toolkit: the face follows the metric scale - #452
Merged
Conversation
Every pixel constant in this toolkit is multiplied by MetricScale. The OpenType face was not — so on a display with two device pixels to the point, an application got bitmap chrome at the right size around type at half of it, and had to notice and re-render the face itself. That is per-application work for a toolkit-wide fact, and it was reported as "small but readable", which is a polite way of saying wrong. UseOpenTypeTextSize now takes LOGICAL pixels, like every other metric here: the face is rendered at sizePx × MetricScale, and re-rendered whenever that scale changes. A caller asking for 16 gets type that reads the same size on every display, which is the point of having a scale at all. SetFont(nil) forgets the requested size as well as the face. Back to the bitmap means back to the bitmap — otherwise the next SetMetricScale would quietly re-install a face nobody wants, which is exactly what broke a neighbouring test and is how this was caught. A failed re-render leaves the working face in place: a toolkit that dropped its text because a resize could not re-render it would be worse than one whose text is briefly the wrong size. The bitmap default already followed the scale — its metrics go through scaled() like everything else. That is what made the face the odd one out. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
A scale small enough to round the size to nothing gives one pixel, not zero: type of size zero is not small type, it is no text at all. And a face that cannot be built leaves the working one in place — dropping the text because a resize could not re-render it would be worse than text briefly the wrong size. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Every pixel constant in this toolkit is multiplied by
MetricScale. The OpenType face was not — so on a display with two device pixels to the point, an application got bitmap chrome at the right size around type at half of it, and had to notice and re-render the face itself. Per-application work for a toolkit-wide fact.UseOpenTypeTextSizenow takes logical pixels, like every other metric here: the face is rendered atsizePx × MetricScaleand re-rendered whenever that scale changes;SetFont(nil)forgets the requested size as well as the face — back to the bitmap means back to the bitmap, otherwise the nextSetMetricScalequietly re-installs a face nobody wants. That is what broke a neighbouring test, and how this was caught;The bitmap default already followed the scale — its metrics go through
scaled()like everything else. That is what made the face the odd one out.🤖 Generated with Claude Code