Early Bird Registration for DrupalCon Portland 2024 is open! Register by 23:59 PST on 31 March 2024, to get $100 off your ticket.
One test to rule them all,
One test to find them,
one test to bring them all and in the darkness of the testbot bind them.
I'm a bit feared that the global time continuum breaks if this code is executed.
Someone else should decide whether this code could/should be executed.
Comment | File | Size | Author |
---|---|---|---|
#47 | views-1777194-47.patch | 6.63 KB | dawehner |
#47 | interdiff.txt | 739 bytes | dawehner |
#44 | views-1777194-44.patch | 6.31 KB | dawehner |
#43 | views-1777194-43.patch | 9.54 KB | dawehner |
#40 | views-1777194-40.patch | 7.33 KB | dawehner |
Comments
Comment #1
xjmhttp://www.wired.com/images_blogs/wiredscience/2011/12/gollum.jpg
Comment #3
aspilicious CreditAttribution: aspilicious commentedSetting to needs review so you can see it explode
Comment #4
aspilicious CreditAttribution: aspilicious commentedWhat I found in the original patch. I changed some of that in my.
should be $base_table =>$info (or something like that)
This is lacking the $base_table. AND default is not always a valid plugin ID is it?
Comment #5
dawehnerYou are so right, let's try this one.
Comment #6
dawehner.
Comment #7
aspilicious CreditAttribution: aspilicious commentedIs still going to fail... It needs 4 items
Comment #8
dawehnerThis one just produces an usual exception ;)
Comment #10
dawehnerSpin off: #1777512: UidRevision filter handler uses undefined variable
Comment #11
dawehnerAs this got in let's retest it.
Comment #12
aspilicious CreditAttribution: aspilicious commented#8: one_test_rule_them_all-8.patch queued for re-testing.
Comment #14
aspilicious CreditAttribution: aspilicious commentedOffcourse this fails ;)
viewsObjectTypes...
Comment #15
dawehnerHe!
Comment #17
aspilicious CreditAttribution: aspilicious commentedI did some serious debugging and I think this test catches a couple of bugs.
I used the "debug" script on the bottom.
Everything goes wrong in $view->execute.
Something is wrong with the filter plugins.
In my test I changed:
if (isset($field_info[$type]['id'])) {
to
if (isset($field_info[$type]['id']) && $type != 'filter') {
That makes the system happy but some other errors appear (with base table: users)
1) 'Missing handler: users langcode argument'
2) Undefined offset: 1 Notice GroupwiseMax.php 210 Drupal\views\Plugin\views\relationship\GroupwiseMax->left_query()
3) 'Missing handler: sort'
It's posisble they are all related
Comment #18
aspilicious CreditAttribution: aspilicious commentedOk one bug found:
The node_language argument plugin doesn't exist.
That line is added in: http://drupalcode.org/project/views.git/commitdiff/5203f9c5ddf2e5703a97e...
But I'm not sure I understand why it has to be there if there is no actual plugin handling it...
Comment #19
aspilicious CreditAttribution: aspilicious commented2) and 3) are related to the same bug. That bug is caused by this piece of code in users.views.inc
In "function left_query($options) {" the printed $options array equals
subquery sort is NULL which is bad as we do:
I hope this enough info :)
Comment #20
dawehner#15: one_test_rule_them_all-8.patch queued for re-testing.
Comment #22
dawehnerHacked around the fact that INOperator filters require some way of valid input, even this maybe be a real bug.
If there is no input there should be nothing adding to the query.
Comment #24
dawehnerI love this test!!!
#1783006: Fix some bugs in system.views.inc/file.views.inc
Comment #26
dawehnerSome more found bugs:
Comment #27
dawehner#24: views-1777194-23.patch queued for re-testing.
Comment #29
dawehner#24: views-1777194-23.patch queued for re-testing.
Comment #31
dawehner#24: views-1777194-23.patch queued for re-testing.
Comment #32
dawehnerRerole the patch without the debug message.
Comment #33
aspilicious CreditAttribution: aspilicious commentedFor me this is rtbc but dawehner wants input from the others. As this is a duplicate of #1754206: Add one test that enables all modules and adds all filters Tim must be into this as well. I ran these tests locally and they go lightning fast so I don't see any performance issues on this one.
I'll rtbc this in 7 days if no one else did it.
Comment #34
aspilicious CreditAttribution: aspilicious commented#32: views-1777194-32.patch queued for re-testing.
Comment #36
dawehnerRereole even something else is broken.
Comment #38
dawehnerLet's try this one.
Comment #40
dawehnerWe really should have got this one in earlier :( In general this will still not run
Comment #42
Lars Toomre CreditAttribution: Lars Toomre commentedSmall nit from reading proposed patch...
Simple question... Does core documentation allow for in-line comments at the end of a line? I thought that the comment was required to go on the line above, ideally be a sentence and end in a period.
Also, my understanding is that 'eg' should be written as 'e.g.'.
Comment #43
dawehnerWe should get this in, to find more bugs, before they get committed.
Comment #44
dawehnerLet's remove the fixes for the groupwise handler and fix it there: #1799040: Fixed groupwise max relationship
Comment #45
aspilicious CreditAttribution: aspilicious commentedLet's do this
Comment #46
damiankloip CreditAttribution: damiankloip commentedThis looks good, a nice partner to the PluginInstanceTests!
Comment #47
dawehnerJust had an idea to check for the class ... but this adds 360 additional assertions.
Comment #48
damiankloip CreditAttribution: damiankloip commentedEven so, I think this is still a good idea. Tests should maybe have some assertions?! :)
Comment #49
tim.plunkett#47: views-1777194-47.patch queued for re-testing.
Comment #50
tim.plunkettCommitted!
http://drupalcode.org/project/views.git/commit/7152053