use protobufs

Author: sqs

diff/diff.go

@@ -1,27 +1,12 @@
 package diff
 
-import (
-	"bytes"
-	"time"
-)
-
-// A FileDiff represents a unified diff for a single file.
-//
-// A file unified diff has a header that resembles the following:
-//
-//  --- oldname	2009-10-11 15:12:20.000000000 -0700
-//  +++ newname	2009-10-11 15:12:30.000000000 -0700
-type FileDiff struct {
-	OrigName string     // the original name of the file
-	OrigTime *time.Time // the original timestamp (nil if not present)
-
-	NewName string     // the new name of the file (often same as OrigName)
-	NewTime *time.Time // the new timestamp (nil if not present)
+import "bytes"
 
-	Extended []string // extended header lines (e.g., git's "new mode <mode>", "rename from <path>", etc.)
+// NOTE: types are code-generated in diff.pb.go.
 
-	Hunks []*Hunk // hunks that were changed from orig to new
-}
+//go:generate protoc -I../../../../github.com/gogo/protobuf/protobuf -I../../../../github.com/gogo/protobuf -I../../../../sourcegraph.com/sqs/pbtypes -I. --gogo_out=. diff.proto
+//go:generate sed -i "s#timestamp\\.pb#sourcegraph.com/sqs/pbtypes#g" diff.pb.go
+//go:generate sed -i "s#vcs\\.pb#sourcegraph.com/sourcegraph/go-vcs/vcs#g" diff.pb.go
 
 // Stat computes the number of lines added/changed/deleted in all
 // hunks in this file's diff.
@@ -33,26 +18,6 @@ func (d *FileDiff) Stat() Stat {
 	return total
 }
 
-// A Hunk represents a series of changes (additions or deletions) in a
-// file's unified diff.
-type Hunk struct {
-	OrigStartLine   int // starting line number in original file
-	OrigLines       int // number of lines the hunk applies to in the original file
-	OrigNoNewlineAt int // if > 0, then the original file had a 'No newline at end of file' mark at this offset
-
-	NewStartLine int // starting line number in new file
-	NewLines     int // number of lines the hunk applies to in the new file
-
-	Section string // optional section heading
-
-	// 0-indexed line offset in unified file diff (including section headers);
-	// this is only set when Hunks are read from entire file diff (i.e., when ReadAllHunks is called)
-	// This accounts for hunk headers, too, so the StartPosition of the first hunk will be 1.
-	StartPosition int
-
-	Body []byte // hunk body (lines prefixed with '-', '+', or ' ')
-}
-
 // Stat computes the number of lines added/changed/deleted in this
 // hunk.
 func (h *Hunk) Stat() Stat {
@@ -101,12 +66,6 @@ const hunkHeader = "@@ -%d,%d +%d,%d @@"
 // http://www.gnu.org/software/diffutils/manual/html_node/Detailed-Unified.html.
 const diffTimeFormat = "2006-01-02 15:04:05.000000000 -0700"
 
-// A Stat is a diff stat that represents the number of lines
-// added/changed/deleted.
-type Stat struct {
-	Added, Changed, Deleted int // numbers of lines
-}
-
 func (s *Stat) add(o Stat) {
 	s.Added += o.Added
 	s.Changed += o.Changed

diff/diff.pb.go

@@ -0,0 +1,77 @@
+syntax = "proto3";
+package diff;
+
+import "gogoproto/gogo.proto";
+import "timestamp.proto";
+
+
+// A FileDiff represents a unified diff for a single file.
+// 
+// A file unified diff has a header that resembles the following:
+// 
+//   --- oldname	2009-10-11 15:12:20.000000000 -0700
+//   +++ newname	2009-10-11 15:12:30.000000000 -0700
+message FileDiff {
+	// the original name of the file
+	string orig_name = 1;
+
+	// the original timestamp (nil if not present)
+	pbtypes.Timestamp orig_time = 2;
+
+	// the new name of the file (often same as OrigName)
+	string new_name = 3;
+
+	// the new timestamp (nil if not present)
+	pbtypes.Timestamp new_time = 4;
+
+	// extended header lines (e.g., git's "new mode <mode>", "rename from <path>", etc.)
+	repeated string extended = 5;
+
+	// hunks that were changed from orig to new
+	repeated Hunk hunks = 6;
+}
+
+
+// A Hunk represents a series of changes (additions or deletions) in a file's
+// unified diff.
+message Hunk {
+	// starting line number in original file
+	int32 orig_start_line = 1 [(gogoproto.customtype) = "int"];
+
+	// number of lines the hunk applies to in the original file
+	int32 orig_lines = 2 [(gogoproto.customtype) = "int"];
+
+	// if > 0, then the original file had a 'No newline at end of file' mark at this offset
+	int32 orig_no_newline_at = 3 [(gogoproto.customtype) = "int"];
+
+	// starting line number in new file
+	int32 new_start_line = 4 [(gogoproto.customtype) = "int"];
+
+	// number of lines the hunk applies to in the new file
+	int32 new_lines = 5 [(gogoproto.customtype) = "int"];
+
+	// optional section heading
+	string section = 6;
+
+	// 0-indexed line offset in unified file diff (including section headers); this is
+	// only set when Hunks are read from entire file diff (i.e., when ReadAllHunks is
+	// called) This accounts for hunk headers, too, so the StartPosition of the first
+	// hunk will be 1.
+	int32 start_position = 7 [(gogoproto.customtype) = "int"];
+
+	// hunk body (lines prefixed with '-', '+', or ' ')
+	bytes body = 8;
+}
+
+// A Stat is a diff stat that represents the number of lines added/changed/deleted.
+message Stat {
+	// number of lines added
+	int32 added = 1 [(gogoproto.customtype) = "int"];
+
+	// number of lines changed
+	int32 changed = 2 [(gogoproto.customtype) = "int"];
+
+	// number of lines deleted
+	int32 deleted = 3 [(gogoproto.customtype) = "int"];
+}
+

diff/diff.proto

@@ -8,6 +8,8 @@ import (
 	"io"
 	"strings"
 	"time"
+
+	"sourcegraph.com/sqs/pbtypes"
 )
 
 // ParseMultiFileDiff parses a multi-file unified diff. It returns an error if parsing failed as a whole, but does its
@@ -172,10 +174,19 @@ func (r *FileDiffReader) ReadAllHeaders() (*FileDiff, error) {
 		return nil, err
 	}
 
-	fd.OrigName, fd.NewName, fd.OrigTime, fd.NewTime, err = r.ReadFileHeaders()
+	var origTime, newTime *time.Time
+	fd.OrigName, fd.NewName, origTime, newTime, err = r.ReadFileHeaders()
 	if err != nil {
 		return nil, err
 	}
+	if origTime != nil {
+		ts := pbtypes.NewTimestamp(*origTime)
+		fd.OrigTime = &ts
+	}
+	if newTime != nil {
+		ts := pbtypes.NewTimestamp(*newTime)
+		fd.NewTime = &ts
+	}
 
 	return fd, nil
 }

diff/parse.go

@@ -5,6 +5,8 @@ import (
 	"fmt"
 	"io"
 	"time"
+
+	"sourcegraph.com/sqs/pbtypes"
 )
 
 // PrintMultiFileDiff prints a multi-file diff in unified diff format.
@@ -34,10 +36,10 @@ func PrintFileDiff(d *FileDiff) ([]byte, error) {
 		}
 	}
 
-	if err := printFileHeader(&buf, "--- ", d.OrigName, d.OrigTime); err != nil {
+	if err := printFileHeader(&buf, "--- ", d.OrigName, timePtr(d.OrigTime)); err != nil {
 		return nil, err
 	}
-	if err := printFileHeader(&buf, "+++ ", d.NewName, d.NewTime); err != nil {
+	if err := printFileHeader(&buf, "+++ ", d.NewName, timePtr(d.NewTime)); err != nil {
 		return nil, err
 	}
 
@@ -52,6 +54,14 @@ func PrintFileDiff(d *FileDiff) ([]byte, error) {
 	return buf.Bytes(), nil
 }
 
+func timePtr(ts *pbtypes.Timestamp) *time.Time {
+	if ts == nil {
+		return nil
+	}
+	t := ts.Time()
+	return &t
+}
+
 func printFileHeader(w io.Writer, prefix string, filename string, timestamp *time.Time) error {
 	if _, err := fmt.Fprint(w, prefix, filename); err != nil {
 		return err