Keep delimiters from pairing across runs in one paragraph

[AbhinavMir]

Runs that still have content after them now escape their pairable delimiters, so at most one raw delimiter survives per paragraph and nothing can pair with it.

For issue #45.

---

^{Need help on this PR? Tag @codesmith-bot with what you need. Autofix is disabled.}

[abimaelmartell]

@cubic-dev-ai

[cubic-dev-ai]

@cubic-dev-ai

@abimaelmartell I have started the AI code review. It will take a few minutes to complete.

[cubic-dev-ai]

1 issue found across 3 files

Prompt for AI agents (unresolved issues)

Check if these issues are valid — if so, understand the root cause of each and fix them. If appropriate, use sub-agents to investigate and fix each issue separately.


<file name="src/render/markdown/escape.rs">

<violation number="1" location="src/render/markdown/escape.rs:73">
P2: The new `pairs_ahead` in `paired` does not escape a `*`, `_`, or `~` that sits at the very end of a run and is followed by more of the paragraph (a line break or anchor then more plain text). For those characters the escape is gated by `next_nonspace`, which at run end falls back to `trailing_active` (false when the next run is plain), so the trailing delimiter stays raw and can pair with a later run's raw delimiter — e.g. ["a *", LineBreak, "b *"] renders two unescaped `*` in one paragraph, defeating the fix. Consider making the run-end `next_nonspace` fall back to `trailing_active || pairs_ahead` (or short-circuit on `pairs_ahead`) so a delimiter before more paragraph content is always escaped.</violation>
</file>

<sub>Reply with feedback, questions, or to request a fix.

Fix all with cubic | Re-trigger cubic</sub>

[cubic-dev-ai]

P2: The new pairs_ahead in paired does not escape a *, _, or ~ that sits at the very end of a run and is followed by more of the paragraph (a line break or anchor then more plain text). For those characters the escape is gated by next_nonspace, which at run end falls back to trailing_active (false when the next run is plain), so the trailing delimiter stays raw and can pair with a later run's raw delimiter — e.g. ["a *", LineBreak, "b *"] renders two unescaped * in one paragraph, defeating the fix. Consider making the run-end next_nonspace fall back to trailing_active || pairs_ahead (or short-circuit on pairs_ahead) so a delimiter before more paragraph content is always escaped.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. At src/render/markdown/escape.rs, line 73:

<comment>The new `pairs_ahead` in `paired` does not escape a `*`, `_`, or `~` that sits at the very end of a run and is followed by more of the paragraph (a line break or anchor then more plain text). For those characters the escape is gated by `next_nonspace`, which at run end falls back to `trailing_active` (false when the next run is plain), so the trailing delimiter stays raw and can pair with a later run's raw delimiter — e.g. ["a *", LineBreak, "b *"] renders two unescaped `*` in one paragraph, defeating the fix. Consider making the run-end `next_nonspace` fall back to `trailing_active || pairs_ahead` (or short-circuit on `pairs_ahead`) so a delimiter before more paragraph content is always escaped.</comment>

<file context>
@@ -64,7 +69,8 @@ pub(crate) fn escape_text(text: &str, ctx: InlineContext, opts: EscapeOpts) -> S
         let next_nonspace = next.map_or(trailing_active, |n| !n.is_whitespace());
-        let paired = |slot: usize| trailing_active || last[slot].is_some_and(|j| j > i);
+        let paired =
+            |slot: usize| trailing_active || pairs_ahead || last[slot].is_some_and(|j| j > i);
         let escape = match c {
             '\\' => true,
</file context>