feat: gemtext parser, pre tag

Cargo.lock

@@ -8,6 +8,33 @@ version = "1.0.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "f26201604c87b1e01bd3d98f8d5d9a8fcbb815e8cedb41ffccbeb4bf593a35fe"
 
+[[package]]
+name = "ahash"
+version = "0.8.11"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e89da841a80418a9b391ebaea17f5c112ffaaa96f621d2c285b5174da76b9011"
+dependencies = [
+ "cfg-if",
+ "once_cell",
+ "version_check",
+ "zerocopy",
+]
+
+[[package]]
+name = "aho-corasick"
+version = "1.1.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8e60d3430d3a69478ad0993f19238d2df97c507009a52b3c10addcd7f6bcb916"
+dependencies = [
+ "memchr",
+]
+
+[[package]]
+name = "allocator-api2"
+version = "0.2.18"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5c6cb57a04249c6480766f7f7cef5467412af1490f8d1e243141daddada3264f"
+
 [[package]]
 name = "anstream"
 version = "0.6.15"
@@ -94,6 +121,19 @@ version = "1.0.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "baf1de4339761588bc0619e3cbc0120ee582ebb74b53b4efbf79117bd2da40fd"
 
+[[package]]
+name = "chumsky"
+version = "1.0.0-alpha.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c7b80276986f86789dc56ca6542d53bba9cda3c66091ebbe7bd96fc1bdf20f1f"
+dependencies = [
+ "hashbrown",
+ "regex-automata",
+ "serde",
+ "stacker",
+ "unicode-ident",
+]
+
 [[package]]
 name = "clap"
 version = "4.5.13"
@@ -154,6 +194,7 @@ name = "dalet"
 version = "1.0.0-pre4"
 dependencies = [
  "bincode",
+ "chumsky",
  "clap",
  "flate2",
  "num_enum",
@@ -184,6 +225,10 @@ name = "hashbrown"
 version = "0.14.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "e5274423e17b7c9fc20b6e7e208532f9b19825d82dfd615708b70edd83df41f1"
+dependencies = [
+ "ahash",
+ "allocator-api2",
+]
 
 [[package]]
 name = "heck"
@@ -267,6 +312,12 @@ dependencies = [
  "syn",
 ]
 
+[[package]]
+name = "once_cell"
+version = "1.19.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3fdb12b2476b595f9358c5161aa467c2438859caa136dec86c26fdd2efe17b92"
+
 [[package]]
 name = "paste"
 version = "1.0.15"
@@ -297,6 +348,15 @@ dependencies = [
  "unicode-ident",
 ]
 
+[[package]]
+name = "psm"
+version = "0.1.21"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5787f7cda34e3033a72192c018bc5883100330f362ef279a8cbccfce8bb4e874"
+dependencies = [
+ "cc",
+]
+
 [[package]]
 name = "quote"
 version = "1.0.36"
@@ -306,6 +366,23 @@ dependencies = [
  "proc-macro2",
 ]
 
+[[package]]
+name = "regex-automata"
+version = "0.3.9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "59b23e92ee4318893fa3fe3e6fb365258efbfe6ac6ab30f090cdcbb7aa37efa9"
+dependencies = [
+ "aho-corasick",
+ "memchr",
+ "regex-syntax",
+]
+
+[[package]]
+name = "regex-syntax"
+version = "0.7.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "dbb5fb1acd8a1a18b3dd5be62d25485eb770e05afb408a9627d14d451bae12da"
+
 [[package]]
 name = "rmp"
 version = "0.8.14"
@@ -359,6 +436,19 @@ dependencies = [
  "syn",
 ]
 
+[[package]]
+name = "stacker"
+version = "0.1.15"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c886bd4480155fd3ef527d45e9ac8dd7118a898a46530b7b94c3e21866259fce"
+dependencies = [
+ "cc",
+ "cfg-if",
+ "libc",
+ "psm",
+ "winapi",
+]
+
 [[package]]
 name = "strsim"
 version = "0.11.1"
@@ -405,6 +495,34 @@ version = "0.2.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "06abde3611657adf66d383f00b093d7faecc7fa57071cce2578660c9f1010821"
 
+[[package]]
+name = "version_check"
+version = "0.9.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0b928f33d975fc6ad9f86c8f283853ad26bdd5b10b7f1542aa2fa15e2289105a"
+
+[[package]]
+name = "winapi"
+version = "0.3.9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5c839a674fcd7a98952e593242ea400abe93992746761e38641405d28b00f419"
+dependencies = [
+ "winapi-i686-pc-windows-gnu",
+ "winapi-x86_64-pc-windows-gnu",
+]
+
+[[package]]
+name = "winapi-i686-pc-windows-gnu"
+version = "0.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ac3b87c63620426dd9b991e5ce0329eff545bccbbb34f3be09ff6fb6ab51b7b6"
+
+[[package]]
+name = "winapi-x86_64-pc-windows-gnu"
+version = "0.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "712e227841d057c1ee1cd2fb22fa7e5a5461ae8e48fa2ca79ec42cfc1931183f"
+
 [[package]]
 name = "windows-sys"
 version = "0.52.0"
@@ -487,6 +605,26 @@ dependencies = [
  "memchr",
 ]
 
+[[package]]
+name = "zerocopy"
+version = "0.7.35"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1b9b4fd18abc82b8136838da5d50bae7bdea537c574d8dc1a34ed098d6c166f0"
+dependencies = [
+ "zerocopy-derive",
+]
+
+[[package]]
+name = "zerocopy-derive"
+version = "0.7.35"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "fa4f8080344d4671fb4e831a13ad1e68092748387dfc4f55e356242fae12ce3e"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn",
+]
+
 [[package]]
 name = "zstd"
 version = "0.13.2"

Cargo.toml

@@ -17,6 +17,7 @@ num_enum = "0.7.3"
 serde = { version = "1.0", features = ["derive"] }
 serde_repr = "0.1"
 zstd = "0.13.2"
+chumsky = { version = "=1.0.0-alpha.7", features = ["label"], optional = true }
 
 [dev-dependencies]
 rmp-serde = { version = "1.3.0" }
@@ -25,6 +26,7 @@ flate2 = "1.0"
 
 
 [features]
-default = ["types", "daletpack"]
+default = ["types", "daletpack", "parsers"]
+parsers = ["dep:chumsky", "types"]
 types = []
 daletpack = ["types"]

src/abstractions.rs

@@ -36,6 +36,7 @@ pub enum Tag {
     Bl(NotNullBody, AlignArgument),
     Carousel(Vec<Tag>),
     Code(String, TextOrNullArgument),
+    Pre(String),
 }
 
 pub trait ToDaletl {
@@ -78,6 +79,7 @@ impl ToDaletlTag for Tag {
             Tag::Bl(b, a) => t_new(Tid::Bl, b.to_daletl_body(), a.to_daletl_argument()),
             Tag::Carousel(b) => t_new(Tid::Carousel, b.to_daletl_body(), NA),
             Tag::Code(s, a) => t_new(Tid::Code, s.to_daletl_body(), a.to_daletl_argument()),
+            Tag::Pre(s) => t_new(Tid::Pre, s.to_daletl_body(), NA),
         }
     }
 }

src/daletl.rs

@@ -92,4 +92,5 @@ pub enum Tid {
     Bl,
     Carousel,
     Code,
+    Pre,
 }

src/lib.rs

@@ -6,3 +6,6 @@ pub mod abstractions;
 
