New module Text.Parsing.Parser.String.Basic

[jamesdbrock]

Description of the change

New module Text.Parsing.Parser.String.Basic

New functions

• Parser.String.Basic.number
• Parser.String.Basic.intDecimal

Moved the Parser.Token parsers digit, hexDigit, octDigit, upper, space, letter, alphaNum into the new module Parser.String.Basic.

Resolves the issues #73 #115 . Actually it doesn't resolve them directly because we still have the float token parser which was failing to parse correctly. We're keeping the float token parser which is documented to conform to “the grammar rules of the Haskell report,” but who knows if that's really true or not. Instead we've added a new number parser for parsing a Number, which I think is what most people want anyway.

---

Checklist:

• Added the change to the changelog's "Unreleased" section with a link to this PR and your username
• Linked any existing issues or proposals that this pull request should close
• Updated or added relevant documentation in the README and/or documentation directory
• Added a test for the contribution (if applicable)

[jamesdbrock]

The float parser was introduced in #27

[JordanMartinez]

I thought we weren't removing these until v0.15.0 comes out?

[jamesdbrock]

Oh? Thanks @JordanMartinez good catch.

[jamesdbrock]

I can't find a reference for this, can you post a link please @JordanMartinez ?

[JordanMartinez]

See purescript/purescript-lists#176

We have a couple of MonadZero issues like that scattered throughout the core and contrib libs.

[thomashoneyman]

There’s a strict requirement that we don’t remove the warning in core libraries, and we’ve largely stuck to that in non-core libraries too. But there’s not a requirement that we keep these in place other than the core org (I’ve removed it elsewhere when making breaking changes 😅)

[thomashoneyman]

We're keeping the float token parser which is documented to conform to “the grammar rules of the Haskell report,” but who knows if that's really true or not.

Personally speaking, I don't know why we even have the Haskell and Java parsers included with this library. They feel like they belong in some companion package, like purescript-parsing-language-haskell, if they're desired or used at all. It feels weird to make decisions in this library based on the grammar rules of the Haskell report. I'd be happy removing the language parsers altogether (do they even work?).

[thomashoneyman]

@jamesdbrock I noticed this released as 8.2.0, but this looks like a breaking change because of functions changing modules. Was it supposed to be 9.0.0?

[jamesdbrock]

@thomashoneyman

I'd be happy removing the language parsers altogether

Yeah I've been thinking that too.

but this looks like a breaking change because of functions changing modules

I'm re-exporting the functions from the old modules so there's no breaking change.

[jamesdbrock]

purescript/package-sets#1060