add CollapseWhitespace for flattening runs of whitespace

CollapseWhitespace turns every run of whitespace into a single space and
trims the ends, so a block of text becomes one tidy line. Comes with a doc
comment and table-driven tests. (closes #7)

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")
-	}
-}

whitespace.go

@@ -0,0 +1,12 @@
+package util
+
+import "strings"
+
+// CollapseWhitespace replaces every run of whitespace in s with a single space
+// and removes any whitespace at the start and end. Tabs, newlines and spaces
+// outside ASCII such as the non-breaking space all count as whitespace, so a
+// block of text comes back as one line with single spaces between the words.
+// A string that is nothing but whitespace comes back empty.
+func CollapseWhitespace(s string) string {
+	return strings.Join(strings.Fields(s), " ")
+}

whitespace_test.go

@@ -0,0 +1,36 @@
+package util
+
+import "testing"
+
+// Written as a code point so that it is visible in the source rather than
+// looking like an ordinary space.
+const nonBreakingSpace = string(rune(0x00a0))
+
+func TestCollapseWhitespace(t *testing.T) {
+	tests := []struct {
+		name     string
+		input    string
+		expected string
+	}{
+		{"nothing to change", "one two", "one two"},
+		{"leading and trailing spaces", "  one two  ", "one two"},
+		{"run in the middle", "one    two", "one two"},
+		{"tabs and newlines", "one\t\ttwo\nthree", "one two three"},
+		{"windows line endings", "one\r\ntwo", "one two"},
+		{"non-breaking space", "one" + nonBreakingSpace + "two", "one two"},
+		{"mixed whitespace run", "one \t" + nonBreakingSpace + " two", "one two"},
+		{"only whitespace", " \t\n ", ""},
+		{"empty string", "", ""},
+		{"single word with padding", "\tword\n", "word"},
+		{"paragraph becomes one line", "first line\n\nsecond line\n", "first line second line"},
+	}
+
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			got := CollapseWhitespace(test.input)
+			if got != test.expected {
+				t.Errorf("expected %q got %q", test.expected, got)
+			}
+		})
+	}
+}