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
The acceptsTableEdit is run on every row. This assumes, that the row belongs to a table, before actually checking if it is part of table.
This actually exposes a faulty behavior in Advanced Tables Obsidian plugin, and noticed in platers/obsidian-lint#827. AT relies on the info provided in the MetadataCache, but by the time the format command is executed, cache is already invalidated. The code in this project implicitly relies on cache, but it doesn't have to.
I think, the correct behavior would be to move this check a bit later, when all the lines that belong to a table already found, and execute acceptsTableEdit on one of the lines. Is there some deeper meaning to the current implementation, that I didn't notice?
The text was updated successfully, but these errors were encountered:
When reviewing the sources, I found this statement.
md-advanced-tables/src/table-editor.ts
Line 1060 in 609c512
The
acceptsTableEdit
is run on every row. This assumes, that the row belongs to a table, before actually checking if it is part of table.This actually exposes a faulty behavior in Advanced Tables Obsidian plugin, and noticed in platers/obsidian-lint#827. AT relies on the info provided in the
MetadataCache
, but by the time the format command is executed, cache is already invalidated. The code in this project implicitly relies on cache, but it doesn't have to.I think, the correct behavior would be to move this check a bit later, when all the lines that belong to a table already found, and execute
acceptsTableEdit
on one of the lines. Is there some deeper meaning to the current implementation, that I didn't notice?The text was updated successfully, but these errors were encountered: