Add basic spec tests

[chtenb]

This project could do with a more structured and comprehensive set of tests. This might especially come in handy when we want to make some fundamental changes in the near future.

@jamesdbrock I encountered some things I've flagged with TODO in the diff. Should we

• not expose chainr1' and chainl1'?
• change sepEndBy and endBy such that the ";" case in the following test cases succeed?

  , { name: "sepEndBy"
    , parser: mkAnyParser $ sepEndBy anyLetter (char ';')
    -- TODO: ";" should be parsed successfully
    , inputs: { successes: [ "", "a", "a;b", "a;b;c", "a;" ], failures: [ ";", ";a", "ab", "a;ab" ] }
    }
  , { name: "endBy"
    , parser: mkAnyParser $ endBy anyLetter (char ';')
    -- TODO: ";" should be parsed successfully
    , inputs: { successes: [ "", "a;", "a;b;", "a;b;c;" ], failures: [ "a", ";", ";a", "ab", "a;b", "a;b;c" ] }
    }

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]

I agree that chainr1' and chainl1' should not be exported. That was recently fixed in parsing too. purescript-contrib/purescript-parsing#131

[jamesdbrock]

I’m not sure about the ";" case for endBy and sepEndBy... I think it’s correct that it should fail to parse?

[chtenb]

I’m not sure about the ";" case for endBy and sepEndBy... I think it’s correct that it should fail to parse?

The docs respectively say

Parse zero or more separated values, ending with a separator.

and

Parse zero or more separated values, optionally ending with a separator.

Since zero values is allowed, and the ending is also allowed (mandatory even in one of them), my reasoning is that ; should succeed.

[jamesdbrock]

I just tested Parsec and the Parsec implementation agrees with your interpretation.

parse (sepEndBy letter (char ';')) "srcname" ";"

Right ""

[chtenb]

And what about endBy anyLetter (char ';') on the input string ""? Since the sep is mandatory for endBy I would expect this to fail instead of succeed. Does that align with Parsec too?