ByteString parser improvements

[LinuxUser404]

ByteString parser seems to be not usable because there are 0 parser functions operating on ByteString. When we operate on strings of chars we have useful parsers like "oneOf", "digit", "spaces". It would be nice to have something like "zeros"(for bytes filled with zeros), "bigEndiandInt32", "byte" to consume bytes from the ByteString stream and write the results into appropriate data types. Something similar was done in "HCodecs"(and many other places where people try to handle some specific file formats). Parsec would be a more natural place to look for such convenient general functions.

[int-index]

There is a well-maintained for of Parsec — https://hackage.haskell.org/package/megaparsec — that has support for ByteString and even custom input stream types.

[LinuxUser404]

Thanks for your reply. So basically if I need some extra functionality I should use a different package... Is there any reason why those are different packages while they are meant to do the same thing?

[int-index]

No technical reason, just a communication issue (I guess merging these changes upstream was difficult). I think if @mrkkrp got commit bits, he would not be opposed to having all these changes directly in parsec.

[hvr]

It's not really a communication issue; it's more like that parsec is more conservative at this point given it has to carry a lot of weight in the ecosystem, and any misstep may have significant cost; while megaparsec is free to explore the design space and can move faster.

I intend to improve parsec by stealing ideas from megaparsec/trifecta/parsers, while avoiding bigger changes that would completely break API compatibility. However, we do have a diversity of parsers, and there is not going to be "the one" parser to rule them all. Parsec just happens to be one of the oldest ones and tries to keep its dependency footprint light enough to be consumable by lib:Cabal. Personally, I often use and recommend megaparsec or attoparsec for projects for which Parsec is not a good fit; or you can just go for https://hackage.haskell.org/package/parsers which abstracts over parser libraries.

Long story short: PRs which add (popular) functionality to Parsec are quite welcome. :-)