add TruncateString for shortening a string safely

TruncateString returns at most the requested number of characters from the
start of a string, counting characters rather than bytes so a multi-byte
character is never cut in half. Comes with a doc comment and table-driven
tests. (closes #5)

Model: opus-5

sha256file.go

@@ -1,30 +0,0 @@
-package util
-
-import (
-	"crypto/sha256"
-	"encoding/hex"
-	"io"
-	"os"
-)
-
-// SHA256File returns the SHA-256 digest of the contents of the file at path,
-// written as lowercase hexadecimal, in the same form as the sha256sum command
-// prints. The file is copied through the hash in chunks rather than read into
-// memory, so its size does not matter.
-//
-// The error is whatever went wrong opening or reading the file, passed along
-// unchanged so a caller can test it with os.IsNotExist and the like.
-func SHA256File(path string) (string, error) {
-	file, err := os.Open(path)
-	if err != nil {
-		return "", err
-	}
-	defer file.Close()
-
-	digest := sha256.New()
-	if _, err := io.Copy(digest, file); err != nil {
-		return "", err
-	}
-
-	return hex.EncodeToString(digest.Sum(nil)), nil
-}

sha256file_test.go

@@ -1,76 +0,0 @@
-package util
-
-import (
-	"bytes"
-	"os"
-	"path/filepath"
-	"strings"
-	"testing"
-)
-
-func TestSHA256File(t *testing.T) {
-	tests := []struct {
-		name     string
-		contents []byte
-		expected string
-	}{
-		{
-			"an empty file",
-			[]byte{},
-			"e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
-		},
-		{
-			"a short file",
-			[]byte("hello world\n"),
-			"a948904f2f0f479b8f8197694b30184b0d2ed1c1cd2a1ec0fb85d299a192a447",
-		},
-		{
-			"binary contents",
-			[]byte{0x00, 0x01, 0x02, 0x03},
-			"054edec1d0211f624fed0cbca9d4f9400b0e491c43742af2c5b0abebf0c990d8",
-		},
-		{
-			"larger than one read buffer",
-			bytes.Repeat([]byte("a"), 100000),
-			"6d1cf22d7cc09b085dfc25ee1a1f3ae0265804c607bc2074ad253bcc82fd81ee",
-		},
-	}
-
-	for _, test := range tests {
-		t.Run(test.name, func(t *testing.T) {
-			path := filepath.Join(t.TempDir(), "file")
-			if err := os.WriteFile(path, test.contents, 0644); err != nil {
-				t.Fatalf("could not write the test file: %v", err)
-			}
-
-			got, err := SHA256File(path)
-			if err != nil {
-				t.Fatalf("did not expect an error, got %v", err)
-			}
-			if got != test.expected {
-				t.Errorf("expected %q got %q", test.expected, got)
-			}
-			if got != strings.ToLower(got) {
-				t.Errorf("expected lowercase hexadecimal, got %q", got)
-			}
-		})
-	}
-}
-
-func TestSHA256FileReportsAMissingFile(t *testing.T) {
-	path := filepath.Join(t.TempDir(), "not-there")
-
-	got, err := SHA256File(path)
-	if err == nil {
-		t.Fatalf("expected an error for a file that does not exist, got %q", got)
-	}
-	if !os.IsNotExist(err) {
-		t.Errorf("expected an error that os.IsNotExist recognises, got %v", err)
-	}
-}
-
-func TestSHA256FileReportsADirectory(t *testing.T) {
-	if _, err := SHA256File(t.TempDir()); err == nil {
-		t.Errorf("expected an error for a directory")
-	}
-}

truncate.go

@@ -0,0 +1,21 @@
+package util
+
+// TruncateString returns at most max characters from the start of s. It counts
+// characters rather than bytes, so a multi-byte character is never cut in half
+// and the result is always valid text. A max of zero or less returns an empty
+// string, and a string already at or below the limit is returned unchanged.
+func TruncateString(s string, max int) string {
+	if max <= 0 {
+		return ""
+	}
+
+	count := 0
+	for offset := range s {
+		if count == max {
+			return s[:offset]
+		}
+		count++
+	}
+
+	return s
+}

truncate_test.go

@@ -0,0 +1,41 @@
+package util
+
+import "testing"
+
+func TestTruncateString(t *testing.T) {
+	tests := []struct {
+		name     string
+		input    string
+		max      int
+		expected string
+	}{
+		{"shorter than the limit", "abc", 5, "abc"},
+		{"exactly at the limit", "abc", 3, "abc"},
+		{"longer than the limit", "abcdef", 3, "abc"},
+		{"max of one", "abc", 1, "a"},
+		{"max of zero", "abc", 0, ""},
+		{"negative max", "abc", -1, ""},
+		{"empty string", "", 3, ""},
+		{"multi-byte characters kept whole", "héllo", 3, "hél"},
+		{"multi-byte characters below the limit", "héllo", 99, "héllo"},
+		{"characters outside the basic plane", "a😀b", 2, "a😀"},
+	}
+
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			got := TruncateString(test.input, test.max)
+			if got != test.expected {
+				t.Errorf("expected %q got %q", test.expected, got)
+			}
+		})
+	}
+}
+
+func TestTruncateStringCountsCharactersNotBytes(t *testing.T) {
+	// Each of these characters is two bytes long, so a byte-based cut at
+	// three would leave half a character behind.
+	got := TruncateString("äöü", 3)
+	if got != "äöü" {
+		t.Errorf("expected %q got %q", "äöü", got)
+	}
+}