+
+ blocks when their
-first line indicates that it contains a REPL calculation, which the
-style sheet can pick up and present accordingly.
-
-\begin{code}
-data DocMode
- = Outside -- not inside a begin{code} ... \end{code} block
- | Beginning -- at the start of a begin{code} ... \end{code} block
- | InsideRepl -- inside a REPL code block
- | InsideProg -- inside a non-REPL code block
- deriving (Eq,Show)
-
-type DocState = IORef DocMode
-
-genMode :: IO MODE
-genMode = Gen <$> newIORef []
-\end{code}
-
-\begin{code}
--- | the state is the accumulated test function identifiers for
--- generating the list of them gets added to the end of the programme
-type GenState = IORef [String]
-
-docMode :: IO MODE
-docMode = Doc <$> newIORef Outside
-\end{code}
-
-
-\begin{code}
-loop :: MODE -> FilePath -> FilePath -> IO ()
-loop mode =
- sed $ Select
- [ (,) [re|^%include ${file}(@{%string}) ${rex}(@{%string})$|] $ EDIT_fun TOP $ include mode
- , (,) [re|^%main ${arg}(top|bottom)$|] $ EDIT_gen $ main_ mode
- , (,) [re|^\\begin\{code\}$|] $ EDIT_gen $ begin mode
- , (,) [re|^${fn}(evalme@{%id}) = checkThis ${arg}(@{%string}).*$|] $ EDIT_fun TOP $ evalme mode
- , (,) [re|^\\end\{code\}$|] $ EDIT_fun TOP $ end mode
- , (,) [re|^.*$|] $ EDIT_fun TOP $ other mode
- ]
-\end{code}
-
-\begin{code}
-include, evalme, end,
- other :: MODE
- -> LineNo
- -> Match LBS.ByteString
- -> Location
- -> Capture LBS.ByteString
- -> IO (Maybe LBS.ByteString)
-
-main_,
- begin :: MODE
- -> LineNo
- -> Matches LBS.ByteString
- -> IO (LineEdit LBS.ByteString)
-
-include (Doc _ ) = includeDoc
-include (Gen _ ) = passthru
-
-main_ (Doc _ ) = mainDoc
-main_ (Gen gs) = mainGen gs
-
-begin (Doc ds) = beginDoc ds
-begin (Gen _ ) = passthru_g
-
-evalme (Doc ds) = evalmeDoc ds
-evalme (Gen gs) = evalmeGen gs
-
-end (Doc ds) = endDoc ds
-end (Gen _ ) = passthru
-
-other (Doc ds) = otherDoc ds
-other (Gen _ ) = passthru
-
-passthru :: LineNo
- -> Match LBS.ByteString
- -> Location
- -> Capture LBS.ByteString
- -> IO (Maybe LBS.ByteString)
-passthru _ _ _ _ = return Nothing
-
-passthru_g :: LineNo
- -> Matches LBS.ByteString
- -> IO (LineEdit LBS.ByteString)
-passthru_g _ _ = return NoEdit
-\end{code}
-
-
-Script to Generate All Tutorial Tests and Docs
-----------------------------------------------
-
-\begin{code}
-gen_all :: IO ()
-gen_all = do
- -- prepare HTML docs for the (literate) tools
- pd "re-gen-modules"
- pd "re-include"
- pd "re-nginx-log-processor"
- pd "re-pp"
- pd "re-tests"
- pd "TestKit"
- pd "RE/Capture"
- pd "RE/Edit"
- pd "RE/IsRegex"
- pd "RE/Options"
- pd "RE/Replace"
- pd "RE/TestBench"
- pd "RE/Tools/Grep"
- pd "RE/Tools/Lex"
- pd "RE/Tools/Sed"
- pd "RE/Internal/NamedCaptures"
- -- render the tutorial in HTML
- dm <- docMode
- loop dm "examples/re-tutorial-master.lhs" "tmp/re-tutorial.lhs"
- createDirectoryIfMissing False "tmp"
- pandoc'
- "re-tutorial.lhs"
- "examples/re-tutorial.lhs"
- "tmp/re-tutorial.lhs"
- "docs/re-tutorial.html"
- -- generate the tutorial-based tests
- gm <- genMode
- loop gm "examples/re-tutorial-master.lhs" "examples/re-tutorial.lhs"
- putStrLn ">> examples/re-tutorial.lhs"
- where
- pd fnm = case captureTextMaybe [cp|mnm|] $ fnm T.?=~ [re|^RE/(Tools/|Internal/)?${mnm}(@{%id})|] of
- Just mnm -> pandoc fnm ("Text/"<>fnm<>".lhs") ("docs/"<>mnm<>".html")
- Nothing -> pandoc fnm ( "examples/"<>fnm<>".lhs") ("docs/"<>fnm<>".html")
-\end{code}
-
-
-Generating the Tutorial
------------------------
-
-\begin{code}
-includeDoc :: LineNo
- -> Match LBS.ByteString
- -> Location
- -> Capture LBS.ByteString
- -> IO (Maybe LBS.ByteString)
-includeDoc _ mtch _ _ = fmap Just $
- extract fp =<< compileRegex () re_s
- where
- fp = prs_s $ captureText [cp|file|] mtch
- re_s = prs_s $ captureText [cp|rex|] mtch
-
- prs_s = fromMaybe (error "includeDoc") . parseString
-\end{code}
-
-\begin{code}
-mainDoc :: LineNo
- -> Matches LBS.ByteString
- -> IO (LineEdit LBS.ByteString)
-mainDoc _ _ = return Delete
-\end{code}
-
-\begin{code}
-beginDoc :: DocState
- -> LineNo
- -> Matches LBS.ByteString
- -> IO (LineEdit LBS.ByteString)
-beginDoc ds _ _ = writeIORef ds Beginning >> return Delete
-\end{code}
-
-\begin{code}
-evalmeDoc, endDoc, otherDoc :: DocState
- -> LineNo
- -> Match LBS.ByteString
- -> Location
- -> Capture LBS.ByteString
- -> IO (Maybe LBS.ByteString)
-
-evalmeDoc ds lno _ _ _ = do
- dm <- readIORef ds
- when (dm/=Beginning) $
- bad_state "evalme" lno dm
- writeIORef ds InsideRepl
- return $ Just $ ""
+ , prg "pages"
+ , prg "all"
+ , prg "doc (-|) (-|)"
+ , prg "gen (-|) (-|)"
+ ]
+\end{code}
+
+
+The Sed Script
+--------------
+
+\begin{code}
+-- | the MODE determines whether we are generating documentation
+-- or a Haskell testsuite and includes any IO-accessible state
+-- needed by the relevant processor
+data MODE
+ = Doc DocState -- ^ document-generation state
+ | Gen GenState -- ^ adjusting-the-program-for-testing state
+\end{code}
+
+The `DocState` is initialised to `Outside` and flips though the different
+states as it traverses a code block, so that we can wrap code
+blocks in special
\n"<>begin_code
-
-endDoc ds lno _ _ _ = do
- dm <- readIORef ds
- case dm of
- Outside -> bad_state "end" lno dm
- Beginning -> return $ Just $ begin_code <> "\n" <> end_code
- InsideRepl -> return $ Just $ end_code <> "\n
"
- InsideProg -> return Nothing
-
-otherDoc ds _ mtch _ _ = do
- dm <- readIORef ds
- case dm of
- Beginning -> do
- writeIORef ds InsideProg
- return $ Just $ begin_code <> "\n" <> matchSource mtch
- _ -> return Nothing
-
-bad_state :: String -> LineNo -> DocMode -> IO a
-bad_state lab lno dm = error $
- printf "Bad document syntax: %s: %d: %s" lab (getLineNo lno) $ show dm
-\end{code}
-
-
-Generating the Tests
---------------------
-
-\begin{code}
-evalmeGen :: GenState
- -> LineNo
- -> Match LBS.ByteString
- -> Location
- -> Capture LBS.ByteString
- -> IO (Maybe LBS.ByteString)
-evalmeGen gs _ mtch0 _ _ = Just <$>
- replaceCapturesM replace_ ALL f mtch0
- where
- f mtch loc cap = case _loc_capture loc of
- 2 -> do
- modifyIORef gs (ide:)
- return $ Just $ LBS.pack $ show ide
- where
- ide = LBS.unpack $ captureText [cp|fn|] mtch
- _ -> return $ Just $ capturedText cap
-\end{code}
-
-How are we doing?
-
-\begin{code}
-mainGen :: GenState
- -> LineNo
- -> Matches LBS.ByteString
- -> IO (LineEdit LBS.ByteString)
-mainGen gs _ mtchs = case allMatches mtchs of
- [mtch] ->
- case captureText [cp|arg|] $ mtch of
- "top" -> return $ ReplaceWith $ LBS.unlines $
- [ begin_code
- , "module Main(main) where"
- , end_code
- , ""
- , "*********************************************************"
- , "*"
- , "* WARNING: this is generated from pp-tutorial-master.lhs "
- , "*"
- , "*********************************************************"
- ]
- "bottom" -> do
- fns <- readIORef gs
- return $ ReplaceWith $ LBS.unlines $
- [ begin_code
- , "main :: IO ()"
- , "main = runTests"
- ] ++ mk_list fns ++
- [ end_code
- ]
- _ -> error "mainGen (b)"
- _ -> error "mainGen (a)"
-\end{code}
-
-We cannot place these strings inline without confusing pandoc so we
-use these definitions instead.
-
-\begin{code}
-begin_code, end_code :: LBS.ByteString
-begin_code = "\\"<>"begin{code}"
-end_code = "\\"<>"end{code}"
-\end{code}
-
-
-
-\begin{code}
-mk_list :: [String] -> [LBS.ByteString]
-mk_list [] = ["[]"]
-mk_list (ide0:ides) = f "[" ide0 $ foldr (f ",") [" ]"] ides
- where
- f pfx ide t = (" "<>pfx<>" "<>LBS.pack ide) : t
-\end{code}
-
-
-Extracting a Literate Fragment from a Haskell Program Text
-----------------------------------------------------------
-
-\begin{code}
-extract :: FilePath -> RE -> IO LBS.ByteString
-extract fp rex = extr . LBS.lines <$> LBS.readFile fp
- where
- extr lns =
- case parse $ scan rex lns of
- Nothing -> oops
- Just (lno,n) -> LBS.unlines $ (hdr :) $ (take n $ drop i lns) ++ [ftr]
- where
- i = getZeroBasedLineNo lno
-
- oops = error $ concat
- [ "failed to locate fragment matching "
- , show $ reSource rex
- , " in file "
- , show fp
- ]
-
- hdr = ""
- ftr = "
"
-\end{code}
-
-\begin{code}
-parse :: [Token] -> Maybe (LineNo,Int)
-parse [] = Nothing
-parse (tk:tks) = case (tk,tks) of
- (Bra b_ln,Hit:Ket k_ln:_) -> Just (b_ln,count_lines_incl b_ln k_ln)
- _ -> parse tks
-\end{code}
-
-\begin{code}
-count_lines_incl :: LineNo -> LineNo -> Int
-count_lines_incl b_ln k_ln =
- getZeroBasedLineNo k_ln + 1 - getZeroBasedLineNo b_ln
-\end{code}
-
-\begin{code}
-data Token = Bra LineNo | Hit | Ket LineNo deriving (Show)
-\end{code}
-
-\begin{code}
-scan :: RE -> [LBS.ByteString] -> [Token]
-scan rex = grepScript
- [ (,) [re|\\begin\{code\}|] $ \i -> chk $ Bra i
- , (,) rex $ \_ -> chk Hit
- , (,) [re|\\end\{code\}|] $ \i -> chk $ Ket i
- ]
- where
- chk x mtchs = case anyMatches mtchs of
- True -> Just x
- False -> Nothing
-\end{code}
-
-
-pandoc
-------
-
-\begin{code}
-pandoc :: T.Text -> T.Text -> T.Text -> IO ()
-pandoc title in_file = pandoc' title in_file in_file
-
-pandoc' :: T.Text -> T.Text -> T.Text -> T.Text -> IO ()
-pandoc' title repo_path in_file out_file = do
- writeFile "tmp/bc.html" bc
- fmap (const ()) $
- SH.shelly $ SH.verbosely $
- SH.run "pandoc"
- [ "-f", "markdown+lhs+grid_tables"
- , "-t", "html"
- , "-s"
- , "-B", "tmp/bc.html"
- , "-c", "lib/styles.css"
- , "-c", "lib/bs.css"
- , "-o", out_file
- , in_file
- ]
- where
- bc = concat
- [ ""
- ]
-
- repo_url = concat
- [ "https://github.com/iconnect/regex/blob/master/"
- , T.unpack repo_path
- ]
-\end{code}
-
-testing
--------
-
-\begin{code}
-test :: IO ()
-test = do
- dm <- docMode
- test_pp "pp-doc" (loop dm) "data/pp-test.lhs" "data/pp-result-doc.lhs"
- gm <- genMode
- test_pp "pp-gen" (loop gm) "data/pp-test.lhs" "data/pp-result-gen.lhs"
- putStrLn "tests passed"
-\end{code}
diff --git a/examples/re-prep.lhs b/examples/re-prep.lhs
new file mode 100644
--- /dev/null
+++ b/examples/re-prep.lhs
@@ -0,0 +1,828 @@
+\begin{code}
+{-# LANGUAGE NoImplicitPrelude #-}
+{-# LANGUAGE RecordWildCards #-}
+{-# LANGUAGE GeneralizedNewtypeDeriving #-}
+{-# LANGUAGE TemplateHaskell #-}
+{-# LANGUAGE QuasiQuotes #-}
+{-# LANGUAGE FlexibleContexts #-}
+{-# LANGUAGE OverloadedStrings #-}
+
+module Main
+ ( main
+ ) where
+
+import Control.Applicative
+import qualified Control.Monad as M
+import qualified Data.ByteString.Lazy.Char8 as LBS
+import Data.IORef
+import Data.Maybe
+import Data.Monoid
+import qualified Data.Text as T
+import qualified Data.Text.Encoding as TE
+import Network.HTTP.Conduit
+import Prelude.Compat
+import qualified Shelly as SH
+import System.Directory
+import System.Environment
+import TestKit
+import Text.Heredoc
+import Text.Printf
+import Text.RE.Edit
+import Text.RE.TDFA.ByteString.Lazy
+import qualified Text.RE.TDFA.Text as TT
+\end{code}
+
+\begin{code}
+main :: IO ()
+main = do
+ as <- getArgs
+ case as of
+ [] -> test
+ ["test"] -> test
+ ["doc",fn,fn'] | is_file fn -> doc fn fn'
+ ["gen",fn,fn'] | is_file fn -> gen fn fn'
+ ["badges"] -> badges
+ ["bump-version",vrn] -> bumpVersion vrn
+ ["pages"] -> pages
+ ["all"] -> gen_all
+ _ -> usage
+ where
+ is_file = not . (== "--") . take 2
+
+ doc fn fn' = docMode >>= \dm -> loop dm fn fn'
+ gen fn fn' = genMode >>= \gm -> loop gm fn fn'
+
+ usage = do
+ pnm <- getProgName
+ let prg = ((" "++pnm++" ")++)
+ putStr $ unlines
+ [ "usage:"
+ , prg "--help"
+ , prg "[test]"
+ , prg "badges"
+ , prg "bump-version blocks when their
+first line indicates that it contains a REPL calculation, which the
+style sheet can pick up and present accordingly.
+
+\begin{code}
+data DocMode
+ = Outside -- not inside a begin{code} ... \end{code} block
+ | Beginning -- at the start of a begin{code} ... \end{code} block
+ | InsideRepl -- inside a REPL code block
+ | InsideProg -- inside a non-REPL code block
+ deriving (Eq,Show)
+
+type DocState = IORef DocMode
+
+genMode :: IO MODE
+genMode = Gen <$> newIORef []
+\end{code}
+
+\begin{code}
+-- | the state is the accumulated test function identifiers for
+-- generating the list of them gets added to the end of the programme
+type GenState = IORef [String]
+
+docMode :: IO MODE
+docMode = Doc <$> newIORef Outside
+\end{code}
+
+
+\begin{code}
+loop :: MODE -> FilePath -> FilePath -> IO ()
+loop mode =
+ sed $ Select
+ [ (,) [re|^%include ${file}(@{%string}) ${rex}(@{%string})$|] $ EDIT_fun TOP $ inclde mode
+ , (,) [re|^%main ${arg}(top|bottom)$|] $ EDIT_gen $ main_ mode
+ , (,) [re|^\\begin\{code\}$|] $ EDIT_gen $ begin mode
+ , (,) [re|^${fn}(evalme@{%id}) = checkThis ${arg}(@{%string}).*$|] $ EDIT_fun TOP $ evalme mode
+ , (,) [re|^\\end\{code\}$|] $ EDIT_fun TOP $ end mode
+ , (,) [re|^.*$|] $ EDIT_fun TOP $ other mode
+ ]
+\end{code}
+
+\begin{code}
+inclde, evalme, end,
+ other :: MODE
+ -> LineNo
+ -> Match LBS.ByteString
+ -> Location
+ -> Capture LBS.ByteString
+ -> IO (Maybe LBS.ByteString)
+
+main_,
+ begin :: MODE
+ -> LineNo
+ -> Matches LBS.ByteString
+ -> IO (LineEdit LBS.ByteString)
+
+inclde (Doc _ ) = includeDoc
+inclde (Gen _ ) = passthru
+
+main_ (Doc _ ) = mainDoc
+main_ (Gen gs) = mainGen gs
+
+begin (Doc ds) = beginDoc ds
+begin (Gen _ ) = passthru_g
+
+evalme (Doc ds) = evalmeDoc ds
+evalme (Gen gs) = evalmeGen gs
+
+end (Doc ds) = endDoc ds
+end (Gen _ ) = passthru
+
+other (Doc ds) = otherDoc ds
+other (Gen _ ) = passthru
+
+passthru :: LineNo
+ -> Match LBS.ByteString
+ -> Location
+ -> Capture LBS.ByteString
+ -> IO (Maybe LBS.ByteString)
+passthru _ _ _ _ = return Nothing
+
+passthru_g :: LineNo
+ -> Matches LBS.ByteString
+ -> IO (LineEdit LBS.ByteString)
+passthru_g _ _ = return NoEdit
+\end{code}
+
+
+Script to Generate the Whole Web Site
+-------------------------------------
+
+\begin{code}
+gen_all :: IO ()
+gen_all = do
+ -- prepare HTML docs for the (literate) tools
+ pd "re-gen-cabals"
+ pd "re-gen-modules"
+ pd "re-include"
+ pd "re-nginx-log-processor"
+ pd "re-prep"
+ pd "re-tests"
+ pd "TestKit"
+ pd "RE/Capture"
+ pd "RE/Edit"
+ pd "RE/IsRegex"
+ pd "RE/Options"
+ pd "RE/Replace"
+ pd "RE/TestBench"
+ pd "RE/Tools/Grep"
+ pd "RE/Tools/Lex"
+ pd "RE/Tools/Sed"
+ pd "RE/Internal/NamedCaptures"
+ -- render the tutorial in HTML
+ dm <- docMode
+ loop dm "examples/re-tutorial-master.lhs" "tmp/re-tutorial.lhs"
+ createDirectoryIfMissing False "tmp"
+ pandoc_lhs'
+ "re-tutorial.lhs"
+ "examples/re-tutorial.lhs"
+ "tmp/re-tutorial.lhs"
+ "docs/re-tutorial.html"
+ -- generate the tutorial-based tests
+ gm <- genMode
+ loop gm "examples/re-tutorial-master.lhs" "examples/re-tutorial.lhs"
+ putStrLn ">> examples/re-tutorial.lhs"
+ pages
+ where
+ pd fnm = case (mtch !$$? [cp|fdr|],mtch !$$? [cp|mnm|]) of
+ (Nothing ,Just mnm) -> pandoc_lhs ("Text.RE." <>mnm) ("Text/" <>fnm<>".lhs") ("docs/"<>mnm<>".html")
+ (Just fdr,Just mnm) -> pandoc_lhs ("Text.RE."<>fdr<>"."<>mnm) ("Text/" <>fnm<>".lhs") ("docs/"<>mnm<>".html")
+ _ -> pandoc_lhs ("examples/"<>fnm<>".lhs" ) ("examples/"<>fnm<>".lhs") ("docs/"<>fnm<>".html")
+ where
+ mtch = fnm TT.?=~ [re|^RE/(${fdr}(Tools|Internal)/)?${mnm}(@{%id})|]
+\end{code}
+
+
+Generating the Tutorial
+-----------------------
+
+\begin{code}
+includeDoc :: LineNo
+ -> Match LBS.ByteString
+ -> Location
+ -> Capture LBS.ByteString
+ -> IO (Maybe LBS.ByteString)
+includeDoc _ mtch _ _ = fmap Just $
+ extract fp =<< compileRegex () re_s
+ where
+ fp = prs_s $ captureText [cp|file|] mtch
+ re_s = prs_s $ captureText [cp|rex|] mtch
+
+ prs_s = maybe (error "includeDoc") T.unpack . parseString
+\end{code}
+
+\begin{code}
+mainDoc :: LineNo
+ -> Matches LBS.ByteString
+ -> IO (LineEdit LBS.ByteString)
+mainDoc _ _ = return Delete
+\end{code}
+
+\begin{code}
+beginDoc :: DocState
+ -> LineNo
+ -> Matches LBS.ByteString
+ -> IO (LineEdit LBS.ByteString)
+beginDoc ds _ _ = writeIORef ds Beginning >> return Delete
+\end{code}
+
+\begin{code}
+evalmeDoc, endDoc, otherDoc :: DocState
+ -> LineNo
+ -> Match LBS.ByteString
+ -> Location
+ -> Capture LBS.ByteString
+ -> IO (Maybe LBS.ByteString)
+
+evalmeDoc ds lno _ _ _ = do
+ dm <- readIORef ds
+ M.when (dm/=Beginning) $
+ bad_state "evalme" lno dm
+ writeIORef ds InsideRepl
+ return $ Just $ "
\n"<>begin_code
+
+endDoc ds lno _ _ _ = do
+ dm <- readIORef ds
+ case dm of
+ Outside -> bad_state "end" lno dm
+ Beginning -> return $ Just $ begin_code <> "\n" <> end_code
+ InsideRepl -> return $ Just $ end_code <> "\n
"
+ InsideProg -> return Nothing
+
+otherDoc ds _ mtch _ _ = do
+ dm <- readIORef ds
+ case dm of
+ Beginning -> do
+ writeIORef ds InsideProg
+ return $ Just $ begin_code <> "\n" <> matchSource mtch
+ _ -> return Nothing
+
+bad_state :: String -> LineNo -> DocMode -> IO a
+bad_state lab lno dm = error $
+ printf "Bad document syntax: %s: %d: %s" lab (getLineNo lno) $ show dm
+\end{code}
+
+
+Generating the Tests
+--------------------
+
+\begin{code}
+evalmeGen :: GenState
+ -> LineNo
+ -> Match LBS.ByteString
+ -> Location
+ -> Capture LBS.ByteString
+ -> IO (Maybe LBS.ByteString)
+evalmeGen gs _ mtch0 _ _ = Just <$>
+ replaceCapturesM replace_ ALL f mtch0
+ where
+ f mtch loc cap = case _loc_capture loc of
+ 2 -> do
+ modifyIORef gs (ide:)
+ return $ Just $ LBS.pack $ show ide
+ where
+ ide = LBS.unpack $ captureText [cp|fn|] mtch
+ _ -> return $ Just $ capturedText cap
+\end{code}
+
+How are we doing?
+
+\begin{code}
+mainGen :: GenState
+ -> LineNo
+ -> Matches LBS.ByteString
+ -> IO (LineEdit LBS.ByteString)
+mainGen gs _ mtchs = case allMatches mtchs of
+ [mtch] ->
+ case captureText [cp|arg|] $ mtch of
+ "top" -> return $ ReplaceWith $ LBS.unlines $
+ [ begin_code
+ , "module Main(main) where"
+ , end_code
+ , ""
+ , "*********************************************************"
+ , "*"
+ , "* WARNING: this is generated from pp-tutorial-master.lhs "
+ , "*"
+ , "*********************************************************"
+ ]
+ "bottom" -> do
+ fns <- readIORef gs
+ return $ ReplaceWith $ LBS.unlines $
+ [ begin_code
+ , "main :: IO ()"
+ , "main = runTests"
+ ] ++ mk_list fns ++
+ [ end_code
+ ]
+ _ -> error "mainGen (b)"
+ _ -> error "mainGen (a)"
+\end{code}
+
+We cannot place these strings inline without confusing pandoc so we
+use these definitions instead.
+
+\begin{code}
+begin_code, end_code :: LBS.ByteString
+begin_code = "\\"<>"begin{code}"
+end_code = "\\"<>"end{code}"
+\end{code}
+
+
+
+\begin{code}
+mk_list :: [String] -> [LBS.ByteString]
+mk_list [] = ["[]"]
+mk_list (ide0:ides) = f "[" ide0 $ foldr (f ",") [" ]"] ides
+ where
+ f pfx ide t = (" "<>pfx<>" "<>LBS.pack ide) : t
+\end{code}
+
+
+Extracting a Literate Fragment from a Haskell Program Text
+----------------------------------------------------------
+
+\begin{code}
+extract :: FilePath -> RE -> IO LBS.ByteString
+extract fp rex = extr . LBS.lines <$> LBS.readFile fp
+ where
+ extr lns =
+ case parse $ scan rex lns of
+ Nothing -> oops
+ Just (lno,n) -> LBS.unlines $ (hdr :) $ (take n $ drop i lns) ++ [ftr]
+ where
+ i = getZeroBasedLineNo lno
+
+ oops = error $ concat
+ [ "failed to locate fragment matching "
+ , show $ reSource rex
+ , " in file "
+ , show fp
+ ]
+
+ hdr = ""
+ ftr = "
"
+\end{code}
+
+\begin{code}
+parse :: [Token] -> Maybe (LineNo,Int)
+parse [] = Nothing
+parse (tk:tks) = case (tk,tks) of
+ (Bra b_ln,Hit:Ket k_ln:_) -> Just (b_ln,count_lines_incl b_ln k_ln)
+ _ -> parse tks
+\end{code}
+
+\begin{code}
+count_lines_incl :: LineNo -> LineNo -> Int
+count_lines_incl b_ln k_ln =
+ getZeroBasedLineNo k_ln + 1 - getZeroBasedLineNo b_ln
+\end{code}
+
+\begin{code}
+data Token = Bra LineNo | Hit | Ket LineNo deriving (Show)
+\end{code}
+
+\begin{code}
+scan :: RE -> [LBS.ByteString] -> [Token]
+scan rex = grepScript
+ [ (,) [re|\\begin\{code\}|] $ \i -> chk $ Bra i
+ , (,) rex $ \_ -> chk Hit
+ , (,) [re|\\end\{code\}|] $ \i -> chk $ Ket i
+ ]
+ where
+ chk x mtchs = case anyMatches mtchs of
+ True -> Just x
+ False -> Nothing
+\end{code}
+
+
+badges
+------
+
+\begin{code}
+badges :: IO ()
+badges = do
+ mapM_ collect
+ [ (,) "license" "https://img.shields.io/badge/license-BSD3-brightgreen.svg"
+ , (,) "unix-build" "https://img.shields.io/travis/iconnect/regex.svg?label=Linux%2BmacOS"
+ , (,) "windows-build" "https://img.shields.io/appveyor/ci/engineerirngirisconnectcouk/regex.svg?label=Windows"
+ , (,) "coverage" "https://img.shields.io/coveralls/iconnect/regex.svg"
+ , (,) "build-status" "https://img.shields.io/travis/iconnect/regex.svg?label=Build%20Status"
+ , (,) "maintainers-contact" "https://img.shields.io/badge/email-maintainers%40regex.uk-blue.svg"
+ , (,) "feedback-contact" "https://img.shields.io/badge/email-feedback%40regex.uk-blue.svg"
+ ]
+ where
+ collect (nm,url) = do
+ putStrLn $ "updating badge: " ++ nm
+ simpleHttp url >>= LBS.writeFile (badge_fn nm)
+
+ badge_fn nm = "docs/badges/"++nm++".svg"
+\end{code}
+
+
+pages
+-----
+
+\begin{code}
+pages :: IO ()
+pages = do
+ prep_page MM_hackage "lib/md/index.md" "README.markdown"
+ prep_page MM_github "lib/md/index.md" "README.md"
+ mapM_ pandoc_page [minBound..maxBound]
+\end{code}
+
+\begin{code}
+data Page
+ = PG_index
+ | PG_about
+ | PG_contact
+ | PG_build_status
+ | PG_installation
+ | PG_tutorial
+ | PG_examples
+ | PG_roadmap
+ | PG_macros
+ | PG_directory
+ | PG_changelog
+ deriving (Bounded,Enum,Eq,Ord,Show)
+
+page_root :: Page -> String
+page_root = map tr . drop 3 . show
+ where
+ tr '_' = '-'
+ tr c = c
+
+page_master_file, page_docs_file :: Page -> FilePath
