--  Copyright (C) 2004-2009 David Roundy, Eric Kow, Simon Michael, Tomas Caithaml
--
--  This program is free software; you can redistribute it and/or modify
--  it under the terms of the GNU General Public License as published by
--  the Free Software Foundation; either version 2, or (at your option)
--  any later version.
--
--  This program is distributed in the hope that it will be useful,
--  but WITHOUT ANY WARRANTY; without even the implied warranty of
--  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
--  GNU General Public License for more details.
--
--  You should have received a copy of the GNU General Public License
--  along with this program; see the file COPYING.  If not, write to
--  the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
--  Boston, MA 02110-1301, USA.

module Darcs.Commands.ShowAuthors ( showAuthors ) where

import qualified Ratified( readFile )
import Control.Arrow ((&&&), (***))
import Data.List ( isInfixOf, sortBy, groupBy, group, sort)
import Data.Ord (comparing)
import Data.Char ( toLower, isSpace )
import Data.Maybe( isJust )
import Text.ParserCombinators.Parsec hiding (lower, count, Line)
import Text.ParserCombinators.Parsec.Error
import Text.Regex ( Regex, mkRegexWithOpts, matchRegex )

import Darcs.Arguments ( DarcsFlag(..), workingRepoDir )
import Darcs.Commands ( DarcsCommand(..), nodefaults, putWarning )
import Darcs.External ( viewDoc )
import Darcs.Patch.PatchInfoAnd ( info )
import Darcs.Repository ( amInRepository, readRepo, withRepository, RepoJob(..) )
import Darcs.Patch.Info ( piAuthor )
import Darcs.Patch.Set ( newset2RL )
import Darcs.Witnesses.Ordered ( mapRL )
import Printer ( text )
import Data.Function (on)

data Spelling = Spelling String String [Regex] -- name, email, regexps
type ParsedLine = Maybe Spelling -- Nothing for blank lines


showAuthorsDescription :: String
showAuthorsDescription = "List authors by patch count."

showAuthorsHelp :: String
showAuthorsHelp =
 "The `darcs show authors' command lists the authors of the current\n" ++
 "repository, sorted by the number of patches contributed.  With the\n" ++
 "--verbose option, this command simply lists the author of each patch\n" ++
 "(without aggregation or sorting).\n" ++
 "\n" ++
 "An author's name or email address may change over time.  To tell Darcs\n" ++
 "when multiple author strings refer to the same individual, create an\n" ++
 "`.authorspellings' file in the root of the working tree.  Each line in\n" ++
 "this file begins with an author's canonical name and address, and may\n" ++
 "be followed by a comma separated list of extended regular expressions.\n" ++
 "Blank lines and lines beginning with two hyphens are ignored.\n" ++
 "The format of .authorspelling can be described by this pattern:\n" ++
 "\n" ++
 " name <address> [, regexp ]*\n" ++
 "\n" ++
 "There are some pitfalls concerning special characters:\n" ++
 "Whitespaces are stripped, if you need space in regexp use [ ]. \n" ++
 "Because comma serves as a separator you have to escape it if you want\n" ++
 "it in regexp. Note that .authorspelingfile use extended regular\n" ++
 "expressions so +, ? and so on are metacharacters and you need to \n" ++
 "escape them to be interpreted literally.\n" ++
 "\n" ++
 "Any patch with an author string that matches the canonical address or\n" ++
 "any of the associated regexps is considered to be the work of that\n" ++
 "author.  All matching is case-insensitive and partial (it can match a\n" ++
 "substring). Use ^,$ to match the whole string in regexps\n" ++
 "\n" ++
 "Currently this canonicalization step is done only in `darcs show\n" ++
 "authors'.  Other commands, such as `darcs changes' use author strings\n" ++
 "verbatim.\n" ++
 "\n" ++
 "An example .authorspelling file is:\n" ++
 "\n" ++
 "    -- This is a comment.\n" ++
 "    Fred Nurk <fred@example.com>\n" ++
 "    John Snagge <snagge@bbc.co.uk>, John, snagge@, js@(si|mit).edu\n" ++
 "    Chuck Jones\\, Jr. <chuck@pobox.com>, cj\\+user@example.com\n"

