From 074ae1b24cf721ed3d86427795b34a4aefa1290c Mon Sep 17 00:00:00 2001
From: Lars Wirzenius <liw@liw.fi>
Date: Sat, 28 Nov 2020 17:03:04 +0200
Subject: feat: add SHA256 checksum of contents of regular files

---
 Cargo.toml         |  1 +
 src/lib.rs         | 29 +++++++++++++++++++++++++++++
 subplot/summain.py |  4 ++++
 summain.md         |  5 +++++
 4 files changed, 39 insertions(+)

diff --git a/Cargo.toml b/Cargo.toml
index 3435e8c..73924e3 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -11,6 +11,7 @@ edition = "2018"
 
 [dependencies]
 anyhow = "1"
+digest = "0.9"
 serde = { version = "1", features = ["derive"] }
 serde_yaml = "0.8"
 sha2 = "0.9"
diff --git a/src/lib.rs b/src/lib.rs
index ccd652f..9f46f54 100644
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -32,10 +32,15 @@
 //! ~~~
 
 use serde::Serialize;
+use sha2::{Digest, Sha256};
 use std::fs::symlink_metadata;
+use std::fs::File;
+use std::io::{BufReader, Read};
 use std::os::linux::fs::MetadataExt;
 use std::path::{Path, PathBuf};
 
+const BUF_SIZE: usize = 1024 * 1024;
+
 /// An entry in a file manifest.
 #[derive(Serialize, Debug)]
 pub struct ManifestEntry {
@@ -48,6 +53,7 @@ pub struct ManifestEntry {
     mtime_nsec: i64,
     nlink: u64,
     size: Option<u64>,
+    sha256: Option<String>,
 }
 
 impl ManifestEntry {
@@ -59,6 +65,11 @@ impl ManifestEntry {
     /// The structure can be serialized using serde.
     pub fn new(path: &Path) -> std::io::Result<Self> {
         let m = symlink_metadata(path)?;
+        let hash = if m.is_file() {
+            Some(file_checksum(path)?)
+        } else {
+            None
+        };
         Ok(Self {
             path: path.to_path_buf(),
             atime: m.st_atime(),
@@ -68,10 +79,28 @@ impl ManifestEntry {
             mtime_nsec: m.st_mtime_nsec(),
             nlink: m.st_nlink(),
             size: if m.is_dir() { None } else { Some(m.st_size()) },
+            sha256: hash,
         })
     }
 }
 
+fn file_checksum(path: &Path) -> std::io::Result<String> {
+    let mut hasher = Sha256::new();
+
+    let file = File::open(path)?;
+    let mut reader = BufReader::new(file);
+    let mut buf = vec![0; BUF_SIZE];
+    loop {
+        let n = reader.read(&mut buf)?;
+        if n == 0 {
+            break;
+        }
+        hasher.update(&buf[..n]);
+    }
+    let hash = hasher.finalize();
+    Ok(format!("{:x}", hash))
+}
+
 mod mode {
     use serde::{self, Serializer};
 
diff --git a/subplot/summain.py b/subplot/summain.py
index 3ea6188..98eecf2 100644
--- a/subplot/summain.py
+++ b/subplot/summain.py
@@ -1,3 +1,4 @@
+import logging
 import os
 
 
@@ -33,4 +34,7 @@ def output_matches_file(ctx, filename=None):
 
     actual = runcmd_get_stdout(ctx)
     expected = get_file(filename).decode("UTF-8")
+    logging.debug("output_matches:")
+    logging.debug(f"  actual: {actual!r}")
+    logging.debug(f"  expect: {expected!r}")
     assert_eq(actual, expected)
diff --git a/summain.md b/summain.md
index 7c2c5fa..e6a4d15 100644
--- a/summain.md
+++ b/summain.md
@@ -87,6 +87,7 @@ mtime: 456
 mtime_nsec: 0
 nlink: 2
 size: ~
+sha256: ~
 ```
 
 ## Writeable file
@@ -112,6 +113,7 @@ mtime: 22
 mtime_nsec: 0
 nlink: 1
 size: 0
+sha256: e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855
 ```
 
 ## Read-only file
@@ -136,6 +138,7 @@ mtime: 44
 mtime_nsec: 0
 nlink: 1
 size: 0
+sha256: e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855
 ```
 
 ## Two files sorted
@@ -163,6 +166,7 @@ mtime: 44
 mtime_nsec: 0
 nlink: 1
 size: 0
+sha256: e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855
 ---
 path: bbb
 atime: 33
@@ -172,6 +176,7 @@ mtime: 44
 mtime_nsec: 0
 nlink: 1
 size: 0
+sha256: e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855
 ```
 
 ---
-- 
cgit v1.2.1