| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
| |
This library no longer supports Python 2, thus usage of six can be
removed.
Change-Id: I8d0c1cfd6dff375b3b7756a5b36c95a2df3f04c6
|
|
|
|
|
|
|
|
|
|
| |
most of our client major function require stack_id. We should
provide more specific information (at lest in function description)
to clearify that it's allowed to feed in stack name to `stack_id`
too. For example, both `heat.stacks.get($Stack_UUID)` and
`heat.stacks.get($Stack_Name)` works.
Change-Id: I4428097140e0391a77679f5c69e00966f249efd7
|
|
|
|
|
|
|
|
|
|
| |
When calling stacks.get, mark the returning stacks as loaded, so that
attribute retrieval works right away. If not, any getattr on the Stack
will call get() which will request the stack again (and ignore things
like resolve_outputs).
Change-Id: I3a3f25aa0c6f88e7e4274107c0578fd30dc9f155
Related-Bug: #1774958
|
|
|
|
|
|
|
|
|
|
|
|
| |
1.As mentioned in [1], we should avoid using six.iteritems to achieve
iterators. We can use dict.items instead, as it will return iterators
in PY3 as well. And dict.items/keys will more readable.
2.In py2, the performance about list should be negligible, see the
link [2].
[1] https://wiki.openstack.org/wiki/Python3
[2] http://lists.openstack.org/pipermail/openstack-dev/2015-June/066391.html
Change-Id: I3913e8876b34e112140788d7a405da6eedfb5f29
|
|
|
|
|
|
|
|
| |
Heat supports exporting data of stack preparing
for abandon stack. This change supports
'openstack stack export' in client.
Change-Id: I5d3778f307d10174dbb738548282f7495bd9c825
|
|
|
|
|
|
|
|
|
|
|
|
| |
Move the required module to heatclient/common.
- heatclient/openstack/common/apiclient/base.py
It also moves some required functions and exceptions from
heatclient/openstack/common/cliutils.py and
heatclient/openstack/common/exceptions.py.
Change-Id: I68704c7fab9417492d8390ad05a9797f78d46907
|
|
|
|
|
|
|
|
|
| |
Graduate from Oslo Incubator to oslo.i18n library.
Cleanup of unused Oslo Incubator utils.
Added optional enable_lazy() usage.
Change-Id: I5ec02477f3f52f4d7097669120581e82cced3748
Signed-off-by: Chuck Short <chuck.short@canonical.com>
|
|
|
|
|
|
|
| |
According to the rule in http://docs.openstack.org/developer/hacking/#imports
I modify some irregular import format in heatclinet.
Change-Id: I0200ddad88ee833d7de9d6df2c1519e4f4560db7
|
|
|
|
|
|
|
|
| |
This shows the files map for a given stack
Depends-On: I3198b6a7dc06648af24c198d39470f3b0d5d6f11
Change-Id: Ib7d033a660fe294bf3f7b42e4aa7020149ce24c1
blueprint: files-show
|
|
|
|
|
|
|
|
|
| |
This allows redirects to use fully qualified location
from response header and removes the check to prohibit
redirects to different project(required for super admin
actions).
Change-Id: I46876189e96375922c4f0bf30452b7c4980fe39c
|
|
|
|
|
|
| |
Change-Id: I84aa5ed9dc2b8aee089de3b1e1810552b8f2dd96
Depends-On: I7e3577dfc854018245d79afdfee45a9d250d73a7
Implements: blueprint environment-show
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| | |
template-validate --ignore-errors <ERR1,ERR2>
implements blueprint heat-template-validate-improvements
Change-Id: Ice4f623bc06366ddd600d5cb72d236e5a4459e19
|
|/
|
|
|
|
|
|
|
| |
Currently update preview provides no way to show the updates
for nested stack resources, so this adds a flag --show-nested,
which is similar to other interfaces e.g template-validate.
Closes-Bug: #1521971
Change-Id: I06f3b52d5d48dd5e6e266321e58ca8e6116d6017
|
|
|
|
|
|
|
|
|
|
| |
Also add --no-resolve-outputs parameter to heat stack-show command
to let user get stack without outputs.
@APIImpact, @DOCImpact
Depends-On: I03104efc535a7bd4326cbec56c01cd887db38ab6
Change-Id: I6c715fe90e42259e744b76924673194d0a81c07e
|
|
|
|
|
|
|
|
|
| |
Use stack lookup without redirecting to stack show when we need
to get only stack id. This significantly reduces load on engine
for large stacks, because, for example, we doesn't resolve outputs
for variety of api calls.
Change-Id: Ifcd61f8b039c930a4c2d20d31b5ccb8b11864e5b
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Add output APIs support for output-show and output-list
commands. Now output-show and output-list uses output
calls instead of requesting stack get.
implements bp api-call-output
Change-Id: I1bc1cee1c78ecf0c9a18ecc0a871d38e4141e0f7
Depends on: Ia24b24f59e2c592801e4e670ba5510f642ecf45c
|
|/
|
|
|
|
|
| |
Enabled H405 multi line docstring summary not separated with an empty
line in tox testing.
