- Haskell-Cafe - Haskell.org

Fwd: [Haskell-beginners] HaXml.SAX successfully parses a malformed XML document
by David Frey 17 Oct '08

17 Oct '08

I asked this question on the haskell-beginners list, but I didn't get a response, so I am forwarding it on to the cafe list. David --- Original Message --- Date: 10/16/2008 From: "David Frey" <dpfrey(a)shaw.ca> Subject: [Haskell-beginners] HaXml.SAX successfully parses a malformed XML document It seems that saxParse from Text.XML.HaXml.SAX will successfully parse a malformed XML document. Notice that the SubElem opening and closing tag are not matched in the XML document below. --- input.xml --- <RootNode> <ElemOne> <SubElem attr="foo">element data</SubElemBroken> <NoDataElem/> <ElemOne> </RootNode> --- end input.xml --- The Haskell code below runs without error. It prints out the type of elements found during the parse. --- Haskell Code --- module Main where import qualified Text.XML.HaXml.SAX as SAX import Data.Maybe main = let inputFilename = "input.xml" in do content <- readFile inputFilename let (elements, error) = SAX.saxParse inputFilename content if isNothing error then mapM_ putStrLn (summarizeElements elements) else putStrLn $ "ERROR:" ++ (fromJust error) summarizeElements :: [SAX.SaxElement] -> [String] summarizeElements elements = map summarizeElement elements summarizeElement :: SAX.SaxElement -> String summarizeElement element = case element of (SAX.SaxDocTypeDecl d) -> "DocType" (SAX.SaxProcessingInstruction p) -> "Processing Instruction" (SAX.SaxComment s) -> "Comment" (SAX.SaxElementOpen name attrs) -> "Element Open" (SAX.SaxElementClose name) -> "Element Close" (SAX.SaxElementTag name attrs) -> "No Content Element" (SAX.SaxCharData charData) -> "Character Data" (SAX.SaxReference reference) -> "Reference" --- End Haskell Code --- The Python code below throws an exception when parsing the same input document. --- Python Code --- from xml.sax import make_parser from xml.sax.handler import ContentHandler def main(): c = ContentHandler() p = make_parser() p.setContentHandler(c) p.parse(open('input.xml')) if __name__ == '__main__': main() -- End Python Code --- Is this a bug in the HaXml SAX parser? Thanks, David _______________________________________________ Beginners mailing list Beginners(a)haskell.org http://www.haskell.org/mailman/listinfo/beginners

1 0

How to deal with pointers to pointers in the FFI
by Jefferson Heard 17 Oct '08

17 Oct '08

I have the following functions in C: OGRErr OGR_G_CreateFromWkb (unsigned char *, OGRSpatialReferenceH, OGRGeometryH *, int) OGRErr OGR_G_CreateFromWkt (char **, OGRSpatialReferenceH, OGRGeometryH *) void OGR_G_DestroyGeometry (OGRGeometryH) OGRGeometryH OGR_G_CreateGeometry (OGRwkbGeometryType) The normal sequence of calls is OGRGeometryH handle = OGR_G_CreateGeometry(SOME_TYPE); // do stuff OGR_G_DestroyGeometry(handle); OR OGR_G_CreateFromWkb(blob, ref, &handle, 0); // do stuff OGR_G_DestroyGeometry(handle); As you can see, in one case, I have to pass in a pointer to the returned handle, and not just the handle. How can I accomplish this feat using a single type? I had data OGRGeometryH type Geometry = Ptr OGRGeometryH but then can I declare that a function returns a Ptr (Geometry) ? -- I try to take things like a crow; war and chaos don't always ruin a picnic, they just mean you have to be careful what you swallow. -- Jessica Edwards

3 4

Troubles getting isA of ArrowIf from HXT 8.10 to work as expected
by Corey O'Connor 17 Oct '08

17 Oct '08

I'm attempting to use ifA to conditionalize generated XHTML. ifA (arr test_status >>> arr (/= "Running")) (mkelem "a" [attr "href" (arr test_log_URL >>> mkText)] [txt "log"]) (mkelem "a" [attr "href" (arr test_log_URL >>> mkText)] [txt "no log"]) The arrow type at the point of the ifA is (I think) "XmlArrow a => a XmlTree TestRun" where TestRun is defined as: data TestRun = TestRun { test_log_URL :: String, test_status :: String } The issue I'm seeing is that regardless of the predicate arrow the result is always the same: The XmlTree for the True branch is always generated. In other words, I get the same behavior using this code: ifA (arr (const False)) (mkelem "a" [attr "href" (arr test_log_URL >>> mkText)] [txt "log"]) (mkelem "a" [attr "href" (arr test_log_URL >>> mkText)] [txt "no log"]) Which, I thought, the "arr (const False)" would force the ifA to always descend the False path. This is my first attempt at using HXT and Arrows (Which is otherwise going great!) so I figure I'm missing something. Any ideas? Cheers, Corey O'Connor

1 0

Temlpate Haskell: [d| ... |]
by Achim Schneider 17 Oct '08

17 Oct '08

> [d| ... |], where the "..." is a list of top-level declarations; the > quotation has type Q [Dec]. (http://www.haskell.org/ghc/docs/latest/html/users_guide/template-haskell.ht…) Can someone elaborate on what a list means here? Neither declarations = [d| foo = "bar" bar = "foo" |] nor declarations = [d| [ foo = "bar" , bar = "foo" ] |] work. I'm trying to automatically generate a bunch of functions out of an xml file. -- (c) this sig last receiving data processing entity. Inspect headers for copyright history. All rights reserved. Copying, hiring, renting, performance and/or quoting of this signature prohibited.

3 4

Re: What I wish someone had told
by John Lato 17 Oct '08

17 Oct '08

Richard O'Keefe wrote: > On 17 Oct 2008, at 9:53 am, Daryoush Mehrtash wrote: > >> So does this mean that the reason for complexity of generics is the >> Java inheritance? > > No. The reason for the complexity of generics in Java is that > they weren't designed into the language in the first place. > It took several attempts and quite a lot of work to come up with > a version of generics that was 100% interoperable with earlier > JVMs *and* still worth having. > > You could have something very like Java generics but without > the strangeness if you designed it into the language from the > beginning, as Betrand Meyer did with Eiffel. Of course, what > _he_ didn't design in from the beginning was lambdas, now > present as "agents". Eiffel has its own kinds of strangeness. > Or C# generics, which are built into the IL (in .Net 2.0 and above). In many ways, C# is a very nice language, and having generics fully supported in reflection is pretty cool. I do have some words to say about various ASP.Net libraries, however...

1 0

howto mix <- within do?
by Larry Evans 17 Oct '08

17 Oct '08

The attached code produces error: <-- cut here -- runghc -dcore-lint do_with_assignment.proto.hs do_with_assignment.proto.hs:30:2: Couldn't match expected type `[]' against inferred type `IO' Expected type: [t] Inferred type: IO () In the expression: putStr "v0=" In a 'do' expression: putStr "v0=" >-- cut here-- However, similar code here: http://www.nabble.com/List-as-input-p19987726.html apparently does work. Is there some compiler option I need to use? The compiler is ghc: ghc --version The Glorious Glasgow Haskell Compilation System, version 6.8.2

4 6

Invalid declaration names in TH that typecheck
by Sean Leather 17 Oct '08

17 Oct '08

[I posted this on the Template Haskell list, but nobody seemed interested. Thus, I'm re-posting it here for higher coverage.] Apparently, I can create a function declaration (FunD) with a name containing an invalid sequence of characters. It passes typechecking and compiles. It even shows up in GHCi! The function name is "ep:#:", and the ":#:" is derived from an infix type constructor. In GHCi (6.8.3), I type "ep", hit <Tab> to get the list of completions, and, sure enough, there it is! When I try to use it, GHCi interprets it as two distinct symbols, "ep" and ":#:", as it should. Template Haskell's claim to fame is that it allows you to do "type-safe compile-time meta-programming," and while this issue is perhaps considered lexical rather than type-level, I'm wondering if there should be some additional error checking (somewhere) to account for this. I admit though, that I wouldn't know how to do this easily within the bounds of the TH API. The Name type is common to functions, infix operators, and everything else. That which is valid for one is not necessarily so for the other. Regards, Sean

2 1

List as input
by leledumbo 17 Oct '08

17 Oct '08

module Main where import Data.List -- quicksort of any list qsort [] = [] qsort (x:xs) = qsort(filter(<x) xs) ++ [x] ++ qsort(filter(>=x) xs) -- optimized quicksort, uses middle element as pivot qsortOpt [] = [] qsortOpt x = qsortOpt less ++ [pivot] ++ qsortOpt greater where pivot = x !! ((length x) `div` 2) less = filter (<pivot) (delete pivot x) greater = filter (>=pivot) (delete pivot x) main = do putStr "Enter a list: " l <- readLn print (qsortOpt l) -- end of code Why do I get ERROR "qsort.hs":17 - Unresolved top-level overloading *** Binding : main *** Outstanding context : (Read b, Show b, Ord b) ? -- View this message in context: http://www.nabble.com/List-as-input-tp19987726p19987726.html Sent from the Haskell - Haskell-Cafe mailing list archive at Nabble.com.

10 16

Terminal-like Application Design
by Jeff Wheeler 17 Oct '08

17 Oct '08

Hi, I'm a slight Haskell newbie, but I'm trying to write a terminal-like application that accepts simple commands with optional arguments, and can then execute them. Most of these commands will need IO, as later I will want to communicate over USB for most of them. I was hoping, though, that I could get some comments on the initial architecture I've been playing with [1]. I suspect I should be using some sort of monad to represent the commands, but I don't fully understand monads, and am not sure how it would apply in this context. Should I be using a monad here, and if so, how? Thanks in advance, Jeff Wheeler [1] http://media.nokrev.com/junk/cli/

6 5

[fun] HaskellDB Talk trailer
by Ahn, Ki Yung 17 Oct '08

17 Oct '08

There is an impressive HaskellDB Talk trailer on the web. http://www.vimeo.com/1983774 Cheers to the HaskellDB developers :-)

3 2