You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Link parsing is based on raw text, not on the displayed text without markup #127
When links are parsed their beginning and end indices are found in the input text which contains the raw pango markup, but the link regions are determined by using the same index in the final parsed text. This means that as markup is added, including the markup used in styling the links, the link region ends up in the wrong spot.
Tracked it down: parseText stores each link's begin/end as offsets into the raw marked-up string, but recheckTextBoxes uses them against pango's layout text, which is the markup-stripped version. So the link's own <u><span> wrapper (and any styling before it) pushes the region off the glyphs, worst on the second link onward. Fix in #158 counts displayed bytes for begin/end instead, with a couple of regression tests pinning the slice to the link text.
Activity
UncleJ4ck commentedon Nov 9, 2026
Tracked it down:
parseTextstores each link's begin/end as offsets into the raw marked-up string, butrecheckTextBoxesuses them against pango's layout text, which is the markup-stripped version. So the link's own<u><span>wrapper (and any styling before it) pushes the region off the glyphs, worst on the second link onward. Fix in #158 counts displayed bytes for begin/end instead, with a couple of regression tests pinning the slice to the link text.