New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

RFC: Close all the "roast" RT tickets #2099

Open
zoffixznet opened this Issue Jul 19, 2018 · 4 comments

Comments

Projects
None yet
3 participants
@zoffixznet
Contributor

zoffixznet commented Jul 19, 2018

There's a whole bunch of what I think are pointless tickets in RT, like RT#124766, RT#124766, RT#124772, RT#124751, etc.

The tickets have no bodies and their titles just reference a location of a fudge in roast.

I'd like to close all of those tickets because:

  1. I can find all the fudges without tickets, by just running grep -Firn '#?rakudo' .
  2. There are more fudges than these tickets, because some fudges are just your generic "JVM doesn't support Unicode well" type of fudges, so the list of tickets isn't an exhaustive list of fudges
  3. When a fudge is resolved, you have to go through the extra steps to close the corresponding ticket. Since RT is not regularly used, personally, for me that'd mean going through the extra hassle of logging into RT
  4. These tickets give useless search results when trying to find other tickets. For example, I was trying to find other tickets for adverb ticket R#2098 and I got the search results littered with RT#124754, RT#124673, RT#124672, RT#124553, RT#124996, RT#124995, RT#124991, RT#124989 all of which just reference a line number of some NYI fudge instead of a proper ticket

@zoffixznet zoffixznet added the RFC label Jul 19, 2018

@coke

This comment has been minimized.

Show comment
Hide comment
@coke

coke Jul 19, 2018

Contributor

Part of the purpose of having those tickets was to force a discussion About whether these were actually intended to be part of the language or just cruft or whatever. We can have that discussion on the individual fudges versus the tickets that’s fine I’m OK with you closing them out. I would even be OK with removing any fudged test from roast - They can always be pulled from get history and add it back in if it’s decided that with they were actually desirable.

Contributor

coke commented Jul 19, 2018

Part of the purpose of having those tickets was to force a discussion About whether these were actually intended to be part of the language or just cruft or whatever. We can have that discussion on the individual fudges versus the tickets that’s fine I’m OK with you closing them out. I would even be OK with removing any fudged test from roast - They can always be pulled from get history and add it back in if it’s decided that with they were actually desirable.

@zoffixznet

This comment has been minimized.

Show comment
Hide comment
@zoffixznet

zoffixznet Jul 19, 2018

Contributor

having those tickets was to force a discussion

That would work if tickets actually had any content rather than just be a filename/line number for files that keep changing.

OK with removing any fudged test from roast

IMO:

  • Bug cover fudges remain in roast master
  • All the NYI fudges that don't have at least one working implementation get removed from roast and filed as proper tickets that can be discussed/implemented/rejected.

I mentioned it elsewhere: as of 6.d, the language spec will require a POV (Proof-Of-Viability) implementation for a feature to exist in at least one largely-compliant Perl 6 implementation for that feature to be included in a released spec. This way we don't spec untried language features that may conflict with something else or be a real-bad-idea once implemented and tried out.

Contributor

zoffixznet commented Jul 19, 2018

having those tickets was to force a discussion

That would work if tickets actually had any content rather than just be a filename/line number for files that keep changing.

OK with removing any fudged test from roast

IMO:

  • Bug cover fudges remain in roast master
  • All the NYI fudges that don't have at least one working implementation get removed from roast and filed as proper tickets that can be discussed/implemented/rejected.

I mentioned it elsewhere: as of 6.d, the language spec will require a POV (Proof-Of-Viability) implementation for a feature to exist in at least one largely-compliant Perl 6 implementation for that feature to be included in a released spec. This way we don't spec untried language features that may conflict with something else or be a real-bad-idea once implemented and tried out.

@AlexDaniel

This comment has been minimized.

Show comment
Hide comment
@AlexDaniel

AlexDaniel Jul 30, 2018

Member

Is there any way we can have an autogenerated list of fudges? Like something fancier than just grep? IMO if we had it, we'd be able to close all these tickets without even thinking.

Member

AlexDaniel commented Jul 30, 2018

Is there any way we can have an autogenerated list of fudges? Like something fancier than just grep? IMO if we had it, we'd be able to close all these tickets without even thinking.

@zoffixznet

This comment has been minimized.

Show comment
Hide comment
@zoffixznet

zoffixznet Jul 30, 2018

Contributor

Like something fancier than just grep?

What extra information/layout is needed?

Contributor

zoffixznet commented Jul 30, 2018

Like something fancier than just grep?

What extra information/layout is needed?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment