On Fri, Jan 17, 2020 at 10:36 AM Daniel Shahaf wrote:
> Julian and I discussed this on IRC today; the conclusion is:
>
> - With the behaviour of the test with seed=0 having been manually
> reviewed, we're satisfied that the test achieves adequate coverage.
>
> - The properties that the magic num
Julian Foad wrote on Thu, 16 Jan 2020 09:18 +00:00:
> I inspected the debug prints whizzing by, and noted that it included
> each of the different kinds of input cases I expected (e.g. empty
> ranges, reversed ranges, duplicate ranges, overlapping ranges, etc.) and
> that the results included ex
Daniel Shahaf wrote:
PRNG makes the test code harder to read and to maintain, makes it
more difficult to diagnose FAILs, and increases the risk of a bug
PRNG use makes for reasonably compact and readable test generation, and
totally easy enough to reproduce and diagnose failures, IMO.
On the
Nathan Hartman wrote:
On Wed, Jan 8, 2020 at 9:24 AM Daniel Shahaf wrote:
> Julian Foad wrote on Wed, Jan 08, 2020 at 10:14:42 +:
> > The suggestion was that we should prefer the regression test suite to be
> > deterministic, [...]
> But unlike nearly all of the test suite, we don't actually
Julian, would you like to answer the question in the second quoted paragraph?
Daniel Shahaf wrote on Wed, Jan 08, 2020 at 14:24:39 +:
> Julian Foad wrote on Wed, Jan 08, 2020 at 10:14:42 +:
> > What?
>
> Using a PRNG makes the test code harder to read and to maintain, makes it
> more diff
On Wed, Jan 8, 2020 at 9:24 AM Daniel Shahaf wrote:
> Julian Foad wrote on Wed, Jan 08, 2020 at 10:14:42 +:
> > The suggestion was that we should prefer the regression test suite to be
> > deterministic, running the same fixed set of tests each time. Repeating by
> > default, not just repeata
Julian Foad wrote on Wed, Jan 08, 2020 at 10:14:42 +:
> Daniel Shahaf wrote:
> > Julian Foad wrote on Tue, Jan 07, 2020 at 09:47:48 +:
> > > For now, I propose to make each test use a repeatable sequence,
> > > independent
> > > of the other tests. I think that will be enough; the options
Daniel Shahaf wrote:
Julian Foad wrote on Tue, Jan 07, 2020 at 09:47:48 +:
For now, I propose to make each test use a repeatable sequence, independent
of the other tests. I think that will be enough; the options I mentioned
can be added if and when there is a demand for them.
I don't unde
Julian Foad wrote on Tue, Jan 07, 2020 at 09:47:48 +:
> For now, I propose to make each test use a repeatable sequence, independent
> of the other tests. I think that will be enough; the options I mentioned
> can be added if and when there is a demand for them.
>
> How does that sound?
I don
Nathan Hartman wrote:
Random input tests are not deterministic in the sense that one never
knows in advance which inputs will be tested.
> While that's great for finding bugs that a deterministic test won't
surface, I wonder if we should try to keep the regression test suite
deterministic (as mu
On Mon, Jan 6, 2020 at 12:05 PM Julian Foad wrote:
>
> Nathan Hartman wrote:
> > I assume that's the one added in r1872121,
> > test_mergeinfo_merge_random_non_validated_inputs().
> >
> > I got a fail on it.
>
> Thanks for the report. I made a bad assumption in the test code.
>
> http://svn.apach
Nathan Hartman wrote:
I assume that's the one added in r1872121,
test_mergeinfo_merge_random_non_validated_inputs().
I got a fail on it.
Thanks for the report. I made a bad assumption in the test code.
http://svn.apache.org/r1872388 should fix it.
- Julian
On Mon, Dec 30, 2019 at 12:28 PM Julian Foad wrote:
> I have added thorough random-input testing for svn_rangelist_merge2().
I assume that's the one added in r1872121,
test_mergeinfo_merge_random_non_validated_inputs().
I got a fail on it.
Quoth the function's docstring: "Unlike the tests with
Now I have made some progress. The relevant commits are noted in the
issue tracker.
> Ill-defined canonical form for a rangelist.
Not yet tackled. It doesn't seem to be an immediate blocker.
> Rangelist "merge" is a set-union operation.
Not re-written yet. This is one of the next steps; se
Here are the three patches resulting from my work so far:
merge-4840-backtrace-4.patch
merge-4840-print-empty-range.patch
merge-4840-test-1.patch
Together, anyone who wants to follow along can reproduce the errors.
These are not quite in a clean enough state to commit, but I expect to
commit s
15 matches
Mail list logo
