You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
(format"/users/%s/gists" (or username (gh-api-get-username api)))
;; or
(format"/repos/%s/%s/pulls/%s" user repo id)
This approach is useful if the repo has already been fetched with gh-repos-repo-get, but is more difficulty to fetch from scratch. It also makes it much more difficult to explore the API when starting out, as the need to pass in the object instead of the string is undocumented. It appears that a number of the other apis just take the user repo pair, and allow a stub to update if needed (gh-issues, and gh-pulls in particular take this approach).
I'm happy to swap over gh-repos to use that approach, but wasn't sure if there was a historical reason, and what code depended on it. Is there a preferred, consistent approach?
The text was updated successfully, but these errors were encountered:
In a number of places in gh-repos the following forms are used to access the user / repo to fetch.
instead of
This approach is useful if the repo has already been fetched with
gh-repos-repo-get
, but is more difficulty to fetch from scratch. It also makes it much more difficult to explore the API when starting out, as the need to pass in the object instead of the string is undocumented. It appears that a number of the other apis just take the user repo pair, and allow a stub to update if needed (gh-issues, and gh-pulls in particular take this approach).I'm happy to swap over gh-repos to use that approach, but wasn't sure if there was a historical reason, and what code depended on it. Is there a preferred, consistent approach?
The text was updated successfully, but these errors were encountered: