-
Notifications
You must be signed in to change notification settings - Fork 39.9k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Move cached discovery clients to their own packages #72214
Conversation
dd36022
to
c2c0c3f
Compare
/lgtm |
@liggitt could you help approve the change? Thanks. |
this is a nicer structure, but I have mixed feelings about whether the external churn for callers is worth the change. do we have a good sense for the amount of direct usage these have and what the impact on client-go users would be? cc @kubernetes/sig-api-machinery-pr-reviews |
If we trust the google search engine, there are not many imports. Less than 20 results for the memcached one. 2 results for the disk cached one. I tried github code search as well, but it doesn't support exact match. I can send a breaking change waring to the kubernetes-dev mailing list if that helps move this forward. |
seems reasonable, would like a second ack from @kubernetes/sig-api-machinery-pr-reviews /assign @deads2k |
Gentle ping @deads2k |
c2c0c3f
to
c748262
Compare
c748262
to
24d5fb9
Compare
/retest |
24d5fb9
to
171d9ec
Compare
171d9ec
to
1f2e2e6
Compare
The structure is much nicer. Can you leave type aliases behind so as to not break people? https://golang.org/doc/go1.9#language |
I can't do that for the disk cached one, otherwise there will be cyclic imports. It only has two public importers anyway. For the mememory cached one, I didn't do the type aliase, because the package doesn't define any public type. I did add a glue package for the legacy code in the last commit. |
Hm, can you think of a variation on the directory structure that doesn't
cause cyclic imports?
…On Tue, Feb 12, 2019 at 7:20 PM Chao Xu ***@***.***> wrote:
Can you leave type aliases behind so as to not break people?
I can't do that for the disk cached one, otherwise there will be cyclic
imports.
For the mememory cached one, I didn't do the type aliase, because the
package doesn't define any public type. I did add a glue package for the
legacy code in the last commit.
—
You are receiving this because you are on a team that was mentioned.
Reply to this email directly, view it on GitHub
<#72214 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AAngluje_nFwITJoGzemRThXSd10qEn_ks5vM4SGgaJpZM4ZbF_W>
.
|
I don't feel strongly about an alias. This change lgtm /lgtm holding in case @lavalamp really wants an alias. |
@lavalamp no, the alias will introduce cycle no matter how we arrange the directory now, because of these two facts:
|
/approve I guess we break some interface every release, why stop now? :/ |
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: caesarxuchao, deads2k, lavalamp The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
/retest |
/retest Review the full test history for this PR. Silence the bot with an |
Fix #68854.
This pull tries to rearrange the discovery clients in a more logical way.
[update] a good by-product is the removal of a few vendored packages from the staging repositories.
/sig api-machinery
/kind cleanup
/assign @liggitt @mborsz