Fix datetime assume UTC allow ranges

[danlooo]

This PR aims to fix the error of parsing datetime e.g. at CDSE. Consider the README example:

using STAC, Dates

collections = ["sentinel-2-l1c"]
time_range = (DateTime(2026,2,9,10,20), DateTime(2026,2,9,10,30))
lon_range = (0, 20)  # west, east
lat_range = (32, 45)  # south, north

catalog = STAC.Catalog("https://stac.dataspace.copernicus.eu/v1/")

search_results = collect(search(catalog, collections, lon_range, lat_range, time_range))

returns error

search_results = collect(search(catalog, collections, lon_range, lat_range, time_range))
77-element Vector{STAC.Item}:
Error showing value of type Vector{STAC.Item}:

SYSTEM (REPL): showing an error caused an error
ERROR: 1-element ExceptionStack:
MethodError: no method matching parseDT(::Type{DateTime}, ::String)
The function `parseDT` exists, but no method is defined for this combination of argument types.

Closest candidates are:
  parseDT(::Type{Tuple}, ::Any)
   @ CFTime ~/.julia/packages/CFTime/5SQZS/src/conversions.jl:106

According to STAC Spec 1.1, acquisition time of STAC items is mainly stored in filed datetime, SHOULD be in UTC, and may be an interval instead from start_datetime to end_datetime.

Therefore, I used Dates.dateformat"yyyy-mm-ddTHH:MM:SS.sZ", also tried start of datetime range, and kept returning nothing otherwise.

[codecov]

Codecov Report

❌ Patch coverage is 72.72727% with 6 lines in your changes missing coverage. Please review.
✅ Project coverage is 75.00%. Comparing base (6a7f562) to head (5f1f39f).
⚠️ Report is 20 commits behind head on main.

Files with missing lines	Patch %	Lines
src/item.jl	72.72%	6 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #31      +/-   ##
==========================================
- Coverage   75.77%   75.00%   -0.78%     
==========================================
  Files           9        9              
  Lines         322      336      +14     
==========================================
+ Hits          244      252       +8     
- Misses         78       84       +6     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[Alexander-Barth]

Thanks a lot for the PR! (I removed CFTime.parseDT but I did not remember that it was used here)

Concerning the start_datetime and end_datetime: would it be better to return a tuple with both (rather than the first)? The function is already now (somewhat) type unstable (Union{DateTime,Nothing}).

Alternatively, there could also be the methods start_datetime(item::Item) and end_datetime(item::Item) (and DateTime would just be nothing if there is a range) ?

Maybe we can have a function like tryparse_utc, in case we want to add also the handle the suffix "+00:00" for UTC.

function tryparse_utc(s) 
    try
        return DateTime(s, Dates.dateformat"yyyy-mm-ddTHH:MM:SS.sZ")
    catch
        return nothing
    end
end

[danlooo]

The method is called DateTime(item::Item) thus I think it would be confusing to also return a tuple of time span. There is also STAC extension timestamps defining the entire life cycle from capture to publishing, motivating new calls like get_datetime(item, :published). How about having the general method get_datetime and maybe add the exposure duration to the show method if a range is present?

[Alexander-Barth]

Thanks a lot ! Would you be ok to call the function datetime(item, symbol) instead of get_datetime(item, symbol).
get_foo/set_foo is quite often used in C++ and Java, but not that much in julia. We have already several method called title, description, keywords, doi... that get the corresponding info with the prefix get_.

[Alexander-Barth]

This is great, thank you!