Repository navigation
refactor(deploy): consolidate deploy creation logic - #8605
Merged
Merged
Conversation
Contributor
|
Important Review skippedReview was skipped as selected files did not have any reviewable changes. ⚙️ Run configuration
You can disable this status message by setting the Use the checkbox below for a quick retry:
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info
📝 Summary
Merge Risk: ⚪ Minimal · up to
|
| Check name | Status | Explanation |
|---|---|---|
| Title check |  |
The title clearly and concisely describes the main change: consolidating deploy creation logic across deployment paths. |
| Description check | |
The description is directly related to the changeset and explains the deploy helper consolidation, optional-chaining changes, and reported behavior changes. |
| Docstring Coverage | |
No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff. Analyzed 0 functions across 1… |
| Linked Issues check | |
Check skipped because no linked issues were found for this pull request. |
| Out of Scope Changes check | |
Check skipped because no linked issues were found for this pull request. |
✨ Finishing Touches
-
📝 Generate docstrings
-
- Commit to this branch
- Create a new PR
🧪 Generate unit tests (beta)
-
- Commit to this branch
- Create a new PR
-
- Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts
Comment @coderabbitai help to get the list of available commands.
📊 Benchmark resultsComparing with 336e09e
|
commit: |
- Create deploys through a single helper, instead of duplicating across `--build` and `--no-build`
- Behaviour change (technically): with `--build`, a deploy response without an `id` no longer
falls back to an empty-string deploy ID; the deploy ID is passed through as returned by the API,
matching the `--no-build` path.
- Use optional chaining where we now can.
- Behaviour change (technically): a non-object `published_deploy` (for example a
function) with a truthy `locked` property now counts as locked.
- Remove optional chaining now revealed to be unnecessary.
- Behaviour change (technically) if the types are wrong (they aren't).
serhalp
force-pushed
the
serhalp/deploy-types-refactors-batch1
branch
from
October 9, 2026 17:28
d5abf07 to
a5c7b12
Compare
serhalp
marked this pull request as ready for review
October 9, 2026 17:41
|
You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard. |
seancdavis
approved these changes
Oct 14, 2026
serhalp
enabled auto-merge (squash)
October 14, 2026 14:09
serhalp
added a commit
that referenced
this pull request
Oct 16, 2026
Batch 2 of 3 of follow-ups to #8599 (batch 1 here: #8605). Out of an abundance of caution, that PR was kept purely focused on type changes. - Drop runtime checks for options the types now require - Remove the unused `assetType` option from file hashing - Only upload in-memory bodies for static files - Every upload built its body with `body ?? fs.createReadStream(filepath)`, although only the generated `netlify.toml` (an `InlineUploadFile`) ever carries an in-memory body. Static files still upload their inline `body` when present and stream `filepath` otherwise; functions, edge functions and servers now always stream from `filepath`. - Build the functions config with `Object.fromEntries` - The per-function config payload sent to the API was accumulated with a spreading `reduce`. It is now built by a named `toFunctionConfigPayload` mapper and `Object.fromEntries`, which also gives the payload a derived type.
serhalp
added a commit
that referenced
this pull request
Oct 16, 2026
Batch 2 of 3 of follow-ups to #8599 (batch 1 here: #8605). Out of an abundance of caution, that PR was kept purely focused on type changes. - Drop runtime checks for options the types now require - Remove the unused `assetType` option from file hashing - Only upload in-memory bodies for static files - Every upload built its body with `body ?? fs.createReadStream(filepath)`, although only the generated `netlify.toml` (an `InlineUploadFile`) ever carries an in-memory body. Static files still upload their inline `body` when present and stream `filepath` otherwise; functions, edge functions and servers now always stream from `filepath`. - Build the functions config with `Object.fromEntries` - The per-function config payload sent to the API was accumulated with a spreading `reduce`. It is now built by a named `toFunctionConfigPayload` mapper and `Object.fromEntries`, which also gives the payload a derived type.
serhalp
added a commit
that referenced
this pull request
Oct 16, 2026
Batch 3 of 3 of follow-ups to #8599 ([batch 1](#8605), [batch 2](#8626)). Out of an abundance of caution, that PR was kept purely focused on type changes. - Share one poller between `waitForDiff` and `waitForDeploy` instead of duplicating code - Resolve the drop deploy poll with the ready deploy - `waitForDropDeploy` captured the ready deploy in an outer variable and returned it with a non-null assertion after polling. The poll now resolves with it directly via `pWaitFor.resolveWith`. - Rethrow the final drop upload error from inside the retry loop - The drop file upload retry loop remembered the last error, skipped the backoff on the final attempt, and threw the remembered error after the loop. It now throws directly on a permanent failure or on the last attempt, and otherwise backs off and retries. The status is read through an `instanceof Error` guard instead of a cast. - Don't return the upload promise to the backoff emitter. It wasn't used, resulting in lint errors about promise misuse.
serhalp
added a commit
that referenced
this pull request
Oct 16, 2026
Batch 2 of 3 of follow-ups to #8599 (batch 1 here: #8605). Out of an abundance of caution, that PR was kept purely focused on type changes. - Drop runtime checks for options the types now require - Remove the unused `assetType` option from file hashing - Only upload in-memory bodies for static files - Every upload built its body with `body ?? fs.createReadStream(filepath)`, although only the generated `netlify.toml` (an `InlineUploadFile`) ever carries an in-memory body. Static files still upload their inline `body` when present and stream `filepath` otherwise; functions, edge functions and servers now always stream from `filepath`. - Build the functions config with `Object.fromEntries` - The per-function config payload sent to the API was accumulated with a spreading `reduce`. It is now built by a named `toFunctionConfigPayload` mapper and `Object.fromEntries`, which also gives the payload a derived type.
serhalp
added a commit
that referenced
this pull request
Oct 16, 2026
Batch 3 of 3 of follow-ups to #8599 ([batch 1](#8605), [batch 2](#8626)). Out of an abundance of caution, that PR was kept purely focused on type changes. - Share one poller between `waitForDiff` and `waitForDeploy` instead of duplicating code - Resolve the drop deploy poll with the ready deploy - `waitForDropDeploy` captured the ready deploy in an outer variable and returned it with a non-null assertion after polling. The poll now resolves with it directly via `pWaitFor.resolveWith`. - Rethrow the final drop upload error from inside the retry loop - The drop file upload retry loop remembered the last error, skipped the backoff on the final attempt, and threw the remembered error after the loop. It now throws directly on a permanent failure or on the last attempt, and otherwise backs off and retries. The status is read through an `instanceof Error` guard instead of a cast. - Don't return the upload promise to the backoff emitter. It wasn't used, resulting in lint errors about promise misuse.
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
Summary
Batch 1 of 3 of follow-ups to #8599. Out of an abundance of caution, that PR was kept purely focused on type changes.
--buildand--no-build--build, a deploy response without anidno longer falls back to an empty-string deploy ID; the deploy ID is passed through as returned by the API, matching the--no-buildpath. ✅ Verified not possible in backend code.published_deploy(for example a function) with a truthylockedproperty now counts as locked. ✅ Verified not possible in backend code.