[Haskell-cafe] TagSoup 0.9

Sun May 23 22:24:40 EDT 2010

Thanks Neil,

Using Network.HTTP worked.

However something else I have just run into concerns some web pages that
start with:

<?xml version="1.0" encoding="iso-8859-1"?>

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">

I get the following bad result:

TagText "HTTP/1.1 200 OK\r\nContent-Type: text/html\r\nLast-Modified: Tue,
27 Oct 2009 19:30:40 GMT\r\nETag: \"6f248cf73b57ca1:25e2\"\r\nDate: Sun, 23
May 2010 22:46:41 GMT\r\nTransfer-Encoding:  chunked\r\nConnection:
close\r\nConnection:
Transfer-Encoding\r\n\r\n00004000\r\n\255\254<\NUL?\NULx\NULm\NULl\NUL
\NULv\NULe\NULr\NULs\NULi\NULo\NULn\NUL=\NUL\"\NUL1\NUL.\NUL0\NUL\"\NUL
\NULe\NULn\NULc\NULo\NULd\NULi\NULn\NULg\NUL=\NUL\"\NULi\NULs\NULo\NUL-\NUL8
\NUL8\NUL5\NUL9\NUL-\NUL1\NUL\"\NUL

etc etc

Is this an easy thing to fix? I've started to look over the code.

-----Original Message-----
From: Neil Mitchell [mailto:ndmitchell at gmail.com] 
Sent: Wednesday, May 19, 2010 12:19 PM
To: Ralph Hodgson
Cc: Daniel Fischer; haskell-cafe at haskell.org; Don Stewart
Subject: Re: [Haskell-cafe] TagSoup 0.9

Hi Ralph,

> I was using TagSoup 0.8 with great success. On upgrading to 0.9 I have
this error:

> 

> TQ\TagSoup\TagSoupExtensions.lhs:29:17:

>    `Tag' is not applied to enough type arguments

>    Expected kind `*', but `Tag' has kind `* -> *'

>    In the type synonym declaration for `Bundle'

> Failed, modules loaded: TQ.Common.TextAndListHandling.

My change notes have this being a change between 0.6 and 0.8. As

Malcolm says, any old uses of "Tag" should become "Tag String". The

reason is that Tag is now parameterised, and you can use Tag

ByteString etc. However, I should point out that Tag ByteString won't

be any faster than Tag String in this version (it's in the future work

pile).

>> > Forgot to add: I now need to understand the following warnings on this

>> > line "> import Text.HTML.Download":

Everyone's comments have been right. I previously included

Text.HTML.Download so that it was easy to test tagsoup against the

web. Since I first wrote that snippet the HTTP downloading libraries

have improved substantially, so people should use those in favour of

the version in tagsoup - you'll be able to connect to more websites in

more reliable ways, go through proxies etc. I don't intend to remove

the Download module any time soon, but I will do eventually.

Thanks, Neil

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.haskell.org/pipermail/haskell-cafe/attachments/20100523/2336f383/attachment.html