| 1 |
On Sat, 22 Aug 2015 16:57:41 hw wrote: |
| 2 |
> Am 22.08.2015 um 15:43 schrieb Alan McKinnon: |
| 3 |
> > On 22/08/2015 15:26, hw wrote: |
| 4 |
> >> Hi, |
| 5 |
> >> |
| 6 |
> >> I have the following in a perl script: |
| 7 |
> >> if ($a != $b) { |
| 8 |
> >> |
| 9 |
> >> print "e: '$a', t: '$b'\n"; |
| 10 |
> >> |
| 11 |
> >> } |
| 12 |
> >> |
| 13 |
> >> That will print: |
| 14 |
> >> |
| 15 |
> >> e: '69.99', t: '69.99' |
| 16 |
> >> |
| 17 |
> >> |
| 18 |
> >> When I replace != with ne (if ($a ne $a) {), it doesn't print. |
| 19 |
> >> |
| 20 |
> >> |
| 21 |
> >> Is that a bug or a feature? And if it's a feature, what's the |
| 22 |
> >> explanation? |
| 23 |
|
| 24 |
|
| 25 |
> > != does a numerical comparison. More on this below |
| 26 |
> > ne does a string comparison. When viewed as a bunch of |
| 27 |
characters, 69.99 |
| 28 |
> > and 69.99 are identical. |
| 29 |
|
| 30 |
> > Now, your comparisons are NOT random. They are entirely |
| 31 |
predictable, as |
| 32 |
> > long as you know what is going on; you are running into floating |
| 33 |
point |
| 34 |
> > numbers. And as it turns out, computers never represent these |
| 35 |
things |
| 36 |
> > exactly (they are NOT integers). Even though they look identical |
| 37 |
> > on-screen, in RAM they will not be (this must be so for perl to do the |
| 38 |
> > print). Maybe they actually resolve to 69.990000001 and |
| 39 |
69.99000000. You |
| 40 |
> > see them as close-as-dammit equal, perl sees them as entirely |
| 41 |
different. |
| 42 |
> |
| 43 |
> Why can't it print the number as it is, or at least as it is compared, |
| 44 |
> like it should? If it would, one could see at once what the problem is. |
| 45 |
|
| 46 |
|
| 47 |
Your print values are coming from the original variables. The numeric |
| 48 |
comparison does an internal conversion to two temporary variables, |
| 49 |
and compares those, WITHOUT affecting the original data. |
| 50 |
|
| 51 |
This string to float conversion appears not to be deterministic, if it |
| 52 |
converts two identical strings into two different float values, but I don't |
| 53 |
know enough about the internals to comment any further. |
| 54 |
|
| 55 |
|
| 56 |
-- |
| 57 |
Reverend Paul Colquhoun, ULC. http://andor.dropbear.id.au/ |
| 58 |
Asking for technical help in newsgroups? Read this first: |
| 59 |
http://catb.org/~esr/faqs/smart-questions.html#intro |
Archives