Add handling added/deleted files in diff (#50)

Author: sofiia-tesliuk

diff/diff.go

@@ -54,10 +54,12 @@ func (h *Hunk) Stat() Stat {
 }
 
 var (
-	hunkPrefix = []byte("@@ ")
+	hunkPrefix          = []byte("@@ ")
+	onlyInMessagePrefix = []byte("Only in ")
 )
 
 const hunkHeader = "@@ -%d,%d +%d,%d @@"
+const onlyInMessage = "Only in %s: %s\n"
 
 // diffTimeParseLayout is the layout used to parse the time in unified diff file
 // header timestamps.

diff/diff_test.go

@@ -496,6 +496,123 @@ func TestParseMultiFileDiffHeaders(t *testing.T) {
 				},
 			},
 		},
+		{
+			filename: "sample_contains_added_deleted_files.diff",
+			wantDiffs: []*FileDiff{
+				{
+					OrigName: "source_a/file_1.txt",
+					OrigTime: nil,
+					NewName:  "source_b/file_1.txt",
+					NewTime:  nil,
+					Extended: []string{
+						"diff -u source_a/file_1.txt  source_b/file_1.txt",
+					},
+				},
+				{
+					OrigName: "source_a/file_2.txt",
+					OrigTime: nil,
+					NewName:  "",
+					NewTime:  nil,
+					Extended: nil,
+				},
+				{
+					OrigName: "source_b/file_3.txt",
+					OrigTime: nil,
+					NewName:  "",
+					NewTime:  nil,
+					Extended: nil,
+				},
+			},
+		},
+		{
+			filename: "sample_contains_only_added_deleted_files.diff",
+			wantDiffs: []*FileDiff{
+				{
+					OrigName: "source_a/file_1.txt",
+					OrigTime: nil,
+					NewName:  "",
+					NewTime:  nil,
+					Extended: nil,
+				},
+				{
+					OrigName: "source_a/file_2.txt",
+					OrigTime: nil,
+					NewName:  "",
+					NewTime:  nil,
+					Extended: nil,
+				},
+				{
+					OrigName: "source_b/file_3.txt",
+					OrigTime: nil,
+					NewName:  "",
+					NewTime:  nil,
+					Extended: nil,
+				},
+			},
+		},
+		{
+			filename: "sample_onlyin_line_isnt_a_file_header.diff",
+			wantDiffs: []*FileDiff{
+				{
+					OrigName: "source_a/file_1.txt",
+					OrigTime: nil,
+					NewName:  "source_b/file_1.txt",
+					NewTime:  nil,
+					Extended: []string{
+						"diff -u source_a/file_1.txt  source_b/file_1.txt",
+					},
+				},
+				{
+					OrigName: "source_a/file_2.txt",
+					OrigTime: nil,
+					NewName:  "",
+					NewTime:  nil,
+					Extended: []string{
+						"Only in universe!",
+					},
+				},
+				{
+					OrigName: "source_b/file_3.txt some unrelated stuff here.",
+					OrigTime: nil,
+					NewName:  "",
+					NewTime:  nil,
+					Extended: nil,
+				},
+				{
+					OrigName: "source_b/file_3.txt",
+					OrigTime: nil,
+					NewName:  "",
+					NewTime:  nil,
+					Extended: nil,
+				},
+			},
+		},
+		{
+			filename: "sample_onlyin_complex_filenames.diff",
+			wantDiffs: []*FileDiff{
+				{
+					OrigName: "internal/trace/foo bar/bam",
+					OrigTime: nil,
+					NewName:  "",
+					NewTime:  nil,
+					Extended: nil,
+				},
+				{
+					OrigName: "internal/trace/foo bar/bam: bar",
+					OrigTime: nil,
+					NewName:  "",
+					NewTime:  nil,
+					Extended: nil,
+				},
+				{
+					OrigName: "internal/trace/hello/world: bazz",
+					OrigTime: nil,
+					NewName:  "",
+					NewTime:  nil,
+					Extended: nil,
+				},
+			},
+		},
 	}
 	for _, test := range tests {
 		diffData, err := ioutil.ReadFile(filepath.Join("testdata", test.filename))
@@ -574,6 +691,10 @@ func TestParseMultiFileDiffAndPrintMultiFileDiff(t *testing.T) {
 		{filename: "long_line_multi.diff", wantFileDiffs: 3},
 		{filename: "empty.diff", wantFileDiffs: 0},
 		{filename: "empty_multi.diff", wantFileDiffs: 2},
+		{filename: "sample_contains_added_deleted_files.diff", wantFileDiffs: 3},
+		{filename: "sample_contains_only_added_deleted_files.diff", wantFileDiffs: 3},
+		{filename: "sample_onlyin_line_isnt_a_file_header.diff", wantFileDiffs: 4},
+		{filename: "sample_onlyin_complex_filenames.diff", wantFileDiffs: 3},
 	}
 	for _, test := range tests {
 		diffData, err := ioutil.ReadFile(filepath.Join("testdata", test.filename))

diff/parse.go

@@ -6,6 +6,7 @@ import (
 	"errors"
 	"fmt"
 	"io"
+	"path/filepath"
 	"strconv"
 	"strings"
 	"time"
@@ -72,6 +73,12 @@ func (r *MultiFileDiffReader) ReadFile() (*FileDiff, error) {
 		}
 	}
 
+	// FileDiff is added/deleted file
+	// No further collection of hunks needed
+	if fd.NewName == "" {
+		return fd, nil
+	}
+
 	// Before reading hunks, check to see if there are any. If there
 	// aren't any, and there's another file after this file in the
 	// diff, then the hunks reader will complain ErrNoHunkHeader. It's
@@ -223,8 +230,16 @@ func (r *FileDiffReader) HunksReader() *HunksReader {
 
 // ReadFileHeaders reads the unified file diff header (the lines that
 // start with "---" and "+++" with the orig/new file names and
-// timestamps).
+// timestamps). Or which starts with "Only in " with dir path and filename.
+// "Only in" message is supported in POSIX locale: https://pubs.opengroup.org/onlinepubs/9699919799/utilities/diff.html#tag_20_34_10
 func (r *FileDiffReader) ReadFileHeaders() (origName, newName string, origTimestamp, newTimestamp *time.Time, err error) {
+	if r.fileHeaderLine != nil {
+		if isOnlyMessage, source, filename := parseOnlyInMessage(r.fileHeaderLine); isOnlyMessage {
+			return filepath.Join(string(source), string(filename)),
+				"", nil, nil, nil
+		}
+	}
+
 	origName, origTimestamp, err = r.readOneFileHeader([]byte("--- "))
 	if err != nil {
 		return "", "", nil, nil, err
@@ -330,6 +345,12 @@ func (r *FileDiffReader) ReadExtendedHeaders() ([]string, error) {
 			return xheaders, nil
 		}
 
+		// Reached message that file is added/deleted
+		if isOnlyInMessage, _, _ := parseOnlyInMessage(line); isOnlyInMessage {
+			r.fileHeaderLine = line // pass to readOneFileHeader (see fileHeaderLine field doc)
+			return xheaders, nil
+		}
+
 		r.line++
 		r.offset += int64(len(line))
 		xheaders = append(xheaders, string(line))
@@ -403,6 +424,10 @@ var (
 
 	// ErrExtendedHeadersEOF is when an EOF was encountered while reading extended file headers, which means that there were no ---/+++ headers encountered before hunks (if any) began.
 	ErrExtendedHeadersEOF = errors.New("expected file header while reading extended headers, got EOF")
+
+	// ErrBadOnlyInMessage is when a file have a malformed `only in` message
+	// Should be in format `Only in {source}: {filename}`
+	ErrBadOnlyInMessage = errors.New("bad 'only in' message")
 )
 
 // ParseHunks parses hunks from a unified diff. The diff must consist
@@ -612,6 +637,19 @@ func (r *HunksReader) ReadAllHunks() ([]*Hunk, error) {
 	}
 }
 
+// parseOnlyInMessage checks if line is a "Only in {source}: {filename}" and returns source and filename
+func parseOnlyInMessage(line []byte) (bool, []byte, []byte) {
+	if !bytes.HasPrefix(line, onlyInMessagePrefix) {
+		return false, nil, nil
+	}
+	line = line[len(onlyInMessagePrefix):]
+	idx := bytes.Index(line, []byte(": "))
+	if idx < 0 {
+		return false, nil, nil
+	}
+	return true, line[:idx], line[idx+2:]
+}
+
 // A ParseError is a description of a unified diff syntax error.
 type ParseError struct {
 	Line   int   // Line where the error occurred

diff/print.go

@@ -4,6 +4,7 @@ import (
 	"bytes"
 	"fmt"
 	"io"
+	"path/filepath"
 	"time"
 
 	"sourcegraph.com/sqs/pbtypes"
@@ -36,6 +37,16 @@ func PrintFileDiff(d *FileDiff) ([]byte, error) {
 		}
 	}
 
+	// FileDiff is added/deleted file
+	// No further hunks printing needed
+	if d.NewName == "" {
+		_, err := fmt.Fprintf(&buf, onlyInMessage, filepath.Dir(d.OrigName), filepath.Base(d.OrigName))
+		if err != nil {
+			return nil, err
+		}
+		return buf.Bytes(), nil
+	}
+
 	if d.Hunks == nil {
 		return buf.Bytes(), nil
 	}

diff/testdata/sample_contains_added_deleted_files.diff

@@ -0,0 +1,11 @@
+diff -u source_a/file_1.txt  source_b/file_1.txt
+--- source_a/file_1.txt
++++ source_b/file_1.txt
+@@ -2,3 +3,4 @@
+ To be, or not to be, that is the question:
+-Whether 'tis nobler in the mind to suffer
++The slings and arrows of outrageous fortune,
++Or to take arms against a sea of troubles
+ And by opposing end them. To die—to sleep,
+Only in source_a: file_2.txt
+Only in source_b: file_3.txt

diff/testdata/sample_contains_only_added_deleted_files.diff

@@ -0,0 +1,3 @@
+Only in source_a: file_1.txt
+Only in source_a: file_2.txt
+Only in source_b: file_3.txt

diff/testdata/sample_onlyin_complex_filenames.diff

@@ -0,0 +1,3 @@
+Only in internal/trace/foo bar: bam
+Only in internal/trace/foo bar: bam: bar
+Only in internal/trace/hello: world: bazz

diff/testdata/sample_onlyin_line_isnt_a_file_header.diff

@@ -0,0 +1,13 @@
+diff -u source_a/file_1.txt  source_b/file_1.txt
+--- source_a/file_1.txt
++++ source_b/file_1.txt
+@@ -2,3 +3,4 @@
+ To be, or not to be, that is the question:
+-Whether 'tis nobler in the mind to suffer
++The slings and arrows of outrageous fortune,
++Or to take arms against a sea of troubles
+ And by opposing end them. To die—to sleep,
+Only in universe!
+Only in source_a: file_2.txt
+Only in source_b: file_3.txt some unrelated stuff here.
+Only in source_b: file_3.txt