| 17:08 | <Michael Ficarra> | ljharb: Yulia had done a lot of work on it in a time when I was very busy and couldn't follow, so she would be the best person to ask about what remains |
| 17:09 | <yulia> | Right, there are a few small unresolved topics that i need to get to |
| 17:09 | <yulia> | the major one is this: https://github.com/tc39/proposal-iterator-helpers/issues/122 |
| 17:10 | <yulia> | i haven't had time to dig into this properly due to other priorities |
| 17:10 | <Michael Ficarra> | yulia: welcome back! |
| 17:10 | <yulia> | Thanks |
| 17:10 | <yulia> | I am happy to sit down with folks and hammer this out + update firefox's implementation to match |
| 17:11 | <yulia> | the other issues in the repo, last i checked, were less significant |
| 17:11 | <yulia> | I can see about getting it done by the end of the week, its been on my list for way too long |
| 17:12 | <yulia> | Though, feedback / opinions would be very welcome... |
| 17:12 | <yulia> | maybe, Michael Ficarra -- I summarize the proposed change, bring it to committee, and then the following meeting we try to advance? |
| 17:13 | <Michael Ficarra> | the following meeting isn't until June, so I think that gives us plenty of time, yes |
| 17:26 | <ljharb> | the issue title still confuses me, i continue to have no idea what "passing the protocol" means |
| 17:42 | <Michael Ficarra> | ljharb: it's just whether return/throw on an iterator produced by iterator helpers behave the same as if they were called on the original iterator |
| 17:43 | <Michael Ficarra> | see https://github.com/tc39/proposal-iterator-helpers/issues/122#issuecomment-909432348 |
| 17:43 | <ljharb> | ah k, thanks |
| 17:43 | <ljharb> | so like, forwarding the full iterator API (ie, return/throw)? |
| 17:44 | <Michael Ficarra> | yep |
| 17:44 | <Michael Ficarra> | it's harder to do than just making next work |
| 17:44 | <Michael Ficarra> | the unfortunate bit is that a whole class of usages only care that next works |
| 17:45 | <yulia> | the argument against passing the full iterator api is that most of these methods are not intended to be communicating generators, and in this case does it make sense to pass return and throw |
| 17:48 | <ljharb> | ugh, return and throw are gross |
| 17:52 | <Michael Ficarra> | archival footage of TC39 adding generators to the language |
| 17:54 | <ljharb> | please don't besmirch the good name of shoehorns |
| 18:00 | <yulia> | hm, it may get tricky in relation to the other issues like this one: https://github.com/tc39/proposal-iterator-helpers/issues/174 |
| 18:00 | <yulia> | I'll do a more detailed review later this week |
| 19:10 | <bakkot> | Michael Ficarra: I think we were all agreed that return and throw should get forwarded and the remaining question is whether .next forwards its argument |
| 19:10 | <bakkot> | return needs to get forwarded so that the underlying iterator is properly closed |
| 19:11 | <bakkot> | (and throw doesn't really matter because nothing in the language calls it) |
| 19:23 | <Mathieu Hofman> | hm, it may get tricky in relation to the other issues like this one: https://github.com/tc39/proposal-iterator-helpers/issues/174 |
| 19:47 | <yulia> | Mathieu Hofman: great, i will take you up on that |
| 19:48 | <yulia> | maybe this won't make it to the end of the week, it is hairier than i remember |
| 19:48 | <yulia> | but if you have time maybe we can sit down towards thursday/fri? |
| 19:48 | <yulia> | or early next week |
| 19:48 | <yulia> | i guess i can put a discussion item on the agenda anyway |