Revert removing deprecated exercises#832
Conversation
petertseng
left a comment
There was a problem hiding this comment.
have a pre-emptive approval for when you are satisfied with what Travis CI tells you, and potentially removing the given implementations. feel free to ask for another review if you want it though.
I will put forth the option of excluding deprecated exercises from the README up-to-date check, but I note the following:
- Anyone who uses
configlet generateto generate READMEs would still cause README updates on the deprecated exercises - Increased complexity in the scripts
I think I do not support my own suggestion, but I welcome it if others can find a way to make it work well.
| @@ -1 +1,33 @@ | |||
| fn parse_hex_digit(c: char) -> Option<i64> { | |||
| match c { | |||
There was a problem hiding this comment.
although it does not hurt to have it since it's deprecated, this lib.rs doesn't necessarily need to have a complete implementation (and we would want it not to have one if it were non-deprecated)
|
|
||
| impl<'a> CodonInfo<'a> { | ||
| pub fn name_for(&self, codon: &str) -> Result<&'a str, &'static str> { | ||
| if codon.len() != 3 { |
There was a problem hiding this comment.
although it does not hurt to have it since it's deprecated, this lib.rs doesn't necessarily need to have a complete implementation (and we would want it not to have one if it were non-deprecated)
same
| @@ -1,3 +1,6 @@ | |||
| pub fn twofer(name: &str) -> String { | |||
| unimplemented!("One for {}, one for me.", name); | |||
| match name { | |||
There was a problem hiding this comment.
although it does not hurt to have it since it's deprecated, this lib.rs doesn't necessarily need to have a complete implementation (and we would want it not to have one if it were non-deprecated)
same
|
I copied the full implementations over the three stub As you've already approved this (thank you!), and Travis now passes, I will now merge this. |
|
Given that the documentation claims that students who have not already started a deprecated exercise will not be able to do so, it is OK to have stubs that already provide a solution. For the purposes of our track upholding its documented obligations, it is irrelevant that the aforementioned claim is false (I can currently download hexadecimal and submit an implementation to it, not having started it previously); that issue must be solved elsewhere, so this track need not take specific action on that. |
Per new documentation in exercism/docs#126, it is actually not safe to remove deprecated exercises. As such, #773 should not have been approved. Without #773, we don't need #830.
This reverts both of those PRs.