Defining and Catching Exceptions - exception

I'm still getting the hang of haskell, and am trying to build my first "real" coding project, i.e. one where I'm going to including testing, write up documentation, etc. I'm sufficiently new to haskell where I know I don't have a lot of the knowledge of the language necessary so that everything I want to do is immediately within reach, but that's kind of the idea, so that the act of finishing will require that I touch most of the major pieces of the language.
Anyway, so the current issue I'm having is regarding throwing and catching exceptions in the language, something that I understand can be done with quite varied approaches. I have a function here, toLower:
toLower :: String -> String
toLower plaintext =
if (catch (nonAlpha plaintext) handlerNonAlpha)
then map charToLower plaintext
else exitFailure
Which will take a string, throw an exception and exit if the string includes any non-alpha characters (so if not A-Z or a-z), or if not convert the string to lowercase. So what I have for the nonAlpha function is:
--- detect non-alpha character - throw error if existant
data NonNumericException = NonNumException
instance Exception NonNumericException
handlerNonAlpha :: NonNumericException -> IO()
handlerNonAlpha ex =
putStrLn "Caught Exception: " ++ (show ex) ++ " - A non-alpha character was included in the plaintext."
nonAlpha :: String -> Bool
nonAlpha str =
let nonalphas = [x | x <- str, (ord x) < 65 || (90 < (ord x) && (ord x) < 97) || 123 < (ord x)]
in if (length nonalphas) == 0
then True
else throw NonNumException
As I said I'm pretty new to haskell, so I'm a little vague on how this data/instance structure works, but as I understand it I'm defining an parent NonNumericException, of which NonNumException is a child (and I could have more), and in the instance line defining them to be Exceptions. The catch structure, if it detects an exception (for instance, when one is thrown at the end of nonAlpha if there is a non-alpha character), then calls the handler.
So here are the compile errors that I get:
utilities.hs:61:3:
Couldn't match expected type `[Char]' with actual type `IO ()'
In the return type of a call of `putStrLn'
In the first argument of `(++)', namely
`putStrLn "Caught Exception: "'
In the expression:
putStrLn "Caught Exception: "
++
(show ex)
++ " - A non-alpha character was included in the plaintext."
utilities.hs:61:3:
Couldn't match expected type `IO ()' with actual type `[Char]'
In the expression:
putStrLn "Caught Exception: "
++
(show ex)
++ " - A non-alpha character was included in the plaintext."
In an equation for `handlerNonAlpha':
handlerNonAlpha ex
= putStrLn "Caught Exception: "
++
(show ex)
++ " - A non-alpha character was included in the plaintext."
utilities.hs:73:7:
Couldn't match expected type `Bool' with actual type `IO ()'
In the return type of a call of `catch'
In the expression: (catch (nonAlpha plaintext) handlerNonAlpha)
In the expression:
if (catch (nonAlpha plaintext) handlerNonAlpha) then
map charToLower plaintext
else
exitFailure
utilities.hs:73:14:
Couldn't match expected type `IO ()' with actual type `Bool'
In the return type of a call of `nonAlpha'
In the first argument of `catch', namely `(nonAlpha plaintext)'
In the expression: (catch (nonAlpha plaintext) handlerNonAlpha)
utilities.hs:75:8:
Couldn't match type `IO a0' with `[Char]'
Expected type: String
Actual type: IO a0
In the expression: exitFailure
In the expression:
if (catch (nonAlpha plaintext) handlerNonAlpha) then
map charToLower plaintext
else
exitFailure
In an equation for `toLower':
toLower plaintext
= if (catch (nonAlpha plaintext) handlerNonAlpha) then
map charToLower plaintext
else
exitFailure
So I guess my two question are, a) what's going wrong with the types for the handler (the line 61 errors), and b) how do I properly set the types for the functions that may throw an exception or exit with failure, but otherwise will return a bool or a string?
EDIT: I guess I should note. I do see the similarities between this question and a number of others that have been asked. Part of what I'm looking for that I don't see is a description of what the structures here are actually doing, and what is best practice and why.

