Add ATLAS metadata filesystem source

.gitignore

@@ -0,0 +1,3 @@
+target/
+.DS_Store
+data/*.tsv

Cargo.toml

@@ -8,3 +8,4 @@ fuser = "0.14"
 libc = "0.2"
 env_logger = "0.11"
 log = "0.4"
+serde_json = "1"

README.md

@@ -1,20 +1,83 @@
 # CS-FileSystem
-User Space File System to populate Cybershuttle Data Sources
 
+This is a user-space filesystem for exposing Cybershuttle data sources.
 
-sudo apt install cargo
+This version loads ATLAS metadata from a TSV file and exposes each protein entry
+as a directory containing a `metadata.json` file.
+
+```text
+/tmp/atlas_mount/
+  atlas/
+    1r6w_A/
+      metadata.json
+    2y44_A/
+      metadata.json
+```
+
+## Requirements
 
+### Linux
+
+```bash
+sudo apt install cargo
 sudo apt install -y libfuse3-dev libfuse-dev pkg-config
+```
+
+### macOS
+
+```bash
+brew install pkgconf
+brew install --cask macfuse
+```
+
+macFUSE may require approval in `System Settings -> Privacy & Security`.
+
+## ATLAS TSV
+
+Place the ATLAS metadata TSV somewhere local. The examples below assume the TSV is at:
+
+```text
+data/2024_11_18_ATLAS_info.tsv
+```
+
+The TSV is not committed to the repository. Create the `data` directory and copy
+or download the file there:
+
+```bash
+mkdir -p data
+cp /path/to/2024_11_18_ATLAS_info.tsv data/
+```
+
+## Run
+
+Build and mount the filesystem:
+
+```bash
+mkdir -p /tmp/atlas_mount
+cargo run --release -- data/2024_11_18_ATLAS_info.tsv /tmp/atlas_mount
+```
+
+Leave that command running while the filesystem is mounted.
+
+In another terminal:
 
-cargo build
+```bash
+ls /tmp/atlas_mount
+ls /tmp/atlas_mount/atlas | head
+ls /tmp/atlas_mount/atlas/1r6w_A
+cat /tmp/atlas_mount/atlas/1r6w_A/metadata.json
+```
 
-mkdir /tmp/myfs
-cargo run --release -- /tmp/myfs
+## Unmount
 
+Linux:
 
-In a different terminal
-ls /tmp/myfs
+```bash
+fusermount -u /tmp/atlas_mount
+```
 
+macOS:
 
-To unmount
-fusermount -u /tmp/myfs
+```bash
+diskutil unmount /tmp/atlas_mount
+```

src/atlas.rs

@@ -0,0 +1,37 @@
+use std::fs;
+use std::io::{BufRead, BufReader};
+
+pub struct AtlasEntry {
+    pub id: String,
+    pub metadata_json: String, // pre-serialized JSON for this entry
+}
+
+pub fn load_atlas(tsv_path: &str) -> Vec<AtlasEntry> {
+    let file = fs::File::open(tsv_path).expect("Failed to open TSV");
+    let reader = BufReader::new(file);
+    let mut lines = reader.lines();
+
+    let header_line = lines.next().unwrap().unwrap();
+    let headers: Vec<&str> = header_line.split('\t').collect();
+    let mut entries = Vec::new();
+
+    for line in lines {
+        let line = line.unwrap();
+        let fields: Vec<&str> = line.split('\t').collect();
+
+        let mut map = serde_json::Map::new();
+
+        for (i, header) in headers.iter().enumerate() {
+            if let Some(value) = fields.get(i) {
+                map.insert(
+                    header.to_string(),
+                    serde_json::Value::String(value.to_string()),
+                );
+            }
+        }
+        let id = fields.get(0).unwrap_or(&"unknown").to_string();
+        let metadata_json = serde_json::to_string_pretty(&serde_json::Value::Object(map)).unwrap();
+        entries.push(AtlasEntry { id, metadata_json });
+    }
+    entries
+}