showAuthors :: DarcsCommand
showAuthors = DarcsCommand {
  commandProgramName = "darcs",
  commandName = "authors",
  commandHelp = showAuthorsHelp,
  commandDescription = showAuthorsDescription,
  commandExtraArgs = 0,
  commandExtraArgHelp = [],
  commandCommand = authorsCmd,
  commandPrereq = amInRepository,
  commandGetArgPossibilities = return [],
  commandArgdefaults = nodefaults,
  commandAdvancedOptions = [],
  commandBasicOptions = [workingRepoDir] }

authorsCmd :: [DarcsFlag] -> [String] -> IO ()
authorsCmd opts _ = withRepository opts $ RepoJob $ \repository -> do
  patches <- readRepo repository
  spellings <- compiledAuthorSpellings opts
  let authors = mapRL (piAuthor . info) $ newset2RL patches
  viewDoc $ text $ unlines $
   if Verbose `elem` opts
    then authors
    else -- A list of the form ["<count> <canonical name>"]...

      -- Turn the final result into a list of strings.
      map (\ (count, name) -> show count ++ "\t" ++ name) $
      -- Sort by descending patch count.
      reverse $ sortBy (comparing fst) $
      -- Combine duplicates from a list [(count, canonized name)]
      -- with duplicates canonized names (see next comment).
      map ((sum *** head) . unzip) $
      groupBy ((==) `on` snd) $
      sortBy  (comparing snd) $
      -- Because it would take a long time to canonize "foo" into
      -- "foo <foo@bar.baz>" once per patch, the code below
      -- generates a list [(count, canonized name)].
      map (length &&& (canonizeAuthor spellings . head)) $
      group $ sort authors

canonizeAuthor :: [Spelling] -> String -> String
canonizeAuthor spells author = getName canonicals
 where
   getName [] = author
   getName ((Spelling name email _):_) = name ++ " <" ++ email ++ ">"
   canonicals = filter (ismatch author) spells
   ismatch s (Spelling _ mail regexps) =
     s `correspondsTo` mail || any (s `contains_regex`) regexps
   contains_regex a r = isJust $ matchRegex r a
   correspondsTo a b = lower b `isInfixOf` lower a
   lower = map toLower

compiledAuthorSpellings :: [DarcsFlag] -> IO [Spelling]
compiledAuthorSpellings opts = do
  let as_file = ".authorspellings"
  contents <- (Ratified.readFile -- never unlinked from within darcs
               as_file `catch` (\_ -> return ""))
  let parse_results = map (parse sentence as_file) $ lines contents
  clean 1 parse_results
 where
  clean :: Int -> [Either ParseError ParsedLine] -> IO [Spelling]
  clean _ [] = return []
  -- print parse error
  clean n ((Left err):xs) = do
    let npos = setSourceLine (errorPos err) n
    putWarning opts . text . show $ setErrorPos npos err
    clean (n+1) xs
  -- skip blank line
  clean n ((Right Nothing):xs)  = clean (n+1) xs
  -- unwrap Spelling
  clean n ((Right (Just a):xs)) = do
    as <- clean (n+1) xs
    return (a:as)

----------
-- PARSERS

sentence :: Parser ParsedLine
sentence = spaces >> (comment <|> blank <|> addressline)
  where
    comment = string "--" >> return Nothing
    blank = eof >> return Nothing

addressline :: Parser ParsedLine
addressline = do
  name <- canonicalName <?> "Canonical name"
  addr <- between (char '<') (char '>') (many1 (noneOf ">")) <?> "Address"
  spaces
  rest <- option [] (char ',' >> regexp `sepBy` char ',') <?> "List of regexps"
  return $ Just $ Spelling (strip name) addr (compile rest)
  where
    strip = dropWhile isSpace . reverse . dropWhile isSpace . reverse
    makeRegex s = mkRegexWithOpts s True False
    compile = map makeRegex . filter (not . null) . map strip

regexp :: Parser String
regexp = many1 p <?> "Regular expression"
  where p = try (string "\\," >> return ',')
            <|> noneOf ","

canonicalName :: Parser String
canonicalName = many1 p
  where p = try (string "\\," >> return ',')
            <|> noneOf ",<"