On Sun, Jan 20, 2013 at 08:21:32AM -0800, Mark Lentczner wrote:

On Mon, Jan 14, 2013 at 9:43 AM, Gregory Collins wrote:

...

The bytestring instances are what it's most often used for! The haddocks warn about this issue. And: yes, we do want an ASCII case fold. If we didn't, we'd be using Text.

Note: case-insensitive provides ASCII case folding when used on ByteString. However, HTTP headers are in ISO-8859-1. Hence, using this for case folding HTTP headers isn't technically correct: It will fail for headers and other case-insensitive tokens with ISO-8859-1 accented characters., though admittedly in practice one doesn't see them.

That bit of pedantry out of the way....

Interestingly, case-insensitive may well support this case by accident! The case-folding it uses is just that from Data.ByteString.Char8, which is Unicode code points 0-255, which includes (if I recall correctly) latin-1 as a subset. Testing this may be misleading because Data.ByteString.Char8.putStrLn doesn't do anything clever, but you should find that packing, foldCase, unpacking does the right thing. Of course, the documentation should be amended to state this if we are interested in making it a guarantee. regards, Ben Millwood PS. Can I ask that people remove bm380 at srcf dot net from their CC lists in this thread? It's an artifact of when I accidentally replied from the wrong address and it's meaning I'm getting some stuff twice.

On 21 January 2013 00:29, Mark Lentczner wrote:

In 0.4.0.2 and earlier, it was mapping Data.Char's toLower for all types except Text (where it was using Text's toCaseFold) From 0.4.0.3 and on, it is using Text's toCaseFold for all types, except ByteString, where it uses it's own toLower function optimized for just ISO-8859-1.

The output should be identical between 0.4.0.2 and 0.4.0.3. In the hp branch I changed the documentation to mention ISO-8859-1 instead of ASCII: https://github.com/basvandijk/case-insensitive/tree/hp Feel free to send me pull requests for better phrasing. Bas