summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* version bump to 11.4.411.4.411.4-patchdanielsdeleo2013-04-2349-97/+97
|
* forcibly turn diffing back on after diff tests11.4.4.rc.0backport-diff-disabledanielsdeleo2013-04-231-2/+1
|
* avoid clobbering original diff disable state when nesteddanielsdeleo2013-04-231-1/+1
|
* Disable diffs during file functional testsdanielsdeleo2013-04-235-0/+27
| | | | | No functional tests actually use the diff behavior and this speeds them up quite a bit.
* Merge branch 'CHEF-4117'danielsdeleo2013-04-233-1/+9
|
* bump version to 11.4.2 for patch release11.4.2danielsdeleo2013-04-2249-130/+130
|
* remove accidental :focus in testsdanielsdeleo2013-04-221-1/+1
|
* Merge branch 'CHEF-3367'11.4.2.rc.0Bryan McLellan2013-04-222-4/+50
|
* Merge branch 'CHEF-3432'danielsdeleo2013-04-224-30/+70
|
* Release 11.4.011.4.0Bryan McLellan2013-02-121-1/+1
|
* CHEF-1292 Lower json stack depth tests to 301Bryan McLellan2013-02-123-9/+11
| | | | | 1001 was causing stack level too deep errors on 1.8.7. This is still a huge stack depth and 1001 was arbitrary.
* Merge branch 'more-runlock-fixes' into 11-stableBryan McLellan2013-02-121-5/+23
|\
| * remove setup message, it's a race either waydanielsdeleo2013-02-121-2/+0
| |
| * fix race in test setupdanielsdeleo2013-02-121-1/+3
| | | | | | | | | | | | the heart of the test is confirming that the last 2 messages are p1 releasing lock and then p2 acquiring it. The two messages confirm that the test case is set up correctly.
| * set sync=true on IPC in run lock testsdanielsdeleo2013-02-121-3/+16
| |
| * ensure writes are flushed in run lock IPCdanielsdeleo2013-02-121-0/+5
|/
* 11.4.0.rc.0Bryan McLellan2013-02-121-1/+1
|
* CHEF-3863: Return to not trusting the JSON gem at allBryan McLellan2013-02-121-1/+3
| | | | | | Within days of relaxing to a pessmistic constraint on patch release there was a patch release with breaking changes. We'll have to go back to testing the latest version by hand again.
* [CHEF-3863] improve error compat with old impl.danielsdeleo2013-02-121-3/+17
|
* rdoc json compat methodsdanielsdeleo2013-02-112-0/+12
|
* switch from JSON.parse to JSONCompatdanielsdeleo2013-02-112-2/+2
|
* allow lookup of resources without symbol creationdanielsdeleo2013-02-112-2/+14
|
* add fallback deserialization to hash for Resourcesdanielsdeleo2013-02-111-0/+2
|
* mutate the incoming hashdanielsdeleo2013-02-111-3/+2
| | | | | I'd rather not, but this object should be local to the caller at this point so it seems less bad than creating a bunch more objects
* Use consts for class stringsdanielsdeleo2013-02-111-10/+22
|
* Add json_class support w/o JSON for most typesdanielsdeleo2013-02-111-1/+61
|
* Merge branch 'CHEF-3836' into 11-stable11.2.0Bryan McLellan2013-02-071-31/+145
|\
| * add additional synchronization to run lock specdanielsdeleo2013-02-071-31/+145
|/
* [CHEF-3809] Attached missing win32 CloseHandle functionKeith Shook2013-02-072-1/+9
|
* Fix up knife user public key test for windows pathsBryan McLellan2013-02-071-1/+1
|
* CHEF-3840: remove ronn from the gemspecBryan McLellan2013-02-061-1/+1
| | | | Use bundler or magic to install it
* [CHEF-3840] ci: don't install document generation gemssdelano2013-02-062-2/+2
|
* [CHEF-3840] fix rdiscount compile on windows by excluding groupdanielsdeleo2013-02-061-1/+4
| | | | | This requires an update to the scripts for the test stage of the omnibus Ci pipeline, but is probably the better long term solution
* Release 11.2.0-clientBryan McLellan2013-02-0649-126/+126
|
* 11.2.0.rc.111.2.0.rc.1Bryan McLellan2013-02-0649-127/+127
|
* development version bumpBryan McLellan2013-02-061-1/+1
|
* Merge branch 'CHEF-3707'Bryan McLellan2013-02-0616-39/+1114
|\
| * Ensure a Chef::User object is returned from #reregisterSteven Danna2013-02-061-0/+1
| |
| * Add validation of password length to the user-create plugin.Steven Danna2013-02-062-0/+15
| | | | | | | | | | | | Ideally this type of validation would be done somewhere in the model. Currently our use of the set_or_return method doesn't easily allow for these types of validations.
| * User tests are about users, avoid '@client' as a variable name.Steven Danna2013-02-061-27/+27
| |
| * Use same name validations as used by the server in user model.Steven Danna2013-02-062-1/+15
| |
| * Chef::User.list should handle responses from Hosted/Private ChefSteven Danna2013-02-062-6/+46
| | | | | | | | | | | | The API responses from a GET to the "users" endpoint currently returns a different response on Hosted and Private Chef. Chef::User.list transforms this response into form returned by the open source server.
| * Update knife-configure to create user rather than client.Steven Danna2013-02-062-39/+39
| | | | | | | | | | | | | | | | Chef 11 now has users as a first-class, authenticatable actors. knife configure -i now creates a user rather than a client.
| * Add user commands to knife.Steven Danna2013-02-0612-0/+623
| | | | | | | | | | | | | | | | | | | | | | | | | | This commit adds the following commands to knife: knife user list knife user show USERNAME knife user create USERNAME knife user delete USERNAME knife user reregister USERNAME Note that since Chef::User objects do not contain a json_class, the edit object method will not work as expecated.
| * Add Chef::User classSteven Danna2013-02-062-0/+382
|/ | | | | | | | This commit adds a Chef::User class similar to the Chef::ApiClient class. Notable differences are the presence of a password attribute and the fact that the API will not return chef_type or json_type attributes. The latter means that the Chef::REST class will not automatically inflate API responses to Chef::User objects.
* CHEF-2960 - Increase JSON constraintBryan McLellan2013-02-061-1/+1
|
* CHEF-3156 - Be more optimistic about net-sshBryan McLellan2013-02-061-1/+1
|
* bump version and regen docs11.0.0danielsdeleo2013-02-0149-97/+97
|
* bump to 11.0.0 rc 0, regen docs11.0.0.rc.0danielsdeleo2013-02-0149-121/+121
|
* version bump to 11.0.0.beta.2, regen docs11.0.0.beta.2danielsdeleo2013-01-2949-97/+97
|
View Lorry Controller Status