Change-Id: I89d063133154ae343315872b677fa06cc901cf3b
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Previously we just passed the "existing" kwarg directly to the PUT,
which was then ignored by the API, so PATCH update preview commands
wouldn't work. Now heat supports PATCH preview_update_stack, we
can revise this to use the PATCH method when appropriate.
Requires this heat patch for --existing --dry-run to work when used
together: https://review.openstack.org/#/c/230054
Change-Id: Ib07c0a83edfbef7a08b4af344850d0a791b7704d
Closes-Bug: #1501207
|
|
|
|
|
|
|
|
|
| |
This is the client-side change to support
Idb51c983fc07e49d371b8ccb3cea6f7e92f99903
Change-Id: I83744bd552bcb5cee183336bec97ab616bd82a91
Implements-blueprint: nested-validation
Closes-Bug: #1495695
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When performing a stack-update using the --dry-run option will change
the output behavior. Instead of printing a list of stacks, the --dry-run
option will print the name, type, and action to be taken on resources in
the updated stack.
Docs change here https://review.openstack.org/132870/
Engine change here https://review.openstack.org/#/c/122473/
Implements: partial-blueprint update-dry-run
Co-Authored-By: Jason Dunsmore <jasondunsmore@gmail.com>
Change-Id: I3dd54084ce7bd5140b28e81b6eb2349123494de9
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A TypeError raised when run cmd 'heat resource-show':
TypeError: request() got an unexpected keyword argument
'follow_redirects'
This problem was introduced by the
change I90bc9d9fcd1058c3eca288f08dd134810f1b0000 to move
SessionClient from HTTPClient to adapter, due there is no
keyword argument 'follow_redirects' in keystoneclient
session.request() method.
Change-Id: I91f13da740bf1e66dffd68cdf666a24d68aa543d
Closes-Bug: #1471650
|
|
|
|
|
|
|
|
| |
HTTPClient has methods get/post/put and others, which
use json_request and raw_request, but other code uses
requests method instead of them.
Change-Id: I74ee34430afdae17f83ccdc877cc882f94651bce
|
|
|
|
|
|
|
|
|
| |
This adds new methods to manage snapshot restoration.
blueprint stack-snapshot
Co-Authored-By: ala.rezmerita@cloudwatt.com
Change-Id: I917c5d89f586110c0614e05e134bebde27d4db55
|
|
|
|
|
|
|
|
|
|
| |
This adds the various commands corresponding to the new snasphot
methods.
blueprint stack-snapshot
Co-Authored-By: ala.rezmerita@cloudwatt.com
Change-Id: I8803c54c5cd76700f1591c8d28650bf60a2f5167
|
|
|
|
|
|
|
|
|
|
| |
Change I30a2b223d7dd19a717ed989b419b1e6cb2ac143b
introduced 'existing' keyword to stack-udpate command that became effectively
required argument to stack.update method of the client's Python API,
breaking backward compatibility.
Change-Id: Icda3313805c29ab36773a4edc8c9f8c874b71ca3
Closes-Bug: #1371167
|
|
|
|
|
|
|
|
|
|
| |
A new argument named existing for stack-update will be
supported by the heat engine to use the parameters stored
in the Db.
Partially-implements: blueprint troubleshooting-low-level-control
Partial-Bug: 1224828
Change-Id: I30a2b223d7dd19a717ed989b419b1e6cb2ac143b
|
|\ |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Module py3kcompat was removed from oslo-incubator, we can use
six directly.
* Sync commit of removing py3kcompat from oslo
* use six replace usage of py3kcompat
Change-Id: If0f9397588b2180fe198e804cbbb5b5c8420fe76
Closes-Bug: #1280033
|
|\ \ |
|
| |/
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Currently when we lookup a stack ID (fully qualified name/id) via
stacks:lookup, we also follow the redirect to stacks:show, which
is not allowed for stack users. We don't actually need to follow
the redirect, as we already have the required ID, so instead we
return the redirect response, so we can just allow all users to
do stacks:lookup in the policy.
This is useful for clients who want to call API actions which are
allowed for stack users, e.g resource:signal without having to know
both the name and ID of the stack.
Change-Id: I6f079cbc7ac3c680f15941668b4efbe4c3c5d60d
Partial-Bug: #1291097
|
|/
|
|
|
|
| |
Co-Authored-By: Richard Lee <rblee88@gmail.com>
Change-Id: I4d7b12f6da09411d45dd6022dc11b8ece76b4932
Implements: blueprint preview-stack
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Module apiclient.base has several updates in oslo, so we need to sync
latest code to have ability to reuse new functional.
Updates in oslo which related to usage of apiclient.base in heatclient:
- I491f9b54c7074546cf738c6150666bfcc3966a78
- I1db6c12a1f798de7f7fafd0c34fb0ef523610153
Related to bp common-client-library-2
Change-Id: I509d6efb0c8eef87ba2e4aa02ebb44ba149a9860
|
|
|
|
|
|
| |
This is compatible with both Python 2 and 3.
Change-Id: If8c3a4a6fe028a3a1dc6899f0f4e65c89403d267
|
|
|
|
|
|
|
| |
Adding stack-abandon support for python-heatclient.
Partial-Bug: #1262267
Change-Id: Ifaec08558103137fe8288d00ba8d7188f01afe03
|
|
|
|
|
|
|
|
|
|
| |
Since python requests has builtins SSL verification we can get ride of
our own implementation.
Increase tests coverage of heatclient.common.http along the way.
Partial Implements: blueprint python-requests-port
Change-Id: I04a169da2334acc91e538ca02cba79d9765752b5
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Add ability to use filters on stack-list by passing a ``--filters``
options to the command line. This option can appear multiple times or
be used once with several key=value filters separated by semicolons. If
the same key appears more than once, both will be taken into account
when sending the request to the API.
e.g.:
The option:
``--filters status=COMPLETE;status=FAILED``
Will issue the call:
``/stacks?status=COMPLETE&status=FAILED``
Closes-Bug: 1268816
Change-Id: I553d6b5e3ceebdb24dcd4e4c8e409a4375d72333
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Add support for ``limit`` and ``marker`` on stack-list and remove
unnecessary ``page_size`` parameter.
When using ``limit`` and/or ``marker``, stack-list will keep previous
behavior and list all stacks starting from ``marker`` up to ``limit``,
even if limit exceeds API limit restrictions, by using multiple calls
to /stacks endpoint.
Partial-Bug: 1268816
Change-Id: Ifb08c1aadd3629d60ddc56ffc507c99b9dba80ce
|
|
|
|
|
|
|
|
|
| |
In the process of unification of the clients code we should
reuse common functionality from Oslo.
Related to blueprint common-client-library-2
Change-Id: I44a1344e3eb40220d8b795fd05e44623195116a2
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When we create or update a stack, when deferred_auth_method=password
(which is the default in heat.conf), the client must pass the username
and password, even if they don't need to for authentication (ie
they're passing a token). Otherwise there's no username and password
in the context to store, and the engine validation in
heat/engine/service.py::_validate_deferred_auth_context fails.
No such issue exists when deferred_auth_method=trusts, which is why
this problem was missed during testing of the patch being reverted.
Note some fixup in addition to the revert was required to make the
tests pass, as ShellTestToken and ShellTestStandaloneToken now require
username/password set in the fake env.
This reverts commit a7ba3c323b16227e0ba2527f21bc89625f125234
Change-Id: I0e7bc3f748022c2bf5ef7ef6dfe4d97953e3fcab
|
|
|
|
|
|
|
|
|
|
|
|
| |
We don't need to call credentials_headers in the stack create/update
code, as it's already called in the HTTPClient. This has the side-
effect of simplifying the shell test, which currently has to mock
header arguments to json_request, which are incorrect anyway because
HTTPClient always sets X-Auth-Token if a token is passed in (as
evidenced by bug #1250731)
Change-Id: I6212cf5a50f5a8d919f743e23ece4bd1d0e1ad1c
Partial-Bug: #1252248
|
|
|
|
|
|
|
|
|
| |
This allows the downstream service access to the data coming
back from the REST API, primarily the stack id.
implements blueprint return-id-on-stack-create
Change-Id: I24db9c28fca1f50f5ef1452679f444095d22d103
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In Python 3.3, hash randomization is enabled by default. It causes the
iteration order of dicts and sets to be unpredictable and differ across
Python runs.
In the test case, the fixed expecting string will not match the test
result, it is relying on the dict order.
This change transforms the input dict to a sequence of two-element list,
with fixed order, and exchange the order of "limit" and "marker" for
expecting string.
Partial implement: blueprint py33-support
Change-Id: If9c25fb0d457b31034111e7e879f1f95a576a78f
|
|
|
|
|
|
|
|
|
|
| |
Replace dict.iteritems() with six.iteritems(dict)
Import six for StringIO
Import urlutils for urlencode
Partial implement: blueprint py33-support
Change-Id: Ieae1299915168ef10ff54e0a9e9a346350e9d1fb
|
|
|
|
|
| |
Change-Id: I8f2697ca99192078d5db2adb4db98287520d9895
Fixes-Bug: #1214176
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The identifier property formats the stack name and ID for
REST API request paths.
This has the following consequences for each action method:
- get, same behaviour but using the identifier property
- delete, using the full identifier avoids one http redirect
- update, this method was non-functional before this change, and is
now fixed
Change-Id: Ib2b12826243e25772acccb8ca56039e54d3857d5
|
|
|
|
| |
Change-Id: Ia3fce29a58957b84f2eba33ac1b2ecfb04543939
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Pagination on the server has not yet been implemented, however
the client assumes it is, causing infinite recursion when paginate
is called.
With this change, no pagination is attempted unless page_size is
specified. However the 'limit' parameter is still honored, so it
is still possible to cap the total number of results returned.
heatclient users should not set page_size until pagination has been
implemented on the server. The stack list in Horizon currently
does not attempt to paginate, so is not affected.
Fixes: bug #1207839
Change-Id: I4087d3a8af48206d6ebe3edc441469464e4a401a
|
|
|
|
| |
Change-Id: I15dd4e9168cba6fd05371fed097d52cbf33cb15d
|