From eb527ea758b49adbc1e311638cc6999190ff1124 Mon Sep 17 00:00:00 2001
From: sigoden <sigoden@gmail.com>
Date: Sat, 8 Feb 2025 08:46:34 +0800
Subject: feat: enhance `.file` for loading resources from diverse sources
 (#1155)

---
 src/utils/loader.rs | 47 +++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 45 insertions(+), 2 deletions(-)

(limited to 'src/utils/loader.rs')

diff --git a/src/utils/loader.rs b/src/utils/loader.rs
index 2ac671d..b36bf4a 100644
--- a/src/utils/loader.rs
+++ b/src/utils/loader.rs
@@ -1,17 +1,19 @@
 use super::*;
 
-use anyhow::{Context, Result};
+use anyhow::{anyhow, Context, Result};
 use indexmap::IndexMap;
+use serde::{Deserialize, Serialize};
 use std::collections::HashMap;
 
 pub const EXTENSION_METADATA: &str = "__extension__";
 
 pub type DocumentMetadata = IndexMap<String, String>;
 
-#[derive(Debug, Clone)]
+#[derive(Debug, Clone, Serialize, Deserialize)]
 pub struct LoadedDocument {
     pub path: String,
     pub contents: String,
+    #[serde(default)]
     pub metadata: DocumentMetadata,
 }
 
@@ -80,3 +82,44 @@ fn load_with_command(path: &str, extension: &str, loader_command: &str) -> Resul
     metadata.insert(EXTENSION_METADATA.into(), DEFAULT_EXTENSION.to_string());
     Ok(LoadedDocument::new(path.into(), contents, metadata))
 }
+
+pub fn is_loader_protocol(loaders: &HashMap<String, String>, path: &str) -> bool {
+    match path.split_once(':') {
+        Some((protocol, _)) => loaders.contains_key(protocol),
+        None => false,
+    }
+}
+
+pub fn load_protocol_path(
+    loaders: &HashMap<String, String>,
+    path: &str,
+) -> Result<Vec<LoadedDocument>> {
+    let (protocol, loader_command, new_path) = path
+        .split_once(':')
+        .and_then(|(protocol, path)| {
+            let loader_command = loaders.get(protocol)?;
+            Some((protocol, loader_command, path))
+        })
+        .ok_or_else(|| anyhow!("No document loader for '{}'", path))?;
+    let contents = run_loader_command(new_path, protocol, loader_command)?;
+    let output = if let Ok(list) = serde_json::from_str::<Vec<LoadedDocument>>(&contents) {
+        list.into_iter()
+            .map(|mut v| {
+                if v.path.starts_with(path) {
+                } else if v.path.starts_with(new_path) {
+                    v.path = format!("{}:{}", protocol, v.path);
+                } else {
+                    v.path = format!("{}/{}", path, v.path);
+                }
+                v
+            })
+            .collect()
+    } else {
+        vec![LoadedDocument::new(
+            path.into(),
+            contents,
+            Default::default(),
+        )]
+    };
+    Ok(output)
+}
-- 
cgit v1.2.3