From 51fd9286b98d606a94d3b1ba25b218e56772299b Mon Sep 17 00:00:00 2001
From: megamaths <megamaths001@gmail.com>
Date: Fri, 1 Aug 2025 10:00:05 +0100
Subject: [PATCH 1/3] added basic markdown attrib and parsing

---
 src/elements/element.rs              |   2 +-
 src/elements/text.rs                 | 569 +++++++++++++++++++++++++--
 tests/integration_tests/text_attr.rs |  17 +
 3 files changed, 550 insertions(+), 38 deletions(-)

diff --git a/src/elements/element.rs b/src/elements/element.rs
index 1a52d5f..5cbc01a 100644
--- a/src/elements/element.rs
+++ b/src/elements/element.rs
@@ -357,7 +357,7 @@ impl SvgElement {
         // TODO: refactor this method to handle text event gen better
         let phantom = matches!(self.name(), "point" | "box");
 
-        if self.has_attr("text") {
+        if self.has_attr("text") || self.has_attr("md") {
             let (orig_elem, text_elements) = process_text_attr(self)?;
             if orig_elem.name != "text" && !phantom {
                 // We only care about the original element if it wasn't a text element
diff --git a/src/elements/text.rs b/src/elements/text.rs
index c90b049..efc6441 100644
--- a/src/elements/text.rs
+++ b/src/elements/text.rs
@@ -11,6 +11,279 @@ fn get_text_value(element: &mut SvgElement) -> String {
     text_string(&text_value)
 }
 
+fn get_md_value(element: &mut SvgElement) -> (Vec<String>, Vec<u32>) {
+    let text_value = element
+        .pop_attr("md")
+        .expect("no md attr in process_text_attr");
+
+    let (parsed_string, sections) = md_parse(&text_value);
+
+    let mut state_per_char = vec![0; parsed_string.len()];
+
+    for i in 0..sections.len() {
+        let bit = sections[i].code_bold_italic;
+        for j in sections[i].start_ind..sections[i].end_ind {
+            state_per_char[j] |= 1 << bit;
+        }
+    }
+
+    let mut strings = vec![];
+    let mut states = vec![];
+    for i in 0..parsed_string.len() {
+        if i == 0 || states[states.len() - 1] != state_per_char[i] {
+            strings.push(String::new());
+            states.push(state_per_char[i])
+        }
+        strings
+            .last_mut()
+            .expect("filled from i == 0")
+            .push(parsed_string[i]);
+    }
+
+    return (strings, states);
+}
+
+#[derive(Debug)]
+struct SectionData {
+    start_ind: usize,
+    end_ind: usize,
+    code_bold_italic: u32,
+}
+
+// based on the commonmarkdown implementation
+#[derive(Debug)]
+struct DelimiterData {
+    ind: usize, // goes just before this char
+    char_type: char,
+    num_delimiters: u32,
+    is_active: bool,
+    could_open: bool,
+    could_close: bool,
+}
+
+fn md_parse(text_value: &str) -> (Vec<char>, Vec<SectionData>) {
+    let mut sections = vec![];
+    let mut result = vec![];
+    let mut delimiters = vec![DelimiterData {
+        ind: 0,
+        char_type: ' ',
+        num_delimiters: 0,
+        is_active: false,
+        could_open: false,
+        could_close: false,
+    }];
+    let mut escaped = false;
+
+    // first pass process \ and find delimiters
+    for c in text_value.chars() {
+        let mut add = true;
+        if c == '\\' {
+            if !escaped {
+                add = false;
+                escaped = true;
+            } else {
+                escaped = false;
+            }
+        }
+        // the delimiters
+        else if c == '`' || c == '_' || c == '*' {
+            if !escaped {
+                let last = delimiters.last_mut().expect("garenteed not to be empty");
+                if c == last.char_type && last.ind == result.len() {
+                    // is a continuation
+                    last.num_delimiters += 1;
+                } else {
+                    delimiters.push(DelimiterData {
+                        ind: result.len(),
+                        char_type: c,
+                        num_delimiters: 1,
+                        is_active: true,
+                        could_open: true,
+                        could_close: true,
+                    });
+                }
+                add = false;
+            } else {
+                escaped = true;
+            }
+        } else if escaped {
+            if c == 'n' {
+                add = false;
+                result.push('\n');
+            } else {
+                // was not an escape
+                result.push('\\');
+            }
+            escaped = false;
+        }
+
+        if add {
+            result.push(c);
+        }
+    }
+
+    // set could open/close
+    for i in 0..delimiters.len() {
+        let prev_char;
+        let next_char;
+        if i != 0 && delimiters[i - 1].ind == delimiters[i].ind {
+            prev_char = delimiters[i - 1].char_type;
+        } else if delimiters[i].ind == 0 {
+            prev_char = ' ';
+        } else {
+            prev_char = result[delimiters[i].ind - 1];
+        }
+
+        if i != delimiters.len() - 1 && delimiters[i + 1].ind == delimiters[i].ind {
+            next_char = delimiters[i + 1].char_type;
+        } else if delimiters[i].ind == result.len() {
+            next_char = ' ';
+        } else {
+            next_char = result[delimiters[i].ind];
+        }
+
+        if next_char.is_whitespace() {
+            delimiters[i].could_open = false;
+        }
+        if prev_char.is_whitespace() {
+            delimiters[i].could_close = false;
+        }
+        if !next_char.is_whitespace()
+            && !prev_char.is_whitespace()
+            && delimiters[i].char_type == '_'
+        {
+            delimiters[i].could_open = false;
+            delimiters[i].could_close = false;
+        }
+
+        if next_char.is_ascii_punctuation()
+            && (!prev_char.is_whitespace() || !prev_char.is_ascii_punctuation())
+        {
+            delimiters[i].could_open = false;
+        }
+        if prev_char.is_ascii_punctuation()
+            && (!next_char.is_whitespace() || !next_char.is_ascii_punctuation())
+        {
+            delimiters[i].could_close = false;
+        }
+    }
+
+    let stack_bottom = 0; // because I have a null element in it
+    let mut current_position = stack_bottom + 1;
+    let mut opener_a = [stack_bottom; 3];
+    let mut opener_d = [stack_bottom; 3];
+    let mut opener_t = [stack_bottom; 3];
+
+    loop {
+        while current_position != delimiters.len()
+            && !delimiters[current_position].could_close
+            && delimiters[current_position].is_active
+        {
+            current_position += 1;
+        }
+        if current_position == delimiters.len() {
+            break;
+        }
+        let opener_min = match delimiters[current_position].char_type {
+            '*' => &mut opener_a,
+            '_' => &mut opener_d,
+            '`' => &mut opener_t,
+            _ => panic!(),
+        };
+        println!("{} {:?}", current_position, delimiters);
+
+        let min = opener_min[(delimiters[current_position].num_delimiters % 3) as usize]
+            .max(stack_bottom);
+        let mut opener_ind = current_position - 1;
+        while opener_ind > min {
+            // found opener
+            if delimiters[opener_ind].is_active
+                && delimiters[opener_ind].could_open
+                && delimiters[opener_ind].char_type == delimiters[current_position].char_type
+            {
+                if (delimiters[opener_ind].could_close || delimiters[current_position].could_open)
+                    && delimiters[opener_ind].num_delimiters % 3
+                        != delimiters[current_position].num_delimiters % 3
+                {
+                } else {
+                    break;
+                }
+            }
+            opener_ind -= 1;
+        }
+
+        if opener_ind == min {
+            // not found a opener
+            opener_min[(delimiters[current_position].num_delimiters % 3) as usize] =
+                current_position - 1;
+            current_position += 1;
+        } else {
+            delimiters[current_position].could_open = false;
+            delimiters[opener_ind].could_close = false;
+            // did
+            let code = delimiters[current_position].char_type == '`';
+            let strong = !code
+                && delimiters[opener_ind].num_delimiters >= 2
+                && delimiters[current_position].num_delimiters >= 2;
+            sections.push(SectionData {
+                start_ind: delimiters[opener_ind].ind,
+                end_ind: delimiters[current_position].ind,
+                code_bold_italic: if code {
+                    0
+                } else if strong {
+                    1
+                } else {
+                    2
+                },
+            });
+
+            println!("{} {} {}", opener_ind, current_position, strong);
+            delimiters[opener_ind].num_delimiters -= 1 + (strong as u32);
+            delimiters[current_position].num_delimiters -= 1 + (strong as u32);
+
+            if delimiters[opener_ind].num_delimiters == 0 {
+                delimiters[opener_ind].is_active = false;
+            }
+            if delimiters[current_position].num_delimiters == 0 {
+                delimiters[current_position].is_active = false;
+                current_position += 1;
+            }
+
+            for i in (opener_ind + 1)..current_position {
+                delimiters[i].is_active = false;
+            }
+        }
+    }
+    println!();
+
+    let mut final_result = vec![];
+
+    // work from the back to avoid index invalidation
+    for i in (0..delimiters.len()).rev() {
+        while delimiters[i].ind < result.len() {
+            if let Some(thing) = result.pop() {
+                final_result.push(thing);
+            }
+        }
+
+        for j in 0..sections.len() {
+            // if start needs to be after or equal
+            if sections[j].start_ind >= delimiters[i].ind {
+                sections[j].start_ind += delimiters[i].num_delimiters as usize;
+            }
+            if sections[j].end_ind > delimiters[i].ind {
+                // if end needs to be after
+                sections[j].end_ind += delimiters[i].num_delimiters as usize;
+            }
+        }
+        for _ in 0..delimiters[i].num_delimiters {
+            final_result.push(delimiters[i].char_type);
+        }
+    }
+
+    return (final_result.into_iter().rev().collect(), sections);
+}
+
 /// Convert unescaped r"\n" into newline characters for multi-line text
 fn text_string(text_value: &str) -> String {
     let mut result = String::new();
@@ -150,6 +423,27 @@ fn get_text_position(element: &mut SvgElement) -> Result<(f32, f32, bool, LocSpe
     Ok((tdx, tdy, outside, text_anchor, text_classes))
 }
 
+fn get_text_len(mono: bool, text: String) -> f32 {
+    if mono {
+        return 0.6 * text.len() as f32;
+    }
+    let mut length = 0.0;
+
+    let long = ['m', 'w'];
+    let short = ['f', 'i', 'j', 'l', 'r', 't'];
+    for i in text.chars() {
+        if long.contains(&i) {
+            length += 0.8;
+        } else if short.contains(&i) {
+            length += 0.33;
+        } else {
+            length += 0.6;
+        }
+    }
+
+    return length;
+}
+
 pub fn process_text_attr(element: &SvgElement) -> Result<(SvgElement, Vec<SvgElement>)> {
     // Different conversions from line count to first-line offset based on whether
     // top, center, or bottom justification.
@@ -162,22 +456,80 @@ pub fn process_text_attr(element: &SvgElement) -> Result<(SvgElement, Vec<SvgEle
 
     let mut orig_elem = element.clone();
 
-    let text_value = get_text_value(&mut orig_elem);
+    let text_values;
+    let state_values;
+    if let (Some(_), Some(_)) = (orig_elem.get_attr("text"), orig_elem.get_attr("md")) {
+        return Err(SvgdxError::ParseError(
+            "has both attributes of text and md".to_owned(),
+        ));
+    } else if let Some(_) = orig_elem.get_attr("text") {
+        text_values = vec![get_text_value(&mut orig_elem)];
+        state_values = vec![0];
+    } else {
+        // as to call must have one of them
+        (text_values, state_values) = get_md_value(&mut orig_elem);
+    }
+    let mut full_text_parsed_string = "".to_string();
+    for t in text_values.iter() {
+        full_text_parsed_string.push_str(&t);
+    }
 
     let (tdx, tdy, outside, text_loc, mut text_classes) = get_text_position(&mut orig_elem)?;
 
     let x_str = fstr(tdx);
     let y_str = fstr(tdy);
     let mut text_elements = Vec::new();
-    let mut lines: Vec<_> = text_value.lines().collect();
+    let mut lines = vec![vec![]];
+    let mut line_types = vec![vec![]];
+    for i in 0..text_values.len() {
+        let mut segments = text_values[i].lines();
+
+        if let Some(first) = segments.next() {
+            if first != "" {
+                lines
+                    .last_mut()
+                    .expect("added item not removed")
+                    .push(first);
+                line_types
+                    .last_mut()
+                    .expect("added item not removed")
+                    .push(state_values[i]);
+            } else if i != 0 {
+                lines.push(vec![]);
+                line_types.push(vec![]);
+            }
+        }
+
+        for s in segments {
+            lines.push(vec![s]);
+            line_types.push(vec![state_values[i]]);
+        }
+
+        if let Some(last_char) = text_values[i].chars().last() {
+            if last_char == '\n' && i != text_values.len() - 1 {
+                lines.push(vec![]);
+                line_types.push(vec![]);
+            }
+        }
+    }
+
+    for i in 0..lines.len() {
+        if lines[i].len() == 0 {
+            lines[i].push("");
+            line_types[i].push(0);
+        }
+    }
     let line_count = lines.len();
+    println!("{:?}", text_values[0].lines().collect::<Vec<_>>());
+    println!("{:?}", text_values);
+    println!("{:?}", lines);
 
-    let multiline = line_count > 1;
+    let multielement = line_count > 1 || text_values.len() > 1;
     let vertical = orig_elem.has_class("d-text-vertical");
     // Whether text is pre-formatted (i.e. spaces are not collapsed)
     let text_pre = orig_elem.has_class("d-text-pre");
 
-    // There will always be a text element; if not multiline this is the only element.
+    // There will always be a text element; if not multielement this is the only element.
     let mut text_elem = if orig_elem.name() == "text" {
         orig_elem.clone()
     } else {
@@ -276,9 +628,9 @@ pub fn process_text_attr(element: &SvgElement) -> Result<(SvgElement, Vec<SvgEle
             text_elem.set_attr(text_attr, &attr);
         }
     }
-    text_elem.text_content = Some(text_value.clone());
+    text_elem.text_content = Some(full_text_parsed_string.clone());
     text_elements.push(text_elem);
-    if multiline {
+    if multielement {
         // Determine position of first text line; others follow this based on line spacing
         let first_line_offset = match (outside, vertical, text_loc) {
             // shapes - text 'inside'
@@ -300,40 +652,66 @@ pub fn process_text_attr(element: &SvgElement) -> Result<(SvgElement, Vec<SvgEle
         }
         tspan_elem.src_line = orig_elem.src_line;
         if vertical {
-            tspan_elem.set_attr("y", &y_str);
             lines = lines.into_iter().rev().collect();
-        } else {
-            tspan_elem.set_attr("x", &x_str);
         }
-        for (idx, text_fragment) in lines.into_iter().enumerate() {
-            let mut text_fragment = text_fragment.to_string();
-            let mut tspan = tspan_elem.clone();
-            let line_offset = if idx == 0 {
-                first_line_offset(line_count, line_spacing)
-            } else {
-                line_spacing
-            };
-
-            if text_pre {
-                // Replace spaces with non-breaking spaces so they aren't collapsed
-                // by XML processing. This allows pre-formatted multi-line text (e.g. for
-                // code listings)
-                text_fragment = text_fragment.replace(' ', NBSP);
-            }
 
