Recreate Sprache Csv Parser in Superpower about superpower HOT 7 CLOSED

Closing this as stale since we haven't looped back to it in a while, but hope all's going well!

from superpower.

Hi!

The problem looks to be that at the end of the input, Cell will still match via the LiteralCellContent.Many() rule, and RecordTerminator will also match, because of the Parse.Return("").AtEnd() rule, meaning there would be an essentially infinite procession of empty records at the end of the input.

A quick fix (not sure about error message quality implications?) might be to remove the terminator part from Record, and make the final rule:

    public static readonly TextParser<IEnumerable<IEnumerable<string>>> Csv =
        Record.ManyDelimitedBy(NewLine).AtEnd().Select(x => x.AsEnumerable());

from superpower.

I figured the Parse.Return("").AtEnd() rule could potentially be hurting me.

Believe it or not, in my experimenting with this conversion, I have attempted what you suggested and it still fails with the same exception.

I am going to keep working on this today, any other suggestions appreciated.

from superpower.

Looks like it was the NewLine field/prop was causing my problem, even in my simpler non-quoted Csv parser.

I changed this line

static readonly TextParser<string> NewLine = Parse.Return(Environment.NewLine);

to

static readonly TextParser<TextSpan> NewLine = Span.EqualTo(Environment.NewLine);

It now works as intended and I am going to clean it up.

@nblumhardt Any chance you want a test or sample PR submitted?

from superpower.

Aha, great! I missed that one :-)

It'd be awesome if you can post the example somewhere we can link, but I don't think we should currently add more text-parser examples to this repo, since the general not-migrating-from-Sprache usage should probably be via TokenParser.

from superpower.

It'd be awesome if you can post the example somewhere we can link, but I don't think we should currently add more text-parser examples to this repo

No problem and makes sense. I will respond to this issue when I have a sample Gist created using TextParser method/style.

since the general not-migrating-from-Sprache usage should probably be via TokenParser.

Curious, are you saying the proper way to use Superpower is to preferably utilize the TokenParser instead of the built in TextParser's? Is that because Token parsers have better error messaging and reporting? Or?

I am going to keep practicing with this library and also make a simple TokenParser to handle CSV/TSV files. Perhaps when I feel comfortable with it you could review it one day? I know you're busy and I will never ask for immediate review. Just a big fan of your work, style, design and architecture.

from superpower.

Thanks! :D

I might have overstated the suggestion to start with token parsers; for simple things that stay simple, text parsers are probably just as good from the perspective of this library. Token parsers are able to make use of more features of Superpower, though, including the nicer error reporting and the easy handling of whitespace and comments. Not a big difference, but probably enough that I'd try starting with a token parser in most cases.

Keen to hear how you go, making the API as usable/approachable as Sprache's is an ongoing project :-) Cheers!

from superpower.