On Oct 12, 2011, at 3:13 AM, Meikel Brandmeyer (kotarak) wrote:
> So the library has several options: Since the numeric changes in this area
> are breaking changes, the library has to either a) take overflow policy into
> its contract
Midje's contract for 1.3 numerics is: if there are overflow
Hi,
yes. In 1.2 the ' will be not part of the symbol +, but it will be attached
to the following 1, quoting it. The reader of 1.3 will consume the ' as part
of the + symbol. To clarify this a little:
1.2 sees: [+ '1] (<=> [+ (quote 1)])
1.3 sees: [+' 1]
Too "outrageously clever" for my taste,
I feel myself silly, but could you explain how does this work?
(def +M (first [+' 1]))
Is it some kind of reader level trick?
Cheers,
Ivan.
On 11 October 2011 04:25, Alan Malloy wrote:
> (def +M (first [+' 1])) seems like it should work in both versions?
>
> On Oct 10, 4:28 pm, Brian Marick
Hi Chas,
someone specifying (roughly 9223372036854775804 5) also accepts
9223372036854775808 as a possible value. So she has to at least think a tiny
bit about overflow. However you are right that she should not be concerned
about the specific implementation of roughly.
So the library has seve
Hi,
Besides the hack by Alan you might also want to think about an overflow save
implementation for roughly:
user=> (defn roughly
[expected-value delta]
(let [exp-pos? (pos? expected-value)
upper(if (neg? delta) (- delta) delta)
lower(- upp
On Oct 11, 2011, at 3:17 PM, Meikel Brandmeyer wrote:
> Maybe I'm a bit bold, but shouldn't a 1.3 user be aware, that his limits do
> overflow and specify a bigint to begin with?
Consider someone who passes in a regular integer to the `roughly` checker. It's
near the overflow boundary. That pl
On Oct 11, 2011, at 4:17 PM, Meikel Brandmeyer wrote:
> Hi,
>
> Am 11.10.2011 um 01:28 schrieb Brian Marick:
>
>> If the 1.3 user has taken care to use promoting-to-bignum arithmetic, it
>> would be rude to blow up if `actual` happens to be a regular integer but
>> adding `delta` to it causes
Hi,
Am 11.10.2011 um 01:28 schrieb Brian Marick:
> If the 1.3 user has taken care to use promoting-to-bignum arithmetic, it
> would be rude to blow up if `actual` happens to be a regular integer but
> adding `delta` to it causes an overflow.
Maybe I'm a bit bold, but shouldn't a 1.3 user be aw
On Oct 10, 2011, at 7:50 PM, Kevin Downey wrote:
> (defmacro plus [] (if … (resolve (symbol "clojure.core/+")) (resolve
> (symbol "clojure.core/+'"
>
> ((plus) actual delta)
Thanks.
On Oct 10, 2011, at 9:25 PM, Alan Malloy wrote:
> (def +M (first [+' 1])) seems like it should work in both
(def +M (first [+' 1])) seems like it should work in both versions?
On Oct 10, 4:28 pm, Brian Marick wrote:
> I may be missing something obvious.
>
> Midje has a checker that lets users say things like
>
> (fact (my-computation) => (roughly 15 2))
>
> Where the first number in `roughly` is th
(defmacro plus [] (if … (resolve (symbol "clojure.core/+")) (resolve
(symbol "clojure.core/+'"
((plus) actual delta)
On Mon, Oct 10, 2011 at 4:28 PM, Brian Marick wrote:
> I may be missing something obvious.
>
> Midje has a checker that lets users say things like
>
> (fact (my-computation)
I may be missing something obvious.
Midje has a checker that lets users say things like
(fact (my-computation) => (roughly 15 2))
Where the first number in `roughly` is the target number and the second is an
acceptable range around that target. Part of what the checker does in 1.2 is
this:
12 matches
Mail list logo