-            tspan.set_attr(
-                if vertical { "dx" } else { "dy" },
-                &format!("{}em", fstr(line_offset)),
-            );
-            tspan.text_content = Some(if text_fragment.is_empty() {
-                // Empty tspans don't take up vertical space, so use a zero-width space.
-                // Without this "a\n\nb" would render three tspans, but it would appear
-                // to have 'b' immediately below 'a' without a blank line between them.
-                ZWSP.to_string()
-            } else {
-                text_fragment.to_string()
-            });
-            text_elements.push(tspan);
+        for (idx, line) in lines.into_iter().enumerate() {
+            //let mut dist_along = 0.0;
+            for (idn, text_fragment) in line.into_iter().enumerate() {
+                let mut text_fragment = text_fragment.to_string();
+                let mut tspan = tspan_elem.clone();
+                if idn == 0 {
+                    if vertical {
+                        tspan.set_attr("y", &y_str);
+                    } else {
+                        tspan.set_attr("x", &x_str);
+                    }
+                }
+
+                if line_types[idx][idn] & (1 << 0) != 0 {
+                    tspan.add_class("d-text-monospace");
+                }
+                if line_types[idx][idn] & (1 << 1) != 0 {
+                    tspan.add_class("d-text-bold");
+                }
+                if line_types[idx][idn] & (1 << 2) != 0 {
+                    tspan.add_class("d-text-italic");
+                }
+
+                let line_offset = if idx == 0 {
+                    first_line_offset(line_count, line_spacing)
+                } else {
+                    line_spacing
+                };
+
+                if text_pre {
+                    // Replace spaces with non-breaking spaces so they aren't collapsed
+                    // by XML processing. This allows pre-formatted multi-line text (e.g. for
+                    // code listings)
+                    text_fragment = text_fragment.replace(' ', NBSP);
+                }
+
+                tspan.set_attr(
+                    if vertical { "dx" } else { "dy" },
+                    &format!("{}em", fstr(line_offset)),
+                );
+
+                //tspan.set_attr(
+                //    if vertical { "dy" } else { "dx" },
+                //    &format!("{}em", fstr(dist_along)),
+                //);
+                tspan.text_content = Some(if text_fragment.is_empty() {
+                    // Empty tspans don't take up vertical space, so use a zero-width space.
+                    // Without this "a\n\nb" would render three tspans, but it would appear
+                    // to have 'b' immediately below 'a' without a blank line between them.
+                    ZWSP.to_string()
+                } else {
+                    text_fragment.to_string()
+                });
+                text_elements.push(tspan);
+
+                //dist_along += get_text_len(line_types[idx][idn] & (1 << 0) != 0, text_fragment);
+            }
         }
     }
     Ok((orig_elem, text_elements))
@@ -356,4 +734,121 @@ mod tests {
         let text = r"Hello, \\nworld!";
         assert_eq!(text_string(text), r"Hello, \nworld!");
     }
+
+    #[test]
+    fn test_md_parse() {
+        // the basic examples no actual md
+
+        let text = r"Hello, \nworld!";
+        assert_eq!(
+            format!("{:?}", md_parse(text)),
+            "(['H', 'e', 'l', 'l', 'o', ',', ' ', '\\n', 'w', 'o', 'r', 'l', 'd', '!'], [])"
+        );
+
+        // when not part of a '\n', '\' is not special
+        let text = r"Hello, world! \1";
+        assert_eq!(format!("{:?}",md_parse(text)), "(['H', 'e', 'l', 'l', 'o', ',', ' ', 'w', 'o', 'r', 'l', 'd', '!', ' ', '\\\\', '1'], [])");
+
+        // when precedes '\n', '\' escapes it.
+        let text = r"Hello, \\nworld!";
+        assert_eq!(
+            format!("{:?}", md_parse(text)),
+            "(['H', 'e', 'l', 'l', 'o', ',', ' ', '\\\\', 'n', 'w', 'o', 'r', 'l', 'd', '!'], [])"
+        );
+
+        fn sd(s: i32, e: i32, i: i32) -> String {
+            "SectionData { start_ind: ".to_owned()
+                + &s.to_string()
+                + ", end_ind: "
+                + &e.to_string()
+                + ", code_bold_italic: "
+                + &i.to_string()
+                + " }"
+        }
+
+        // using the md
+        let text = r"He*ll*o, \nworld!";
+        assert_eq!(
+            format!("{:?}", md_parse(text)),
+            "(['H', 'e', 'l', 'l', 'o', ',', ' ', '\\n', 'w', 'o', 'r', 'l', 'd', '!'], ["
+                .to_owned()
+                + &sd(2, 4, 2)
+                + "])"
+        );
+
+        // mismatched
+        let text = r"*Hello** , \nworld!";
+        assert_eq!(
+            format!("{:?}", md_parse(text)),
+            "(['H', 'e', 'l', 'l', 'o', '*', ' ', ',', ' ', '\\n', 'w', 'o', 'r', 'l', 'd', '!'], ["
+                .to_owned()
+                + &sd(0, 5, 2) + "])"
+        );
+
+        // diff type
+        let text = r"He*llo_, \nworld!";
+        assert_eq!(format!("{:?}",md_parse(text)), "(['H', 'e', '*', 'l', 'l', 'o', '_', ',', ' ', '\\n', 'w', 'o', 'r', 'l', 'd', '!'], [])");
+
+        // multiple diff type
+        let text = r"_hello*";
+        assert_eq!(
+            format!("{:?}", md_parse(text)),
+            "(['_', 'h', 'e', 'l', 'l', 'o', '*'], [])"
+        );
+
+        // multiple same type
+        let text = r"He*ll*o, \nw*or*ld!";
+        assert_eq!(
+            format!("{:?}", md_parse(text)),
+            "(['H', 'e', 'l', 'l', 'o', ',', ' ', '\\n', 'w', 'o', 'r', 'l', 'd', '!'], ["
+                .to_owned()
+                + &sd(2, 4, 2)
+                + ", "
+                + &sd(9, 11, 2)
+                + "])"
+        );
+
+        // space before
+        let text = r"**foo bar **";
+        assert_eq!(
+            format!("{:?}", md_parse(text)),
+            "(['*', '*', 'f', 'o', 'o', ' ', 'b', 'a', 'r', ' ', '*', '*'], [])"
+        );
+
+        // punctuation before alphnum after
+        let text = r"**(**foo)";
+        assert_eq!(
+            format!("{:?}", md_parse(text)),
+            "(['*', '*', '(', '*', '*', 'f', 'o', 'o', ')'], [])"
+        );
+    }
+
+    #[test]
+    fn test_get_md_value() {
+        let mut el = SvgElement::new("text", &[]);
+        let text = r"foo";
+        el.set_attr("md", text);
+        assert_eq!(format!("{:?}", get_md_value(&mut el)), "([\"foo\"], [0])");
+
+        let text = r"**(**foo)";
+        el.set_attr("md", text);
+        assert_eq!(
+            format!("{:?}", get_md_value(&mut el)),
+            "([\"**(**foo)\"], [0])"
+        );
+
+        let text = r"*foo *bar**";
+        el.set_attr("md", text);
+        assert_eq!(
+            format!("{:?}", get_md_value(&mut el)),
+            "([\"foo bar\"], [4])"
+        );
+
+        let text = r"*foo**bar**baz*";
+        el.set_attr("md", text);
+        assert_eq!(
+            format!("{:?}", get_md_value(&mut el)),
+            "([\"foo\", \"bar\", \"baz\"], [4, 6, 4])"
+        );
+    }
 }
diff --git a/tests/integration_tests/text_attr.rs b/tests/integration_tests/text_attr.rs
index 2df86ec..c011d80 100644
--- a/tests/integration_tests/text_attr.rs
+++ b/tests/integration_tests/text_attr.rs
@@ -614,3 +614,20 @@ fn test_multiline_outside() {
         expected.trim()
     );
 }
+
+#[test]
+fn test_md() {
+    let input = r#"
+<rect xy="0" wh="10" md="m*ult*i\nline" text-loc="br" class="d-text-outside"/>
+"#;
+    let expected = r#"
+<rect x=\"0\" y=\"0\" width=\"10\" height=\"10\"/>
+<text x=\"11\" y=\"11\" class=\"d-text d-text-top d-text-left\">
+<tspan x=\"11\" dy=\"0em\">m</tspan><tspan dy=\"0em\" class=\"d-text-italic\">ult</tspan><tspan dy=\"0em\">i</tspan><tspan x=\"11\" dy=\"1.05em\">line</tspan>
+</text>
+"#;
+    assert_eq!(
+        transform_str_default(input).unwrap().trim(),
+        expected.trim()
+    );
+}

