fix raw reader syntax overprocessing bug

Author: cscheid

crates/quarto-markdown-pandoc/src/pandoc/treesitter.rs

@@ -390,7 +390,7 @@ fn process_native_inlines<T: Write>(
                     }))
                 } else {
                     inlines.push(Inline::Str(Str {
-                        text,
+                        text: apply_smart_quotes(text),
                         source_info: SourceInfo::new(
                             if context.filenames.is_empty() {
                                 None

crates/quarto-markdown-pandoc/src/pandoc/treesitter_utils/postprocess.rs

@@ -7,15 +7,13 @@ use crate::filters::{
     Filter, FilterReturn::FilterResult, FilterReturn::Unchanged, topdown_traverse,
 };
 use crate::pandoc::attr::{Attr, is_empty_attr};
-use crate::pandoc::block::{Block, Figure, Plain, RawBlock};
+use crate::pandoc::block::{Block, Figure, Plain};
 use crate::pandoc::caption::Caption;
 use crate::pandoc::inline::{Inline, Inlines, Space, Span, Str, Superscript};
 use crate::pandoc::location::{Range, SourceInfo, empty_range, empty_source_info};
 use crate::pandoc::pandoc::Pandoc;
 use crate::pandoc::shortcode::shortcode_to_span;
 use crate::utils::autoid;
-use once_cell::sync::Lazy;
-use regex::Regex;
 use std::collections::HashMap;
 
 /// Trim leading and trailing spaces from inlines
@@ -171,8 +169,6 @@ pub fn coalesce_abbreviations(inlines: Vec<Inline>) -> (Vec<Inline>, bool) {
 /// Apply post-processing transformations to the Pandoc AST
 pub fn postprocess(doc: Pandoc) -> Result<Pandoc, Vec<String>> {
     let mut errors = Vec::new();
-    let raw_reader_format_specifier: Lazy<Regex> =
-        Lazy::new(|| Regex::new(r"<(?P<reader>.+)").unwrap());
     let result = {
         // Track seen header IDs to avoid duplicates
         let mut seen_ids: HashMap<String, usize> = HashMap::new();
@@ -458,18 +454,6 @@ pub fn postprocess(doc: Pandoc) -> Result<Pandoc, Vec<String>> {
 
                 FilterResult(result, true)
             })
-            .with_raw_block(move |raw_block| {
-                let Some(captures) = raw_reader_format_specifier.captures(&raw_block.text) else {
-                    return Unchanged(raw_block);
-                };
-                return FilterResult(
-                    vec![Block::RawBlock(RawBlock {
-                        format: "pandoc-reader:".to_string() + &captures["reader"],
-                        ..raw_block
-                    })],
-                    false,
-                );
-            })
             .with_attr(|attr| {
                 // TODO in order to do good error messages here, attr will need source mapping
                 errors.push(format!(

crates/quarto-markdown-pandoc/tests/pandoc-match-corpus/markdown/055.qmd

@@ -0,0 +1 @@
+[User's](#users)

crates/quarto-markdown-pandoc/tests/pandoc-match-corpus/markdown/056.qmd

@@ -0,0 +1 @@
+![User's](image.png)

crates/quarto-markdown-pandoc/tests/pandoc-match-corpus/markdown/057.qmd

@@ -0,0 +1,3 @@
+```{=html}
+<iframe class="slide-deck" src="demo/"></iframe>
+```

crates/tree-sitter-qmd/tree-sitter-markdown/grammar.js

@@ -522,15 +522,6 @@ module.exports = grammar({
         $._fenced_code_block_end_backtick,
         $._fenced_code_block_end_tilde,
 
-        $._html_block_1_start,
-        $._html_block_1_end,
-        $._html_block_2_start,
-        $._html_block_3_start,
-        $._html_block_4_start,
-        $._html_block_5_start,
-        $._html_block_6_start,
-        $._html_block_7_start,
-
         // Similarly this is used if the closing of a block is not decided by the external parser.
         // A `$._block_close` will be emitted at the beginning of the next line. Notice that a
         // `$._block_close` can also get emitted if the parent block closes.

crates/tree-sitter-qmd/tree-sitter-markdown/src/grammar.json

@@ -5608,38 +5608,6 @@
       "type": "SYMBOL",
       "name": "_fenced_code_block_end_tilde"
     },
-    {
-      "type": "SYMBOL",
-      "name": "_html_block_1_start"
-    },
-    {
-      "type": "SYMBOL",
-      "name": "_html_block_1_end"
-    },
-    {
-      "type": "SYMBOL",
-      "name": "_html_block_2_start"
-    },
-    {
-      "type": "SYMBOL",
-      "name": "_html_block_3_start"
-    },
-    {
-      "type": "SYMBOL",
-      "name": "_html_block_4_start"
-    },
-    {
-      "type": "SYMBOL",
-      "name": "_html_block_5_start"
-    },
-    {
-      "type": "SYMBOL",
-      "name": "_html_block_6_start"
-    },
-    {
-      "type": "SYMBOL",
-      "name": "_html_block_7_start"
-    },
     {
       "type": "SYMBOL",
       "name": "_close_block"