wip

Makefile

@@ -128,3 +128,8 @@ docker-turso:
 		-p $(DB_TURSO_PORT):8080 \
 		-l goose_test \
 		ghcr.io/tursodatabase/libsql-server:v0.22.10
+
+.PHONY: generate
+generate:
+	cd ./internal/testdata && \
+	find migrations/postgres -type f -exec sha256sum {} + | sha256sum | cut -c 1-32 > ./internal/testdata/migrations

internal/gooseutil/fsyshash.go

@@ -0,0 +1,66 @@
+package gooseutil
+
+import (
+	"crypto/sha256"
+	"encoding/hex"
+	"fmt"
+	"io"
+	"io/fs"
+	"sort"
+)
+
+// Digest is used to generate a unique identifier for the contents of a directory, which can be used
+// to determine if the contents of the directory have changed.
+//
+// The digest is generated by iterating over the files in the directory, skipping directories,
+// sorting them, and then hashing the contents of each file. The hash of each file is then hashed to
+// generate the final digest. The resulting digest is a sha256 hash truncated to 32 characters.
+//
+// It is adapted from the dirhash package. Ref:
+//
+//	https://cs.opensource.google/go/x/mod/+/refs/tags/v0.17.0:sumdb/dirhash/hash.go;l=31-44
+//
+// The output of this function is equivalent to the following shell command:
+//
+//	sha256sum $(find <dir> -type f | sort) | sha256sum | cut -c 1-32
+func Digest(fsys fs.FS, root string) (string, error) {
+	var files []string
+	err := fs.WalkDir(fsys, root, func(path string, d fs.DirEntry, err error) error {
+		if err != nil {
+			return err
+		}
+		if d.IsDir() {
+			if path != root {
+				return fs.SkipDir
+			}
+			// Skip the root directory.
+			return nil
+		}
+		files = append(files, path)
+		return nil
+	})
+	if err != nil {
+		return "", err
+	}
+	if len(files) == 0 {
+		return "", fmt.Errorf("no files found")
+	}
+	h := sha256.New()
+	sort.Strings(files)
+	for _, file := range files {
+		f, err := fsys.Open(file)
+		if err != nil {
+			return "", err
+		}
+		hf := sha256.New()
+		_, err = io.Copy(hf, f)
+		f.Close()
+		if err != nil {
+			return "", err
+		}
+		fmt.Fprintf(h, "%x  %s\n", hf.Sum(nil), file)
+		fmt.Printf("%x  %s\n", hf.Sum(nil), file)
+	}
+	digest := hex.EncodeToString(h.Sum(nil))[:32]
+	return digest, nil
+}

internal/gooseutil/fsyshash_test.go

@@ -0,0 +1,8 @@
+package gooseutil
+
+import "testing"
+
+func TestDigest(t *testing.T) {
+	t.Parallel()
+
+}

internal/testdata/embed.go

@@ -0,0 +1,8 @@
+package testdata
+
+//go:generate sha256sum $(find ./migrations/postgres -type f | sort) | sha256sum | cut -c 1-32 > ./migrations/postgres.sha256sum
+
+import "embed"
+
+//go:embed migrations/**/*.sql
+var EmbedMigrations embed.FS

internal/testdata/embed_test.go

@@ -0,0 +1,42 @@
+package testdata
+
+import (
+	"bytes"
+	"os"
+	"testing"
+
+	"github.com/pressly/goose/v3/internal/gooseutil"
+	"github.com/stretchr/testify/require"
+)
+
+func TestEmbededMigrations(t *testing.T) {
+	t.Parallel()
+
+	files, err := EmbedMigrations.ReadDir(".")
+	require.NoError(t, err)
+	require.Len(t, files, 1)
+	require.Equal(t, "migrations", files[0].Name())
+
+	files, err = EmbedMigrations.ReadDir("migrations")
+	require.NoError(t, err)
+	got := make([]string, 0, len(files))
+	for _, file := range files {
+		got = append(got, file.Name())
+	}
+	require.ElementsMatch(t, []string{"postgres"}, got)
+
+	t.Run("postgres", func(t *testing.T) {
+		expected, err := os.ReadFile("migrations/postgres.sha256")
+		require.NoError(t, err)
+		expected = bytes.TrimSpace(expected)
+		files, err := EmbedMigrations.ReadDir("migrations/postgres")
+		require.NoError(t, err)
+		require.Len(t, files, 5)
+
+		digest, err := gooseutil.Digest(EmbedMigrations, "migrations/postgres")
+		require.NoError(t, err)
+
+		require.Equal(t, string(expected), digest)
+	})
+
+}

internal/testdata/migrations/postgres.sha256

@@ -0,0 +1 @@
+3129d29e299423d908d4ad8e597675b1

internal/testdata/migrations/postgres/00001_users_table.sql

@@ -0,0 +1,10 @@
+-- +goose Up
+CREATE TABLE users (
+    id INTEGER PRIMARY KEY,
+    username TEXT NOT NULL,
+    email TEXT NOT NULL,
+    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
+);
+
+-- +goose Down
+DROP TABLE users;

internal/testdata/migrations/postgres/00002_posts_table.sql

@@ -0,0 +1,14 @@
+-- +goose Up
+-- +goose StatementBegin
+CREATE TABLE posts (
+    id INTEGER PRIMARY KEY,
+    title TEXT NOT NULL,
+    content TEXT NOT NULL,
+    author_id INTEGER NOT NULL,
+    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
+    FOREIGN KEY (author_id) REFERENCES users(id)
+);
+-- +goose StatementEnd
+
+-- +goose Down
+DROP TABLE posts;

internal/testdata/migrations/postgres/00003_comments_table.sql

@@ -0,0 +1,13 @@
+-- +goose Up
+CREATE TABLE comments (
+    id INTEGER PRIMARY KEY,
+    post_id INTEGER NOT NULL,
+    user_id INTEGER NOT NULL,
+    content TEXT NOT NULL,
+    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
+    FOREIGN KEY (post_id) REFERENCES posts(id),
+    FOREIGN KEY (user_id) REFERENCES users(id)
+);
+
+-- +goose Down
+DROP TABLE comments;

internal/testdata/migrations/postgres/00004_insert_data.sql

@@ -0,0 +1,23 @@
+-- +goose Up
+INSERT INTO users (id, username, email)
+VALUES
+    (1, 'john_doe', 'john@example.com'),
+    (2, 'jane_smith', 'jane@example.com'),
+    (3, 'alice_wonderland', 'alice@example.com');
+
+INSERT INTO posts (id, title, content, author_id)
+VALUES
+    (1, 'Introduction to SQL', 'SQL is a powerful language for managing databases...', 1),
+    (2, 'Data Modeling Techniques', 'Choosing the right data model is crucial...', 2),
+    (3, 'Advanced Query Optimization', 'Optimizing queries can greatly improve...', 1);
+
+INSERT INTO comments (id, post_id, user_id, content)
+VALUES
+    (1, 1, 3, 'Great introduction! Looking forward to more.'),
+    (2, 1, 2, 'SQL can be a bit tricky at first, but practice helps.'),
+    (3, 2, 1, 'You covered normalization really well in this post.');
+
+-- +goose Down
+DELETE FROM comments;
+DELETE FROM posts;
+DELETE FROM users;

internal/testdata/migrations/postgres/00005_posts_view.sql

@@ -0,0 +1,15 @@
+-- +goose NO TRANSACTION
+
+-- +goose Up
+CREATE VIEW posts_view AS
+    SELECT
+        p.id,
+        p.title,
+        p.content,
+        p.created_at,
+        u.username AS author
+    FROM posts p
+    JOIN users u ON p.author_id = u.id;
+
+-- +goose Down
+DROP VIEW posts_view;

scripts/dirhash.sh

@@ -0,0 +1,32 @@
+#!/bin/bash
+
+set -euo pipefail
+
+# Check if the required argument is provided, a directory path
+if [ $# -lt 1 ]; then
+  echo "Usage: $0 <directory path>"
+  exit 1
+fi
+
+dir_path="$1"
+
+# Check if the directory path exists
+if [ ! -d "$dir_path" ]; then
+  echo "Error: $dir_path does not exist"
+  exit 1
+fi
+
+# Calculate the hash of each directory in the specified directory
+for dir in "$dir_path"/*; do
+  if [ -d "$dir" ]; then
+    # remove the dir_path from the found files
+    all_files=$(find "$dir" -type f | sort | sed "s|$dir_path/||")
+    cd $dir_path
+    sha256sum $all_files
+    digest=$(sha256sum $all_files | sha256sum | cut -c 1-32)
+    echo ""
+    echo "Hash of $dir: $digest"
+    echo $digest > "$(basename "$dir").sha256"
+    echo ""
+  fi
+done