difflib: implement context diffs

Author: pmezard

README.md

@@ -5,8 +5,9 @@ The following class and functions have be ported:
 
 * `SequenceMatcher`
 * `unified_diff()`
+* `context_diff()`
 
-Related doctests have been ported as well.
+Related doctests and unittests have been ported as well.
 
 I have barely used to code yet so do not consider it being production-ready.
 The API is likely to evolve too.

difflib/difflib.go

@@ -626,3 +626,120 @@ func GetUnifiedDiffString(diff UnifiedDiff) (string, error) {
 	err := WriteUnifiedDiff(w, diff)
 	return string(w.Bytes()), err
 }
+
+// Convert range to the "ed" format.
+func formatRangeContext(start, stop int) string {
+	// Per the diff spec at http://www.unix.org/single_unix_specification/
+	beginning := start + 1 // lines start numbering with one
+	length := stop - start
+	if length == 0 {
+		beginning -= 1 // empty ranges begin at line just before the range
+	}
+	if length <= 1 {
+		return fmt.Sprintf("%d", beginning)
+	}
+	return fmt.Sprintf("%d,%d", beginning, beginning+length-1)
+}
+
+type ContextDiff UnifiedDiff
+
+// Compare two sequences of lines; generate the delta as a context diff.
+//
+// Context diffs are a compact way of showing line changes and a few
+// lines of context. The number of context lines is set by diff.Context
+// which defaults to three.
+//
+// By default, the diff control lines (those with *** or ---) are
+// created with a trailing newline.
+//
+// For inputs that do not have trailing newlines, set the diff.Eol
+// argument to "" so that the output will be uniformly newline free.
+//
+// The context diff format normally has a header for filenames and
+// modification times.  Any or all of these may be specified using
+// strings for diff.FromFile, diff.ToFile, diff.FromDate, diff.ToDate.
+// The modification times are normally expressed in the ISO 8601 format.
+// If not specified, the strings default to blanks.
+func WriteContextDiff(writer io.Writer, diff ContextDiff) error {
+	buf := bufio.NewWriter(writer)
+	defer buf.Flush()
+	var diffErr error
+	w := func(format string, args ...interface{}) {
+		_, err := buf.WriteString(fmt.Sprintf(format, args...))
+		if diffErr == nil && err != nil {
+			diffErr = err
+		}
+	}
+
+	if len(diff.Eol) == 0 {
+		diff.Eol = "\n"
+	}
+
+	prefix := map[byte]string{
+		'i': "+ ",
+		'd': "- ",
+		'r': "! ",
+		'e': "  ",
+	}
+
+	started := false
+	m := NewMatcher(diff.A, diff.B)
+	for _, g := range m.GetGroupedOpCodes(diff.Context) {
+		if !started {
+			started = true
+			fromDate := ""
+			if len(diff.FromDate) > 0 {
+				fromDate = "\t" + diff.FromDate
+			}
+			toDate := ""
+			if len(diff.ToDate) > 0 {
+				toDate = "\t" + diff.ToDate
+			}
+			w("*** %s%s%s", diff.FromFile, fromDate, diff.Eol)
+			w("--- %s%s%s", diff.ToFile, toDate, diff.Eol)
+		}
+
+		first, last := g[0], g[len(g)-1]
+		w("***************" + diff.Eol)
+
+		range1 := formatRangeContext(first.I1, last.I2)
+		w("*** %s ****%s", range1, diff.Eol)
+		for _, c := range g {
+			if c.Tag == 'r' || c.Tag == 'd' {
+				for _, cc := range g {
+					if cc.Tag == 'i' {
+						continue
+					}
+					for _, line := range diff.A[cc.I1:cc.I2] {
+						w(prefix[cc.Tag] + line)
+					}
+				}
+				break
+			}
+		}
+
+		range2 := formatRangeContext(first.J1, last.J2)
+		w("--- %s ----%s", range2, diff.Eol)
+		for _, c := range g {
+			if c.Tag == 'r' || c.Tag == 'i' {
+				for _, cc := range g {
+					if cc.Tag == 'd' {
+						continue
+					}
+					for _, line := range diff.B[cc.J1:cc.J2] {
+						w(prefix[cc.Tag] + line)
+					}
+				}
+				break
+			}
+		}
+	}
+	return diffErr
+}
+
+// Like WriteContextDiff but returns the diff a string.
+func GetContextDiffString(diff ContextDiff) (string, error) {
+	w := &bytes.Buffer{}
+	err := WriteContextDiff(w, diff)
+	return string(w.Bytes()), err
+}

difflib/difflib_test.go

@@ -148,6 +148,46 @@ four`
 	}
 }
 
+func TestContextDiff(t *testing.T) {
+	a := `one
+two
+three
+four`
+	b := `zero
+one
+tree
+four`
+	diff := ContextDiff{
+		A:        splitLines(a),
+		B:        splitLines(b),
+		FromFile: "Original",
+		ToFile:   "Current",
+		Context:  3,
+		Eol:      "\n",
+	}
+	result, err := GetContextDiffString(diff)
+	assertEqual(t, err, nil)
+	expected := `*** Original
+--- Current
+***************
+*** 1,4 ****
+  one
+! two
+! three
+  four
+--- 1,4 ----
++ zero
+  one
+! tree
+  four
+`
+	// TABs are a pain to preserve through editors
+	expected = strings.Replace(expected, "\\t", "\t", -1)
+	if expected != result {
+		t.Errorf("unexpected diff result:\n%s", result)
+	}
+}
+
 func rep(s string, count int) string {
 	return strings.Repeat(s, count)
 }
@@ -232,3 +272,68 @@ func TestOutputFormatRangeFormatUnified(t *testing.T) {
 	assertEqual(t, fm(3, 6), "4,3")
 	assertEqual(t, fm(0, 0), "0,0")
 }
+
+func TestOutputFormatRangeFormatContext(t *testing.T) {
+	// Per the diff spec at http://www.unix.org/single_unix_specification/
+	//
+	// The range of lines in file1 shall be written in the following format
+	// if the range contains two or more lines:
+	//     "*** %d,%d ****\n", <beginning line number>, <ending line number>
+	// and the following format otherwise:
+	//     "*** %d ****\n", <ending line number>
+	// The ending line number of an empty range shall be the number of the preceding line,
+	// or 0 if the range is at the start of the file.
+	//
+	// Next, the range of lines in file2 shall be written in the following format
+	// if the range contains two or more lines:
+	//     "--- %d,%d ----\n", <beginning line number>, <ending line number>
+	// and the following format otherwise:
+	//     "--- %d ----\n", <ending line number>
+	fm := formatRangeContext
+	assertEqual(t, fm(3, 3), "3")
+	assertEqual(t, fm(3, 4), "4")
+	assertEqual(t, fm(3, 5), "4,5")
+	assertEqual(t, fm(3, 6), "4,6")
+	assertEqual(t, fm(0, 0), "0")
+}
+
+func TestOutputFormatTabDelimiter(t *testing.T) {
+	diff := UnifiedDiff{
+		A:        splitChars("one"),
+		B:        splitChars("two"),
+		FromFile: "Original",
+		FromDate: "2005-01-26 23:30:50",
+		ToFile:   "Current",
+		ToDate:   "2010-04-12 10:20:52",
+		Eol:      "\n",
+	}
+	ud, err := GetUnifiedDiffString(diff)
+	assertEqual(t, err, nil)
+	assertEqual(t, splitLines(ud)[:2], []string{
+		"--- Original\t2005-01-26 23:30:50\n",
+		"+++ Current\t2010-04-12 10:20:52\n",
+	})
+	cd, err := GetContextDiffString(ContextDiff(diff))
+	assertEqual(t, err, nil)
+	assertEqual(t, splitLines(cd)[:2], []string{
+		"*** Original\t2005-01-26 23:30:50\n",
+		"--- Current\t2010-04-12 10:20:52\n",
+	})
+}
+
+func TestOutputFormatNoTrailingTabOnEmptyFiledate(t *testing.T) {
+	diff := UnifiedDiff{
+		A:        splitChars("one"),
+		B:        splitChars("two"),
+		FromFile: "Original",
+		ToFile:   "Current",
+		Eol:      "\n",
+	}
+	ud, err := GetUnifiedDiffString(diff)
+	assertEqual(t, err, nil)
+	assertEqual(t, splitLines(ud)[:2], []string{"--- Original\n", "+++ Current\n"})
+
+	cd, err := GetContextDiffString(ContextDiff(diff))
+	assertEqual(t, err, nil)
+	assertEqual(t, splitLines(cd)[:2], []string{"*** Original\n", "--- Current\n"})
+}