Files: 
CommentFileSizeAuthor
#11 system-module.2066557-11-a.patch12.71 KBchakrapani
PASSED: [[SimpleTest]]: [MySQL] 63,386 pass(es).
[ View ]
#5 drupal8.system-module.2066557-5.patch10.87 KBmparker17
PASSED: [[SimpleTest]]: [MySQL] 57,937 pass(es).
[ View ]
#5 drupal8.system-module.2066557-5-a.patch11.08 KBmparker17
PASSED: [[SimpleTest]]: [MySQL] 58,098 pass(es).
[ View ]
#5 interdiff.txt2.11 KBmparker17
#1 drupal8.system-module.2066557-1-a.patch11.06 KBmparker17
FAILED: [[SimpleTest]]: [MySQL] 58,176 pass(es), 6 fail(s), and 6 exception(s).
[ View ]
#1 drupal8.system-module.2066557-1.patch10.84 KBmparker17
FAILED: [[SimpleTest]]: [MySQL] 58,085 pass(es), 6 fail(s), and 6 exception(s).
[ View ]
#1 inter-interdiff.txt1.49 KBmparker17

Comments

StatusFileSize
new1.49 KB
new10.84 KB
FAILED: [[SimpleTest]]: [MySQL] 58,085 pass(es), 6 fail(s), and 6 exception(s).
[ View ]
new11.06 KB
FAILED: [[SimpleTest]]: [MySQL] 58,176 pass(es), 6 fail(s), and 6 exception(s).
[ View ]

Okay. After some adventures with fatal errors, I've created two patches.

Turns out the original test code made the assumption that it would only run inside it's test harness. Since my method for testing was "enable the hidden module with Drush and hit the menu paths", the result of $query->execute() would be null and the call to ->fetchCol() would subsequently fatal error.

The -1 patch is a straight conversion. The -1-a improves the test so it doesn't fatal error if the query returns NULL. If the -1-a patch passes all tests, that one should be preferred. If it fails, then just go with the -1.

inter-interdiff.txt shows the differences between two patches.

Status:Active» Needs review

Assigned:mparker17» Unassigned

Status:Needs review» Needs work

The last submitted patch, drupal8.system-module.2066557-1.patch, failed testing.

Status:Needs work» Needs review
Issue tags:+WSCCI-conversion
StatusFileSize
new2.11 KB
new11.08 KB
PASSED: [[SimpleTest]]: [MySQL] 58,098 pass(es).
[ View ]
new10.87 KB
PASSED: [[SimpleTest]]: [MySQL] 57,937 pass(es).
[ View ]

Try these.

The -5 versus -5-a is the same as above.

Since I made the same change to both -5 and -5-a, the interdiff between -1 versus -5 and -1-a versus -5-a is the same, I've included only one interdiff.

+++ b/core/modules/system/tests/modules/database_test/lib/Drupal/database_test/Controller/DatabaseTestJsonController.php
@@ -0,0 +1,149 @@
+class DatabaseTestJsonController {
...
+    $table_name = db_query_temporary('SELECT age FROM {test}', array());
...
+      'row_count' => db_select($table_name)->countQuery()->execute()->fetchField(),
...
+    $query = db_select('test', 't');
...
+    $query = db_select('test_task', 't');
...
+    $query = db_select('test_task', 't');
...
+    $query = db_select('test_task', 't');
+    $query

All this db_select calls could be replaced directly by an injected db connection, so for example extending ControllerBase would just work.

Status:Needs review» Needs work

+++ b/core/modules/system/tests/modules/database_test/lib/Drupal/database_test/Controller/DatabaseTestJsonController.php
@@ -0,0 +1,149 @@
+class DatabaseTestJsonController {
...
+    $table_name = db_query_temporary('SELECT age FROM {test}', array());
...
+      'row_count' => db_select($table_name)->countQuery()->execute()->fetchField(),
...
+    $query = db_select('test', 't');
...
+    $query = db_select('test_task', 't');
...
+    $query = db_select('test_task', 't');
...
+    $query = db_select('test_task', 't');
+    $query

All this db_select calls could be replaced directly by an injected db connection, so for example extending ControllerBase would just work.

Assigned:Unassigned» sushyl

Thanks for your work on this issue! Please see #1971384-43: [META] Convert page callbacks to controllers for an update on the routing system conversion process.

Assigned:sushyl» Unassigned

Unassigned, Couldn't spare enough time to look into it.

Issue summary:View changes
Status:Needs work» Needs review
StatusFileSize
new12.71 KB
PASSED: [[SimpleTest]]: [MySQL] 63,386 pass(es).
[ View ]

There has been some changes to the core and some of the changes from the earlier patches have been already committed(eg: removal of hook_menu).
Re-rolling the patch against latest head based on #5-a.patch.
comments from #6/#7 are not considered yet in the current patch.
The patch passed the tests when run locally.