Thanks for contributing! Do you know if this an official part of the API? I ask as we've had that happen before; see here:

https://blog.johnnyreilly.com/2017/02/under-duck-afternoon-in-open-source.html

Could you check please? @sokra may know..

Does fileTimestamps produce the same value as getTimes? Or does it contain more / less? To what extent? I'm considering the merits of having compilerOptions in there as a fallback

Does fileTimestamps produce the same value as getTimes? Or does it contain more / less?

It's atm the same, but it depends on the implementation of the watcher. The results of watcher.getTimes() are stored in fileTimestamps. See here and here.

In case of the WatchIgnorePlugin it contains more data, as WatchIgnorePlugin enhances the timestamps with all ignored files.

Performance isn't really a topic.

• without WatchIgnorePlugin it should be faster but this in the lower ms range.
• with WatchIgnorePlugin I would expect that it's a bit slower, cause fileTimestamps is bigger but maybe the effect of not collecting the data compensates this..

Thanks @zinserjan - that's really helpful. I've been asking in the webpack slack channel to see if there's any reason not to go with this. No-one has advised that's the case so far. If no-one has given a compelling reason why we shouldn't take this in the next 24 hours then I reckon we're good to merge. Thanks for the contribution by the way, we really appreciate it! 🌻

@sokra (the creator of webpack) has just confirmed this should be safe so I plan to merge. Thanks for the PR!

@johnnyreilly awesome!

Released with v2.3.1 - thanks again! https://github.com/TypeStrong/ts-loader/releases/tag/v2.3.1