From ba656e5c9f58e0440278918e5c6cdf0ac2fe673c Mon Sep 17 00:00:00 2001
From: Andrew Kelley <andrew@ziglang.org>
Date: Mon, 16 Oct 2023 18:15:47 -0700
Subject: [PATCH] zig fetch: add `--debug-hash` argument

This argument causes zig to print verbose hashing information to stdout,
which can be used to diff two different fetches and find out why the
hashes do not equal each other.
---
 src/Package/Fetch.zig | 37 ++++++++++++++++++++++++++++++++++---
 src/main.zig          |  7 +++++++
 2 files changed, 41 insertions(+), 3 deletions(-)

diff --git a/src/Package/Fetch.zig b/src/Package/Fetch.zig
index ee835031a8fc9e1d1b241d32a28fc01873ba40ab..7256c5e15c504c03d549c602b11570094e5d4d46 100644
--- a/src/Package/Fetch.zig
+++ b/src/Package/Fetch.zig
@@ -81,6 +81,10 @@ pub const JobQueue = struct {
     wait_group: WaitGroup = .{},
     global_cache: Cache.Directory,
     recursive: bool,
+    /// Dumps hash information to stdout which can be used to troubleshoot why
+    /// two hashes of the same package do not match.
+    /// If this is true, `recursive` must be false.
+    debug_hash: bool,
     work_around_btrfs_bug: bool,
 
     pub const Table = std.AutoArrayHashMapUnmanaged(Manifest.MultiHashHexDigest, *Fetch);
@@ -1315,7 +1319,7 @@ fn computeHash(
             const kind: HashedFile.Kind = switch (entry.kind) {
                 .directory => unreachable,
                 .file => .file,
-                .sym_link => .sym_link,
+                .sym_link => .link,
                 else => return f.fail(f.location_tok, try eb.printString(
                     "package contains '{s}' which has illegal file type '{s}'",
                     .{ entry.path, @tagName(entry.kind) },
@@ -1399,9 +1403,36 @@ fn computeHash(
     }
 
     if (any_failures) return error.FetchFailed;
+
+    if (f.job_queue.debug_hash) {
+        assert(!f.job_queue.recursive);
+        // Print something to stdout that can be text diffed to figure out why
+        // the package hash is different.
+        dumpHashInfo(all_files.items) catch |err| {
+            std.debug.print("unable to write to stdout: {s}\n", .{@errorName(err)});
+            std.process.exit(1);
+        };
+    }
+
     return hasher.finalResult();
 }
 
+fn dumpHashInfo(all_files: []const *const HashedFile) !void {
+    const stdout = std.io.getStdOut();
+    var bw = std.io.bufferedWriter(stdout.writer());
+    const w = bw.writer();
+
+    for (all_files) |hashed_file| {
+        try w.print("{s}: {s}: {s}\n", .{
+            @tagName(hashed_file.kind),
+            std.fmt.fmtSliceHexLower(&hashed_file.hash),
+            hashed_file.normalized_path,
+        });
+    }
+
+    try bw.flush();
+}
+
 fn workerHashFile(dir: fs.Dir, hashed_file: *HashedFile, wg: *WaitGroup) void {
     defer wg.finish();
     hashed_file.failure = hashFileFallible(dir, hashed_file);
@@ -1427,7 +1458,7 @@ fn hashFileFallible(dir: fs.Dir, hashed_file: *HashedFile) HashedFile.Error!void
                 hasher.update(buf[0..bytes_read]);
             }
         },
-        .sym_link => {
+        .link => {
             const link_name = try dir.readLink(hashed_file.fs_path, &buf);
             if (fs.path.sep != canonical_sep) {
                 // Package hashes are intended to be consistent across
@@ -1480,7 +1511,7 @@ const HashedFile = struct {
         fs.File.StatError ||
         fs.Dir.ReadLinkError;
 
-    const Kind = enum { file, sym_link };
+    const Kind = enum { file, link };
 
     fn lessThan(context: void, lhs: *const HashedFile, rhs: *const HashedFile) bool {
         _ = context;
diff --git a/src/main.zig b/src/main.zig
index fe8c871ce8ae683889c088dc9ebaa16dcd586228..6dd4fb072560e306fdf60fb6ae3ef8e3de0e22dd 100644
--- a/src/main.zig
+++ b/src/main.zig
@@ -5143,6 +5143,7 @@ pub fn cmdBuild(gpa: Allocator, arena: Allocator, args: []const []const u8) !voi
                 .thread_pool = &thread_pool,
                 .global_cache = global_cache_directory,
                 .recursive = true,
+                .debug_hash = false,
                 .work_around_btrfs_bug = work_around_btrfs_bug,
             };
             defer job_queue.deinit();
@@ -6991,6 +6992,7 @@ pub const usage_fetch =
     \\Options:
     \\  -h, --help                    Print this help and exit
     \\  --global-cache-dir [path]     Override path to global Zig cache directory
+    \\  --debug-hash                  Print verbose hash information to stdout
     \\
 ;
 
@@ -7004,6 +7006,7 @@ fn cmdFetch(
         std.process.hasEnvVarConstant("ZIG_BTRFS_WORKAROUND");
     var opt_path_or_url: ?[]const u8 = null;
     var override_global_cache_dir: ?[]const u8 = try optionalStringEnvVar(arena, "ZIG_GLOBAL_CACHE_DIR");
+    var debug_hash: bool = false;
 
     {
         var i: usize = 0;
@@ -7019,6 +7022,9 @@ fn cmdFetch(
                     i += 1;
                     override_global_cache_dir = args[i];
                     continue;
+                } else if (mem.eql(u8, arg, "--debug-hash")) {
+                    debug_hash = true;
+                    continue;
                 } else {
                     fatal("unrecognized parameter: '{s}'", .{arg});
                 }
@@ -7057,6 +7063,7 @@ fn cmdFetch(
         .thread_pool = &thread_pool,
         .global_cache = global_cache_directory,
         .recursive = false,
+        .debug_hash = debug_hash,
         .work_around_btrfs_bug = work_around_btrfs_bug,
     };
     defer job_queue.deinit();
-- 
2.54.0