+page_master_file pg = "lib/md/" ++ page_root pg ++ ".md"
+page_docs_file pg = "docs/" ++ page_root pg ++ ".html"
+
+page_address :: Page -> LBS.ByteString
+page_address = LBS.pack . page_root
+
+page_title :: Page -> LBS.ByteString
+page_title pg = case pg of
+ PG_index -> "Home"
+ PG_about -> "About"
+ PG_contact -> "Contact"
+ PG_build_status -> "Build Status"
+ PG_installation -> "Installation"
+ PG_tutorial -> "Tutorial"
+ PG_examples -> "Examples"
+ PG_roadmap -> "Roadmap"
+ PG_macros -> "Macro Tables"
+ PG_directory -> "Directory"
+ PG_changelog -> "Change Log"
+\end{code}
+
+\begin{code}
+pandoc_page :: Page -> IO ()
+pandoc_page pg = do
+ mt_lbs <- LBS.readFile $ page_master_file pg
+ (hdgs,md_lbs) <- prep_page' MM_pandoc mt_lbs
+ LBS.writeFile "tmp/metadata.markdown" $ LBS.unlines ["---","title: "<>page_title pg,"---"]
+ LBS.writeFile "tmp/heading.markdown" $ page_heading pg
+ LBS.writeFile "tmp/page_pre_body.html" $ mk_pre_body_html pg hdgs
+ LBS.writeFile "tmp/page_pst_body.html" pst_body_html
+ LBS.writeFile "tmp/page.markdown" md_lbs
+ fmap (const ()) $
+ SH.shelly $ SH.verbosely $
+ SH.run "pandoc"
+ [ "-f", "markdown+grid_tables+autolink_bare_uris"
+ , "-t", "html5"
+ , "-T", "regex"
+ , "-s"
+ , "-H", "lib/favicons.html"
+ , "-B", "tmp/page_pre_body.html"
+ , "-A", "tmp/page_pst_body.html"
+ , "-c", "lib/styles.css"
+ , "-o", T.pack $ page_docs_file pg
+ , "tmp/metadata.markdown"
+ , "tmp/heading.markdown"
+ , "tmp/page.markdown"
+ ]
+
+data Heading =
+ Heading
+ { _hdg_id :: LBS.ByteString
+ , _hdg_title :: LBS.ByteString
+ }
+ deriving (Show)
+
+data MarkdownMode
+ = MM_github
+ | MM_hackage
+ | MM_pandoc
+ deriving (Eq,Show)
+
+page_heading :: Page -> LBS.ByteString
+page_heading PG_index = ""
+page_heading pg =
+ "Home » **"<>page_title pg<>"**
\n" + +prep_page :: MarkdownMode -> FilePath -> FilePath -> IO () +prep_page mmd in_fp out_fp = do + lbs <- LBS.readFile in_fp + (_,lbs') <- prep_page' mmd lbs + LBS.writeFile out_fp lbs' + +prep_page' :: MarkdownMode -> LBS.ByteString -> IO ([Heading],LBS.ByteString) +prep_page' mmd lbs = do + rf_h <- newIORef [] + rf_t <- newIORef False + lbs1 <- sed' (scr rf_h rf_t) =<< include lbs + lbs2 <- fromMaybe "" <$> fin_task_list' mmd rf_t + hdgs <- reverse <$> readIORef rf_h + return (hdgs,lbs1<>lbs2) + where + scr rf_h rf_t = Select + [ (,) [re|^%heading#${ide}(@{%id}) +${ttl}([^ ].*)$|] $ EDIT_fun TOP $ heading mmd rf_t rf_h + , (,) [re|^- \[ \] +${itm}(.*)$|] $ EDIT_fun TOP $ task_list mmd rf_t False + , (,) [re|^- \[[Xx]\] +${itm}(.*)$|] $ EDIT_fun TOP $ task_list mmd rf_t True + , (,) [re|^.*$|] $ EDIT_fun TOP $ fin_task_list mmd rf_t + ] + +heading :: MarkdownMode + -> IORef Bool + -> IORef [Heading] + -> LineNo + -> Match LBS.ByteString + -> Location + -> Capture LBS.ByteString + -> IO (Maybe LBS.ByteString) +heading mmd rf_t rf_h _ mtch _ _ = do + lbs <- fromMaybe "" <$> fin_task_list' mmd rf_t + modifyIORef rf_h (Heading ide ttl:) + return $ Just $ lbs<>h2 + where + h2 = case mmd of + MM_github -> "## "<>ttl + MM_hackage -> "## "<>ttl + MM_pandoc -> ""<>ttl<>"
" + + ide = mtch !$$ [cp|ide|] + ttl = mtch !$$ [cp|ttl|] + +mk_pre_body_html :: Page -> [Heading] -> LBS.ByteString +mk_pre_body_html pg hdgs = hdr <> LBS.concat (map nav [minBound..maxBound]) <> ftr + where + hdr :: LBS.ByteString + hdr = [here|
+
+|]
+\end{code}
+
+
+Task Lists
+----------
+
+\begin{code}
+-- | replacement function to convert GFM task list line into HTML if we
+-- aren't writing GFM (i.e., generating markdown for GitHub)
+task_list :: MarkdownMode -- ^ what flavour of md are we generating
+ -> IORef Bool -- ^ will contain True iff we have already entered a task list
+ -> Bool -- ^ true if this is a checjed line
+ -> LineNo -- ^ line no of the replacement redex (unused)
+ -> Match LBS.ByteString -- ^ the matched task-list line
+ -> Location -- ^ which match and capure (unused)
+ -> Capture LBS.ByteString -- ^ the capture weare replacing (unsuded)
+ -> IO (Maybe LBS.ByteString) -- ^ the replacement text, or Nothing to indicate no change to this line
+task_list mmd rf chk _ mtch _ _ =
+ case mmd of
+ MM_github -> return Nothing
+ MM_hackage -> return $ Just $ " "<>cb<>" "<>itm<>"\n"
+ MM_pandoc -> do
+ in_tl <- readIORef rf
+ writeIORef rf True
+ return $ tl_line in_tl chk
+ where
+ tl_line in_tl enbl = Just $ LBS.concat
+ [ if in_tl then "" else "
+
+
+ |] <> branding <> [here|
+
+
+
+ 
Code
+
+ Code
+
+ 
Issues
+
+
+ Issues
+
+
+ 
+
+
+
+
+
+
+ 
+
+
+
+
+
+
+
+
+
+
+
+
+
+ 
+
+
+
+
+
+
+|]
+
+pst_body_html :: LBS.ByteString
+pst_body_html = [here|
+ - \n"
+ , "
- " + , "" + , itm + , " " + ] + + cb = if chk then "☒" else "☐" + + itm = mtch !$$ [cp|itm|] + +-- | replacement function used for 'other' lines -- terminate any task +-- list that was being generated +fin_task_list :: MarkdownMode -- ^ what flavour of md are we generating + -> IORef Bool -- ^ will contain True iff we have already entered a task list + -> LineNo -- ^ line no of the replacement redex (unused) + -> Match LBS.ByteString -- ^ the matched task-list line + -> Location -- ^ which match and capure (unused) + -> Capture LBS.ByteString -- ^ the capture weare replacing (unsuded) + -> IO (Maybe LBS.ByteString) -- ^ the replacement text, or Nothing to indicate no change to this line +fin_task_list mmd rf_t _ mtch _ _ = + fmap (<>matchSource mtch) <$> fin_task_list' mmd rf_t + +-- | close any task list being processed, returning the closing text +-- as necessary +fin_task_list' :: MarkdownMode -- ^ what flavour of md are we generating + -> IORef Bool -- ^ will contain True iff we have already entered a task list + -> IO (Maybe LBS.ByteString) -- ^ task-list closure HTML, if task-list HTML needs closing +fin_task_list' mmd rf = do + in_tl <- readIORef rf + writeIORef rf False + case mmd==MM_github || not in_tl of + True -> return Nothing + False -> return $ Just $ "
"
+ -- , " " <> branding
+ -- , "
"
+ [ ""
+ , " "
+ , "
"
+ , ""
+ ]
+
+ ft = LBS.concat
+ [ "
"
+ ]
+
+ repo_url = LBS.concat
+ [ "https://github.com/iconnect/regex/blob/master/"
+ , LBS.pack $ T.unpack repo_path
+ ]
+\end{code}
+
+
+simple include processor
+------------------------
+
+\begin{code}
+include :: LBS.ByteString -> IO LBS.ByteString
+include = sed' $ Select
+ [ (,) [re|^%include ${file}(@{%string})$|] $ EDIT_fun TOP incl
+ , (,) [re|^.*$|] $ EDIT_fun TOP $ \_ _ _ _->return Nothing
+ ]
+ where
+ incl _ mtch _ _ = Just <$> LBS.readFile (prs_s $ mtch !$$ [cp|file|])
+ prs_s = maybe (error "include") T.unpack . parseString
+\end{code}
+
+
+branding
+--------
+
+\begin{code}
+branding :: LBS.ByteString
+branding = [here|[re|${gex}(.*)|]|]
+\end{code}
+
+
+testing
+-------
+
+\begin{code}
+test :: IO ()
+test = do
+ dm <- docMode
+ test_pp "pp-doc" (loop dm) "data/pp-test.lhs" "data/pp-result-doc.lhs"
+ gm <- genMode
+ test_pp "pp-gen" (loop gm) "data/pp-test.lhs" "data/pp-result-gen.lhs"
+ putStrLn "tests passed"
+\end{code}
diff --git a/examples/re-tests.lhs b/examples/re-tests.lhs
--- a/examples/re-tests.lhs
+++ b/examples/re-tests.lhs
@@ -1,17 +1,21 @@
\begin{code}
+{-# LANGUAGE NoImplicitPrelude #-}
{-# LANGUAGE RecordWildCards #-}
{-# LANGUAGE OverloadedStrings #-}
{-# LANGUAGE QuasiQuotes #-}
+{-# LANGUAGE MultiParamTypeClasses #-}
+{-# LANGUAGE FlexibleInstances #-}
{-# LANGUAGE TemplateHaskell #-}
+{-# OPTIONS_GHC -fno-warn-orphans #-}
module Main (main) where
-import Control.Applicative
import Control.Exception
import Data.Array
import qualified Data.ByteString.Char8 as B
import qualified Data.ByteString.Lazy.Char8 as LBS
import qualified Data.Foldable as F
+import qualified Data.HashMap.Strict as HM
import Data.Maybe
import Data.Monoid
import qualified Data.Sequence as S
@@ -19,9 +23,14 @@
import qualified Data.Text as T
import qualified Data.Text.Lazy as LT
import Language.Haskell.TH.Quote
+import Prelude.Compat
+import Test.SmallCheck.Series
import Test.Tasty
import Test.Tasty.HUnit
+import Test.Tasty.SmallCheck as SC
+import Text.Heredoc
import qualified Text.Regex.PCRE as PCRE_
+
import qualified Text.Regex.TDFA as TDFA_
import Text.RE
import Text.RE.Internal.NamedCaptures
@@ -65,11 +74,11 @@
where
tc rty m_env =
testCase (show rty) $ do
- dumpMacroTable "prelude" rty m_env
+ dumpMacroTable "macros" rty m_env
assertBool "testMacroEnv" =<< testMacroEnv "prelude" rty m_env
-str, str' :: String
-str = "a bbbb aa b"
+str_, str' :: String
+str_ = "a bbbb aa b"
str' = "foo"
regex_, regex_alt :: RE
@@ -158,35 +167,35 @@
parsing_tests = testGroup "Parsing"
[ testCase "complete check (matchM/ByteString)" $ do
r <- compileRegex () $ reSource regex_
- assertEqual "Match" (B.pack <$> regex_str_match) $ B.pack str ?=~ r
+ assertEqual "Match" (B.pack <$> regex_str_match) $ B.pack str_ ?=~ r
, testCase "matched (matchM/Text)" $ do
r <- compileRegex () $ reSource regex_
- assertEqual "matched" True $ matched $ T.pack str ?=~ r
+ assertEqual "matched" True $ matched $ T.pack str_ ?=~ r
]
core_tests :: TestTree
core_tests = testGroup "Match"
[ testCase "text (=~~Text.Lazy)" $ do
- txt <- LT.pack str =~~ [re|(a+) (b+)|] :: IO (LT.Text)
+ txt <- LT.pack str_ =~~ [re|(a+) (b+)|] :: IO (LT.Text)
assertEqual "text" txt "a bbbb"
, testCase "multi (=~~/String)" $ do
- let sm = str =~ regex_ :: Match String
+ let sm = str_ =~ regex_ :: Match String
m = capture [cp|0|] sm
assertEqual "captureSource" "a bbbb aa b" $ captureSource m
assertEqual "capturedText" "a bbbb" $ capturedText m
assertEqual "capturePrefix" "" $ capturePrefix m
assertEqual "captureSuffix" " aa b" $ captureSuffix m
, testCase "complete (=~~/ByteString)" $ do
- mtch <- B.pack str =~~ regex_ :: IO (Match B.ByteString)
+ mtch <- B.pack str_ =~~ regex_ :: IO (Match B.ByteString)
assertEqual "Match" mtch $ B.pack <$> regex_str_match
, testCase "complete (all,String)" $ do
- let mtchs = str =~ regex_ :: Matches String
+ let mtchs = str_ =~ regex_ :: Matches String
assertEqual "Matches" mtchs regex_str_matches
, testCase "complete (all,reg_alt)" $ do
- let mtchs = str =~ regex_alt :: Matches String
+ let mtchs = str_ =~ regex_alt :: Matches String
assertEqual "Matches" mtchs regex_alt_str_matches
, testCase "complete (=~~,all)" $ do
- mtchs <- str =~~ regex_ :: IO (Matches String)
+ mtchs <- str_ =~~ regex_ :: IO (Matches String)
assertEqual "Matches" mtchs regex_str_matches
, testCase "fail (all)" $ do
let mtchs = str' =~ regex_ :: Matches String
@@ -196,27 +205,27 @@
replace_tests :: TestTree
replace_tests = testGroup "Replace"
[ testCase "String/single" $ do
- let m = str =~ regex_ :: Match String
+ let m = str_ =~ regex_ :: Match String
r = replaceCaptures' ALL fmt m
assertEqual "replaceCaptures'" r "(0:0:(0:1:a) (0:2:bbbb)) aa b"
, testCase "String/alt" $ do
- let ms = str =~ regex_ :: Matches String
+ let ms = str_ =~ regex_ :: Matches String
r = replaceAllCaptures' ALL fmt ms
chk r
, testCase "String" $ do
- let ms = str =~ regex_ :: Matches String
+ let ms = str_ =~ regex_ :: Matches String
r = replaceAllCaptures' ALL fmt ms
chk r
, testCase "ByteString" $ do
- let ms = B.pack str =~ regex_ :: Matches B.ByteString
+ let ms = B.pack str_ =~ regex_ :: Matches B.ByteString
r = replaceAllCaptures' ALL fmt ms
chk r
, testCase "LBS.ByteString" $ do
- let ms = LBS.pack str =~ regex_ :: Matches LBS.ByteString
+ let ms = LBS.pack str_ =~ regex_ :: Matches LBS.ByteString
r = replaceAllCaptures' ALL fmt ms
chk r
, testCase "Seq Char" $ do
- let ms = S.fromList str =~ regex_ :: Matches (S.Seq Char)
+ let ms = S.fromList str_ =~ regex_ :: Matches (S.Seq Char)
f = \_ (Location i j) Capture{..} -> Just $ S.fromList $
"(" <> show i <> ":" <> show_co j <> ":" <>
F.toList capturedText <> ")"
@@ -224,11 +233,11 @@
assertEqual "replaceAllCaptures'" r $
S.fromList "(0:0:(0:1:a) (0:2:bbbb)) (1:0:(1:1:aa) (1:2:b))"
, testCase "Text" $ do
- let ms = T.pack str =~ regex_ :: Matches T.Text
+ let ms = T.pack str_ =~ regex_ :: Matches T.Text
r = replaceAllCaptures' ALL fmt ms
chk r
, testCase "LT.Text" $ do
- let ms = LT.pack str =~ regex_ :: Matches LT.Text
+ let ms = LT.pack str_ =~ regex_ :: Matches LT.Text
r = replaceAllCaptures' ALL fmt ms
chk r
]
@@ -438,4 +447,76 @@
tdfa_prelude_macros :: [PreludeMacro]
tdfa_prelude_macros = [minBound..maxBound]
+\end{code}
+
+
+Testing : FormatToken/Scan Properties
+-------------------------------------
+
+\begin{code}
+namedCapturesTestTree :: TestTree
+namedCapturesTestTree = localOption (SmallCheckDepth 4) $
+ testGroup "NamedCaptures"
+ [ formatScanTestTree
+ , analyseTokensTestTree
+ ]
+\end{code}
+
+\begin{code}
+instance Monad m => Serial m Token
+\end{code}
+
+
+Testing : FormatToken/Scan Properties
+-------------------------------------
+
+\begin{code}
+formatScanTestTree :: TestTree
+formatScanTestTree =
+ testGroup "FormatToken/Scan Properties"
+ [ localOption (SmallCheckDepth 4) $
+ SC.testProperty "formatTokens == formatTokens0" $
+ \tks -> formatTokens tks == formatTokens0 tks
+ , localOption (SmallCheckDepth 4) $
+ SC.testProperty "scan . formatTokens' idFormatTokenOptions == id" $
+ \tks -> all validToken tks ==>
+ scan (formatTokens' idFormatTokenOptions tks) == tks
+ ]
+\end{code}
+
+
+Testing : Analysing [Token] Unit Tests
+--------------------------------------
+
+\begin{code}
+analyseTokensTestTree :: TestTree
+analyseTokensTestTree =
+ testGroup "Analysing [Token] Unit Tests"
+ [ tc [here|foobar|] []
+ , tc [here||] []
+ , tc [here|$([0-9]{4})|] []
+ , tc [here|${x}()|] [(1,"x")]
+ , tc [here|${}()|] []
+ , tc [here|${}()${foo}()|] [(2,"foo")]
+ , tc [here|${x}(${y()})|] [(1,"x")]
+ , tc [here|${x}(${y}())|] [(1,"x"),(2,"y")]
+ , tc [here|${a}(${b{}())|] [(1,"a")]
+ , tc [here|${y}([0-9]{4})-${m}([0-9]{2})-${d}([0-9]{2})|] [(1,"y"),(2,"m"),(3,"d")]
+ , tc [here|@$(@|\{${name}([^{}]+)\})|] [(2,"name")]
+ , tc [here|${y}[0-9]{4}|] []
+ , tc [here|${}([0-9]{4})|] []
+ ]
+ where
+ tc s al =
+ testCase s $ assertEqual "CaptureNames"
+ (xnc s)
+ (HM.fromList
+ [ (CaptureName $ T.pack n,CaptureOrdinal i)
+ | (i,n)<-al
+ ]
+ )
+
+ xnc = either oops fst . extractNamedCaptures
+ where
+ oops = error "analyseTokensTestTree: unexpected parse failure"
\end{code}
diff --git a/examples/re-tutorial.lhs b/examples/re-tutorial.lhs
--- a/examples/re-tutorial.lhs
+++ b/examples/re-tutorial.lhs
@@ -29,7 +29,7 @@
should be enabled.
\begin{code}
-{-# LANGUAGE QuasiQuotes #-}
+{-# LANGUAGE QuasiQuotes #-}
\end{code}
Use this command to configure ghci accordingly (not necessary if you
@@ -38,21 +38,22 @@
:seti -XQuasiQuotes
```
-[Overloaded string
-literals](https://www.schoolofhaskell.com/school/to-infinity-and-beyond/pick-of-the-week/guide-to-ghc-extensions/basic-syntax-extensions#overloadedstrings)
-are usually enabled in modern Haskell and we do so here.
-\begin{code}
-{-# LANGUAGE OverloadedStrings #-}
-\end{code}
-
Because we are mimicking the REPL in this tutorial we will leave off the type
signatures on the example calculations and disable the compiler
warnings about missing type signatures.
\begin{code}
-{-# OPTIONS_GHC -fno-warn-missing-signatures #-}
+{-# OPTIONS_GHC -fno-warn-missing-signatures #-}
\end{code}
+This pragma is a just technical pragma, combined with the
+`Prelude.Compat` import below used to avoid certain warnings while
+comiling against multiple versions of the compiler. It can be safely
+ignored.
+\begin{code}
+{-# LANGUAGE NoImplicitPrelude #-}
+\end{code}
+
Loading Up: The Imports
-----------------------
@@ -107,9 +108,11 @@
import Text.Printf
\end{code}
-And finally we will import a small specail toolkit will be used to help
+And finally we a special edition of the prelude (see the commentary for
+the pragma section above) and a small specail toolkit will be used to help
manage the example calculations.
\begin{code}
+import Prelude.Compat
import TestKit
\end{code}
This allows simple calculations to be defined stylistically
@@ -613,7 +616,7 @@
that matches one of the REs, substituting the text returned from the action
in the output stream. (Used in the [include](re-include.html),
[gen-modules](re-gen-modules.html),
- [log-processor](re-nginx-log-processor.html) and [tutorial-pp](re-pp)
+ [log-processor](re-nginx-log-processor.html) and [tutorial-pp](re-prep)
examples.)
@@ -691,9 +694,9 @@
package's website uses the following 'gen_all' diriver which uses
REs to analyse file paths.
-%include "examples/re-pp.lhs" "^gen_all ::"
+%include "examples/re-prep.lhs" "^gen_all ::"
-See [examples/re-pp.lhs](re-pp.html)
+See [examples/re-prep.lhs](re-prep.html)
Example: parsing RE macros
@@ -751,7 +754,7 @@
Just $ replaceAllCaptures' TOP phi $
tpl *=~ [re|\$${arg}(\$|[0-9]+|\{${name}([^{}]+)\})|]
where
- phi t_mtch _ _ = case captureMaybe [cp|name|] t_mtch of
+ phi t_mtch _ _ = case t_mtch !$? [cp|name|] of
Just cap -> this $ CID_name $ CaptureName txt
where
txt = T.pack $ capturedText cap
@@ -761,7 +764,7 @@
where
t = capturedText $ capture [cp|arg|] t_mtch
- this cid = capturedText <$> captureMaybe cid mtch
+ this cid = capturedText <$> mtch !$? cid
my_replace :: RE -> Template-> String -> String
my_replace rex tpl src = replaceAllCaptures' TOP (parse_tpl_ tpl) $ src *=~ rex
@@ -813,7 +816,7 @@
The preprocessor that converts this literate Haskell program into a web
page and a test suite that makes plenty of use of regex is in
-[examples/re-pp.lhs](re-pp.html).
+[examples/re-prep.lhs](re-prep.html).
Example: gen-modules
diff --git a/regex.cabal b/regex.cabal
--- a/regex.cabal
+++ b/regex.cabal
@@ -1,6 +1,6 @@
Name: regex
-Version: 0.0.0.2
-Synopsis: A Regular Expression Toolkit for regex-base
+Version: 0.1.0.0
+Synopsis: Toolkit for regex-base
Description: A Regular Expression Toolkit for regex-base with
Compile-time checking of RE syntax, data types for
matches and captures, a text replacement toolkit,
@@ -18,23 +18,25 @@
Stability: rfc
bug-reports: https://github.com/iconnect/regex/issues
-extra-source-files:
- README.md
+Extra-Source-Files:
+ README.markdown
changelog
-Cabal-version: >= 1.16
+Cabal-Version: >= 1.16
-Source-repository head
+Source-Repository head
type: git
location: https://github.com/iconnect/regex.git
Source-Repository this
Type: git
Location: https://github.com/iconnect/regex.git
- Tag: 0.0.0.2
+ Tag: 0.1.0.0
+
Library
- Exposed-modules:
+ Hs-Source-Dirs: .