From 6476018acfbaa98a8df05164ecb57e5d2fccaead Mon Sep 17 00:00:00 2001
From: megamaths <megamaths001@gmail.com>
Date: Tue, 5 Aug 2025 14:50:29 +0100
Subject: [PATCH 2/3] clean the printlines and fixed problem with single
 element md not adding styles

---
 src/elements/text.rs | 40 +++++++++++++---------------------------
 1 file changed, 13 insertions(+), 27 deletions(-)

diff --git a/src/elements/text.rs b/src/elements/text.rs
index efc6441..7fd6d04 100644
--- a/src/elements/text.rs
+++ b/src/elements/text.rs
@@ -190,7 +190,6 @@ fn md_parse(text_value: &str) -> (Vec<char>, Vec<SectionData>) {
             '`' => &mut opener_t,
             _ => panic!(),
         };
-        println!("{} {:?}", current_position, delimiters);
 
         let min = opener_min[(delimiters[current_position].num_delimiters % 3) as usize]
             .max(stack_bottom);
@@ -206,6 +205,7 @@ fn md_parse(text_value: &str) -> (Vec<char>, Vec<SectionData>) {
                         != delimiters[current_position].num_delimiters % 3
                 {
                 } else {
+                    // found valid opener
                     break;
                 }
             }
@@ -237,7 +237,6 @@ fn md_parse(text_value: &str) -> (Vec<char>, Vec<SectionData>) {
                 },
             });
 
-            println!("{} {} {}", opener_ind, current_position, strong);
             delimiters[opener_ind].num_delimiters -= 1 + (strong as u32);
             delimiters[current_position].num_delimiters -= 1 + (strong as u32);
 
@@ -254,7 +253,6 @@ fn md_parse(text_value: &str) -> (Vec<char>, Vec<SectionData>) {
             }
         }
     }
-    println!();
 
     let mut final_result = vec![];
 
@@ -423,27 +421,6 @@ fn get_text_position(element: &mut SvgElement) -> Result<(f32, f32, bool, LocSpe
     Ok((tdx, tdy, outside, text_anchor, text_classes))
 }
 
-fn get_text_len(mono: bool, text: String) -> f32 {
-    if mono {
-        return 0.6 * text.len() as f32;
-    }
-    let mut length = 0.0;
-
-    let long = ['m', 'w'];
-    let short = ['f', 'i', 'j', 'l', 'r', 't'];
-    for i in text.chars() {
-        if long.contains(&i) {
-            length += 0.8;
-        } else if short.contains(&i) {
-            length += 0.33;
-        } else {
-            length += 0.6;
-        }
-    }
-
-    return length;
-}
-
 pub fn process_text_attr(element: &SvgElement) -> Result<(SvgElement, Vec<SvgElement>)> {
     // Different conversions from line count to first-line offset based on whether
     // top, center, or bottom justification.
@@ -520,9 +497,6 @@ pub fn process_text_attr(element: &SvgElement) -> Result<(SvgElement, Vec<SvgEle
         }
     }
     let line_count = lines.len();
-    println!("{:?}", text_values[0].lines().collect::<Vec<_>>());
-    println!("{:?}", text_values);
-    println!("{:?}", lines);
 
     let multielement = line_count > 1 || text_values.len() > 1;
     let vertical = orig_elem.has_class("d-text-vertical");
@@ -596,6 +570,18 @@ pub fn process_text_attr(element: &SvgElement) -> Result<(SvgElement, Vec<SvgEle
             text_classes.push(class);
         }
     }
+
+    if !multielement{
+        if line_types[0][0] & (1 << 0) != 0 {
+            text_classes.push("d-text-monospace".to_string());
+        }
+        if line_types[0][0] & (1 << 1) != 0 {
+            text_classes.push("d-text-bold".to_string());
+        }
+        if line_types[0][0] & (1 << 2) != 0 {
+            text_classes.push("d-text-italic".to_string());
+        }
+    }
     text_elem.src_line = orig_elem.src_line;
     text_elem.set_classes(&text_classes);
 

From ea19bb048bf17a67b65dc610d4808a9def9c2b5d Mon Sep 17 00:00:00 2001
From: megamaths <megamaths001@gmail.com>
Date: Tue, 5 Aug 2025 16:25:22 +0100
Subject: [PATCH 3/3] split markdown parser into multiple functions

---
 src/elements/text.rs | 134 ++++++++++++++++++++++++-------------------
 1 file changed, 74 insertions(+), 60 deletions(-)

diff --git a/src/elements/text.rs b/src/elements/text.rs
index 7fd6d04..9ef60b2 100644
--- a/src/elements/text.rs
+++ b/src/elements/text.rs
@@ -20,10 +20,10 @@ fn get_md_value(element: &mut SvgElement) -> (Vec<String>, Vec<u32>) {
 
     let mut state_per_char = vec![0; parsed_string.len()];
 
-    for i in 0..sections.len() {
-        let bit = sections[i].code_bold_italic;
-        for j in sections[i].start_ind..sections[i].end_ind {
-            state_per_char[j] |= 1 << bit;
+    for s in sections {
+        let bit = s.code_bold_italic;
+        for i in s.start_ind..s.end_ind {
+            state_per_char[i] |= 1 << bit;
         }
     }
 
@@ -55,14 +55,13 @@ struct SectionData {
 struct DelimiterData {
     ind: usize, // goes just before this char
     char_type: char,
-    num_delimiters: u32,
+    num_delimiters: usize,
     is_active: bool,
     could_open: bool,
     could_close: bool,
 }
 
-fn md_parse(text_value: &str) -> (Vec<char>, Vec<SectionData>) {
-    let mut sections = vec![];
+fn md_parse_escapes_and_delimiters(text_value: &str) -> (Vec<char>, Vec<DelimiterData>) {
     let mut result = vec![];
     let mut delimiters = vec![DelimiterData {
         ind: 0,
@@ -77,17 +76,16 @@ fn md_parse(text_value: &str) -> (Vec<char>, Vec<SectionData>) {
     // first pass process \ and find delimiters
     for c in text_value.chars() {
         let mut add = true;
-        if c == '\\' {
-            if !escaped {
+        match (c, escaped) {
+            ('\\', false) => {
                 add = false;
                 escaped = true;
-            } else {
-                escaped = false;
             }
-        }
-        // the delimiters
-        else if c == '`' || c == '_' || c == '*' {
-            if !escaped {
+            ('\\', true) => {
+                escaped = true;
+            }
+            // the delimiters
+            ('`', false) | ('_', false) | ('*', false) => {
                 let last = delimiters.last_mut().expect("garenteed not to be empty");
                 if c == last.char_type && last.ind == result.len() {
                     // is a continuation
@@ -103,25 +101,31 @@ fn md_parse(text_value: &str) -> (Vec<char>, Vec<SectionData>) {
                     });
                 }
                 add = false;
-            } else {
-                escaped = true;
             }
-        } else if escaped {
-            if c == 'n' {
+            ('`', true) | ('_', true) | ('*', true) => {
+                escaped = false;
+            }
+            ('n', true) => {
                 add = false;
                 result.push('\n');
-            } else {
+                escaped = false;
+            }
+            (_, true) => {
                 // was not an escape
                 result.push('\\');
+                escaped = false;
             }
-            escaped = false;
+            (_, false) => {}
         }
-
         if add {
             result.push(c);
         }
     }
 
+    return (result, delimiters);
+}
+
+fn md_parse_set_delimiter_open_close(result: &Vec<char>, delimiters: &mut Vec<DelimiterData>) {
     // set could open/close
     for i in 0..delimiters.len() {
         let prev_char;
@@ -142,18 +146,23 @@ fn md_parse(text_value: &str) -> (Vec<char>, Vec<SectionData>) {
             next_char = result[delimiters[i].ind];
         }
 
-        if next_char.is_whitespace() {
-            delimiters[i].could_open = false;
-        }
-        if prev_char.is_whitespace() {
-            delimiters[i].could_close = false;
-        }
-        if !next_char.is_whitespace()
-            && !prev_char.is_whitespace()
-            && delimiters[i].char_type == '_'
-        {
-            delimiters[i].could_open = false;
-            delimiters[i].could_close = false;
+        match (prev_char.is_whitespace(), next_char.is_whitespace()) {
+            (false, false) => {
+                if delimiters[i].char_type == '_' {
+                    delimiters[i].could_open = false;
+                    delimiters[i].could_close = false;
+                }
+            }
+            (true, false) => {
+                delimiters[i].could_close = false;
+            }
+            (false, true) => {
+                delimiters[i].could_open = false;
+            }
+            (true, true) => {
+                delimiters[i].could_open = false;
+                delimiters[i].could_close = false;
+            }
         }
 
         if next_char.is_ascii_punctuation()
@@ -167,7 +176,10 @@ fn md_parse(text_value: &str) -> (Vec<char>, Vec<SectionData>) {
             delimiters[i].could_close = false;
         }
     }
+}
 