 #[cfg(feature = "daletpack")]
 pub mod daletpack;
+
+#[cfg(feature = "parsers")]
+pub mod parsers;

src/parsers/gemtext.rs

@@ -0,0 +1,68 @@
+use crate::abstractions::{Body, HeadingLevel, NotNullBody, Tag};
+
+#[derive(Debug)]
+pub enum GemTextParseError {
+    InvalidLink,
+}
+
+pub fn parse_gemtext(s: String) -> Result<Vec<Tag>, GemTextParseError> {
+    let mut page: Vec<Tag> = Vec::new();
+    let mut preformatted = false;
+    let mut preformatted_text = String::new();
+
+    let mut before_is_ordered_list = false;
+    let mut ordered_list: Vec<Tag> = Vec::new();
+
+    for line in s.lines() {
+        let mut line = line.trim().to_owned();
+
+        if before_is_ordered_list && !line.starts_with("* ") {
+            page.push(Tag::Ul(ordered_list.clone()));
+            before_is_ordered_list = false;
+            ordered_list.clear();
+        } else if preformatted && !line.starts_with("```") {
+            preformatted_text.push_str(&line);
+            preformatted_text.push('\n');
+        } else if line.starts_with("=>") {
+            let body = line.split_off(2);
+            let mut body = body.trim().splitn(2, " ");
+
+            let url = body.next().ok_or(GemTextParseError::InvalidLink)?.trim();
+
+            match body.next() {
+                Some(label) => page.push(Tag::Link(
+                    Body::Text(label.trim().to_owned()),
+                    url.to_owned(),
+                )),
+                None => page.push(Tag::Link(Body::Null, url.to_owned())),
+            };
+        } else if line.starts_with("# ") {
+            let body = line.split_off(2);
+            page.push(Tag::H(body.trim().to_owned(), HeadingLevel::One));
+        } else if line.starts_with("## ") {
+            let body = line.split_off(3);
+            page.push(Tag::H(body.trim().to_owned(), HeadingLevel::Two));
+        } else if line.starts_with("### ") {
+            let body = line.split_off(4);
+            page.push(Tag::H(body.trim().to_owned(), HeadingLevel::Three));
+        } else if line.starts_with("* ") {
+            before_is_ordered_list = true;
+            let body = line.split_off(2);
+            ordered_list.push(Tag::El(NotNullBody::Text(body)));
+        } else if line.starts_with("> ") {
+            let body = line.split_off(2);
+            page.push(Tag::Bq(NotNullBody::Text(body)));
+        } else if line.starts_with("```") {
+            if preformatted {
+                page.push(Tag::Pre(preformatted_text.clone()));
+                preformatted_text.clear();
+            }
+
+            preformatted = !preformatted;
+        } else if !line.is_empty() {
+            page.push(Tag::P(NotNullBody::Text(line)));
+        }
+    }
+
+    Ok(page)
+}

src/parsers/mod.rs

@@ -0,0 +1 @@
+pub mod gemtext;

tests/gemtext.gmi

@@ -0,0 +1,32 @@
+=> https://example.com    A cool website
+=> gopher://example.com   An even cooler gopherhole
+=> gemini://example.com   A supremely cool Gemini capsule
+=> sftp://example.com
+
+Hi
+
+=>https://example.com A cool website
+=>gopher://example.com      An even cooler gopherhole
+=> gemini://example.com A supremely cool Gemini capsule
+=>   sftp://example.com
+
+# Heading
+
+## Sub-heading
+
+### Sub-sub-heading
+
+* Mercury
+* Gemini
+* Apollo
+
+> Gemtext supports blockquotes.  The quoted content is written as a single long line, which begins with a single > character
+
+```
+preformatted
+=> ()
+# false heading
+text
+```
+
+This is paragraph

tests/gemtext.rs

@@ -0,0 +1,10 @@
+use dalet::parsers::gemtext::parse_gemtext;
+
+#[test]
+fn gem_text() {
+    let text = include_str!("./gemtext.gmi");
+
+    let parsed = parse_gemtext(text.to_owned()).unwrap();
+
+    println!("{:#?}", parsed);
+}