add Slugify for making URL-safe slugs out of text

Slugify lowercases text and keeps only ASCII letters and digits, turning every
run of other characters into a single hyphen with none left at either end.
Comes with a doc comment and table-driven tests. (closes #9)

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")
-	}
-}

slugify.go

@@ -0,0 +1,36 @@
+package util
+
+import "strings"
+
+// Slugify returns input as lowercase ASCII letters, digits and hyphens, which
+// is safe to put in a URL path, a filename or a fragment identifier. Every run
+// of other characters becomes a single hyphen, and the result never starts or
+// ends with one.
+//
+// Characters outside ASCII are treated as separators rather than being folded
+// to their nearest ASCII relative, because doing that properly needs a Unicode
+// table that this package does not carry. Text with no ASCII letters or digits
+// in it therefore slugifies to an empty string.
+func Slugify(input string) string {
+	var out strings.Builder
+	pendingHyphen := false
+
+	for _, character := range strings.ToLower(input) {
+		if (character >= 'a' && character <= 'z') || (character >= '0' && character <= '9') {
+			if pendingHyphen {
+				out.WriteByte('-')
+				pendingHyphen = false
+			}
+			out.WriteRune(character)
+			continue
+		}
+
+		// Remember that a separator was seen, but only write a hyphen once
+		// another usable character turns up, so nothing trails off the end.
+		if out.Len() > 0 {
+			pendingHyphen = true
+		}
+	}
+
+	return out.String()
+}

slugify_test.go

@@ -0,0 +1,34 @@
+package util
+
+import "testing"
+
+func TestSlugify(t *testing.T) {
+	tests := []struct {
+		name     string
+		input    string
+		expected string
+	}{
+		{"already a slug", "hello-world", "hello-world"},
+		{"mixed case", "Hello World", "hello-world"},
+		{"punctuation", "What's new?", "what-s-new"},
+		{"run of separators", "one   ---   two", "one-two"},
+		{"leading separators", "///one", "one"},
+		{"trailing separators", "one///", "one"},
+		{"separators at both ends", "  --one two--  ", "one-two"},
+		{"digits are kept", "Go 1.14 release", "go-1-14-release"},
+		{"underscores are separators", "some_name_here", "some-name-here"},
+		{"nothing usable", "!!! ???", ""},
+		{"empty string", "", ""},
+		{"outside ascii", "Grüße, Welt", "gr-e-welt"},
+		{"only characters outside ascii", "日本語", ""},
+	}
+
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			got := Slugify(test.input)
+			if got != test.expected {
+				t.Errorf("expected %q got %q", test.expected, got)
+			}
+		})
+	}
+}