What is best practice in Haskell is to leverage the awesome power of its type system to avoid needing to throw/catch exceptions for pure functions. There are cases where throwing an exception can actually make sense, but for something like your toLower function you can just choose to have a different return type. For example:
-- We can factor out our check for a non-alpha character
isNonAlpha :: Char -> Bool
isNonAlpha c = c' < 65 || (90 < c' && c' < 97) || 123 < c'
where c' = ord c
-- Why throw an exception? Just return False
hasNonAlpha :: String -> Bool
hasNonAlpha str = any isNonAlpha str
-- Renamed to not conflict with Data.Char.toLower
myToLower :: String -> Maybe String
myToLower plaintext =
if hasNonAlpha plaintext
then Nothing
else Just $ map toLower plaintext
Not only is this cleaner code, but now we don't have to worry about error handling at all, and someone else using your code won't get a nasty surprise. Instead, the notion of failure is encoded at the type level. To use this as an "error handling" mechanism, just work in the Maybe monad:
doSomething :: String -> String -> Maybe String
doSomething s1 s2 = do
s1Lower <- myToLower s1
s2Lower <- myToLower s2
return $ s1Lower ++ s2Lower
If either myToLower s1 or myToLower s2 returns Nothing, then doSomething will return Nothing. There is no ambiguity, no chance for an unhandled exception, and no crashing at runtime. Haskell exceptions themselves, those thrown by the function throw, must be caught by catch, which has to execute in the IO monad. Without the IO monad, you can't catch exceptions. In pure functions, you can always represent the concept of failure with another data type without having to resort to throw, so there is no need to over-complicate code with it.
Alternative, you could have even written myToLower monadically as
import Control.Monad
-- Other code
myToLower :: String -> Maybe String
myToLower plaintext = do
guard $ not $ hasNonAlpha plaintext
return $ map toLower plaintext
The guard from Control.Monad acts as a sort of filter for MonadPlus instances. Since Maybe is an instance of MonadPlus (as are lists), this gives us very simple code.
Or, if you want to pass around an error message:
type MyError = String
myToLower :: String -> Either MyError String
myToLower plaintext = if hasNonAlpha plaintext
then Left $ "The string " ++ plaintext ++ " has non-alpha character(s)"
else Right $ map toLower plaintext
Then you can change the type of doSomething to match:
doSomething :: String -> String -> Either MyError String
doSomething s1 s2 = do
s1Lower <- myToLower s1
s2Lower <- myToLower s2
return $ s1Lower ++ s2Lower
If you notice, the monadic syntax lets us change the type signature of our function without even having to change the code! Play around with this implementation to get a feel for how it works.

Learning about exceptions is useful, and they are great for handling exceptional circumstances.
The best place to read about exceptions is Simon Marlow's paper, An Extensible Dynamically-Typed Heirarchy of Exceptions. His book, Parallel Concurrent Programming in Haskell is another good resource on their use.
The following are a few comments on your question.
error on line 61
handlerNonAlpha :: NonNumericException -> IO()
handlerNonAlpha ex =
putStrLn "Caught Exception: " ++ (show ex) ++ ...
Function arguments are consumed eagerly in haskell. You'll have to modify this line as follows to perform string concatenation before calling putStrLn:
putStrLn $ "Caught Exception: " ++ (show ex) ++ ...
comment on nonAlpha
Exceptions can only be caught from an IO computation, and it's best to avoid throwing them from pure functions. Besides this, the problem with nonAlpha is that it claims to return a Bool, but actually returns either True or throws an exception. Why not just return False?
nonAlpha :: String -> Bool
nonAlpha str =
let nonalphas = [x | x <- str, (ord x) < 65 || (90 < (ord x) && (ord x) < 97) || 123 < (ord x)]
in if (length nonalphas) == 0
then True
else False
Pull your exception throwing code out of nonAlpha like so. The name of this function and its lack of return value indicate that it might throw an exception:
trapInvalid :: String -> IO ()
trapInvalid str = unless (nonAlpha str) $ throw NonNumException

