fast-tagsoup 1.0.5 → 1.0.6
raw patch · 2 files changed
+6/−9 lines, 2 files
Files
- Text/HTML/TagSoup/Fast.hs +5/−8
- fast-tagsoup.cabal +1/−1
Text/HTML/TagSoup/Fast.hs view
@@ -260,8 +260,6 @@ -- nextIC p 0 _ = False -- nextIC !p !l ((a,b):xs) = (r p == a || r p == b) && nextIC (pp p) (mm l) xs -{-# INLINE parseTags #-}- -- ord maxBound = 1114111 = 0x10FFFF =   -- maximum 8 characters between '&' and ';' -- entities length are also not longer than 8@@ -377,8 +375,6 @@ poke d x add s (pp d) l xs -{-# INLINE unescapeHtml #-}- -- | Show a list of tags, as they might have been parsed. renderTags = renderTags' escapeHtml B.concat -- | Alternative to 'renderTags' working with 'Text'@@ -399,8 +395,6 @@ renderAtts ((a,v):as) rs = "\"" : escape v : "=\"" : a : " " : renderAtts as rs -{-# INLINE renderTags #-}- -- | Decode XML to UTF-8 using @encoding@ attribute of @\<?xml\>@ tag. ensureUtf8Xml :: B.ByteString -> B.ByteString ensureUtf8Xml s@@ -408,8 +402,8 @@ -- first of all we try decode utf-8. -- Some sites specify non utf-8 encoding, while the text is in utf. | otherwise =- case parseTags $ B.dropWhile isSpace s of- -- ^ sometimes there is a space before ?xml+ case dropWhile isText $ parseTags s of+ -- ^ sometimes there is a space or junk text before <?xml> (TagOpen "?xml" attrs : _) | Just enc <- lookup "encoding" attrs , B.map toLower enc /= "utf-8" ->@@ -429,3 +423,6 @@ -- convertFuzzy works only on GNU _ -> s -- TL.decodeUtf8With (\ _ -> fmap B.w2c) s -- TL.lenientDecode s+ where isText (TagText _) = True+ isText _ = False+wb addr byte = poke (intPtrToPtr addr) (byte :: Word8)
fast-tagsoup.cabal view
@@ -1,6 +1,6 @@ cabal-version: >= 1.6 name: fast-tagsoup-version: 1.0.5+version: 1.0.6 copyright: Vladimir Shabanov 2011-2012 author: Vladimir Shabanov <vshabanoff@gmail.com> maintainer: Vladimir Shabanov <vshabanoff@gmail.com>