Re: [racket-users] Weird problem with bytes-utf-8-length

2017-12-08 Thread Jonathan Simpson
That seems to work. Thank you for the rapid response! I feel so much better now that 1 + 1 = 2 again :) -- Jonathan On Friday, December 8, 2017 at 7:07:26 PM UTC-5, Matthew Flatt wrote: > > This is a bug in the optimizer. The `bytes-utf-8-length` function is > incorrectly marked as always

Re: [racket-users] Weird problem with bytes-utf-8-length

2017-12-08 Thread Matthew Flatt
This is a bug in the optimizer. The `bytes-utf-8-length` function is incorrectly marked as always returning a fixnum. I'll push a repair. Meanwhile, to work around the bug, use something like this: (define non-broken-bytes-utf-8-length bytes-utf-8-length) ;; To confuse and work around the

[racket-users] Weird problem with bytes-utf-8-length

2017-12-08 Thread Jonathan Simpson
Welcome to DrRacket, version 6.11 [3m]. Language: racket, with debugging; memory limit: 128 MB. > (if (bytes-utf-8-length (bytes 255)) 1 2) 1 > (bytes-utf-8-length (bytes 255)) #f > (not (bytes-utf-8-length (bytes 255))) #f Can anyone explain this to me? Surely the output of the last line should