View Issue Details [ Jump to Notes ] | [ Issue History ] [ Print ] | ||||||||||||
ID | Project | Category | View Status | Date Submitted | Last Update | ||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
0000373 | mercury | Bug | public | 2015-01-13 17:40 | 2016-02-24 09:44 | ||||||||
Reporter | wangp | ||||||||||||
Assigned To | |||||||||||||
Priority | normal | Severity | minor | Reproducibility | always | ||||||||
Status | new | Resolution | open | ||||||||||
Product Version | |||||||||||||
Target Version | 15.11 | Fixed in Version | |||||||||||
Summary | 0000373: string.to_float imprecisely specified | ||||||||||||
Description | string.to_float does not explicitly describe the literals that it will accept, and may be backend dependent. As it is, it goes beyond Mercury float literals and that accepted by parsing_utils.float_literal_as_string. Examples: "1" (missing fractional part and exponent) "+1" (leading plus) ".1" (leading dot) "1." "1.e" "1e" | ||||||||||||
Tags | No tags attached. | ||||||||||||
Attached Files |
|
Notes | |
juliensf (administrator) 2015-01-13 18:12 |
There is no "may" about it. It *is* backend dependent. |
juliensf (administrator) 2016-02-24 09:44 |
A fix for this issue should also address the question of whether string.to_float("-0") is equal to 0.0 or -0.0? The C back-ends** and C#/Java back-ends currently disagree on this. ** at least on OS X, with clang and BSD libc etc. |