instead. Presumably this file was renamed at some point without updating its contents. 

Second, I notice that there is not much in this file, in particular nothing testing interactive completion in the minibuffer. I recently wrote a package for simulating interactive input to a command in order to implement tests for my ido-comrpleting-read+ package, and it might be useful for adding more tests here. You can view it here: https://github.com/DarwinAwardWinner/with-simulated-input and view example usage in the ido-cr+ test suite here: https://github.com/DarwinAwardWinner/ido-completing-read-plus/blob/master/tests/test-ido-completing-read%2B.el 

Let me know if there is interest in using such an approach to add automated tests for some of the core interactive Emacs functions like "completing-read".

On Fri, Aug 4, 2017 at 4:11 PM Dmitry Gutov <dgutov@yandex.ru> wrote:

On 8/5/17 2:00 AM, Ryan Thompson wrote:
> Would you also like me
> to provide an example where the old code does the wrong thing?

Adding a test case or two to test/lisp/minibuffer-tests.el would be
ideal (in the same patch that fixes it).

0001-Fix-a-bug-in-completion-table-with-predicate.patch
Description: Binary data