Add Text.substring to allow for an easy short hand of Text.take (start.up_to end) #7913
Conversation
For #7876 adds a Text.substring function which supports negative indexes and returns a part of a string from 0-based index 'start' and continuing for 'length'
For #7876 adds a Text.substring function which supports negative indexes and returns a part of a string from 0-based index 'start' and continuing for 'length'. Also simplified get function as it looped unnecessarily.
|
|
Cassandra-Clark
requested review from
jdunkerley,
radeusgd and
GregoryTravis
as code owners
distribution/lib/Standard/Base/0.0.0-dev/src/Data/Text/Extensions.enso
Outdated
Show resolved
Hide resolved
…ons.enso punctuation corrections Co-authored-by: GregoryTravis <greg.m.travis@gmail.com>
Added test for start index larger than string
…g/enso into wip/cassandrac/substring
distribution/lib/Standard/Base/0.0.0-dev/src/Data/Text/Extensions.enso
Outdated
Show resolved
Hide resolved
distribution/lib/Standard/Base/0.0.0-dev/src/Data/Text/Extensions.enso
Outdated
Show resolved
Hide resolved
…ons.enso updated Arguments: section to use consistent style Co-authored-by: Radosław Waśko <radoslaw.wasko@enso.org>
…ons.enso updated Index_Out_Of_Bounds error to reference cached length Co-authored-by: Radosław Waśko <radoslaw.wasko@enso.org>
There was a problem hiding this comment.
Looks great!
I have one thought regarding length vs end here, wondering if we should try to uniformize this with other similar methods (but also happy to hear why we should not 🙂).
Anyway, I think we can discuss this asynchronously and if necessary do a separate follow up PR - no need to block this one.
| @@ -1791,6 +1786,35 @@ Text.parse_time_of_day self format:Date_Time_Formatter=Date_Time_Formatter.iso_t | |||
| Text.parse_time_zone : Time_Zone ! Time_Error | |||
| Text.parse_time_zone self = Time_Zone.parse self | |||
|
|
|||
| ## ALIAS mid, slice, substring | |||
There was a problem hiding this comment.
Regarding: slice
We already do have a slice method on Vector:
## GROUP Selections
Creates a new vector with the skipping elements until `start` and then
continuing until `end` index.
Arguments:
- start: The index of the first element to include.
- end: The index to stop slicing at.
> Example
Remove the first 2 elements then continue until index 5 from the vector.
[1, 2, 3, 4, 5, 6, 7, 8].slice 2 5 == [3, 4, 5]
slice : Integer -> Integer -> Vector Any
slice self start end = @Builtin_Method "Vector.slice"
I'm slightly worried that with this alias users can get confused as to what the second argument should be - in Vector.slice that is end but in Text.substring (also aliased by slice) it is length.
Shouldn't we unify this? I was thinking that length argument makes more sense for substring, but actually Java's String::substring also takes endIndex and not length.
|
And a small nitpick: please ensure to name the PR before merging it, as the name will be part of the commit history so it's good to make it something clearly readable - e.g. it could be the same/similar to the ticket name.
|
Cassandra-Clark
changed the title
|
Having discussed with @radeusgd - lets mark the @Cassandra-Clark please make these changes and add a changelog entry then we should be ready to mark this PR as ready to merge. |
…g/enso into wip/cassandrac/substring
Per conversation with James, added slice back to substring
Pull Request Description
Implemented substring for #7876. Also simplified get function.
Checklist
Please ensure that the following checklist has been satisfied before submitting the PR: