[ https://issues.apache.org/jira/browse/CASSANDRA-14975?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16745621#comment-16745621 ]
Ariel Weisberg commented on CASSANDRA-14975: -------------------------------------------- bq. It looks like the cassandra branch is testing against your cassandra-dtest cqlsh_tests branch but you linked to your cassandra-dtest 14975 branch. Did you mean to do that? Accidental, I updated it to point to the squashed branch. bq. From what I can tell assertCsvResultEqual is never called with ignore_order False? Can we just remove that option and just always ignore the order? Oops. The default wasn't supposed to be True. I'll need to try it out and see what's going on. bq. I'm slightly confused why the ordering of some of these return values has changed, for example here? Did the COPY operator start returning csv rows in non lexicographic order or some such? So... the answer is I'm not 100% sure where the order is coming from now and originally. In all the cases I looked at the database is returning the results in token order which means it's not guaranteed. There are some changes in Python 3 where dictionaries iterate in insertion order rather then hash order. This has caused a lot of tests to fail and adding order insensitive comparison methods is what we have been doing throughout the dtests in cases where it's known that there isn't a guaranteed order coming from the DB. bq. Looks like you need to tag Ryan McGuire to review? Hah. Yes. [~enigmacurry] wrote that test and I am not sure whether some of the changes I made WRT to going from binary strings to unicode strings break what the test is testing. I am not sure he is working on C* anymore. bq. Did you mean to add a dependency on cql in requirements.txt? From what I can tell it isn't used in cassandra-dtest (it it used in cassandra's pylib or something)? I'm not adding a dependency on cql I'm removing a dependency on cqlsh. We are supposed to use the cqlsh from the Cassandra repo and the one in the venv is getting in the way. Do we need what cql provides? I have no idea. It was what we had before Michael changed it to cqlsh (which also includes cql). Might as well remove it though because you are right I don't see it getting used. bq. It looks like we dep six, but from what I can tell we don't ever use it. In that case we could drop importing it in formatting.py. I'm assuming that we don't actually have six which is why we re-implement e.g. six.StringIO here Where do you see dependencies declared? You mean we import it? bq. You probably want a from __future__ import print_function in wcwidth.py and util.py, I think those print statements are working because of implied continuations... Apparently it's this weirdness https://stackoverflow.com/questions/6182964/why-is-parenthesis-in-print-voluntary-in-python-2-7 bq. You probably want a from __future__ import print_function in wcwidth.py and util.py, I think those print statements are working because of implied continuations ... I'll try that. > cqlsh dtests are not running in CircleCI > ---------------------------------------- > > Key: CASSANDRA-14975 > URL: https://issues.apache.org/jira/browse/CASSANDRA-14975 > Project: Cassandra > Issue Type: Test > Components: Test/dtest > Reporter: Ariel Weisberg > Assignee: Ariel Weisberg > Priority: Major > Fix For: 3.0.x, 3.11.x, 4.0.x > > > Right now the dtests skip these tests because most don't pass. Originally > they weren't being collected because the test files end in_tests.py instead > of _test.py, but I fixed that by adding _tests.py to the list of recognized > patterns. > Get them all running in CircleCI. Nothing special they will be part of the > existing dtest jobs. -- This message was sent by Atlassian JIRA (v7.6.3#76005) --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org For additional commands, e-mail: commits-h...@cassandra.apache.org