The Parsec library code contains string-parsing (with escapes) that you might be able to use or canibalize. See 'StringLiteral' et seq in ParsecToken.hs. The escape handling there is a fair amount of code, bit it covers a lot of escape options. #g -- At 14:54 30/07/03 -0500, Shawn P. Garbett wrote:
In attempting to improve the CSV parser based on comments, I have the following code that's attached. I'm having a heck of a time getting the double quotes = an escaped quote thing to work. There is some commented out code which was my last attempt. As it stands the code works, minus the escaped quotes. Remove the comments and it just hangs.
module CSV (contents, csv) where
import Parsec
---------------------------------------------------------------------- -- CSV Module ----------------------------------------------------------------------
-- Useful common parsers comma :: Parser Char comma = char ','
quote :: Parser Char quote = char '\"'
-- How to handle these buggers.... --esc_quote :: Parser Char --esc_quote = do {char '\"'; char '\"';}
--text :: Parser String --text = do {esc_quote; t<-text; return ('\"':t)} -- <|> do {c <- noneOf "\""; t<-text; return (c:t) } -- <|> return ""
-- A cell can be a quoted value, a number or empty -- Quotes can be embedded by using double quotes "" cell :: Parser String cell = between quote quote (many (noneOf "\"")) -- quoted values --cell = between quote quote text -- quoted values <|> many1 (noneOf "\",\n") -- unquoted values <|> return "" -- give up, Empty cell
-- Group of cells with a newline cells :: Parser [String] cells = do c <- sepBy cell comma newline return c
-- Comma Separated Values, set of rows followed by eof csv :: Parser [[String]] csv = manyTill cells eof
-- Useful For extracting comma delimited values of a cell contents :: Parser [String] contents = sepBy1 (many (noneOf ",")) comma
_______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
-------------------
Graham Klyne