Related

Json parser, incorrectly parsing string as a number

I'm still pretty new to Haskell and functional programming in general, so I'm writing a small program with Parsec to parse JSON and pretty print it as a means of learning basic concepts. This is what I have so far:
import Text.Parsec
import Text.Parsec.String
data JValue = JString String
| JNumber Double
| JBool Bool
| JNull
| JObject [(String, JValue)]
| JArray [JValue]
deriving (Eq, Ord, Show)
parseJString, parseJNumber, parseJBool, parseJNull :: Parser JValue
parseJString = do
str <- between (char '"') (char '"') (many (noneOf "\""))
return . JString $ str
parseJNumber = do
num <- many digit
return . JNumber . read $ num
parseJBool = do
val <- string "true" <|> string "false"
case val of
"true" -> return (JBool True)
"false" -> return (JBool False)
parseJNull = string "null" >> return JNull
parseJValue :: Parser JValue
parseJValue = parseJString
<|> parseJNumber
<|> parseJBool
<|> parseJNull
For now, I'm assuming that the numbers are integers. Individually, parseJString, parseJNumber, parseJBool, and parseJNull work as expected in ghci. Additionally, parseJValue correctly parses strings and numbers.
ghci> parse parseJString "test" "\"test input\""
Right (JString "test input")
ghci> parse parseJNumber "test" "345"
Right (JNumber 345.0)
ghci> parse parseJBool "test" "true"
Right (JBool True)
ghci> parse parseJNull "test" "null"
Right JNull
ghci> parse parseJValue "test" "\"jvalue test\""
Right (JString "jvalue test")
ghci> parse parseJValue "test" "789"
Right (JNumber 789.0)
parseJValue fails, however, when I try to parse true, false, or null, and I get an interesting error.
ghci> parse parseJValue "test" "true"
Right (JNumber *** Exception: Prelude.read: no parse
I get a successful parse, but the parse returns a JNumber followed by an error stating that Prelude.read failed. I feel like I'm missing some core concept in building my parsers, but I can't see where I've gone wrong. Also, am I making any beginner mistakes with my code, i.e. would any of this be considered "bad" haskell?
The problem is the usage of many in parseJNumber. It is also a valid parse, when no character of the following string is consumed ("many p applies the parser p zero or more times. [...]"). What you need is many1:
parseJNumber = do
num <- many1 (oneOf "0123456789")
return $ JNumber (read num :: Double)
Edit:
Somehow, I think your combination of (.) and ($) looks kind of weird. I use (.) when I can get rid of a function parameter (like in the usage of (>>=)) and ($) when I'm to lazy to write parentheses. In your function parseJString you do not need (.) in order to get the right binding precedences. (I did the same transformation in the code above.)
parseJString = do
str <- between (char '"') (char '"') (many (noneOf "\""))
return $ JString str
Additionally, you could eliminate code-repetition by refactoring parseJBool:
parseJBool = do
val <- string "true" <|> string "false"
return (case val of
"true" -> JBool True
"false" -> JBool False)
I would even rewrite the case-construct into a (total) local function:
parseJBool = (string "true" <|> string "false") >>= return . toJBool
where
-- there are only two possible strings to pattern match
toJBool "true" = JBool True
toJBool _ = JBool False
Last but not least, you can easily transform your other functions to use (>>=) instead of do-blocks.
-- additionally, you do not need an extra type signature for `read`
-- the constructor `JNumber` already infers the correct type
parseJNumber =
many1 (oneOf "0123456789") >>= return . JNumber . read
parseJString =
between (char '"') (char '"') (many (noneOf "\"")) >>= return . JString
You should try with many1 digit rather than many digit. A many succeeds on zero occurrences of the argument.
Compare:
ghci> parse (many digit) "test" "true"
Right ""
ghci> parse (many1 digit) "test" "true"
unexpected "t"
expecting digit
So in your case, parseJNumber within parseJValue will succeed and return an empty string which is then passed to read. But read "" :: Double fails.

