Anonymous | Login | Signup for a new account | 2019-02-21 19:38 CET | ![]() |
Main | My View | View Issues | Change Log | Roadmap |
View Issue Details [ Jump to Notes ] | [ Issue History ] [ Print ] | |||||||||||
ID | Project | Category | View Status | Date Submitted | Last Update | |||||||
0007690 | OCaml | standard library | public | 2017-12-13 15:46 | 2017-12-19 15:24 | |||||||
Reporter | oandrieu | |||||||||||
Assigned To | nojebar | |||||||||||
Priority | normal | Severity | minor | Reproducibility | always | |||||||
Status | resolved | Resolution | fixed | |||||||||
Platform | OS | OS Version | ||||||||||
Product Version | 4.03.0 | |||||||||||
Target Version | Fixed in Version | |||||||||||
Summary | 0007690: issues with caml_float_of_hex | |||||||||||
Description | There are some overflow issues with the exponent computations in caml_float_of_hex, which results in some invalid conversions. The integer exp variable overflows, which results in the call to ldexp to overflow when it should underflow (and vice versa). The test "if (e < INT_MIN || e > INT_MAX) return -1;" is ineffective on Win64 since int and long have the same size there: this results in some difference in behavior between win64 and linux. And (minor point), the function accept spaces between the 'p' and the exponent digits, whereas strtod and the C99 syntax do not allow spaces there. | |||||||||||
Steps To Reproduce | $ cat f.ml let hexstring_of_float x = Printf.sprintf "%h" x let () = let s = try hexstring_of_float (float_of_string Sys.argv.(1)) with _ -> "invalid" in print_endline s $ ocamlopt.opt f.ml $ ./a.out 0x1.0p-2147483648 infinity $ ./a.out 0x123456789ABCDEF0p2147483647 0x0p+0 $ ./a.out 0x1p2147483648 -> invalid on linux64 -> infinity on win64 $ ./a.out '0x1p 1' 0x1p+1 | |||||||||||
Tags | No tags attached. | |||||||||||
Attached Files | ||||||||||||
![]() |
|
(0018743) oandrieu (reporter) 2017-12-13 17:12 |
I can provide a GPR if you think these corner cases are worth fixing. |
(0018744) nojebar (developer) 2017-12-13 17:14 |
Definitely the overflow one needs fixing; a GPR would be much appreciated! Regarding the second one, I am not sure if it is a bug or not at the moment. |
(0018745) oandrieu (reporter) 2017-12-13 20:09 |
there, GPR#1528 : https://github.com/ocaml/ocaml/pull/1528 [^] |
(0018752) nojebar (developer) 2017-12-19 15:24 |
This PR has now been merged, so marking this resolved. |
![]() |
|||
Date Modified | Username | Field | Change |
2017-12-13 15:46 | oandrieu | New Issue | |
2017-12-13 17:12 | oandrieu | Note Added: 0018743 | |
2017-12-13 17:14 | nojebar | Note Added: 0018744 | |
2017-12-13 20:09 | oandrieu | Note Added: 0018745 | |
2017-12-19 15:24 | nojebar | Note Added: 0018752 | |
2017-12-19 15:24 | nojebar | Status | new => resolved |
2017-12-19 15:24 | nojebar | Resolution | open => fixed |
2017-12-19 15:24 | nojebar | Assigned To | => nojebar |
Copyright © 2000 - 2011 MantisBT Group |