Add String() methods to parsed types (#48)

This enables clients to move back and forth between parsed objects and
text patches. The generated patches are semantically equal to the parsed
object and should re-parse to the same object, but may not be
byte-for-byte identical to the original input.

In my testing, formatted text patches are usually identical to the
input, but there may be cases where this is not true. Binary patches
always differ. This is because Go's 'compress/flate' package ends
streams with an empty block instead of adding the end-of-stream flag to
the last non-empty block, like Git's C implementation. Since the streams
will always be different for this reason, I chose to also enable default
compression (the test patches I generated with Git used no compression.)

The main tests for this feature involve parsing, formatting, and then
re-parsing a patch to make sure we get equal objects.

Formatting is handled by a new internal formatter type, which allows
writing all data to the same stream. This isn't exposed publicly right
now, but will be useful if there's a need for more flexible formatting
functions in the future, like formatting to a user-provided io.Writer.

Author: bluekeyes

.golangci.yml

@@ -19,6 +19,12 @@ issues:
   exclude-use-default: false
 
 linters-settings:
+  errcheck:
+    exclude-functions:
+      - (*github.com/bluekeyes/go-gitdiff/gitdiff.formatter).Write
+      - (*github.com/bluekeyes/go-gitdiff/gitdiff.formatter).WriteString
+      - (*github.com/bluekeyes/go-gitdiff/gitdiff.formatter).WriteByte
+      - fmt.Fprintf(*github.com/bluekeyes/go-gitdiff/gitdiff.formatter)
   goimports:
     local-prefixes: github.com/bluekeyes/go-gitdiff
   revive:

gitdiff/base85.go

@@ -19,8 +19,8 @@ func init() {
 }
 
 // base85Decode decodes Base85-encoded data from src into dst. It uses the
-// alphabet defined by base85.c in the Git source tree, which appears to be
-// unique. src must contain at least len(dst) bytes of encoded data.
+// alphabet defined by base85.c in the Git source tree. src must contain at
+// least len(dst) bytes of encoded data.
 func base85Decode(dst, src []byte) error {
 	var v uint32
 	var n, ndst int
@@ -50,3 +50,42 @@ func base85Decode(dst, src []byte) error {
 	}
 	return nil
 }
+
+// base85Encode encodes src in Base85, writing the result to dst. It uses the
+// alphabet defined by base85.c in the Git source tree.
+func base85Encode(dst, src []byte) {
+	var di, si int
+
+	encode := func(v uint32) {
+		dst[di+0] = b85Alpha[(v/(85*85*85*85))%85]
+		dst[di+1] = b85Alpha[(v/(85*85*85))%85]
+		dst[di+2] = b85Alpha[(v/(85*85))%85]
+		dst[di+3] = b85Alpha[(v/85)%85]
+		dst[di+4] = b85Alpha[v%85]
+	}
+
+	n := (len(src) / 4) * 4
+	for si < n {
+		encode(uint32(src[si+0])<<24 | uint32(src[si+1])<<16 | uint32(src[si+2])<<8 | uint32(src[si+3]))
+		si += 4
+		di += 5
+	}
+
+	var v uint32
+	switch len(src) - si {
+	case 3:
+		v |= uint32(src[si+2]) << 8
+		fallthrough
+	case 2:
+		v |= uint32(src[si+1]) << 16
+		fallthrough
+	case 1:
+		v |= uint32(src[si+0]) << 24
+		encode(v)
+	}
+}
+
+// base85Len returns the length of n bytes of Base85 encoded data.
+func base85Len(n int) int {
+	return (n + 3) / 4 * 5
+}

gitdiff/base85_test.go

@@ -1,6 +1,7 @@
 package gitdiff
 
 import (
+	"bytes"
 	"testing"
 )
 
@@ -58,3 +59,60 @@ func TestBase85Decode(t *testing.T) {
 		})
 	}
 }
+
+func TestBase85Encode(t *testing.T) {
+	tests := map[string]struct {
+		Input  []byte
+		Output string
+	}{
+		"zeroBytes": {
+			Input:  []byte{},
+			Output: "",
+		},
+		"twoBytes": {
+			Input:  []byte{0xCA, 0xFE},
+			Output: "%KiWV",
+		},
+		"fourBytes": {
+			Input:  []byte{0x0, 0x0, 0xCA, 0xFE},
+			Output: "007GV",
+		},
+		"sixBytes": {
+			Input:  []byte{0x0, 0x0, 0xCA, 0xFE, 0xCA, 0xFE},
+			Output: "007GV%KiWV",
+		},
+	}
+
+	for name, test := range tests {
+		t.Run(name, func(t *testing.T) {
+			dst := make([]byte, len(test.Output))
+			base85Encode(dst, test.Input)
+			for i, b := range test.Output {
+				if dst[i] != byte(b) {
+					t.Errorf("incorrect character at index %d: expected '%c', actual '%c'", i, b, dst[i])
+				}
+			}
+		})
+	}
+}
+
+func FuzzBase85Roundtrip(f *testing.F) {
+	f.Add([]byte{0x2b, 0x0d})
+	f.Add([]byte{0xbc, 0xb4, 0x3f})
+	f.Add([]byte{0xfa, 0x62, 0x05, 0x83, 0x24, 0x39, 0xd5, 0x25})
+	f.Add([]byte{0x31, 0x59, 0x02, 0xa0, 0x61, 0x12, 0xd9, 0x43, 0xb8, 0x23, 0x1a, 0xb4, 0x02, 0xae, 0xfa, 0xcc, 0x22, 0xad, 0x41, 0xb9, 0xb8})
+
+	f.Fuzz(func(t *testing.T, in []byte) {
+		n := len(in)
+		dst := make([]byte, base85Len(n))
+		out := make([]byte, n)
+
+		base85Encode(dst, in)
+		if err := base85Decode(out, dst); err != nil {
+			t.Fatalf("unexpected error decoding base85 data: %v", err)
+		}
+		if !bytes.Equal(in, out) {
+			t.Errorf("decoded data differed from input data:\n   input: %x\n  output: %x\nencoding: %s\n", in, out, string(dst))
+		}
+	})
+}