[
https://issues.apache.org/jira/browse/CASSANDRA-10289?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14934745#comment-14934745
]
Stefania commented on CASSANDRA-10289:
--------------------------------------
bq. I'm attaching a patch that fixes the cqlshlib tests on trunk. There is one
test that fails and that I'm certain is incorrect behavior; I'll make another
ticket for that and link it here.
I assume this is {{TestCqlshOutput.test_prompt}} since it's the only one still
failing? These tests are quite fast to run, I wonder if we should run them on
every developer branch, perhaps by adding them to the dtest suite. We tend to
test cqlsh via the dtests but some use cases would be easier to add to these
tests, e.g. completion.
bq. I'll need a reviewer to look through each test and help me differentiate
bugs from new behavior.
The 'plumbing' changes in _basecase.py_, _cassconnect.py_ and _run_cqlsh.py_
are +1.
The changes in _test_cql_parsing.py_ were required because of CASSANDRA-9232,
we had to drop the K_XXX rules because we reached the maximum python limit for
regex named groups. So these too are +1.
For the following tests instead, it looks like we have broken things and you
probably need to revert the changes in this patch and either fix cqlsh or open
a separate ticket.
In _test_cqlsh_completion.py_:
* {{test_complete_in_delete}}: I don't think keyspaces are a valid completion
after {{DELETE a [}} and after {{DELETE FROM twenty_rows_composite_table USING
TIMESTAMP 0 WHERE TOKEN(a) >=}}. From a quick analysis of _cqlhandling.py_ I
think it comes from {{<term>}}, which picks up {{<functionName>}}, which was
changed to include {{ks.}} by CASSANDRA-7556.
* {{test_complete_in_drop_keyspace}}: the {{;}} after {{DROP KEYSPACE IF}} is
not valid.
* {{create_columnfamily_table_template}}: the {{(}} after {{CREATE ... IF}}
does not look valid to me.
In _test_cqlsh_output.py_:
* {{test_numeric_output}}: it looks weird that we get {{99999.99219}} instead
of {{1e+05}} but I think it's correct because of CASSANDRA-9224, there are 5
digits to the left and so it tries to print 5 decimals, it's similar to
{{test_float_formatting}} in the dtests.
* {{test_timestamp_output}}: we currently have timezone issues, see
CASSANDRA-10313 or CASSANDRA-10397.
> Fix cqlshlib tests
> ------------------
>
> Key: CASSANDRA-10289
> URL: https://issues.apache.org/jira/browse/CASSANDRA-10289
> Project: Cassandra
> Issue Type: Bug
> Components: Tests
> Reporter: Jim Witschey
> Assignee: Jim Witschey
> Labels: cqlsh
> Fix For: 3.0.0 rc2
>
> Attachments: trunk-10289.txt
>
>
> The cqlsh tests in trunk haven't been running for a while:
> http://cassci.datastax.com/view/All_Jobs/job/trunk_cqlshlib/423/testReport/
> This looks like the driver errors that happened because of CASSANDRA-6717.
> Not sure why it's happening now; the driver installation looks normal to me
> on those jobs. [~mshuler]?
> There were also some changes to cqlsh itself that also broke the test
> harness, but I believe those are fixed here:
> https://github.com/mambocab/cassandra/tree/fix-cqlsh-tests
> Once the tests are running successfully on CassCI, I'll test my patch and
> mark as patch available.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
