On Mon, Mar 15, 2010 at 6:37 PM, Greg Fitzgerald
If the community wants to freeze on Parsec 2, then I have no problem renaming the package, otherwise I think it is both unnecessary and a waste of effort.
Reasons one might prefer Parsec 2.1: * Haskell98 implementation * Daan Leijen's whitepapers: http://legacy.cs.uu.nl/daan/download/parsec/parsec.pdf http://legacy.cs.uu.nl/daan/download/papers/parsec-paper.pdf
Reasons one might prefer Parsec 3.1: * Parametric in the input stream type (parses ByteStrings) * Monad transformer that can be stacked on arbitrary monads * More Haddock documentation
Parsec 3 is certainly useful, but there are still good, valid reasons one might prefer Parsec 2.1.
Derek, if you wouldn't mind renaming Parsec 3, I volunteer to become the maintainer of the Parsec package.
Here's the list of packages on Hackage which: - State they do not work with parsec-2 - But whose package description would lead to cabal-install trying to form an install plan with parsec-4 The list isn't too long, but we would want to put an upper bound on their package descriptions before making the switch. (I don't know how many of these truly require features unique to parsec-3) -Antoine nntp-0.0.2.1 (parsec >=3.0) uri-0.1.2 (parsec >=3.0) Scurry-0.0.2 (parsec >=3.0.0) pngload-0.1 (parsec >=3.0.0) natural-sort-0.1.1 (parsec >=3) hmk-0.9.1 (parsec >=3.0.0) piki-0.3.0 (parsec >=3) buster-gtk-2.0 (parsec >=3.0.0) hunp-0.1 (parsec >=3) scc-0.3 (parsec >=3) hgom-0.5.2 (parsec >=3.0.0) sgf-0.1.1 (parsec >=3) dedukti-1.0.3 (parsec >=3.0.0) hpuz-1.0.3 (parsec >=3.0.0) mighttpd-0.1.0 (parsec >=3) hellnet-0.1.1 (parsec >=3.0.0) press-0.1.2 (parsec >=3.0.0) dbus-core-0.3 (parsec >=3.0.0) hssqlppp-0.0.7 (parsec >=3) polyseq-0.1.1 (parsec >=3.0.0) iteratee-parsec-0.0.2 (parsec >=3) ministg-0.1 (parsec >=3) PageIO-0.0.2 (parsec >=3.0) hack-contrib-press-0.1.2 (parsec >=3.0.0) webserver-0.1.0 (parsec >=3) wyvern-0.1 (parsec >=3.0) buster-0.99.5 (parsec >=3.0.0)