This new document describes etags and how they can be used to
avoid the lost update problem. To avoid confusion it contains an
introduction that explains that it does not cover all of the many
use and corner cases for etags.
Change-Id: Iacb7cb7cddef32b3cabd371255ae8443ce5beb80
The steps in the proposal process were interleaved in a way
that made following them confusing. This change extracts the
definitions of terms from the actual process and simplifies the
process.
For the most part the spirit of the staged review process is
maintained but the steps required to reach out to the community
are clarified: Where there is consensus we involve just the CPLs.
Where there is not we reach out to the rest of community. I'm not
certain is the correct way to go. It's a compromise between keeping
things moving with minimal friction and including the broader
community as much as possible.
My hope is that this will help to clarify the term "frozen" and
when freezing is to be done.
This will probably require some iteration.
Change-Id: I416ff1dd35249674fb27f9dd7625786e52001556
Also, probably worth reviewing this guidelines to see if it
reflects the current plans and state of the art.
Change-Id: I54bcd7559b4bc35ded5256403c972592c2f27bc9
* Deal with utf-8 in human names. To avoid requiring six, just
check the python version.
* Allow the review identifier to be a commit hash instead just a
review int as there can be conflicts:
https://code.google.com/p/gerrit/issues/detail?id=1707
Change-Id: Ie89dfd7212f72355ab00c0c8dc9edf07f686488b
One problem with tag names was found during tag
implementation in Nova Ibc44228aeae94c17353af7fccfcfb2c11b2e9190
This problems was connected with URLs. We can't
use '/' in names because in this case we will get a incorrect
URL for tag 'abc/cba': /servers/<server_id>/tags/abc/cba
Even if '/' will be encoded like %2F. Nova will decide
that it is '/'.
But this is not a problem of Nova. Unfortunately some servers
works incorrectly with encoded character '/'. It's a problem of poor
implementation of some servers. To escape problems with handling URLs
character '/' should be forbidden in tag names.
Change-Id: I9832270b3d8210bff9cd2d9732b4d19f757f865a
Provide advice suggesting we not proliferate headers which include
specific meaning in the name of the header.
Note: this is a draft to get the basic idea out for discussion from
which can distill the proper guide.
Change-Id: If3fa2f85c5771b589992ac374ad26626abc9b13c
Most projects contain both their own project name and service name,
and **project name** vs. **service type** guideline is defined for
most cases. However, on current big-tent trend, some projects don't
contain them. So this patch makes the guideline clear for most cases.
Change-Id: I75cabcf0787073d96b6536c4e70b63ed73b814e3
Errors are a crucial part of the developer experience when using an API. As
developers learn the API they inevitably run into errors. The quality and
consistency of the error messages returned to them will play a large part
in how quickly they can learn the API, how they can be more effective with
the API, and how much they enjoy using the API.
We need consistency across all services for the error format returned in
the response body.
Change-Id: Iae6dc4d84c11c95a97d2451668ebffa80c9821ec
We tend to use "404 NotFound" for all nonexistent resource cases
due to the name, but the RFC says its usage and we need to use 400
for some cases. This patch adds these cases for avoiding confusions.
Change-Id: Ie6e461a133a62de1c87b4b6d963c02fa91aca63c
Based on feedback in this review [1] we want to remove the
restriction on voting on guidelines. That restriction was
a hold over from a Summit discussion when we didn't really
know how everything would work out. It's simply not necessary
at this point.
[1] https://review.openstack.org/#/c/213876/3/doc/source/process.rst
Change-Id: I842a365f3bc5d23b39039525d315194b3344ff04
This change replaces the instances of "eg", "e.g.", "ie", and "i.e." in
the http guideline. This conforms with the documenation conventions
about abbreviations.
Change-Id: I801dbb718add90bc5a11521c8aa9f34ec48b6746
Update the process for core reviewers to add exactly one
CodeReview+2 to a review to consider it frozen. This is
preferable to freezing it with a Workflow-1 or CodeReview-2.
Document the use of the add-reviewers.py script.
Change-Id: I9c48bae7b69c5bbeffd2994148b6d02d2150cce5
Previous now-merged patchset [1] allowed for making headers prefixed
only with the service name (Baremetal-FooBar) instead of
OpenStack-Baremetal-FooBar.
[1]: https://review.openstack.org/#/c/186526
Change-Id: I07628e924320c6493ed9a8ac8ae37a7a50be2008
To begin, this guideline only explains the deprecation of X- as a header
naming scheme. In the future we can add more guidance on existing
headers (content-type, etc) as appropriate.
Change-Id: Icf0640ea2259842f15049eddaf1cd684c9c41f7c
