moved to comrak v0.24.1

This commit is contained in:
2024-10-10 20:05:23 -04:00
parent 9cc32fe65a
commit 84d7ba45d3
4 changed files with 302 additions and 30 deletions

View File

@@ -2,9 +2,12 @@ use crate::todo::{File as TodoFile, Status as TaskStatus};
use crate::NaiveDate;
use crate::TaskGroup;
use chrono::Datelike;
use comrak::nodes::{AstNode, NodeValue};
use comrak::parse_document;
use comrak::{Arena, ComrakExtensionOptions, ComrakOptions, ComrakParseOptions};
use comrak::nodes::{Ast, AstNode, LineColumn, NodeHeading, NodeValue};
use comrak::{
format_commonmark, parse_document, Arena, ComrakOptions, ExtensionOptions, ParseOptions,
};
use indexmap::IndexMap;
use regex::Regex;
use std::collections::HashMap;
use std::fs::{read, File};
use std::io::Write;
@@ -62,15 +65,15 @@ pub fn load_file(file: &TodoFile) -> String {
/// Parse contents of markdown file with Comrak ( relaxed tasklist matching is enabled)
pub fn parse_todo_file<'a>(contents: &String, arena: &'a Arena<AstNode<'a>>) -> &'a AstNode<'a> {
let mut extension_options = ExtensionOptions::default();
extension_options.tasklist = true;
let mut parse_options = ParseOptions::default();
parse_options.relaxed_tasklist_matching = true;
let options = &ComrakOptions {
extension: ComrakExtensionOptions {
tasklist: true,
..ComrakExtensionOptions::default()
},
parse: ComrakParseOptions {
relaxed_tasklist_matching: true,
..ComrakParseOptions::default()
},
extension: extension_options,
parse: parse_options,
..ComrakOptions::default()
};
parse_document(arena, contents, options)
@@ -117,10 +120,178 @@ pub fn extract_secitons<'a>(
groups
}
fn remove_heading<'a>(node: &'a AstNode<'a>, level: u8) {
let mut following = node.following_siblings();
let _ = following.next().unwrap();
for sib in following {
let node_ref = sib.data.borrow();
if let NodeValue::Heading(heading) = node_ref.value {
if heading.level == level {
break;
}
} else {
sib.detach();
}
}
node.detach();
}
/// recursively removes nodes from List
fn remove_task_nodes<'a>(root: &'a AstNode<'a>) {
for node in root.children() {
for child_node in node.children() {
remove_task_nodes(child_node)
}
match node.data.borrow().value {
NodeValue::TaskItem(Some(status)) if status == 'x' || status == 'X' => node.detach(),
_ => continue,
}
}
}
fn create_title<'a>(arena: &'a Arena<AstNode<'a>>, date: &str) -> &'a AstNode<'a> {
let mut text = String::new();
text.push_str("Today's tasks ");
text.push_str(date);
create_heading(arena, 1, &text)
}
fn create_heading<'a>(arena: &'a Arena<AstNode<'a>>, level: u8, text: &str) -> &'a AstNode<'a> {
let heading_node = arena.alloc(AstNode::new(
Ast::new(
NodeValue::Heading(NodeHeading {
level,
setext: false,
}),
LineColumn { line: 0, column: 0 },
)
.into(),
));
let text_node = arena.alloc(AstNode::new(
Ast::new(
NodeValue::Text(text.to_string()),
LineColumn { line: 0, column: 2 },
)
.into(),
));
heading_node.append(text_node);
heading_node
}
pub fn create_new_doc<'a>(
arena: &'a Arena<AstNode<'a>>,
new_date: &str,
sections: IndexMap<String, Option<Vec<&'a AstNode<'a>>>>,
) -> &'a AstNode<'a> {
let doc = arena.alloc(AstNode::new(
Ast::new(NodeValue::Document, LineColumn { line: 0, column: 0 }).into(),
));
let title = create_title(&arena, new_date);
doc.append(title);
for (section, value) in sections.iter() {
let heading = create_heading(arena, 2, &section);
doc.append(heading);
match value {
Some(nodes) => {
for node in nodes.iter() {
doc.append(node);
}
}
_ => (),
}
}
doc
}
pub fn extract_sections<'a>(
root: &'a AstNode<'a>,
sections: &Vec<String>,
) -> IndexMap<String, Option<Vec<&'a AstNode<'a>>>> {
let mut section_map: IndexMap<String, Option<Vec<&'a AstNode<'a>>>> = IndexMap::new();
sections.iter().for_each(|section| {
section_map.insert(section.to_string(), None);
});
for node in root.reverse_children() {
let node_ref = node.data.borrow();
match node_ref.value {
NodeValue::Heading(heading) => {
let heading_content_node = if let Some(child) = node.first_child() {
child
} else {
continue;
};
let mut heading_content_ref = heading_content_node.data.borrow_mut();
if let NodeValue::Text(text) = &mut heading_content_ref.value {
if sections.contains(text) {
let mut content = Vec::new();
let mut following = node.following_siblings();
let _ = following.next().unwrap();
for sib in following {
remove_task_nodes(sib);
let node_ref = sib.data.borrow();
if let NodeValue::Heading(inner_heading) = node_ref.value {
if heading.level == inner_heading.level {
break;
}
} else {
content.push(sib);
}
}
section_map.insert(text.to_string(), Some(content));
remove_heading(node, heading.level);
};
}
}
_ => continue,
}
}
section_map
}
pub fn process_doc_tree<'a>(root: &'a AstNode<'a>, new_date: &str, sections: &Vec<String>) {
for node in root.reverse_children() {
let node_ref = node.data.borrow();
match node_ref.value {
NodeValue::Heading(heading) => {
let heading_content_node = if let Some(child) = node.first_child() {
child
} else {
continue;
};
let mut heading_content_ref = heading_content_node.data.borrow_mut();
if let NodeValue::Text(text) = &mut heading_content_ref.value {
let re = Regex::new(r"Today's tasks \d+-\d+-\d+")
.expect("title regex is not parsable");
if matches!(re.find(text), Some(_)) {
text.clear();
text.push_str("Today's tasks ");
text.push_str(new_date);
} else if !sections.contains(text) {
remove_heading(node, heading.level);
};
}
}
NodeValue::List(_list) => remove_task_nodes(node),
_ => continue,
}
}
eprintln!("{:#?}", root);
}
#[cfg(test)]
mod test {
use super::*;
use crate::todo::{Status, Task};
use std::io::BufWriter;
#[test]
fn test_extract_sections() {
@@ -278,4 +449,91 @@ mod test {
";
assert_eq!(result, expected);
}
#[test]
fn test_node_removal() {
let md = "
# Today's tasks 2024-01-01
## Tasks
- [ ] task 1
- [X] task 2
- [x] task 2
- [>] task 3
- [!] task 3
## Long Term
- [ ] task 1
- [X] task 2
- [ ] all of these subtasks should be removed
- [x] subtasks
- [x] sub task to remove
- [!] task 3
- [ ] sub task to keep
- [x] sub task to remove
## Todays Notes
- some notes here
- these can go
";
let new_date = "2024-01-02";
let groups = vec![
"Tasks".to_string(),
"Other".to_string(),
"Long Term".to_string(),
"Last".to_string(),
];
let arena = Arena::new();
let mut extension_options = ExtensionOptions::default();
extension_options.tasklist = true;
let mut parse_options = ParseOptions::default();
parse_options.relaxed_tasklist_matching = true;
let options = &ComrakOptions {
extension: extension_options,
parse: parse_options,
..ComrakOptions::default()
};
let ast = parse_document(&arena, md, options);
let sections = extract_sections(ast, &groups);
let new_doc = create_new_doc(&arena, new_date, sections);
process_doc_tree(ast, new_date, &groups);
let mut output = BufWriter::new(Vec::new());
assert!(format_commonmark(new_doc, options, &mut output).is_ok());
let bytes = output.into_inner().expect("should be a vec");
let text = String::from_utf8(bytes).expect("should be convertable to string");
assert_eq!(
"\
# Today's tasks 2024-01-02
## Tasks
- [ ] task 1
- [>] task 3
- [!] task 3
## Other
## Long Term
- [ ] task 1
- [!] task 3
- [ ] sub task to keep
## Last
",
text
);
}
}