by

Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
  • 01:23

    cquiroz on master

    Update sbt to 1.3.12 Merge pull request #192 from sc… (compare)

  • 01:23
    cquiroz closed #192
  • 01:09
    cquiroz closed #191
  • 00:53
    scala-steward opened #192
  • May 29 20:59
    scala-steward opened #191
  • May 26 14:58

    cquiroz on master

    Update scalafmt-core to 2.5.3 Merge pull request #190 from sc… (compare)

  • May 26 14:58
    cquiroz closed #190
  • May 26 02:31
    scala-steward opened #190
  • May 16 02:02

    cquiroz on master

    Update sbt-scalajs, scalajs-com… Merge pull request #189 from sc… (compare)

  • May 16 02:02
    cquiroz closed #189
  • May 15 22:17
    scala-steward opened #189
  • May 14 15:23

    cquiroz on master

    Update sbt-scalafmt to 2.4.0 Merge pull request #188 from sc… (compare)

  • May 14 15:23
    cquiroz closed #188
  • May 14 14:38
    scala-steward opened #188
  • May 11 15:55

    cquiroz on master

    Update scalafmt Signed-off-by:… (compare)

  • May 11 15:48

    cquiroz on master

    Update scalatest to 3.1.2 Merge pull request #186 from sc… (compare)

  • May 11 15:48
    cquiroz closed #186
  • May 11 15:48
    cquiroz closed #184
  • May 11 15:48
    cquiroz closed #185
  • May 11 15:48

    cquiroz on master

    Update sbt-microsites to 1.2.1 Merge pull request #187 from sc… (compare)

Logan Nelson
@a1kemist
@cquiroz could you publish another incremental RC that includes scala 2.13.1 and scalajs 0.6.31 support? Alternatively, if 2.0.0 GA is coming soon I could probably just wait for that. :)
Carlos Quiroz
@cquiroz
@a1kemist why do you need that? I’m just curious since there are no important code changes to publish
To avoid eviction warnings?
Logan Nelson
@a1kemist
I’m in the process of updating most of my projects to 2.13 and saw that there are several regression fixes in 2.13.1 vs 2.13.0. aside from the eviction warnings (which I try to minimize when possible), that was my main concern.
nafg
@nafg
@a1kemist since 2.13.0 classfiles are binary compatible with 2.13.1, a library doesn't need to be republished. If the library was published for 2.13.0 and you're using 2.13.1 it's fine.
Alex Henning Johannessen
@ahjohannessen
@cquiroz How is it going moving to ScalaJS 1.0 ? :)
Carlos Quiroz
@cquiroz
Unfortunately I discovered a bug
I may need to wait for 1.0.1
Carlos Quiroz
@cquiroz
I’ll publish as soon as I can
Alex Henning Johannessen
@ahjohannessen
@cquiroz Ok, thanks for the heads up
Alex Henning Johannessen
@ahjohannessen

@cquiroz with regards to

I may need to wait for 1.0.1

Have you gotten any wiser? Is there some workaround or similar that can be done?

Carlos Quiroz
@cquiroz
The issue is being tracked at scala-js/scala-js#3950
I tried a few workarounds but it seems that it simply can no handle nested objects which are need to in the locales api
Alex Henning Johannessen
@ahjohannessen
Can you solve it by making breaking changes? Or is it out of your influence?
Alex Henning Johannessen
@ahjohannessen
Carlos Quiroz
@cquiroz
Yeah saw that
I could solve it by not following exactly the locales API
but that seems like a bad idea
Alex Henning Johannessen
@ahjohannessen

@cquiroz Could you give an example wrt:

I could solve it by not following exactly the locales API

Carlos Quiroz
@cquiroz
In particular there is this nested static class Locale.Category that is breaking on scala.js 1.0
I could just make Category top level and that would solve it
but then the API would not match the original java API
Which I guess is mostly a non issue
But still
Carlos Quiroz
@cquiroz
I published 2.0.0-RC4 with support for scala.js 1.0 and sbt-locales
If nothing major is found this will become 2.0.0 next week
Carlos Quiroz
@cquiroz
RC4 ended up incorrectly bulit. I’ll get RC5 asap
Alex Henning Johannessen
@ahjohannessen
@cquiroz :+1:
Carlos Quiroz
@cquiroz
RC5 is on its way to maven central
Carlos Quiroz
@cquiroz
I’m removing the custom serialization format on
I doubt this will affect many people but just in case feel free to comment
Andriy Plokhotnyuk
@plokhotnyuk
Hi, All! Thanks to contributors of the scala-java-time library!
Last week jsoniter-scala got Scala.js support only because your great product exists!
Andriy Plokhotnyuk
@plokhotnyuk
Would you be interested in porting some efficient routines for parsing and serialization of java.time._ values from/to the string representation?
Carlos Quiroz
@cquiroz
Thanks @plokhotnyuk
I’m not sure what do you mean? There are as part of the API methods to read/write strings
Andriy Plokhotnyuk
@plokhotnyuk
Yep, and jsoniter-scala has much more efficient version of parse and toString routines that outperform the original implementation from 10x to 100x times depending on data type and version of JVM.
Please, see results of benchmarks for java.time._ types here to estimate a possible speed up.
Carlos Quiroz
@cquiroz
Parsing and formatting are some of the most complicated parts of scala-java-time so any improvement would be great
Note that parsing and formatting are greatly influenced by the locale
Also I’m very interested on the js size as the library is already quite good, I’d prefer a smaller code size over faster speed
But as I said any improvements would be creatly appreciated
Andriy Plokhotnyuk
@plokhotnyuk
How do you measure 'js size' of the library? I mean which exact command-line or sbt command can print the size in bytes to make it checked in the CI build... Is the timezone DB included in it?
Carlos Quiroz
@cquiroz
it is not really sience :D I have a demo project that makes a few calls into the API
and check the size as fullOptJS
it is not accurate as it depends on what you actually call and the size of your db but it gives you a notion
Andriy Plokhotnyuk
@plokhotnyuk
So, the idea is to use as much java.time._ classes as possible to make them picked by the linker to the result *-fastopt.js and *-opt.js files, isn't it?
Carlos Quiroz
@cquiroz
yes
Andriy Plokhotnyuk
@plokhotnyuk
First results from the Chrome browser: