METRON-1877: Nested IF ELSE statements can cause parse errors in Stellar #1268
Conversation
|
@cestella You're the mastermind of this, therefore you own it forever. This look reasonable to you (both in cause and approach)? |
|
This looks good to me. Good catch, +1 by inspection. |
ottobackwards
left a comment
ottobackwards
left a comment
There was a problem hiding this comment.
Couple of comments
| Token<?> t = it.next(); | ||
| if (t.getUnderlyingType() == IfExpr.class) { | ||
| innerIfCount++; | ||
| } |
There was a problem hiding this comment.
wouldn't if, else if be more clear here?
There was a problem hiding this comment.
I might have addressed this? If the current change isn't what you wanted, let me know.
| @@ -916,6 +916,53 @@ public void testShortCircuit_conditional() throws Exception { | |||
| Assert.assertEquals("foo", run("if NOT('metron' in [ 'metron', 'metronicus'] ) then THROW('expression') else 'foo'", new HashMap<>())); | |||
| } | |||
|
|
|||
There was a problem hiding this comment.
These tests are good, but these evaluations should be tested with other things in the things in the if s
IF SOME_MORE_COMPLEX_FUNCTION THEN IF LAMBDA THEN MATCH ELSE B ELSE C
To make sure we are not messing the stack up.
There was a problem hiding this comment.
if you look at #814, which was when I did checkpoint work for match, I had a lot of fallout to account for. Maybe you can take a peak at the match tests.
There was a problem hiding this comment.
Added test cases for statement following the pattern
// IF TO_UPPER('foo') == 'FOO'
// THEN IF GET_FIRST(MAP(['test_true'], x -> TO_UPPER(x))) == 'TEST_TRUE'
// THEN match{ var1 < 12 => 'less', var1 >= 10 => 'more', default => 'default'}
// ELSE 'b'
// ELSE 'c'
Does this meet what you're looking for? I'm not sure how deep you want to go down the rabbit hole, but this seems like it's hitting a pretty good amount of stuff on the inner IF (multiple functions, list syntax, lambda, ==, along with a match in the then.
There was a problem hiding this comment.
yeah, that is the idea. I had to add more and more permutations ( I added tests for issues as @jjmeyer0 found them, or as @cestella found them )
If you have one mega case that works that is fine. I started out differently, because I had things that did not work at the beginning and I added later ( MAP support etc )
|
+1, great job |
|
Kicking Travis. The test failure is in metron-elasticsearch, and my personal Travis passed (https://travis-ci.org/justinleet/metron/builds/456098365). @nickwallen Is it possible that the tests are more stable if we get in #1247? |
I am not sure. There might be an issue with ES's Bulk Writer refresh policy (or the way that we use it) that we are relying on in those tests. I ended up not relying on it for #1247 which may help, but may not solve the entire issue if that is the root cause. That being said, this is all speculation on my part. |
|
Hm, this looks possibly related to, but not the same as what was of concern in 1247. In the client migration, there was some uncertainty in the ElasticsearchSearchIntegrationTest. But this failure appears to be in the indexing integration test. It's actually an error, not a normal failure. And it appears to happen before any tests can start to run. Can you open a ticket? We should track the intermittent failure, but I think it's unrelated to this PR. Incidentally, I caught something in the logs I never noticed before. |
|
@mmiklavc I have no idea where that comes from. The one jar is just a test jar, and the downloading is from the rat plugin (that's been fixed, but I haven't pulled locally yet). |
5 participants
Contributor Comments
When we're dealing with if-else, the existing code would simply say we need to skip an else. In the case of nested if-else, the outer else would say to skip the inner else and cause issues. By tracking the number of else's to skip, we can manage the nesting properly.
Short-circuiting also needed updating, because it was also experiencing a similar issue.
A variety of unit tests has been added to address both of these, and ensure slightly more complicated nestings are tested.
Pull Request Checklist
Thank you for submitting a contribution to Apache Metron.
Please refer to our Development Guidelines for the complete guide to follow for contributions.
Please refer also to our Build Verification Guidelines for complete smoke testing guides.
In order to streamline the review of the contribution we ask you follow these guidelines and ask you to double check the following:
For all changes:
For code changes:
Have you included steps to reproduce the behavior or problem that is being changed or addressed?
Have you included steps or a guide to how the change may be verified and tested manually?
Have you ensured that the full suite of tests and checks have been executed in the root metron folder via:
Have you written or updated unit tests and or integration tests to verify your changes?
If adding new dependencies to the code, are these dependencies licensed in a way that is compatible for inclusion under ASF 2.0?
Have you verified the basic functionality of the build by building and running locally with Vagrant full-dev environment or the equivalent?
For documentation related changes:
Have you ensured that format looks appropriate for the output in which it is rendered by building and verifying the site-book? If not then run the following commands and the verify changes via
site-book/target/site/index.html:Note:
Please ensure that once the PR is submitted, you check travis-ci for build issues and submit an update to your PR as soon as possible.
It is also recommended that travis-ci is set up for your personal repository such that your branches are built there before submitting a pull request.