add timezone support

[knicola]

Issue https://github.com/breejs/bree/issues/72

Changelog

Added:

• Timezone support for scheduled job runs. Supports cron, interval and timeout schedules.
• A timezone option to instance options. Default value is local
• A timezone option to job options.

Notes

This PR does not support timezone conversion for job.date yet because bree uses safe-timers for that instead of @breejs/later, where most of the tz work was done. It's fairly easy to support job.date as well, but I'm wondering whether it's necessary since Date covers that.

[knicola]

@niftylettuce @shadowgate15 thoughts?

[shadowgate15]

Okay so got a few things:

1. Anywhere you use FakeTimers, make sure to add .serial to the test call or it will mess up other tests and become flaky.
2. Any test that has a chance of not calling an assertion add t.plan(n) at the beginning of the test. (That would be anywhere you are adding it to the setTimeout function.) This just makes sure that no matter what happens if the expected assertions dont get called tests will fail.
3. When you are using FakeTimers you need to overwrite the clock setTimeout() not global. FakeTimers is already removing global so it won't work. See here..

[shadowgate15]

This is looking really good so far!!!!

[knicola]

The CI run for v10 and v12, might be failing because of const next = (() => { ... })();. I'll test my theory later today with those 2 versions and if that's not the case I'll try adding a delay.

But the run for node v14 seems to be failing because of poor coverage. This might be a good time to talk about the copied later.setTimeout() and later.setInterval() code. Write/copy over tests from @breejs/later to bree, exclude from coverage or should we talk about migrating timezone changes back to later?

[knicola]

yup, so date.toLocaleString() behaves differently in earlier versions of node. might have to find a different way to get the target zone's offset to support v10 and v12. any objections on using a 3rd party lib instead? it'll be the first later dep, if it's decided to move tz changes over.

[codecov-commenter]

Codecov Report

Merging #126 (cf43928) into master (ad91421) will not change coverage. The diff coverage is 100.00%.

:exclamation: Current head cf43928 differs from pull request most recent head a629519. Consider uploading reports for the commit a629519 to get more accurate results

@@            Coverage Diff            @@
##            master      #126   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files            4         5    +1     
  Lines          397       455   +58     
=========================================
+ Hits           397       455   +58     

Impacted Files	Coverage Δ
src/index.js	100.00% <100.00%> (ø)
src/job-builder.js	100.00% <100.00%> (ø)
src/job-validator.js	100.00% <100.00%> (ø)
src/later-tz-patch.js	100.00% <100.00%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update ad91421...a629519. Read the comment docs.

[shadowgate15]

sorry for the delay @knicola I would say we should go ahead and move to @breejs/later. I think it is a better long term solution.

[niftylettuce]

@shadowgate15 should we close this?

[Asheboy]

For anyone checking on this, it doesn't look like an equivalent PR has been opened against breejs/later yet.

[knicola]

my bad. I haven't been able to successfully migrate the tests over to breejs/later, then life happened. I'll take another shot at it this week

[knicola]

@shadowgate15 tz patch needs to be replaced with @breejs/later 4.1.0

[shadowgate15]

yep currently working on that

[knicola]

cool, thanks!