First attempt at enable noUncheckedIndexedAccess TS compiler option
#4039
Conversation
| const major = parseInt(groups.major!); | ||
| const minor = parseInt(groups.minor!); | ||
| const patch = parseInt(groups.patch!); |
There was a problem hiding this comment.
Ignoring the lint error with ! as parseInt(undefined) will parse to NaN and I've added a new isNaN check below.
| if (i >= 0) path = pathSegments[i]; | ||
| else { | ||
| if (i >= 0) { | ||
| path = pathSegments[i]!; |
There was a problem hiding this comment.
i starts at pathSegments.length and decrements down to -1, however this statement is protected with i >= 0 so the i should always be inbounds.
| for (; newLength > 1 && c.seps.includes(glob[newLength - 1]!); newLength--); | ||
| glob = glob.slice(0, newLength); |
There was a problem hiding this comment.
It looks like there might be a bug here, the semicolon at the end of the for statement means line below is only ever run once. The for loop itself does nothing.
|
Great to see this being worked on! Implementing this feature will require a bit of work across the entire codebase. We should create a tracking issue for this to break PRs up. Perhaps, this PR can focus on the sub-modules it already touches. WDYT? |
|
Never mind, there are many more issues than I first thought. Will spend some more time on this before I submit a PR. |
|
To elaborate, I think the tracking issue should list each sub-module as a to-do list and include instructions on how to work on the issue. It'd be good to get this done before the Standard Library reaches stability. |
syhol
changed the title
noUncheckedIndexedAccess TS compiler optionnoUncheckedIndexedAccess TS compiler option
Implementing as requested in the issue #937