+fn md_parse_eval_sections(delimiters: &mut Vec<DelimiterData>) -> Vec<SectionData> {
+    let mut sections = vec![];
     let stack_bottom = 0; // because I have a null element in it
     let mut current_position = stack_bottom + 1;
     let mut opener_a = [stack_bottom; 3];
@@ -188,11 +200,10 @@ fn md_parse(text_value: &str) -> (Vec<char>, Vec<SectionData>) {
             '*' => &mut opener_a,
             '_' => &mut opener_d,
             '`' => &mut opener_t,
-            _ => panic!(),
+            _ => panic!("this cant happen as current_position starts at 0 and all other delimiters are of above types"),
         };
 
-        let min = opener_min[(delimiters[current_position].num_delimiters % 3) as usize]
-            .max(stack_bottom);
+        let min = opener_min[delimiters[current_position].num_delimiters % 3].max(stack_bottom);
         let mut opener_ind = current_position - 1;
         while opener_ind > min {
             // found opener
@@ -214,8 +225,7 @@ fn md_parse(text_value: &str) -> (Vec<char>, Vec<SectionData>) {
 
         if opener_ind == min {
             // not found a opener
-            opener_min[(delimiters[current_position].num_delimiters % 3) as usize] =
-                current_position - 1;
+            opener_min[delimiters[current_position].num_delimiters % 3] = current_position - 1;
             current_position += 1;
         } else {
             delimiters[current_position].could_open = false;
@@ -228,17 +238,15 @@ fn md_parse(text_value: &str) -> (Vec<char>, Vec<SectionData>) {
             sections.push(SectionData {
                 start_ind: delimiters[opener_ind].ind,
                 end_ind: delimiters[current_position].ind,
-                code_bold_italic: if code {
-                    0
-                } else if strong {
-                    1
-                } else {
-                    2
+                code_bold_italic: match (code, strong) {
+                    (true, _) => 0,
+                    (_, true) => 1,
+                    (_, _) => 2,
                 },
             });
 
-            delimiters[opener_ind].num_delimiters -= 1 + (strong as u32);
-            delimiters[current_position].num_delimiters -= 1 + (strong as u32);
+            delimiters[opener_ind].num_delimiters -= 1 + (strong as usize);
+            delimiters[current_position].num_delimiters -= 1 + (strong as usize);
 
             if delimiters[opener_ind].num_delimiters == 0 {
                 delimiters[opener_ind].is_active = false;
@@ -248,35 +256,41 @@ fn md_parse(text_value: &str) -> (Vec<char>, Vec<SectionData>) {
                 current_position += 1;
             }
 
-            for i in (opener_ind + 1)..current_position {
-                delimiters[i].is_active = false;
+            for d in &mut delimiters[(opener_ind + 1)..current_position] {
+                d.is_active = false;
             }
         }
     }
+    return sections;
+}
+
+fn md_parse(text_value: &str) -> (Vec<char>, Vec<SectionData>) {
+    let (mut result, mut delimiters) = md_parse_escapes_and_delimiters(text_value);
+    md_parse_set_delimiter_open_close(&result, &mut delimiters);
+    let mut sections = md_parse_eval_sections(&mut delimiters);
 
     let mut final_result = vec![];
 
     // work from the back to avoid index invalidation
-    for i in (0..delimiters.len()).rev() {
-        while delimiters[i].ind < result.len() {
+    for d in delimiters.into_iter().rev() {
+        while d.ind < result.len() {
             if let Some(thing) = result.pop() {
                 final_result.push(thing);
             }
         }
 
-        for j in 0..sections.len() {
+        for s in sections.iter_mut() {
             // if start needs to be after or equal
-            if sections[j].start_ind >= delimiters[i].ind {
-                sections[j].start_ind += delimiters[i].num_delimiters as usize;
+            if s.start_ind >= d.ind {
+                s.start_ind += d.num_delimiters as usize;
             }
-            if sections[j].end_ind > delimiters[i].ind {
+            if s.end_ind > d.ind {
                 // if end needs to be after
-                sections[j].end_ind += delimiters[i].num_delimiters as usize;
+                s.end_ind += d.num_delimiters as usize;
             }
         }
-        for _ in 0..delimiters[i].num_delimiters {
-            final_result.push(delimiters[i].char_type);
-        }
+        let mut temp = vec![d.char_type; d.num_delimiters];
+        final_result.append(&mut temp);
     }
 
     return (final_result.into_iter().rev().collect(), sections);
@@ -571,7 +585,7 @@ pub fn process_text_attr(element: &SvgElement) -> Result<(SvgElement, Vec<SvgEle
         }
     }
 
-    if !multielement{
+    if !multielement {
         if line_types[0][0] & (1 << 0) != 0 {
             text_classes.push("d-text-monospace".to_string());
         }