| Commit-Queue | +1 |
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
If 25 was previously used for *both* FailedMaxDepth and FailedBigText, then should we actually renumber both of them? Otherwise the histograms will show an unexplained drop in FailedMaxDepth when this lands
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
If 25 was previously used for *both* FailedMaxDepth and FailedBigText, then should we actually renumber both of them? Otherwise the histograms will show an unexplained drop in FailedMaxDepth when this lands
[responding just to reset the 'resolved' bit]
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
If 25 was previously used for *both* FailedMaxDepth and FailedBigText, then should we actually renumber both of them? Otherwise the histograms will show an unexplained drop in FailedMaxDepth when this lands
Sure, we can do that, though there's going to be an unexplained discontinuity regardless. Do we want to bump the histogram version as well? I defer to the code owners here on their preferred solution.
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
Andrew PaseltinerIf 25 was previously used for *both* FailedMaxDepth and FailedBigText, then should we actually renumber both of them? Otherwise the histograms will show an unexplained drop in FailedMaxDepth when this lands
Sure, we can do that, though there's going to be an unexplained discontinuity regardless. Do we want to bump the histogram version as well? I defer to the code owners here on their preferred solution.
Re-numbered `kFailedMaxDepth` as well. Waiting for a determination on bumping the metric itself.
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
Andrew PaseltinerIf 25 was previously used for *both* FailedMaxDepth and FailedBigText, then should we actually renumber both of them? Otherwise the histograms will show an unexplained drop in FailedMaxDepth when this lands
Andrew PaseltinerSure, we can do that, though there's going to be an unexplained discontinuity regardless. Do we want to bump the histogram version as well? I defer to the code owners here on their preferred solution.
Re-numbered `kFailedMaxDepth` as well. Waiting for a determination on bumping the metric itself.
I'm happy with this, but I don't own the metric itself; if you're worried, you could reach out to html/parser OWNERS
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |