Skip to content

Japanese bold patterns like 太字**「…」**直後に文字 are not converted (CommonMark flanking rules) #263

Description

@katanumahotori

Summary

In Japanese prose, bold spans wrapped in corner brackets and immediately followed by a character — e.g.

ですがあるとき、**「掴んだ」**という感覚がありました。

— are not converted to <strong> by markdown-it-py. The raw ** asterisks appear in the published article.

This is spec-compliant CommonMark behavior (the closing ** is preceded by punctuation and followed by a letter, so it is not right-flanking), not a bug in this MCP. Filing as a feature request because the pattern is extremely common in Japanese writing and the failure is silent and ugly on note.com.

Hit 9 occurrences in a single real article (2026-06-10).

Possible approaches

  1. Post-process unconverted **...** spans after markdown-it conversion (fallback to <strong>), similar to what Japanese blog tooling commonly does.
  2. Or document the limitation in README and recommend inline <strong>…</strong> (which works today and is what I used as a workaround).

Happy to contribute option 1 if you think it fits the project.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions