feat: link parsing and basic cli setup

2024-03-26 15:13:33 -03:00
parent ba3497e0cf
commit 2e04cbee23
7 changed files with 400 additions and 2 deletions
--- a/.gitignore
+++ b/.gitignore
@@ -1,3 +1,4 @@
 /target
 .vault-backup/
+test.md
 .direnv
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -166,6 +166,21 @@ version = "0.7.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "98cc8fbded0c607b7ba9dd60cd98df59af97e84d24e49c8557331cfc26d301ce"

+[[package]]
+name = "clio"
+version = "0.3.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b7fc6734af48458f72f5a3fa7b840903606427d98a710256e808f76a965047d9"
+dependencies = [
+ "cfg-if",
+ "clap",
+ "is-terminal",
+ "libc",
+ "tempfile",
+ "walkdir",
+ "windows-sys 0.42.0",
+]
+
 [[package]]
 name = "colorchoice"
 version = "1.0.0"
@@ -314,6 +329,12 @@ dependencies = [
 "regex",
 ]

+[[package]]
+name = "fastrand"
+version = "2.0.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "25cbce373ec4653f1a01a31e8a5e5ec0c622dc27ff9c4e6606eefef5cbbed4a5"
+
 [[package]]
 name = "flate2"
 version = "1.0.28"
@@ -342,6 +363,12 @@ version = "0.5.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "2304e00983f87ffb38b55b444b5e3b60a884b5d30c0fca7d82fe33449bbe55ea"

+[[package]]
+name = "hermit-abi"
+version = "0.3.9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d231dfb89cfffdbc30e7fc41579ed6066ad03abda9e567ccafae602b97ec5024"
+
 [[package]]
 name = "ident_case"
 version = "1.0.1"
@@ -358,6 +385,17 @@ dependencies = [
 "hashbrown",
 ]

+[[package]]
+name = "is-terminal"
+version = "0.4.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f23ff5ef2b80d608d61efee834934d862cd92461afc0560dedf493e4c033738b"
+dependencies = [
+ "hermit-abi",
+ "libc",
+ "windows-sys 0.52.0",
+]
+
 [[package]]
 name = "itoa"
 version = "1.0.10"
@@ -402,6 +440,7 @@ name = "mdparser"
 version = "0.1.0"
 dependencies = [
 "clap",
+ "clio",
 "comrak",
 "serde",
 "serde_json",
@@ -690,6 +729,18 @@ dependencies = [
 "yaml-rust",
 ]

+[[package]]
+name = "tempfile"
+version = "3.10.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "85b77fafb263dd9d05cbeac119526425676db3784113aa9295c88498cbf8bff1"
+dependencies = [
+ "cfg-if",
+ "fastrand",
+ "rustix",
+ "windows-sys 0.52.0",
+]
+
 [[package]]
 name = "terminal_size"
 version = "0.3.0"
@@ -876,6 +927,21 @@ version = "0.4.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "712e227841d057c1ee1cd2fb22fa7e5a5461ae8e48fa2ca79ec42cfc1931183f"

+[[package]]
+name = "windows-sys"
+version = "0.42.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5a3e1820f08b8513f676f7ab6c1f99ff312fb97b553d30ff4dd86f9f15728aa7"
+dependencies = [
+ "windows_aarch64_gnullvm 0.42.2",
+ "windows_aarch64_msvc 0.42.2",
+ "windows_i686_gnu 0.42.2",
+ "windows_i686_msvc 0.42.2",
+ "windows_x86_64_gnu 0.42.2",
+ "windows_x86_64_gnullvm 0.42.2",
+ "windows_x86_64_msvc 0.42.2",
+]
+
 [[package]]
 name = "windows-sys"
 version = "0.48.0"
@@ -924,6 +990,12 @@ dependencies = [
 "windows_x86_64_msvc 0.52.4",
 ]

+[[package]]
+name = "windows_aarch64_gnullvm"
+version = "0.42.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "597a5118570b68bc08d8d59125332c54f1ba9d9adeedeef5b99b02ba2b0698f8"
+
 [[package]]
 name = "windows_aarch64_gnullvm"
 version = "0.48.5"
@@ -936,6 +1008,12 @@ version = "0.52.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "bcf46cf4c365c6f2d1cc93ce535f2c8b244591df96ceee75d8e83deb70a9cac9"

+[[package]]
+name = "windows_aarch64_msvc"
+version = "0.42.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e08e8864a60f06ef0d0ff4ba04124db8b0fb3be5776a5cd47641e942e58c4d43"
+
 [[package]]
 name = "windows_aarch64_msvc"
 version = "0.48.5"
@@ -948,6 +1026,12 @@ version = "0.52.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "da9f259dd3bcf6990b55bffd094c4f7235817ba4ceebde8e6d11cd0c5633b675"

+[[package]]
+name = "windows_i686_gnu"
+version = "0.42.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c61d927d8da41da96a81f029489353e68739737d3beca43145c8afec9a31a84f"
+
 [[package]]
 name = "windows_i686_gnu"
 version = "0.48.5"
@@ -960,6 +1044,12 @@ version = "0.52.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "b474d8268f99e0995f25b9f095bc7434632601028cf86590aea5c8a5cb7801d3"

+[[package]]
+name = "windows_i686_msvc"
+version = "0.42.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "44d840b6ec649f480a41c8d80f9c65108b92d89345dd94027bfe06ac444d1060"
+
 [[package]]
 name = "windows_i686_msvc"
 version = "0.48.5"
@@ -972,6 +1062,12 @@ version = "0.52.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "1515e9a29e5bed743cb4415a9ecf5dfca648ce85ee42e15873c3cd8610ff8e02"

+[[package]]
+name = "windows_x86_64_gnu"
+version = "0.42.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8de912b8b8feb55c064867cf047dda097f92d51efad5b491dfb98f6bbb70cb36"
+
 [[package]]
 name = "windows_x86_64_gnu"
 version = "0.48.5"
@@ -984,6 +1080,12 @@ version = "0.52.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "5eee091590e89cc02ad514ffe3ead9eb6b660aedca2183455434b93546371a03"

+[[package]]
+name = "windows_x86_64_gnullvm"
+version = "0.42.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "26d41b46a36d453748aedef1486d5c7a85db22e56aff34643984ea85514e94a3"
+
 [[package]]
 name = "windows_x86_64_gnullvm"
 version = "0.48.5"
@@ -996,6 +1098,12 @@ version = "0.52.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "77ca79f2451b49fa9e2af39f0747fe999fcda4f5e241b2898624dca97a1f2177"

+[[package]]
+name = "windows_x86_64_msvc"
+version = "0.42.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9aec5da331524158c6d1a4ac0ab1541149c0b9505fde06423b02f5ef0106b9f0"
+
 [[package]]
 name = "windows_x86_64_msvc"
 version = "0.48.5"
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -7,6 +7,7 @@ edition = "2021"

 [dependencies]
 clap = "4.5.3"
+clio = { version = "0.3.5", features = ["clap-parse"] }
 comrak = "0.21.0"
 serde = { version = "1.0.197", features = ["derive"] }
 serde_json = "1.0.114"
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -0,0 +1,2 @@
+pub mod links;
+pub mod utils;
--- a/src/links.rs
+++ b/src/links.rs
@@ -0,0 +1,143 @@
+use std::io::Write;
+use std::{cell::RefCell, collections::HashMap};
+use std::{fs, io, path::PathBuf};
+
+use comrak::nodes::{Ast, NodeLink, NodeValue};
+use comrak::{arena_tree::Node, Arena};
+
+pub struct ParseOptions {
+    pub alias_prop: Option<String>,
+    pub path_root: PathBuf,
+    pub to_complete_paths: bool,
+    pub remove_unalised: bool,
+    pub remove_unfound: bool,
+}
+impl Default for ParseOptions {
+    fn default() -> Self {
+        ParseOptions {
+            alias_prop: None,
+            path_root: PathBuf::new(),
+            to_complete_paths: true,
+            remove_unalised: true,
+            remove_unfound: true,
+        }
+    }
+}
+
+#[derive(Debug)]
+pub enum ParsingError {
+    AliasNotFound { file: String },
+    AliasErr(GetAliasError),
+    IoErr(io::Error),
+}
+pub fn parse<'a>(
+    node: &'a Node<'a, RefCell<Ast>>,
+    link: &mut NodeLink,
+    opts: &ParseOptions,
+) -> Result<(), ParsingError> {
+    let path = match find_path(link, &opts.path_root) {
+        Ok(p) => p,
+        Err(err) => {
+            if opts.remove_unfound {
+                node.children().for_each(|n| node.insert_before(n));
+                node.detach();
+                return Ok(());
+            } else {
+                return Err(ParsingError::IoErr(err));
+            }
+        }
+    };
+
+    if opts.to_complete_paths {
+        link.url = String::from(path.to_string_lossy())
+    }
+
+    if let Some(a) = &opts.alias_prop {
+        let alias = match get_alias(&path, &a) {
+            Ok(a) => a,
+            Err(err) => return Err(ParsingError::AliasErr(err)),
+        };
+        if let Some(v) = alias {
+            link.url = v;
+        } else if opts.remove_unalised {
+            node.children().for_each(|n| node.insert_before(n));
+            node.detach();
+        } else {
+            return Err(ParsingError::AliasNotFound {
+                file: link.url.clone(),
+            });
+        }
+    }
+    Ok(())
+}
+
+#[derive(Debug)]
+pub enum GetAliasError {
+    IoErr(io::Error),
+    YamlErr(serde_yaml::Error),
+}
+pub fn get_alias(path: &PathBuf, alias_prop: &String) -> Result<Option<String>, GetAliasError> {
+    let file = match fs::read_to_string(path) {
+        Ok(f) => f,
+        Err(err) => return Err(GetAliasError::IoErr(err)),
+    };
+
+    let arena = Arena::new();
+    let ast = comrak::parse_document(&arena, &file, &crate::utils::default_options());
+
+    let alias = crate::utils::iter_nodes_r(ast, &|node| {
+        if let NodeValue::FrontMatter(f) = &node.data.borrow().value {
+            // Removes starting and trailing "---" delimiters from frontmatter's string.
+            let f = String::from(f.split("---").collect::<Vec<&str>>()[1]);
+
+            let map = match serde_yaml::from_str::<HashMap<String, String>>(&f) {
+                Ok(m) => m,
+                Err(err) => {
+                    return Some(Err::<String, GetAliasError>(GetAliasError::YamlErr(err)));
+                }
+            };
+
+            match map.get(alias_prop) {
+                // The hashmap will be dropped anyways and free up space,
+                // so whatever
+                Some(r) => Some(Ok(r.clone())),
+                None => None,
+            }
+        } else {
+            None
+        }
+    });
+
+    match alias {
+        Some(r) => match r {
+            Ok(s) => Ok(Some(s)),
+            Err(err) => Err(err),
+        },
+        None => Ok(None),
+    }
+}
+
+pub fn find_path(link: &NodeLink, path_root: &PathBuf) -> Result<PathBuf, io::Error> {
+    find_file(&link.url, &path_root)
+}
+
+fn find_file(file: &String, path: &PathBuf) -> Result<PathBuf, io::Error> {
+    match fs::read_dir(path)?.find_map(|e| match e {
+        Ok(e) => {
+            if e.path().is_dir() {
+                match find_file(&file, &e.path()) {
+                    Ok(r) => Some(r),
+                    Err(_) => None,
+                }
+            } else if file == &e.file_name().to_string_lossy().to_string() {
+                Some(e.path())
+            } else {
+                None
+            }
+        }
+        _ => None,
+    }) {
+        Some(r) => Ok(r),
+        None => Err(io::Error::new(io::ErrorKind::NotFound, "File not found")),
+    }
+}
--- a/src/main.rs
+++ b/src/main.rs
@@ -1,3 +1,108 @@
-fn main() {
-    println!("Hello, world!");
+use clap::{Parser, Subcommand};
+use clio::*;
+use comrak::nodes::NodeValue;
+
+use mdparser::{links, utils};
+
+#[derive(Parser, Debug)]
+#[command(version = "0.1", about = "", long_about = None, propagate_version = true)]
+struct Cli {
+    #[command(subcommand)]
+    command: Commands,
+
+    #[arg(short, long, default_value = "-")]
+    input: Input,
+
+    #[arg(short, long, default_value = "-")]
+    output: Output,
+
+    #[arg(long)]
+    surpress_errors: bool,
+}
+
+#[derive(Debug, Subcommand)]
+enum Commands {
+    Links {
+        #[arg(short, long)]
+        path_root: clio::ClioPath,
+
+        #[arg(short, long, default_value = "x_alias_url")]
+        alias_prop: String,
+
+        #[arg(long)]
+        to_absolute_paths: bool,
+
+        #[arg(long)]
+        not_remove_unalised: bool,
+
+        #[arg(long)]
+        not_remove_unfound: bool,
+    },
+    Not {},
+}
+
+fn main() {
+    let mut cli = Cli::parse();
+
+    let file = std::io::read_to_string(&mut cli.input).unwrap_or_else(|err| panic!("{err:#?}"));
+    let arena = comrak::Arena::new();
+    let ast = comrak::parse_document(&arena, &file, &mdparser::utils::default_options());
+
+    // println!("{ast:#?}");
+
+    match &cli.command {
+        Commands::Links {
+            path_root,
+            alias_prop,
+            to_absolute_paths,
+            not_remove_unalised,
+            not_remove_unfound,
+        } => utils::iter_nodes(&ast, &|node| {
+            if let NodeValue::Link(ref mut link) = &mut node.data.borrow_mut().value {
+                match links::parse(
+                    node,
+                    link,
+                    &links::ParseOptions {
+                        path_root: path_root.to_path_buf(),
+                        alias_prop: Some(String::from(alias_prop)),
+                        to_complete_paths: *to_absolute_paths,
+                        remove_unalised: !*not_remove_unalised,
+                        remove_unfound: !*not_remove_unfound,
+                    },
+                ) {
+                    Ok(_) => (),
+                    Err(err) => {
+                        if !&cli.surpress_errors {
+                            panic!("{err:#?}\n");
+                        } else {
+                            eprint!("{err:#?}\n");
+                        }
+                    }
+                };
+
+                /*
+                if link.url.starts_with("https://") || link.url.starts_with("http://") {
+                    return;
+                }
+
+                let file = match mdparser::links::find_path(link, &vault_root.to_path_buf()) {
+                    Ok(f) => f,
+                    Err(_) => {
+                        if !not_remove_unfound {
+                            node.children().for_each(|t| node.insert_before(t));
+                            node.detach();
+                        }
+                        return;
+                    }
+                };
+
+                let n = node.parent();
+                println!("{n:#?}");
+                */
+            }
+        }),
+        _ => (),
+    };
+
+    let _ = comrak::format_commonmark(&ast, &mdparser::utils::default_options(), &mut cli.output);
 }
--- a/src/utils.rs
+++ b/src/utils.rs
@@ -0,0 +1,38 @@
+pub fn default_options() -> comrak::Options {
+    let mut opts = comrak::Options::default();
+
+    opts.render.width = 100;
+    opts.render.hardbreaks = false;
+
+    opts.extension.strikethrough = true;
+    opts.extension.front_matter_delimiter = Some("---".to_owned());
+
+    opts
+}
+
+pub fn iter_nodes<'a, F>(node: &'a comrak::nodes::AstNode<'a>, f: &F)
+where
+    F: Fn(&'a comrak::nodes::AstNode<'a>),
+{
+    f(node);
+    for c in node.children() {
+        iter_nodes(c, f)
+    }
+}
+
+pub fn iter_nodes_r<'a, F, T>(node: &'a comrak::nodes::AstNode<'a>, f: &F) -> Option<T>
+where
+    F: Fn(&'a comrak::nodes::AstNode<'a>) -> Option<T>,
+{
+    let result = f(node);
+    if let Some(r) = result {
+        return Some(r);
+    }
+    for c in node.children() {
+        let result = iter_nodes_r(c, f);
+        if let Some(r) = result {
+            return Some(r);
+        }
+    }
+    None
+}