+ Exposed-Modules:
Text.RE
Text.RE.Capture
Text.RE.CaptureID
@@ -66,326 +68,491 @@
Text.RE.Tools.Lex
Text.RE.Tools.Sed
- Other-modules:
+ Other-Modules:
Text.RE.Internal.AddCaptureNames
+ Default-Language: Haskell2010
+ GHC-Options:
+ -Wall
+ -fwarn-tabs
+
Build-depends:
- array >= 0.4 ,
- bytestring >= 0.10.2.0 ,
- base >= 4 && < 5 ,
- containers >= 0.4 ,
- hashable >= 1.2.3.3 ,
- heredoc >= 0.2.0.0 ,
- regex-base >= 0.93.2 ,
- regex-tdfa >= 1.2.1 ,
- regex-tdfa-text >= 1.0.0.3 ,
- regex-pcre-builtin >= 0.94.4.8.8.35 ,
- smallcheck >= 1.1.1 ,
- tasty >= 0.10.1.2 ,
- tasty-hunit >= 0.9.2 ,
- tasty-smallcheck >= 0.8.0.1 ,
- template-haskell >= 2.7 ,
- transformers >= 0.2.2 ,
- text >= 1.2.1.3 ,
- time >= 1.5.0.1 ,
- unordered-containers >= 0.2.5.1
+ array >= 0.4
+ , base >= 4 && < 5
+ , base-compat >= 0.6.0
+ , bytestring >= 0.10.2.0
+ , containers >= 0.4
+ , hashable >= 1.2.3.3
+ , heredoc >= 0.2.0.0
+ , regex-base >= 0.93.2
+ , regex-pcre-builtin >= 0.94.4.8.8.35
+ , regex-tdfa >= 1.2.0
+ , regex-tdfa-text >= 1.0.0.3
+ , template-haskell >= 2.7
+ , text >= 1.2.0.6
+ , time >= 1.4.2
+ , time-locale-compat >= 0.1.0.1
+ , transformers >= 0.2.2
+ , unordered-containers >= 0.2.5.1
+
+Executable re-gen-cabals
+ Hs-Source-Dirs: examples
+
+ Main-Is: re-gen-cabals.lhs
+
+ Other-Modules:
+ TestKit
+
Default-Language: Haskell2010
GHC-Options:
- -Wall
- -fwarn-tabs
+ -Wall
+ -fwarn-tabs
+ Build-depends:
+ regex
+ , array >= 0.4
+ , base >= 4 && < 5
+ , base-compat >= 0.6.0
+ , bytestring >= 0.10.2.0
+ , containers >= 0.4
+ , directory >= 1.2.1.0
+ , regex-base >= 0.93.2
+ , regex-tdfa >= 1.2.0
+ , shelly >= 1.6.1.2
+ , text >= 1.2.0.6
+
+Test-Suite re-gen-cabals-test
+ type: exitcode-stdio-1.0
+ Hs-Source-Dirs: examples
+
+ Main-Is: re-gen-cabals.lhs
+
+ Other-Modules:
+ TestKit
+
+ Default-Language: Haskell2010
+ GHC-Options:
+ -Wall
+ -fwarn-tabs
+
+ Build-depends:
+ regex
+ , array >= 0.4
+ , base >= 4 && < 5
+ , base-compat >= 0.6.0
+ , bytestring >= 0.10.2.0
+ , containers >= 0.4
+ , directory >= 1.2.1.0
+ , regex-base >= 0.93.2
+ , regex-tdfa >= 1.2.0
+ , shelly >= 1.6.1.2
+ , text >= 1.2.0.6
+
+
+
Executable re-gen-modules
Hs-Source-Dirs: examples
- Main-is: re-gen-modules.lhs
+ Main-Is: re-gen-modules.lhs
Default-Language: Haskell2010
GHC-Options:
- -Wall
- -fwarn-tabs
+ -Wall
+ -fwarn-tabs
Build-depends:
- regex ,
- array >= 0.4 ,
- bytestring >= 0.10.2.0 ,
- base >= 4 ,
- directory >= 1.2.1.0 ,
- regex-base >= 0.93.2 ,
- regex-tdfa >= 1.2.0 ,
- shelly >= 1.6.1.2 ,
- text >= 1.2.1.3
+ regex
+ , array >= 0.4
+ , base >= 4 && < 5
+ , base-compat >= 0.6.0
+ , bytestring >= 0.10.2.0
+ , directory >= 1.2.1.0
+ , regex-base >= 0.93.2
+ , regex-tdfa >= 1.2.0
+ , shelly >= 1.6.1.2
+ , text >= 1.2.0.6
Test-Suite re-gen-modules-test
type: exitcode-stdio-1.0
Hs-Source-Dirs: examples
- Main-is: re-gen-modules.lhs
+ Main-Is: re-gen-modules.lhs
Default-Language: Haskell2010
GHC-Options:
- -Wall
- -fwarn-tabs
+ -Wall
+ -fwarn-tabs
Build-depends:
- regex ,
- array >= 0.4 ,
- bytestring >= 0.10.2.0 ,
- base >= 4 && < 5 ,
- directory >= 1.2.1.0 ,
- regex-base >= 0.93.2 ,
- regex-tdfa >= 1.2.0 ,
- shelly >= 1.6.1.2 ,
- text >= 1.2.1.3
+ regex
+ , array >= 0.4
+ , base >= 4 && < 5
+ , base-compat >= 0.6.0
+ , bytestring >= 0.10.2.0
+ , directory >= 1.2.1.0
+ , regex-base >= 0.93.2
+ , regex-tdfa >= 1.2.0
+ , shelly >= 1.6.1.2
+ , text >= 1.2.0.6
+
Executable re-include
Hs-Source-Dirs: examples
- Main-is: re-include.lhs
+ Main-Is: re-include.lhs
Default-Language: Haskell2010
GHC-Options:
- -Wall
- -fwarn-tabs
+ -Wall
+ -fwarn-tabs
- Other-modules:
+ Other-Modules:
TestKit
Build-depends:
- regex ,
- bytestring >= 0.10.2.0 ,
- base >= 4 && < 5 ,
- directory >= 1.2.1.0 ,
- shelly >= 1.6.1.2 ,
- text >= 1.2.1.3
+ regex
+ , base >= 4 && < 5
+ , base-compat >= 0.6.0
+ , bytestring >= 0.10.2.0
+ , directory >= 1.2.1.0
+ , shelly >= 1.6.1.2
+ , text >= 1.2.0.6
Test-Suite re-include-test
type: exitcode-stdio-1.0
Hs-Source-Dirs: examples
- Main-is: re-include.lhs
+ Main-Is: re-include.lhs
Default-Language: Haskell2010
GHC-Options:
- -Wall
- -fwarn-tabs
+ -Wall
+ -fwarn-tabs
- Other-modules:
+ Other-Modules:
TestKit
Build-depends:
- regex ,
- bytestring >= 0.10.2.0 ,
- base >= 4 ,
- directory >= 1.2.1.0 ,
- shelly >= 1.6.1.2 ,
- text >= 1.2.1.3
+ regex
+ , base >= 4 && < 5
+ , base-compat >= 0.6.0
+ , bytestring >= 0.10.2.0
+ , directory >= 1.2.1.0
+ , shelly >= 1.6.1.2
+ , text >= 1.2.0.6
+
+Executable re-nginx-log-processor
+ Hs-Source-Dirs: examples
+
+ Main-Is: re-nginx-log-processor.lhs
+
+ Default-Language: Haskell2010
+ GHC-Options:
+ -Wall
+ -fwarn-tabs
+
+ Build-depends:
+ regex
+ , array >= 0.4
+ , base >= 4 && < 5
+ , base-compat >= 0.6.0
+ , bytestring >= 0.10.2.0
+ , directory >= 1.2.1.0
+ , regex-base >= 0.93.2
+ , regex-tdfa >= 1.2.0
+ , shelly >= 1.6.1.2
+ , text >= 1.2.0.6
+ , time >= 1.4.2
+ , time-locale-compat >= 0.1.0.1
+ , transformers >= 0.2.2
+ , unordered-containers >= 0.2.5.1
+
+
Test-Suite re-nginx-log-processor-test
type: exitcode-stdio-1.0
Hs-Source-Dirs: examples
- Main-is: re-nginx-log-processor.lhs
+ Main-Is: re-nginx-log-processor.lhs
Default-Language: Haskell2010
GHC-Options:
- -Wall
- -fwarn-tabs
+ -Wall
+ -fwarn-tabs
Build-depends:
- regex ,
- array >= 0.4 ,
- bytestring >= 0.10.2.0 ,
- base >= 4 && < 5 ,
- directory >= 1.2.1.0 ,
- regex-base >= 0.93.2 ,
- regex-tdfa >= 1.2.0 ,
- shelly >= 1.6.1.2 ,
- text >= 1.2.1.3 ,
- time >= 1.5.0.1 ,
- transformers >= 0.3.0.0 ,
- unordered-containers >= 0.2.5.1
+ regex
+ , array >= 0.4
+ , base >= 4 && < 5
+ , base-compat >= 0.6.0
+ , bytestring >= 0.10.2.0
+ , directory >= 1.2.1.0
+ , regex-base >= 0.93.2
+ , regex-tdfa >= 1.2.0
+ , shelly >= 1.6.1.2
+ , text >= 1.2.0.6
+ , time >= 1.4.2
+ , time-locale-compat >= 0.1.0.1
+ , transformers >= 0.2.2
+ , unordered-containers >= 0.2.5.1
-Executable re-nginx-log-processor
+
+Executable re-prep
Hs-Source-Dirs: examples
- Main-is: re-nginx-log-processor.lhs
+ Main-Is: re-prep.lhs
Default-Language: Haskell2010
GHC-Options:
- -Wall
- -fwarn-tabs
+ -Wall
+ -fwarn-tabs
+ Other-Modules:
+ TestKit
+
Build-depends:
- regex ,
- array >= 0.4 ,
- bytestring >= 0.10.2.0 ,
- base >= 4 && < 5 ,
- directory >= 1.2.1.0 ,
- regex-base >= 0.93.2 ,
- regex-tdfa >= 1.2.0 ,
- shelly >= 1.6.1.2 ,
- text >= 1.2.1.3 ,
- time >= 1.5.0.1 ,
- transformers >= 0.3.0.0 ,
- unordered-containers >= 0.2.5.1
+ regex
+ , base >= 4 && < 5
+ , base-compat >= 0.6.0
+ , bytestring >= 0.10.2.0
+ , directory >= 1.2.1.0
+ , heredoc >= 0.2.0.0
+ , http-conduit >= 2.1.7.2
+ , shelly >= 1.6.1.2
+ , text >= 1.2.0.6
-Executable re-pp
+Test-Suite re-prep-test
+ type: exitcode-stdio-1.0
Hs-Source-Dirs: examples
- Main-is: re-pp.lhs
+ Main-Is: re-prep.lhs
Default-Language: Haskell2010
GHC-Options:
- -Wall
- -fwarn-tabs
+ -Wall
+ -fwarn-tabs
- Other-modules:
+ Other-Modules:
TestKit
Build-depends:
- regex ,
- bytestring >= 0.10.2.0 ,
- base >= 4 && < 5 ,
- directory >= 1.2.1.0 ,
- shelly >= 1.6.1.2 ,
- text >= 1.2.1.3
+ regex
+ , base >= 4 && < 5
+ , base-compat >= 0.6.0
+ , bytestring >= 0.10.2.0
+ , directory >= 1.2.1.0
+ , heredoc >= 0.2.0.0
+ , http-conduit >= 2.1.7.2
+ , shelly >= 1.6.1.2
+ , text >= 1.2.0.6
-Test-Suite re-pp-test
- type: exitcode-stdio-1.0
+
+Executable re-tests
Hs-Source-Dirs: examples
- Main-is: re-pp.lhs
+ Main-Is: re-tests.lhs
Default-Language: Haskell2010
GHC-Options:
- -Wall
- -fwarn-tabs
+ -Wall
+ -fwarn-tabs
- Other-modules:
+ Other-Modules:
TestKit
Build-depends:
- regex ,
- bytestring >= 0.10.2.0 ,
- base >= 4 && < 5 ,
- directory >= 1.2.1.0 ,
- shelly >= 1.6.1.2 ,
- text >= 1.2.1.3
+ regex
+ , array >= 0.4
+ , base >= 4 && < 5
+ , base-compat >= 0.6.0
+ , bytestring >= 0.10.2.0
+ , containers >= 0.4
+ , directory >= 1.2.1.0
+ , heredoc >= 0.2.0.0
+ , regex-base >= 0.93.2
+ , regex-pcre-builtin >= 0.94.4.8.8.35
+ , regex-tdfa >= 1.2.0
+ , regex-tdfa-text >= 1.0.0.3
+ , smallcheck >= 1.1.1
+ , tasty >= 0.10.1.2
+ , tasty-hunit >= 0.9.2
+ , tasty-smallcheck >= 0.8.0.1
+ , template-haskell >= 2.7
+ , text >= 1.2.0.6
+ , unordered-containers >= 0.2.5.1
-Test-Suite re-tests
+Test-Suite re-tests-test
type: exitcode-stdio-1.0
Hs-Source-Dirs: examples
- Main-is: re-tests.lhs
+ Main-Is: re-tests.lhs
Default-Language: Haskell2010
GHC-Options:
- -Wall
- -fwarn-tabs
+ -Wall
+ -fwarn-tabs
- Other-modules:
+ Other-Modules:
TestKit
Build-depends:
- regex ,
- array >= 0.4 ,
- bytestring >= 0.10.2.0 ,
- base >= 4 && < 5 ,
- containers >= 0.4 ,
- directory >= 1.2.1.0 ,
- regex-base >= 0.93.2 ,
- regex-tdfa >= 1.2.0 ,
- regex-tdfa-text >= 1.0.0.3 ,
- regex-pcre-builtin >= 0.94.4.8.8.35 ,
- tasty >= 0.10.1.2 ,
- tasty-hunit >= 0.9.2 ,
- template-haskell >= 2.7 ,
- text >= 1.2.1.3
+ regex
+ , array >= 0.4
+ , base >= 4 && < 5
+ , base-compat >= 0.6.0
+ , bytestring >= 0.10.2.0
+ , containers >= 0.4
+ , directory >= 1.2.1.0
+ , heredoc >= 0.2.0.0
+ , regex-base >= 0.93.2
+ , regex-pcre-builtin >= 0.94.4.8.8.35
+ , regex-tdfa >= 1.2.0
+ , regex-tdfa-text >= 1.0.0.3
+ , smallcheck >= 1.1.1
+ , tasty >= 0.10.1.2
+ , tasty-hunit >= 0.9.2
+ , tasty-smallcheck >= 0.8.0.1
+ , template-haskell >= 2.7
+ , text >= 1.2.0.6
+ , unordered-containers >= 0.2.5.1
+
Executable re-tutorial
- Hs-Source-Dirs: examples .
+ Hs-Source-Dirs: examples
- Main-is: re-tutorial.lhs
+ Main-Is: re-tutorial.lhs
- Other-modules:
+ Other-Modules:
TestKit
Default-Language: Haskell2010
Default-Extensions: QuasiQuotes
GHC-Options:
- -Wall
- -fwarn-tabs
+ -Wall
+ -fwarn-tabs
Build-depends:
- regex ,
- array >= 0.4 ,
- bytestring >= 0.10.2.0 ,
- base >= 4 && < 5 ,
- containers >= 0.4 ,
- directory >= 1.2.1.0 ,
- hashable >= 1.2.3.3 ,
- heredoc >= 0.2.0.0 ,
- regex-base >= 0.93.2 ,
- regex-tdfa >= 1.2.1 ,
- regex-tdfa-text >= 1.0.0.3 ,
- regex-pcre-builtin >= 0.94.4.8.8.35 ,
- shelly >= 1.6.1.2 ,
- smallcheck >= 1.1.1 ,
- tasty >= 0.10.1.2 ,
- tasty-hunit >= 0.9.2 ,
- tasty-smallcheck >= 0.8.0.1 ,
- template-haskell >= 2.7 ,
- transformers >= 0.2.2 ,
- text >= 1.2.1.3 ,
- time >= 1.5.0.1 ,
- unordered-containers >= 0.2.5.1
+ regex
+ , array >= 0.4
+ , base >= 4 && < 5
+ , base-compat >= 0.6.0
+ , bytestring >= 0.10.2.0
+ , containers >= 0.4
+ , directory >= 1.2.1.0
+ , hashable >= 1.2.3.3
+ , heredoc >= 0.2.0.0
+ , regex-base >= 0.93.2
+ , regex-pcre-builtin >= 0.94.4.8.8.35
+ , regex-tdfa >= 1.2.0
+ , regex-tdfa-text >= 1.0.0.3
+ , shelly >= 1.6.1.2
+ , smallcheck >= 1.1.1
+ , tasty >= 0.10.1.2
+ , tasty-hunit >= 0.9.2
+ , tasty-smallcheck >= 0.8.0.1
+ , template-haskell >= 2.7
+ , text >= 1.2.0.6
+ , time >= 1.4.2
+ , time-locale-compat >= 0.1.0.1
+ , transformers >= 0.2.2
+ , unordered-containers >= 0.2.5.1
+
Test-Suite re-tutorial-test
type: exitcode-stdio-1.0
- Hs-Source-Dirs: examples .
+ Hs-Source-Dirs: examples
- Main-is: re-tutorial.lhs
+ Main-Is: re-tutorial.lhs
- Other-modules:
+ Other-Modules:
TestKit
Default-Language: Haskell2010
Default-Extensions: QuasiQuotes
GHC-Options:
- -Wall
- -fwarn-tabs
+ -Wall
+ -fwarn-tabs
Build-depends:
- regex ,
- array >= 0.4 ,
- bytestring >= 0.10.2.0 ,
- base >= 4 && < 5 ,
- containers >= 0.4 ,
- directory >= 1.2.1.0 ,
- hashable >= 1.2.3.3 ,
- heredoc >= 0.2.0.0 ,
- regex-base >= 0.93.2 ,
- regex-tdfa >= 1.2.1 ,
- regex-tdfa-text >= 1.0.0.3 ,
- regex-pcre-builtin >= 0.94.4.8.8.35 ,
- shelly >= 1.6.1.2 ,
- smallcheck >= 1.1.1 ,
- tasty >= 0.10.1.2 ,
- tasty-hunit >= 0.9.2 ,
- tasty-smallcheck >= 0.8.0.1 ,
- template-haskell >= 2.7 ,
- transformers >= 0.2.2 ,
- text >= 1.2.1.3 ,
- time >= 1.5.0.1 ,
- unordered-containers >= 0.2.5.1
+ regex
+ , array >= 0.4
+ , base >= 4 && < 5
+ , base-compat >= 0.6.0
+ , bytestring >= 0.10.2.0
+ , containers >= 0.4
+ , directory >= 1.2.1.0
+ , hashable >= 1.2.3.3
+ , heredoc >= 0.2.0.0
+ , regex-base >= 0.93.2
+ , regex-pcre-builtin >= 0.94.4.8.8.35
+ , regex-tdfa >= 1.2.0
+ , regex-tdfa-text >= 1.0.0.3
+ , shelly >= 1.6.1.2
+ , smallcheck >= 1.1.1
+ , tasty >= 0.10.1.2
+ , tasty-hunit >= 0.9.2
+ , tasty-smallcheck >= 0.8.0.1
+ , template-haskell >= 2.7
+ , text >= 1.2.0.6
+ , time >= 1.4.2
+ , time-locale-compat >= 0.1.0.1
+ , transformers >= 0.2.2
+ , unordered-containers >= 0.2.5.1
+
+
+
+Test-Suite re-tutorial-os-test
+ type: exitcode-stdio-1.0
+ Hs-Source-Dirs: examples
+
+ Main-Is: re-tutorial.lhs
+
+ Other-Modules:
+ TestKit
+
+ Default-Language: Haskell2010
+ Extensions: OverloadedStrings
+ Default-Extensions: QuasiQuotes
+ GHC-Options:
+ -Wall
+ -fwarn-tabs
+
+ Build-depends:
+ regex
+ , array >= 0.4
+ , base >= 4 && < 5
+ , base-compat >= 0.6.0
+ , bytestring >= 0.10.2.0
+ , containers >= 0.4
+ , directory >= 1.2.1.0
+ , hashable >= 1.2.3.3
+ , heredoc >= 0.2.0.0
+ , regex-base >= 0.93.2
+ , regex-pcre-builtin >= 0.94.4.8.8.35
+ , regex-tdfa >= 1.2.0
+ , regex-tdfa-text >= 1.0.0.3
+ , shelly >= 1.6.1.2
+ , smallcheck >= 1.1.1
+ , tasty >= 0.10.1.2
+ , tasty-hunit >= 0.9.2
+ , tasty-smallcheck >= 0.8.0.1
+ , template-haskell >= 2.7
+ , text >= 1.2.0.6
+ , time >= 1.4.2
+ , time-locale-compat >= 0.1.0.1
+ , transformers >= 0.2.2
+ , unordered-containers >= 0.2.5.1
+
+
+-- Generated from lib/rgex-master.cabal with re-gen-cabals