Hi, this CL is ready, please help review, thanks~
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
// Note: When invoking BrowserController::ForEachBrowser in
// OnBrowserCreated, the new browser will show up for
// kAscendingCreationTime but not yet for kAscendingActivationTime.
// TODO(crbug.com/369688254): Revisit this behavior.Do not have permission to check this bug. But I think it should gone after https://chromium-review.googlesource.com/c/chromium/src/+/7080234
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
Thanks for the changes, left some comments.
ash::BrowserController::BrowserOrder::kAscendingCreationTime,Curious: how do we know that this ordering isn't an important assumption in any of the touched components?
if (callback(*GetDelegate(browser)) == kBreakIteration) {
return false; // stop iterating
}
return true; // continue iteratingnit, a bit shorter:
```suggestion
return callback(*GetDelegate(browser)) != kBreakIteration;
```
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
ash::BrowserController::BrowserOrder::kAscendingCreationTime,Curious: how do we know that this ordering isn't an important assumption in any of the touched components?
Our current plan is to remove the dependency on the browser creation order. Therefore, I'll first try proposing a CL and running a CQ dry run to see the results. As far as I know, this CL will definitely change some shelf behaviors, so this CL will definitely need owner review.
if (callback(*GetDelegate(browser)) == kBreakIteration) {
return false; // stop iterating
}
return true; // continue iteratingnit, a bit shorter:
```suggestion
return callback(*GetDelegate(browser)) != kBreakIteration;
```
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
| Code-Review | +1 |
Yury could you ptal for ash shelf?
ash::BrowserController::BrowserOrder::kAscendingCreationTime,Curious: how do we know that this ordering isn't an important assumption in any of the touched components?
Qikai is right and we shouldn't be exposing browser ordering directly to clients.
It looks like the shelf context menu is the only client that is specifically interested in this.
Looking over the expectations it's not clear that ordering is strictly important - the comments explicitly mention testing for the presence of the browser (not its position), and the positional assertions may just have just been a convenient way to test for this.
@kh...@chromium.org do you happen to know if creation order is important on the ash shelf context menu?
// Note: When invoking BrowserController::ForEachBrowser in
// OnBrowserCreated, the new browser will show up for
// kAscendingCreationTime but not yet for kAscendingActivationTime.
// TODO(crbug.com/369688254): Revisit this behavior.Do not have permission to check this bug. But I think it should gone after https://chromium-review.googlesource.com/c/chromium/src/+/7080234
Yep! The issue the comment mentions should have been resolved by your CL (the bug itself tracks a more general issue so we can leave that open).
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |