From e4826c62dfeed52ae0cc82b012cdf77a054a474b Mon Sep 17 00:00:00 2001
From: Artemis Tosini <me@artem.ist>
Date: Sat, 13 Jul 2024 21:36:33 +0000
Subject: [PATCH] add nix_hash

---
 rust/nix_hash/Cargo.lock  | 16 ++++++++++
 rust/nix_hash/Cargo.toml  |  7 +++++
 rust/nix_hash/src/main.rs | 64 +++++++++++++++++++++++++++++++++++++++
 3 files changed, 87 insertions(+)
 create mode 100644 rust/nix_hash/Cargo.lock
 create mode 100644 rust/nix_hash/Cargo.toml
 create mode 100644 rust/nix_hash/src/main.rs

diff --git a/rust/nix_hash/Cargo.lock b/rust/nix_hash/Cargo.lock
new file mode 100644
index 0000000..9690c79
--- /dev/null
+++ b/rust/nix_hash/Cargo.lock
@@ -0,0 +1,16 @@
+# This file is automatically @generated by Cargo.
+# It is not intended for manual editing.
+version = 3
+
+[[package]]
+name = "data-encoding"
+version = "2.6.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e8566979429cf69b49a5c740c60791108e86440e8be149bbea4fe54d2c32d6e2"
+
+[[package]]
+name = "nix_hash"
+version = "0.1.0"
+dependencies = [
+ "data-encoding",
+]
diff --git a/rust/nix_hash/Cargo.toml b/rust/nix_hash/Cargo.toml
new file mode 100644
index 0000000..8d163a7
--- /dev/null
+++ b/rust/nix_hash/Cargo.toml
@@ -0,0 +1,7 @@
+[package]
+name = "nix_hash"
+version = "0.1.0"
+edition = "2021"
+
+[dependencies]
+data-encoding = "2.6.0"
diff --git a/rust/nix_hash/src/main.rs b/rust/nix_hash/src/main.rs
new file mode 100644
index 0000000..b197658
--- /dev/null
+++ b/rust/nix_hash/src/main.rs
@@ -0,0 +1,64 @@
+use data_encoding::BitOrder;
+
+fn main() -> Result<(), Box<dyn std::error::Error>> {
+    let mut binary = data_encoding::HEXLOWER
+        .decode(b"3a99595f4daedd5f52d646e413ea1c989abb910cd2ea18fd98ccddeab68f0883")?;
+
+    let mut spec = data_encoding::Specification::new();
+    spec.symbols.push_str("0123456789abcdfghijklmnpqrsvwxyz");
+    spec.bit_order = BitOrder::LeastSignificantFirst;
+
+    println!(
+        "Found:    {}",
+        spec.encoding()?
+            .encode(&binary)
+            .chars()
+            .rev()
+            .collect::<String>()
+    );
+    println!("Expected: 10q8iyvfmpfck3yiisnj1j8vp6lq3km17r26sr95zpdf9mgmk69s");
+    println!("Found:    {}", encode_nix32(&binary));
+
+    Ok(())
+}
+
+static NIX32_CHARS: &[u8; 32] = b"0123456789abcdfghijklmnpqrsvwxyz";
+
+fn encode_nix32(input: &[u8]) -> String {
+    let length = if input.len() == 0 {
+        0
+    } else {
+        // ceil(input.len() * 8 / 5)
+        (input.len() * 8 - 1) / 5 + 1
+    };
+    let mut output = String::with_capacity(length);
+
+    // nix32 hashes feel like they're a bug that stuck
+    // The output is backwards and bits are grouped
+    // from the least significant bit in each byte
+    // instead of the most significant bit.
+    // e.g. encoding "Meow" gives us:
+    // Char:       M (0x4d)          e (0x65)          o (0x6f)          w (0x77)
+    // Value:   0 1 0 0 1 1 0 1 | 0 1 1 0 0 1 0 1 | 0 1 1 0 1 1 1 1 | 0 1 1 1 0 1 1 1
+    // Out No.: 5 5 5 6 6 6 6 6 | 3 4 4 4 4 4 5 5 | 2 2 2 2 3 3 3 3 | 0 0 1 1 1 1 1 2
+    // Out Bit: 2 1 0 4 3 2 1 0 | 0 5 4 3 2 1 4 3 | 3 2 1 0 4 3 2 1 | 1 0 4 3 2 1 0 4
+    //
+    // where "Out No." is the index of the output charater responsible for a given bit,
+    // and 2**"Out Bit" is the value of a given bit for its output character.
+    //
+    // In this example, characters 0 to 6 have values
+    // 0x01, 0x1b, 0x16, 0x1e, 0x19, 0xa, 0xd.
+    // Indexing into the alphabet gives us "1vnyrad"
+
+    for char_no in 0..length {
+        let bit_no = (length - char_no - 1) * 5;
+        let byte_no = bit_no / 8;
+        let offset = bit_no % 8;
+
+        let next_byte = input.get(byte_no + 1).unwrap_or(&0);
+        let value = (input[byte_no] as u16 >> offset) | ((*next_byte as u16) << (8 - offset));
+        output.push(NIX32_CHARS[(value & 0x1f) as usize] as char);
+    }
+
+    output
+}