Iterate on next/previous commits navigation #1960
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
When we need to guess which commit line we should follow, prefer the checked out branch before choosing the newest, freshest ref. This is likely (or hopefully) the better choice.
In most views we use and show short hashes, e.g. the graph. Since we extract these hashes and pass them along, normalize and switch to short hashes (for now) for the purpose of the caches. This would be a good and cheap move *if* we only ever used actual "hashes" when we type "commit_hash". But we might use "refs" here as well under some circumstances which we can't shorten. Either a) find these places where we might introduce "refs" and resolve them or b) maintain a big mapping from short to long hashes, populated during graph and blame parsing.
To what I know, only the `gs_diff` view also gets invoked with branch names. (E.g. from the branches view, or when comparing two branches from the graph view.) From there, its `target_commit` can spread to other views, namely `gs_show_file_at_commit` and the `gs_inline_diff`. Fix that by calling `resolve_commitish` on the caller side.
The main goal is that `[n|p]` must be stable under all circumstances although their implementation differs. Introduce `get_previous_commit` which stores its navigation in the view's `next_commits`. Just like `get_next_commit` does.
For `next_commits`, when we use `git_streaming` we abort the process as soon as we find our needle. The other usages exhaust the iterator but already limit the number of commits git yields.
kaste
force-pushed
the
iter-next-commits
branch
from
62f661b to
6e886cf
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Internally we switch to short_hashes. This can be a temporary switch as we gain more information of the places where we actually uses hashes or allow also revs (e.g. branch names). After all, if we mark all places we could implement an internal map from short to long hashes (e.g. whenever we draw the graph), and switch to long hashes without a big cost. Note: we switch to short hashes here but it actually was wild, no-rules before, so we basically normalize our usages and get a sense of it.
Other than that: there is obviously no simple solution to which is the best newer or older commit of a given commit (as there might be many), so it is likely there are some decision made I don't remember right now.