-
Notifications
You must be signed in to change notification settings - Fork 14
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Update browserify list in transpile #1323
Comments
Discovered while debugging phetsims/ratio-and-proportion#508 with @samreid. |
I'm fine if this changes to something closer to our supported platforms. No particular relevant-for-now reason we have those hard-coded. |
Great, thanks! |
This commit increases the heap size of Geometric Optic Basics Built PhET brand from 38.1MB to 58.7MB. Startup time feels a little slower too. @pixelzoom edit: ... as reported in phetsims/scenery-phet#769 (comment) |
|
According to https://browsersl.ist/
So maybe |
Seems reasonable to me! |
OK pushed. Confirmed the build size is lower. @zepumph or @jonathanolson can you please spot check? By the way, here is my search that led here:
|
Looks good to me, can we confirm that this is working on all of our supported platforms? |
Perhaps an upcoming RC test will help us verify this is OK. @KatieWoe can you please be aware of this issue for the next round of QA testing (that has a fresh RC)? |
Ok. Can you give some details on what we should look for? |
Let's go with this to unblock for now .. This matches what we say on the website. |
@samreid - You can make an issue to consider supported browsers, and potentially update to supported systems. This requires making a change in our website language to match what "defaults" has. And should be communicated to our client partners as well. |
@pixelzoom - With this change, (marla committed above), does it make sense to rebuild the dev tests for GOB? |
Last 30 days (3,872,563 Safari users):
Total ~95% of users on Safari 13+ |
@kathy-phet your call. If you'd like to cancel the current dev test, please note in phetsims/qa#847. That dev version (1.2.0-dev.3) was built when transpiler.ts was configured for "default". Because it's a dev version, there's no release branch to patch, and we'll pick up the transpiler.ts on the next dev version, or when we create the first RC. |
@pixelzoom - Looks like we don't test those browsers in a dev test anyways, so no need to redo the dev. I'm going to close this issue. @samreid @marlitas - reopen if you see a reason. |
Reopening. We had a design meeting on 11/10/2022 where some important decisions were made.
|
I documented the // This list specifies the target browsers for Babel. Its format is described at https://browsersl.ist.
// Note that this is related to System Requirements advertised on the PhET website, so should be modified with care.
// Never remove advertised platforms from this list without a broader discussion. And note that PhET will sometimes
// provide unofficial support for older platforms, so version numbers may be lower than what is advertised on the
// PhET website. For more history, see https://github.com/phetsims/chipper/issues/1323.
const browsers = [ |
@pixelzoom - I didn't think this was for Babel? I thought browserify was broader - back converting code to support use of newer js patterns. |
The format for the
transpiler.js is more specific. It passes the return babel.transform( jsInput, {
...
targets: {
browsers: browsers
} |
Also note the overview documentation of transpiler.js at line 4: * Handles transpilation of code using Babel So I'm going to close this. Reopen if you have more questions. |
Babel is a 3rd party open source library which transpiles code versions, not to be confused with phetsims/babel which contains our sim translations. |
Currently we have:
https://browsersl.ist/#q=%3E+0.5%25%2C+safari+10-11%2C+Firefox+ESR%2C+not+IE+11%2C+ios_saf+11
I recommend:
https://browsersl.ist/#q=%3E+0.25%25%2C+not+dead
It increases our coverage from 89% to 92.5% and removes some very outdated, hard coded browser versions.
We don't have to change our IE11 logic, we can still have
> 0.25%, not dead, not IE 11
or> 0.25%, not dead, IE 11
depending.@jonathanolson, can you think of any reason to keep the current browsers. For example we are hard coding safari 10 and 11, why? I'm happy to implement.
The text was updated successfully, but these errors were encountered: