Request: new gem release

[swrobel]

To use features like oneshot_line that have been added over the last few years, we have to use simplecov from github. A new gem release would be appreciated!

[PragTob]

@sferik is making a hard push for 1.0 (finally). That said we can probably cut a 0.99 or rc.0 type release at some point before that.

[sferik]

I'm hoping to ship 1.0 this week or early next week. I think it's just blocked on #1165.

[beanieboi]

@sferik thank you so much <3

[sferik]

Quick update: this is taking a bit longer than expected but I‘m making good progress.

There are currently ~100 open issues that I’d like to quickly review before shipping 1.0.0, just to ensure we don’t ship it with any known bugs.

If you’re very eager to test 1.0.0 before it’s released, you can specify your simplecov dependency from a git source in your Gemfile:

gem "simplecov", git: "https://github.com/simplecov-ruby/simplecov.git", branch: "main", require: false

If you decide to do this and find any issues, please report them!

[mjankowski]

I've been running from main in a branch for a few weeks and periodically updating and running the suite. The basics of "does this generate coverage output on basic spec runs?" seems to be functioning as normal/expected.

I enabled the eval coverage setting to get view/haml coverage (previously disabled due to all the warning output), and that looks good as well (on single core runs).

Only oddity I've seen is that some combinations of formatters and parallel runners seem to either drop the eval coverage, or maybe not re-combine it correctly when on parallel runs. I'll open an issue for this but want to try to narrow it down slightly better first to see if the issue is in the formatter, the parallel runner, or simplecov itself.

[sferik]

@mjankowski Thanks so much for testing pre-release v1.0.0. Good to hear that it’s been working overall! I’m looking forward to your bug report. No pressure, but I’ll probably hold the v1.0.0 release until your issue is reported/resolved, so please let me know how your investigation goes, even if it turns out not to be a SimpleCov issue.

[mjankowski]

I'm doing some local runs now in various combinations of runners, formatters, etc. Will open issue to summarize when done.

From my runs so far, I suspect that there is NOT a regression from previous release to current main ... but rather something that was already not working (or not configured correctly on my side!), and remains that way after the update.

So, maybe worth addressing, but probably not a release blocker. Will confirm in issue soon.

[sferik]

There are currently ~100 open issues that I’d like to quickly review before shipping 1.0.0, just to ensure we don’t ship it with any known bugs.

I was able to process the number of open issues down to 88. I’ve been using some AI tools to label every issues with a priority (High, Medium, or Low) and a difficulty (Easy, Medium, or Hard), as you can see this issue has been tagged above. Every open issue has also been assigned a type (Bug, Feature, or Task). Any issue with potentially breaking changes has been labeled high priority and is a blocker for the v1.0.0 release. There are currently 15 such high priority issues, including this one. I’m also working to close any issues that are duplicates, stale, or outdated (e.g. issues that only exist on older, unsupported Ruby versions, such as #387).

Based on the past couple days of working on this, my optimistic estimate is that v1.0.0 could be released next week (by May 16), my realistic estimate is that it will be released in two weeks (by May 23), and my pessimistic estimate is that it will be released in 4 weeks (by June 6).