From ef9bf34da0c67ab7958b46eddcf203956db844d2 Mon Sep 17 00:00:00 2001
From: Joe Tsai <joetsai@digital-static.net>
Date: Fri, 30 Apr 2021 14:40:46 -0700
Subject: [PATCH] Avoid diffing by lines if inefficient

Avoid diffing by lines if it turns out to be significantly less
efficient than diffing by bytes.

Before this change:
    (
    	"""
  - 	d5c14bdf6bac81c27afc5429500ed750
  - 	25483503b557c606dad4f144d27ae10b
  - 	90bdbcdbb6ea7156068e3dcfb7459244
  - 	978f480a6e3cced51e297fbff9a506b7
  + 	Xd5c14bdf6bac81c27afc5429500ed750
  + 	X25483503b557c606dad4f144d27ae10b
  + 	X90bdbcdbb6ea7156068e3dcfb7459244
  + 	X978f480a6e3cced51e297fbff9a506b7
    	"""
    )

After this change:
    strings.Join({
  + 	"X",
    	"d5c14bdf6bac81c27afc5429500ed750\n",
  + 	"X",
    	"25483503b557c606dad4f144d27ae10b\n",
  + 	"X",
    	"90bdbcdbb6ea7156068e3dcfb7459244\n",
  + 	"X",
    	"978f480a6e3cced51e297fbff9a506b7\n",
    }, "")
---
 cmp/compare_test.go  |  5 +++++
 cmp/report_slices.go | 19 +++++++++++++++++--
 cmp/testdata/diffs   | 12 ++++++++++++
 3 files changed, 34 insertions(+), 2 deletions(-)

diff --git a/cmp/compare_test.go b/cmp/compare_test.go
index 7eac029..baae591 100644
--- a/cmp/compare_test.go
+++ b/cmp/compare_test.go
@@ -1307,6 +1307,11 @@ using the AllowUnexported option.`, "\n"),
 		x:      "org-4747474747474747,bucket-4242424242424242:m,tag1=a,tag2=aa,#=_value _value=2 11\torg-4747474747474747,bucket-4242424242424242:m,tag1=a,tag2=bb,#=_value _value=2 21\torg-4747474747474747,bucket-4242424242424242:m,tag1=b,tag2=cc,#=_value _value=1 21\torg-4747474747474747,bucket-4242424242424242:m,tag1=a,tag2=dd,#=_value _value=3 31\torg-4747474747474747,bucket-4242424242424242:m,tag1=c,#=_value _value=4 41\t",
 		y:      "org-4747474747474747,bucket-4242424242424242:m,tag1=a,tag2=aa _value=2 11\torg-4747474747474747,bucket-4242424242424242:m,tag1=a,tag2=bb _value=2 21\torg-4747474747474747,bucket-4242424242424242:m,tag1=b,tag2=cc _value=1 21\torg-4747474747474747,bucket-4242424242424242:m,tag1=a,tag2=dd _value=3 31\torg-4747474747474747,bucket-4242424242424242:m,tag1=c _value=4 41\t",
 		reason: "leading/trailing equal spans should not appear in diff lines",
+	}, {
+		label:  label + "/AllLinesDiffer",
+		x:      "d5c14bdf6bac81c27afc5429500ed750\n25483503b557c606dad4f144d27ae10b\n90bdbcdbb6ea7156068e3dcfb7459244\n978f480a6e3cced51e297fbff9a506b7\n",
+		y:      "Xd5c14bdf6bac81c27afc5429500ed750\nX25483503b557c606dad4f144d27ae10b\nX90bdbcdbb6ea7156068e3dcfb7459244\nX978f480a6e3cced51e297fbff9a506b7\n",
+		reason: "all lines are different, so diffing based on lines is pointless",
 	}}
 }
 
diff --git a/cmp/report_slices.go b/cmp/report_slices.go
index f985cc9..bd8ca15 100644
--- a/cmp/report_slices.go
+++ b/cmp/report_slices.go
@@ -98,6 +98,7 @@ func (opts formatOptions) FormatDiffSlice(v *valueNode) textNode {
 	// Auto-detect the type of the data.
 	var isLinedText, isText, isBinary bool
 	var sx, sy string
+	var ssx, ssy []string
 	switch {
 	case t.Kind() == reflect.String:
 		sx, sy = vx.String(), vy.String()
@@ -130,6 +131,22 @@ func (opts formatOptions) FormatDiffSlice(v *valueNode) textNode {
 		}
 		isText = !isBinary
 		isLinedText = isText && numLines >= 4 && maxLineLen <= 1024
+
+		// Avoid diffing by lines if it produces a significantly more complex
+		// edit script than diffing by bytes.
+		if isLinedText {
+			ssx = strings.Split(sx, "\n")
+			ssy = strings.Split(sy, "\n")
+			esLines := diff.Difference(len(ssx), len(ssy), func(ix, iy int) diff.Result {
+				return diff.BoolResult(ssx[ix] == ssy[iy])
+			})
+			esBytes := diff.Difference(len(sx), len(sy), func(ix, iy int) diff.Result {
+				return diff.BoolResult(sx[ix] == sy[iy])
+			})
+			efficiencyLines := float64(esLines.Dist()) / float64(len(esLines))
+			efficiencyBytes := float64(esBytes.Dist()) / float64(len(esBytes))
+			isLinedText = efficiencyLines < 4*efficiencyBytes
+		}
 	}
 
 	// Format the string into printable records.
@@ -139,8 +156,6 @@ func (opts formatOptions) FormatDiffSlice(v *valueNode) textNode {
 	// If the text appears to be multi-lined text,
 	// then perform differencing across individual lines.
 	case isLinedText:
-		ssx := strings.Split(sx, "\n")
-		ssy := strings.Split(sy, "\n")
 		list = opts.formatDiffSlice(
 			reflect.ValueOf(ssx), reflect.ValueOf(ssy), 1, "line",
 			func(v reflect.Value, d diffMode) textRecord {
diff --git a/cmp/testdata/diffs b/cmp/testdata/diffs
index 2d9f9ab..c02df29 100644
--- a/cmp/testdata/diffs
+++ b/cmp/testdata/diffs
@@ -1065,6 +1065,18 @@
   	` _value=4 41	`,
   }, "")
 >>> TestDiff/Reporter/SurroundingEqualElements
+<<< TestDiff/Reporter/AllLinesDiffer
+  strings.Join({
++ 	"X",
+  	"d5c14bdf6bac81c27afc5429500ed750\n",
++ 	"X",
+  	"25483503b557c606dad4f144d27ae10b\n",
++ 	"X",
+  	"90bdbcdbb6ea7156068e3dcfb7459244\n",
++ 	"X",
+  	"978f480a6e3cced51e297fbff9a506b7\n",
+  }, "")
+>>> TestDiff/Reporter/AllLinesDiffer
 <<< TestDiff/EmbeddedStruct/ParentStructA/Inequal
   teststructs.ParentStructA{
   	privateStruct: teststructs.privateStruct{