Use a smarter regexp for openapi tag detection #67
Conversation
k8s-ci-robot
added
cncf-cla: yes
kazel/generator.go
Outdated
| staging = "staging/src/" | ||
| var ( | ||
| // Generator tags are specified using the format "// +k8s:name=value" | ||
| genTagRe = regexp.MustCompile(`//\s*\+k8s:([^\s=]+)?(?:=(\S+))?`) |
There was a problem hiding this comment.
does this have to be on an empty line alone? maybe prepend ^\s*?
There was a problem hiding this comment.
using the multiline flag (so ^ matches at the beginning of a line) slows this down to ~5-6 seconds to run.
adding \n\s* as a prefix makes this take about 2.3s, so a bit more than half a second slower.
adding \s*\n as a suffix doesn't seem to slow things down, though.
optimizing regular expressions is great.
|
Mostly LGTM |
ixdy
force-pushed
the
kazel-openapi-tag-regexp
branch
from
9e4475a
to
89299a8
Compare
|
/lgtm |
k8s-ci-robot
added
the
lgtm
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: ixdy, mikedanese 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 |
Automatic merge from submit-queue (batch tested with PRs 62436, 62499, 62664, 62685, 62660). If you want to cherry-pick this change to another branch, please follow the instructions <a href="https://github.com/kubernetes/community/blob/master/contributors/devel/cherry-picks.md">here</a>. Update kazel to include openapi tag detection fix **What this PR does / why we need it**: includes the fix from kubernetes/repo-infra#67 for kubernetes/repo-infra#66, which was discovered in @thockin's PR https://github.com/kubernetes/kubernetes/pull/57600/files#r178665554. **Special notes for your reviewer**: I'm not sure what's up with the large list of comment diffs in `Godeps.json`; it looks like the `git describe` strings were (unintentionally?) truncated in @brendanburns's PR kubernetes#61955. **Release note**: ```release-note NONE ``` /assign @brendanburns @thockin @mikedanese
I tried a few other options:
go/parserto scan only commentsbut both of these were significantly slower than this approach, taking 5-7s to run.
For comparison, both this implementation and the existing, buggy
Contains()implementation take around 1.5s. (Using the regex seems to be about .1-.2s slower, but that is probably allowable.)I've also this implementation a bit more generic in preparation for kazel learning how to handle additional codegen tags.
Fixes #66.
/assign @mikedanese