From d1c99982c7e6ac9c6a70289ec8db8a48126aa5e2 Mon Sep 17 00:00:00 2001
From: Ermia Behzadifar <potatohater111@gmail.com>
Date: Tue, 12 Nov 2024 16:34:27 +0100
Subject: [PATCH] Start working on cog parser. Codeblock parser seems to work

---
 Cargo.lock                | 26 +++++++++++
 crates/ast/src/lib.rs     |  5 +++
 crates/parser/Cargo.toml  |  2 +
 crates/parser/src/lib.rs  | 90 ++++++++++++++++++++++++++++++++++-----
 crates/parser/src/main.rs |  5 +++
 5 files changed, 117 insertions(+), 11 deletions(-)
 create mode 100644 crates/parser/src/main.rs

diff --git a/Cargo.lock b/Cargo.lock
index 4ff19a7..65d2e31 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -17,3 +17,29 @@ version = "0.1.0"
 [[package]]
 name = "cogs_parser"
 version = "0.1.0"
+dependencies = [
+ "cogs_ast",
+ "nom",
+]
+
+[[package]]
+name = "memchr"
+version = "2.7.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "78ca9ab1a0babb1e7d5695e3530886289c18cf2f87ec19a575a0abdce112e3a3"
+
+[[package]]
+name = "minimal-lexical"
+version = "0.2.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "68354c5c6bd36d73ff3feceb05efa59b6acb7626617f4962be322a825e61f79a"
+
+[[package]]
+name = "nom"
+version = "7.1.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d273983c5a657a70a3e8f2a01329822f3b8c8172b73826411a55751e404a0a4a"
+dependencies = [
+ "memchr",
+ "minimal-lexical",
+]
diff --git a/crates/ast/src/lib.rs b/crates/ast/src/lib.rs
index 90d2fd8..7cbfc53 100644
--- a/crates/ast/src/lib.rs
+++ b/crates/ast/src/lib.rs
@@ -1,23 +1,28 @@
+#[derive(Debug)]
 pub struct Component {
     pub elements: Vec<Element>,
 }
 
+#[derive(Debug)]
 pub enum Element {
     Html(HtmlTag),
     Block(CodeBlock),
 }
 
+#[derive(Debug)]
 pub struct HtmlTag {
     pub tag: String,
     pub attributes: Vec<Attribute>,
     pub content: Vec<Element>,
 }
 
+#[derive(Debug)]
 pub struct Attribute {
     pub name: String,
     pub value: String,
 }
 
+#[derive(Debug)]
 pub struct CodeBlock {
     pub is_async: bool,
     pub content: String,
diff --git a/crates/parser/Cargo.toml b/crates/parser/Cargo.toml
index 1f550d8..78a4b88 100644
--- a/crates/parser/Cargo.toml
+++ b/crates/parser/Cargo.toml
@@ -4,3 +4,5 @@ version = "0.1.0"
 edition = "2021"
 
 [dependencies]
+nom = "7.1.3"
+cogs_ast.path = "../ast"
\ No newline at end of file
diff --git a/crates/parser/src/lib.rs b/crates/parser/src/lib.rs
index b93cf3f..590f6d8 100644
--- a/crates/parser/src/lib.rs
+++ b/crates/parser/src/lib.rs
@@ -1,14 +1,82 @@
-pub fn add(left: u64, right: u64) -> u64 {
-    left + right
+use nom::{
+    IResult,
+    character::{
+        complete::{char, multispace0},
+        streaming::none_of
+    },
+    combinator::{opt, value},
+    branch::alt,
+    multi::{many0, many1},
+    sequence::{delimited, pair, preceded, terminated},
+};
+
+use cogs_ast::{Component, Element, HtmlTag, Attribute, CodeBlock};
+
+pub fn parse_cog(input: &str) -> IResult<&str, Component> {
+    let (input, elements) = many0(parse_element)(input)?;
+    Ok((input, Component { elements }))
 }
 
-#[cfg(test)]
-mod tests {
-    use super::*;
-
-    #[test]
-    fn it_works() {
-        let result = add(2, 2);
-        assert_eq!(result, 4);
-    }
+fn parse_element(input: &str) -> IResult<&str, Element> {
+    alt((parse_html_tag, parse_code_block))(input)
 }
+
+fn parse_html_tag(input: &str) -> IResult<&str, Element> {
+    let (input, (tag, (attributes, content))) = delimited(
+        char('<'),
+        pair(
+            terminated(many1(none_of(" >/\n")), many0(multispace0)),
+            pair(
+                opt(delimited(char(' '), many1(parse_attribute), char('>'))),
+                many0(parse_element),
+            ),
+        ),
+        char('>'),
+    )(input)?;
+
+    println!("got result {tag:?}");
+
+    Ok((
+        input,
+        Element::Html(HtmlTag {
+            tag: tag.iter().collect(),
+            attributes: attributes.unwrap(),
+            content,
+        }),
+    ))
+}
+
+fn parse_attribute(input: &str) -> IResult<&str, Attribute> {
+    println!("Attempting to parse attribute");
+    let (input, (name, value)) = pair(
+        many1(none_of("= ")),
+        preceded(char('='), delimited(char('"'), many0(none_of("\"")), char('"'))),
+    )(input)?;
+
+    Ok((
+        input,
+        Attribute {
+            name: name.iter().collect(),
+            value: value.iter().collect(),
+        },
+    ))
+}
+
+fn parse_code_block(input: &str) -> IResult<&str, Element> {
+    let (input, (is_async, content)) = delimited(
+        char('{'),
+        pair(
+            opt(value(true, preceded(char('a'), char('s')))),
+            many1(none_of("}")),
+        ),
+        char('}'),
+    )(input)?;
+
+    Ok((
+        input,
+        Element::Block(CodeBlock {
+            is_async: is_async.unwrap_or(false),
+            content: content.into_iter().collect(),
+        }),
+    ))
+}
\ No newline at end of file
diff --git a/crates/parser/src/main.rs b/crates/parser/src/main.rs
new file mode 100644
index 0000000..fded43c
--- /dev/null
+++ b/crates/parser/src/main.rs
@@ -0,0 +1,5 @@
+use cogs_parser::parse_cog;
+
+fn main() {
+    println!("{:#?}", parse_cog("{println!(\"Hello\")}\n<h1>hi</h1>"));
+}
\ No newline at end of file