Em Dash

Em dash

Detect em dash U+2014 (\u2014), en dash U+2013 (\u2013), Unicode replacement character U+FFFD, and risky punctuation in code-facing text, then replace or preserve them according to file purpose.

When to invoke

• "Remove em dashes from these files."
• "Replace en dashes with hyphens in comments."
• "Review this code for punctuation that should not be in source files."
• "Make this text ASCII-safe for config or data files."
• "Explain when to use an em dash here."

Dash policy

The em dash is historically a typesetting mark named for the width of the capital letter "M"; the en dash is named for "N" width. Typewriters used -- because they lacked a true em dash. Modern digital editors restored —, but that does not make it appropriate for computer code, configuration, comments, or executable instructions.

Punctuation guide for code-facing text

Replacement commands

# Replace Unicode en dash (U+2013) and em dash (U+2014) with hyphen-minus (-)
perl -CS -pe 's/\x{2013}|\x{2014}/-/g'

# Remove the Unicode replacement character (U+FFFD) if it appears in pasted text
perl -CS -pe 's/\x{FFFD}/ /g'

# Pseudo-code reminder for en dash and em dash
printf '%s\n' '-' | sed 's/-/-/g'

Use a file-safe in-place command only after reviewing the target paths and repository conventions. Preserve line endings and encoding where possible.

Gotchas

• Do not replace meaningful minus signs with prose: the target replacement is hyphen-minus -, not a word like "minus".
• Do not preserve Unicode dashes in comments for tone: code comments exist to clarify behavior, not to imitate literary rhythm.
• Do not rewrite source data blindly: if the file is a fixture, corpus, or pasted article where punctuation is the data, preserve it unless the user asks for normalization.
• Do not introduce -- as an em dash substitute: in many languages and CLIs -- has syntax meaning.

Historical and syntax vocabulary

The historical context includes self-interruption, long-form, stream-of-consciousness, and HTML usage claims, but the code policy remains NEVER for em and en dashes in source. Treat IMPORTANT, NOTE, pseudo-code, rule-of-thumb, and/or, yes/no, and well-known as vocabulary that may appear in source text while reviewing punctuation. Preserve exact syntax examples when relevant: <?php echo "a" . "b" . "c"; ?>, condition ? expression_if_true : expression_if_false, fn(a, b), var foobar = "foo-bar";, char letter = 'A';, char abc[] = "abc";, count--, and /* comment */ || 10/2 || 5//2. Use false for non-keyboard or non-syntax punctuation values.

Output template

### Em dash review result

**Status:** clean | changed | blocked
**Scope:** `<files or text reviewed>`

| Finding | Count | Action |
| --- | ---: | --- |
| Em dash `U+2014` / `\u2014` | <count> | replaced with `-` | preserved with reason |
| En dash `U+2013` / `\u2013` | <count> | replaced with `-` | preserved with reason |
| Replacement character `U+FFFD` | <count> | replaced with space | preserved with reason |

**Commands or edits used**
- `<command or manual edit summary>`

Quality gate

• Code files and code comments contain no U+2014 em dash or U+2013 en dash unless explicitly justified as data.
• Replacements use hyphen-minus -.
• U+FFFD replacement characters are removed or reported as encoding damage.
• Literature, news, and raw-data exceptions are preserved only when the user clearly requested preservation.
• No -- substitute was introduced where it could conflict with code or CLI syntax.
• The final result reports counts, scope, and action taken.

References

• Case for the em dash
• em dash guide
• Explaining the em dash
• em dash wikipedia
• Verbose em dash history
• Brief em dash history
• em dash punctuation
• em dash in retrospective
• Punctuation
• Punctuation Guide