returning two different types from one function

How can I return values of multiple types from a single function?
I want to do something like:
take x y | x == [] = "error : empty list"
| x == y = True
| otherwise = False
I have a background in imperative languages.
There is a type constructor called Either that lets you create a type that could be one of two types. It is often used for handling errors, just like in your example. You would use it like this:
take x y | x == [] = Left "error : empty list"
| x == y = Right True
| otherwise = Right False
The type of take would then be something like Eq a => [a] -> [a] -> Either String Bool. The convention with Either for error handling is that Left represents the error and Right represents the normal return type.
When you have an Either type, you can pattern match against it to see which value it contains:
case take x y of
Left errorMessage -> ... -- handle error here
Right result -> ... -- do what you normally would
There is several solutions to your problem, depending on your intention : do you want to make manifest in your type that your function can fail (and in this case do you want to return the cause of the failure, which may be unnecessary if there is only one mode of failure like here) or do you estimate that getting an empty list in this function shouldn't happen at all, and so want to fail immediately and by throwing an exception ?
So if you want to make explicit the possibility of failure in your type, you can use Maybe, to just indicate failure without explanation (eventually in your documentation) :
take :: (Eq a) => [a] -> [a] -> Maybe Bool
take [] _ = Nothing
take x y = x == y
Or Either to register the reason of the failure (note that Either would be the answer to "returning two types from one function" in general, though your code is more specific) :
take :: (Eq a) => [a] -> [a] -> Either String Bool
take [] _ = Left "Empty list"
take x y = Right $ x == y
Finally you can signal that this failure is completely abnormal and can't be handled locally :
take :: (Eq a) => [a] -> [a] -> Bool
take [] _ = error "Empty list"
take x y = x == y
Note that with this last way, the call site don't have to immediately handle the failure, in fact it can't, since exceptions can only be caught in the IO monad. With the first two ways, the call site have to be modified to handle the case of failure (and can), if only to itself call "error".
There is one final solution that allows the calling code to choose which mode of failure you want (using the failure package http://hackage.haskell.org/package/failure ) :
take :: (Failure String m, Eq a) => [a] -> [a] -> m Bool
take [] _ = failure "Empty list"
take x y = return $ x == y
This can mimics the Maybe and the Either solution, or you can use take as an IO Bool which will throw an exception if it fails. It can even works in a [Bool] context (returns an empty list in case of failure, which is sometimes useful).
You can use the error functions for exceptions:
take :: Eq a => [a] -> [a] -> Bool
take [] _ = error "empty list"
take x y = x == y
The three answers you've gotten so far (from Tikhon Jelvis, Jedai and Philipp) cover the three conventional ways of handling this sort of situation:
Use the error function signal an error. This is often frowned upon, however, because it makes it hard for programs that use your function to recover from the error.
Use Maybe to indicate the case where no Boolean answer can be produced.
Use Either, which is often used to do the same thing as Maybe, but can additionally include more information about the failure (Left "error : empty list").
I'd second the Maybe and Either approach, and add one tidbit (which is slightly more advanced, but you might want to get to eventually): both Maybe and Either a can be made into monads, and this can be used to write code that is neutral between the choice between those two. This blog post discusses eight different ways to tackle your problem, which includes the three mentioned above, a fourth one that uses the Monad type class to abstract the difference between Maybe and Either, and yet four others.
The blog entry is from 2007 so it looks a bit dated, but I managed to get #4 working this way:
{-# LANGUAGE FlexibleInstances #-}
take :: (Monad m, Eq a) => [a] -> [a] -> m Bool
take x y | x == [] = fail "error : empty list"
| x == y = return True
| otherwise = return False
instance Monad (Either String) where
return = Right
(Left err) >>= _ = Left err
(Right x) >>= f = f x
fail err = Left err
Now this take function works with both cases:
*Main> Main.take [1..3] [1..3] :: Maybe Bool
Just True
*Main> Main.take [1] [1..3] :: Maybe Bool
Just False
*Main> Main.take [] [1..3] :: Maybe Bool
Nothing
*Main> Main.take [1..3] [1..3] :: Either String Bool
Right True
*Main> Main.take [1] [1..3] :: Either String Bool
Right False
*Main> Main.take [] [1..3] :: Either String Bool
Left "error : empty list"
Though it's important to note that fail is controversial, so I anticipate reasonable objections to this approach. The use of fail here is not essential, though—it could be replaced with any function f :: (Monad m, ErrorClass m) => String -> m a such that f err is Nothing in Maybe and Left err in Either.

haskell word searching program development

hello I am making some word searching program
for example
when "text.txt" file contains "foo foos foor fo.. foo fool"
and search "foo"
then only number 2 printed
and search again and again
but I am haskell beginner
my code is here
:module +Text.Regex.Posix
putStrLn "type text file"
filepath <- getLine
data <- readFile filepath
--1. this makes <interactive>:1:1: parse error on input `data' how to fix it?
parsedData =~ "[^- \".,\n]+" :: [[String]]
--2. I want to make function and call it again and again
searchingFunc = do putStrLn "search for ..."
search <- getLine
result <- map (\each -> if each == search then count = count + 1) data
putStrLn result
searchingFunc
}
sorry for very very poor code
my development environment is Windows XP SP3 WinGhci 1.0.2
I started the haskell several hours ago sorry
thank you very much for reading!
edit: here's original scheme code
thanks!
#lang scheme/gui
(define count 0)
(define (search str)
(set! count 0)
(map (λ (each) (when (equal? str each) (set! count (+ count 1)))) data)
(send msg set-label (format "~a Found" count)))
(define path (get-file))
(define port (open-input-file path))
(define data '())
(define (loop [line (read-line port)])
(when (not (eof-object? line))
(set! data (append data
(regexp-match* #rx"[^- \".,\n]+" line)))
(loop)))
(loop)
(define (cb-txt t e) (search (send t get-value)))
(define f (new frame% (label "text search") (min-width 300)))
(define txt (new text-field% (label "type here to search") (parent f) (callback (λ (t e) (cb-txt t e)))))
(define msg (new message% (label "0Found ") (parent f)))
(send f show #t)
I should start by iterating what everyone would (and should) say: Start with a book like Real World Haskell! That said, I'll post a quick walkthrough of code that compiles, and hopefully does something close to what you originally intended. Comments are inline, and hopefully should illustrate some of the shortcomings of your approach.
import Text.Regex.Posix
-- Let's start by wrapping your first attempt into a 'Monadic Action'
-- IO is a monad, and hence we can sequence 'actions' (read as: functions)
-- together using do-notation.
attemptOne :: IO [[String]]
-- ^ type declaration of the function 'attemptOne'
-- read as: function returning value having type 'IO [[String]]'
attemptOne = do
putStrLn "type text file"
filePath <- getLine
fileData <- readFile filePath
putStrLn fileData
let parsed = fileData =~ "[^- \".,\n]+" :: [[String]]
-- ^ this form of let syntax allows us to declare that
-- 'wherever there is a use of the left-hand-side, we can
-- substitute it for the right-hand-side and get equivalent
-- results.
putStrLn ("The data after running the regex: " ++ concatMap concat parsed)
return parsed
-- ^ return is a monadic action that 'lifts' a value
-- into the encapsulating monad (in this case, the 'IO' Monad).
-- Here we show that given a search term (a String), and a body of text to
-- search in, we can return the frequency of occurrence of the term within the
-- text.
searchingFunc :: String -> [String] -> Int
searchingFunc term
= length . filter predicate
where
predicate = (==)term
-- ^ we use function composition (.) to create a new function from two
-- existing ones:
-- filter (drop any elements of a list that don't satisfy
-- our predicate)
-- length: return the size of the list
-- Here we build a wrapper-function that allows us to run our 'pure'
-- searchingFunc on an input of the form returned by 'attemptOne'.
runSearchingFunc :: String -> [[String]] -> [Int]
runSearchingFunc term parsedData
= map (searchingFunc term) parsedData
-- Here's an example of piecing everything together with IO actions
main :: IO ()
main = do
results <- attemptOne
-- ^ run our attemptOne function (representing IO actions)
-- and save the result
let searchResults = runSearchingFunc "foo" results
-- ^ us a 'let' binding to state that searchResults is
-- equivalent to running 'runSearchingFunc'
print searchResults
-- ^ run the IO action that prints searchResults
print (runSearchingFunc "foo" results)
-- ^ run the IO action that prints the 'definition'
-- of 'searchResults'; i.e. the above two IO actions
-- are equivalent.
return ()
-- as before, lift a value into the encapsulating Monad;
-- this time, we're lifting a value corresponding to 'null/void'.
To load this code, save it into a .hs file (I saved it into 'temp.hs'), and run the following from ghci. Note: the file 'f' contains a few input words:
*Main Text.Regex.Posix> :l temp.hs
[1 of 1] Compiling Main ( temp.hs, interpreted )
Ok, modules loaded: Main.
*Main Text.Regex.Posix> main
type text file
f
foo foos foor fo foo foo
The data after running the regex: foofoosfoorfofoofoo
[1,0,0,0,1,1]
[1,0,0,0,1,1]
There is a lot going on here, from do notation to Monadic actions, 'let' bindings to the distinction between pure and impure functions/values. I can't stress the value of learning the fundamentals from a good book!
Here is what I made of it. It doesn't does any error checking and is as basic as possible.
import Text.Regex.Posix ((=~))
import Control.Monad (when)
import Text.Printf (printf)
-- Calculates the number of matching words
matchWord :: String -> String -> Int
matchWord file word = length . filter (== word) . concat $ file =~ "[^- \".,\n]+"
getInputFile :: IO String
getInputFile = do putStrLn "Enter the file to search through:"
path <- getLine
readFile path -- Attention! No error checking here
repl :: String -> IO ()
repl file = do putStrLn "Enter word to search for (empty for exit):"
word <- getLine
when (word /= "") $
do print $ matchWord file word
repl file
main :: IO ()
main = do file <- getInputFile
repl file
Please start step by step. IO in Haskell is hard, so you shouldn't start with file manipulation. I would suggest to write a function that works properly on a given String. That way you can learn about syntax, pattern matching, list manipulation (maps, folds) and recursion without beeing distracted by the do notation (which kinda looks imperative, but isn't, and really needs a deeper understanding).
You should check out Learn you a Haskell or Real World Haskell to get a sound foundation. What you do now is just stumbling in the dark - which may work if you learn languages that are similar to the ones you know, but definitely not for Haskell.

Interaction between optimizations and testing for error calls

I have a function in a module that looks something like this:
module MyLibrary (throwIfNegative) where
throwIfNegative :: Integral i => i -> String
throwIfNegative n | n < 0 = error "negative"
| otherwise = "no worries"
I could of course return Maybe String or some other variant, but I think it's fair to say that it's a programmer error to call this function with a negative number so using error is justified here.
Now, since I like having my test coverage at 100% I want to have a test case that checks this behavior. I have tried this
import Control.Exception
import Test.HUnit
import MyLibrary
case_negative =
handleJust errorCalls (const $ return ()) $ do
evaluate $ throwIfNegative (-1)
assertFailure "must throw when given a negative number"
where errorCalls (ErrorCall _) = Just ()
main = runTestTT $ TestCase case_negative
and it sort of works, but it fails when compiling with optimizations:
$ ghc --make -O Test.hs
$ ./Test
### Failure:
must throw when given a negative number
Cases: 1 Tried: 1 Errors: 0 Failures: 1
I'm not sure what's happening here. It seems like despite my use of evaluate, the function does not get evaluated. Also, it works again if I do any of these steps:
Remove HUnit and call the code directly
Move throwIfNegative to the same module as the test case
Remove the type signature of throwIfNegative
I assume this is because it causes the optimizations to be applied differently. Any pointers?
Optimizations, strictness, and imprecise exceptions can be a bit tricky.
The easiest way to reproduce this problem above is with a NOINLINE on throwIfNegative (the function isn't being inlined across module boundaries either):
import Control.Exception
import Test.HUnit
throwIfNegative :: Int -> String
throwIfNegative n | n < 0 = error "negative"
| otherwise = "no worries"
{-# NOINLINE throwIfNegative #-}
case_negative =
handleJust errorCalls (const $ return ()) $ do
evaluate $ throwIfNegative (-1)
assertFailure "must throw when given a negative number"
where errorCalls (ErrorCall _) = Just ()
main = runTestTT $ TestCase case_negative
Reading the core, with optimizations on, the GHC inlines evaluate properly (?):
catch#
# ()
# SomeException
(\ _ ->
case throwIfNegative (I# (-1)) of _ -> ...
and then floats out the call to throwIfError, outside of the case scrutinizer:
lvl_sJb :: String
lvl_sJb = throwIfNegative lvl_sJc
lvl_sJc = I# (-1)
throwIfNegative =
\ (n_adO :: Int) ->
case n_adO of _ { I# x_aBb ->
case <# x_aBb 0 of _ {
False -> lvl_sCw; True -> error lvl_sCy
and strangely, at this point, no other code now calls lvl_sJb, so the entire test becomes dead code, and is stripped out -- GHC has determined that it is unused!
Using seq instead of evaluate is happy enough:
case_negative =
handleJust errorCalls (const $ return ()) $ do
throwIfNegative (-1) `seq` assertFailure "must throw when given a negative number"
where errorCalls (ErrorCall _) = Just ()
or a bang pattern:
case_negative =
handleJust errorCalls (const $ return ()) $ do
let !x = throwIfNegative (-1)
assertFailure "must throw when given a negative number"
where errorCalls (ErrorCall _) = Just ()
so I think we should look at the semantics of evaluate:
-- | Forces its argument to be evaluated to weak head normal form when
-- the resultant 'IO' action is executed. It can be used to order
-- evaluation with respect to other 'IO' operations; its semantics are
-- given by
--
-- > evaluate x `seq` y ==> y
-- > evaluate x `catch` f ==> (return $! x) `catch` f
-- > evaluate x >>= f ==> (return $! x) >>= f
--
-- /Note:/ the first equation implies that #(evaluate x)# is /not/ the
-- same as #(return $! x)#. A correct definition is
--
-- > evaluate x = (return $! x) >>= return
--
evaluate :: a -> IO a
evaluate a = IO $ \s -> let !va = a in (# s, va #) -- NB. see #2273
That #2273 bug is a pretty interesting read.
I think GHC is doing something suspicious here, and recommend not using evalaute (instead, use seq directly). This needs more thinking about what GHC is doing with the strictness.
I've filed a bug report to help get a determination from GHC HQ.

Why pattern matching does not throw exception in Maybe monad

My question is simple. Why wrong pattern matching does not throw exception in Maybe monad. For clarity :
data Task = HTTPTask {
getParams :: [B.ByteString],
postParams :: [B.ByteString],
rawPostData :: B.ByteString
} deriving (Show)
tryConstuctHTTPTask :: B.ByteString -> Maybe Task
tryConstuctHTTPTask str = do
case decode str of
Left _ -> fail ""
Right (Object trie) -> do
Object getP <- DT.lookup (pack "getParams") trie
Object postP <- DT.lookup (pack "postParams") trie
String rawData <- DT.lookup (pack "rawPostData") trie
return $ HTTPTask [] [] rawData
Look at tryConstuctHTTPTask function. I think that when the pattern does not match (for example "Object getP") we must get something like "Prelude.Exception", instead i get the "Nothing". I like this behavior but i am not understand why.
Thanks.
Doing pattern <- expression in a do-block, will call fail when the pattern does not match. So it is equivalent to doing
expression >>= \x ->
case x of
pattern -> ...
_ -> fail
Since fail is defined as Nothing in the Maybe monad, you get Nothing for failed pattern matches using <-.