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

atomicwrite.go

@@ -1,56 +0,0 @@
-package util
-
-import (
-	"os"
-	"path/filepath"
-)
-
-// AtomicWriteFile writes data to path without ever leaving a half-written file
-// behind. It writes to a temporary file in the same directory, flushes it to
-// disk, gives it the mode in perm and renames it over path, so anything reading
-// path sees either the previous contents or the complete new contents and
-// never something in between. If any step fails, the temporary file is removed
-// and path is left as it was.
-//
-// The temporary file has to live in the same directory as path, because a
-// rename across two filesystems is not possible and would not be instant if it
-// were.
-//
-// Two details differ from os.WriteFile. The mode is applied after the file is
-// created, so perm is what the finished file ends up with rather than perm with
-// the process umask taken out of it. And the directory holding the file is not
-// flushed, so a machine that loses power immediately after this returns may
-// come back with the rename undone, even though the data itself was written.
-func AtomicWriteFile(path string, data []byte, perm os.FileMode) error {
-	directory := filepath.Dir(path)
-
-	temporary, err := os.CreateTemp(directory, "."+filepath.Base(path)+".tmp")
-	if err != nil {
-		return err
-	}
-	temporaryPath := temporary.Name()
-
-	// Does nothing once the rename below has succeeded, because by then the
-	// temporary name no longer refers to anything.
-	defer os.Remove(temporaryPath)
-
-	if _, err := temporary.Write(data); err != nil {
-		temporary.Close()
-		return err
-	}
-
-	if err := temporary.Sync(); err != nil {
-		temporary.Close()
-		return err
-	}
-
-	if err := temporary.Close(); err != nil {
-		return err
-	}
-
-	if err := os.Chmod(temporaryPath, perm); err != nil {
-		return err
-	}
-
-	return os.Rename(temporaryPath, path)
-}

atomicwrite_test.go

@@ -1,110 +0,0 @@
-package util
-
-import (
-	"os"
-	"path/filepath"
-	"testing"
-)
-
-func TestAtomicWriteFile(t *testing.T) {
-	tests := []struct {
-		name     string
-		existing string
-		data     []byte
-		perm     os.FileMode
-	}{
-		{"a new file", "", []byte("hello"), 0644},
-		{"replacing a shorter file", "old", []byte("a much longer set of contents"), 0644},
-		{"replacing a longer file", "a much longer set of contents", []byte("new"), 0644},
-		{"an empty payload", "something", []byte{}, 0644},
-		{"a nil payload", "something", nil, 0644},
-		{"a private mode", "", []byte("secret"), 0600},
-		{"an executable mode", "", []byte("#!/bin/sh\n"), 0755},
-		{"binary contents", "", []byte{0x00, 0xff, 0x10, 0x00}, 0644},
-	}
-
-	for _, test := range tests {
-		t.Run(test.name, func(t *testing.T) {
-			directory := t.TempDir()
-			path := filepath.Join(directory, "target")
-
-			if test.existing != "" {
-				if err := os.WriteFile(path, []byte(test.existing), 0666); err != nil {
-					t.Fatalf("could not write the file to be replaced: %v", err)
-				}
-			}
-
-			if err := AtomicWriteFile(path, test.data, test.perm); err != nil {
-				t.Fatalf("did not expect an error, got %v", err)
-			}
-
-			written, err := os.ReadFile(path)
-			if err != nil {
-				t.Fatalf("could not read the file back: %v", err)
-			}
-			if string(written) != string(test.data) {
-				t.Errorf("expected contents %q got %q", string(test.data), string(written))
-			}
-
-			info, err := os.Stat(path)
-			if err != nil {
-				t.Fatalf("could not look at the file: %v", err)
-			}
-			if info.Mode().Perm() != test.perm {
-				t.Errorf("expected mode %v got %v", test.perm, info.Mode().Perm())
-			}
-
-			entries, err := os.ReadDir(directory)
-			if err != nil {
-				t.Fatalf("could not list the directory: %v", err)
-			}
-			if len(entries) != 1 {
-				names := make([]string, 0, len(entries))
-				for _, entry := range entries {
-					names = append(names, entry.Name())
-				}
-				t.Errorf("expected only the target file to be left, got %v", names)
-			}
-		})
-	}
-}
-
-func TestAtomicWriteFileReportsAMissingDirectory(t *testing.T) {
-	path := filepath.Join(t.TempDir(), "not-there", "target")
-
-	if err := AtomicWriteFile(path, []byte("hello"), 0644); err == nil {
-		t.Errorf("expected an error for a directory that does not exist")
-	}
-}
-
-func TestAtomicWriteFileLeavesTheOldFileAloneOnFailure(t *testing.T) {
-	directory := t.TempDir()
-	path := filepath.Join(directory, "target")
-
-	if err := os.WriteFile(path, []byte("original"), 0644); err != nil {
-		t.Fatalf("could not write the file to be replaced: %v", err)
-	}
-
-	// A directory that cannot be written to means the temporary file cannot
-	// be created, so the write has to fail before anything is replaced.
-	if err := os.Chmod(directory, 0500); err != nil {
-		t.Fatalf("could not change the directory mode: %v", err)
-	}
-	defer os.Chmod(directory, 0700)
-
-	if os.Geteuid() == 0 {
-		t.Skip("running as root, which ignores the directory mode")
-	}
-
-	if err := AtomicWriteFile(path, []byte("replacement"), 0644); err == nil {
-		t.Fatalf("expected an error for a directory that cannot be written to")
-	}
-
-	existing, err := os.ReadFile(path)
-	if err != nil {
-		t.Fatalf("could not read the file back: %v", err)
-	}
-	if string(existing) != "original" {
-		t.Errorf("expected the original contents to survive, got %q", string(existing))
-	}
-}

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)
+			}
+		})
+	}
+}