Add Date_Range
#6621
Add Date_Range
#6621
Conversation
radeusgd
force-pushed
the
wip/radeusgd/6543-date-range
branch
3 times, most recently
from
125fe9a
to
c6f9e2f
Compare
| ## PRIVATE | ||
| Assumes that the range is not empty. | ||
| compute_length_step_months start end step increasing = | ||
| diff = case increasing of | ||
| True -> Time_Utils.months_between start end | ||
| False -> Time_Utils.months_between end start | ||
| # assert (diff >= 0) | ||
| steps = diff . div step | ||
| exact_fit = case increasing of | ||
| True -> start + Period.new months=steps*step == end | ||
| False -> start - Period.new months=steps*step == end | ||
| if exact_fit then steps else steps+1 |
There was a problem hiding this comment.
I was a bit skeptical about this heuristic, but I added as many test cases trying to breaking as I could imagine and it works, so I think we should be fine. The bright side is that it gives us O(1) time complexity for the case where the Period has days=0. So the brute force O(N) is now only when both days and months are non-zero.
Apparently, the JDK uses a pretty smart way to compute the month difference:
private long monthsUntil(LocalDate end) {
long packed1 = getProlepticMonth() * 32L + getDayOfMonth(); // no overflow
long packed2 = end.getProlepticMonth() * 32L + end.getDayOfMonth(); // no overflow
return (packed2 - packed1) / 32;
}(cited from LocalDate.java)
distribution/lib/Standard/Base/0.0.0-dev/src/Data/Range/Extensions.enso
Outdated
Show resolved
Hide resolved
distribution/lib/Standard/Base/0.0.0-dev/src/Data/Time/Period.enso
Outdated
Show resolved
Hide resolved
distribution/lib/Standard/Base/0.0.0-dev/src/Data/Time/Period.enso
Outdated
Show resolved
Hide resolved
| 1.up_to 1 include_end=True . to_vector . should_equal [1] | ||
| 1.down_to 1 include_end=True . to_vector . should_equal [1] |
There was a problem hiding this comment.
worth testing 1.up_to 0 include_end=True?
| ## PRIVATE | ||
| Never use the constructor directly to construct a range, as it does not | ||
| allow to verify invariants and may lead to unexpected behavior. | ||
| Internal_Constructor (start : Date) (end : Date) (step : Period) (increasing : Boolean) (cached_length : Integer) |
There was a problem hiding this comment.
Should we call it Between (I know we want to avoid use) but it gets rendered sometimes and visible to user.
There was a problem hiding this comment.
I really wanted to mark it Internal as using it ever is going to cause fires.
Where ever can it render? I specifically ensured we override to_text and to_display_text, ooh I guess I should check pretty too - will do and will add tests.
Ah and I guess to_js_object - I will amend it to have a sane name and skip the internal fields.
The only remaining place I can think of is Meta but that is rather advanced and internal stuff anyway so I think we can ignore it there.
distribution/lib/Standard/Base/0.0.0-dev/src/Data/Time/Date_Range.enso
Outdated
Show resolved
Hide resolved
distribution/lib/Standard/Base/0.0.0-dev/src/Data/Time/Date_Range.enso
Outdated
Show resolved
Hide resolved
radeusgd
force-pushed
the
wip/radeusgd/6543-date-range
branch
from
457bbb1
to
fd4541f
Compare
…ength computation
* develop: Implement loading spinner for visualisations. (#6512) Fix blank input port (#6614) Add `Date_Range` (#6621) All Vector operations shall be applicable on java.util.ArrayList (#6642) Fix redirect paths and enable authentication and new dashboard by default (#6605) Fix #6287: wrong nested breadcrumb ordering (#6617) Whitelist AWS Cognito domains (#6643) Revert "Add COOP+COEP+CORP headers (#6597)" (#6647) Fix shortcuts table formatting (#6644) Automatic type based dropdown does not include singleton in a union type (#6629) Make Meta.get_annotation work for constructor (#6633)
Pull Request Description
Closes #6543
Important Notes
Checklist
Please ensure that the following checklist has been satisfied before submitting the PR:
Scala,
Java,
and
Rust
style guides. In case you are using a language not listed above, follow the Rust style guide.
./run ide build.