This isn't a Mac issue.

Refactoring existing code in a way that is useful across more of Chromium's code base is always good.

@a...

Well, it's not easy to create a bug that's not based on crash or a defect in chromium. The only way I know is to use report bug for ChromeOs and then I can write arbitrary text. However, I'm still forced to check operating system.

You might want to give more specifics on what was discussed already here and how you see this being solved.

We have a number of cases where we union/intersect a range of sets.

Standard library provides functions: std::set_union/std::set_intersection that do union/intersection for ordered containers. In this task I suggest to write similar functions that work on many sets. Since often we do not have sets directly but rather they are obtained as the result of some expression, some support for this should be provided.

This functions have to be benchmarked - at some point I did union for many sets like std::set_union (I used std::priority_queue to select minimum on each step) - it was slower than putting everything in a vector and sorting it,

Here is some rough sketch for intersect - https://godbolt.org/g/vgCVmr - if a person who does this task finds it useful.

I think that this task is worth doing because implementing union/intersect well is not easy and we would have one place for it, where it can be done really good if necessary.

I also think that it's ok not to have a very high quality implementation on the first try since we'll have one place to improve it and tests.

This issue has been available for more than 365 days with no owner or cc list, so archiving this bug. Please re-open or file a new bug if this is still an issue.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot