From 2a112a0004f28afcac2c0bb5f8be76703adb9727 Mon Sep 17 00:00:00 2001
From: Daniel Smith <dbsmith@google.com>
Date: Thu, 16 Jul 2015 18:29:38 -0700
Subject: [PATCH] preformat checker

---
 cmd/mungedocs/mungedocs.go    |  1 +
 cmd/mungedocs/preformatted.go | 57 +++++++++++++++++++++++++++++++++++
 cmd/mungedocs/util.go         | 53 +++++++++++++++++++++++---------
 3 files changed, 97 insertions(+), 14 deletions(-)
 create mode 100644 cmd/mungedocs/preformatted.go

diff --git a/cmd/mungedocs/mungedocs.go b/cmd/mungedocs/mungedocs.go
index 50b3aa5240f..f6db77c6c94 100644
--- a/cmd/mungedocs/mungedocs.go
+++ b/cmd/mungedocs/mungedocs.go
@@ -47,6 +47,7 @@ Examples:
 	allMunges = []munge{
 		{"table-of-contents", updateTOC},
 		{"check-links", checkLinks},
+		{"blank-lines-surround-preformatted", checkPreformatted},
 		{"unversioned-warning", updateUnversionedWarning},
 		{"analytics", checkAnalytics},
 		{"kubectl-dash-f", checkKubectlFileTargets},
diff --git a/cmd/mungedocs/preformatted.go b/cmd/mungedocs/preformatted.go
new file mode 100644
index 00000000000..bd42f71a5bd
--- /dev/null
+++ b/cmd/mungedocs/preformatted.go
@@ -0,0 +1,57 @@
+/*
+Copyright 2015 The Kubernetes Authors All rights reserved.
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License.
+*/
+
+package main
+
+import (
+	"bytes"
+)
+
+// Blocks of ``` need to have blank lines on both sides or they don't look
+// right in HTML.
+func checkPreformatted(filePath string, fileBytes []byte) ([]byte, error) {
+	f := splitByPreformatted(fileBytes)
+	f = append(fileBlocks{{false, []byte{}}}, f...)
+	f = append(f, fileBlock{false, []byte{}})
+
+	output := []byte(nil)
+	for i := 1; i < len(f)-1; i++ {
+		prev := &f[i-1]
+		block := &f[i]
+		next := &f[i+1]
+		if !block.preformatted {
+			continue
+		}
+		neededSuffix := []byte("\n\n")
+		for !bytes.HasSuffix(prev.data, neededSuffix) {
+			prev.data = append(prev.data, '\n')
+		}
+		for !bytes.HasSuffix(block.data, neededSuffix) {
+			block.data = append(block.data, '\n')
+			if bytes.HasPrefix(next.data, []byte("\n")) {
+				// don't change the number of newlines unless needed.
+				next.data = next.data[1:]
+				if len(next.data) == 0 {
+					f = append(f[:i+1], f[i+2:]...)
+				}
+			}
+		}
+	}
+	for _, block := range f {
+		output = append(output, block.data...)
+	}
+	return output, nil
+}
diff --git a/cmd/mungedocs/util.go b/cmd/mungedocs/util.go
index 028e427aff9..1a756dec0a9 100644
--- a/cmd/mungedocs/util.go
+++ b/cmd/mungedocs/util.go
@@ -25,7 +25,9 @@ import (
 
 var (
 	// Finds all preformatted block start/stops.
-	preformatRE = regexp.MustCompile("^```.*")
+	preformatRE    = regexp.MustCompile("^[\\s]*```.*")
+	notPreformatRE = regexp.MustCompile("^[\\s]*```.*```.*")
+	preformatEndRE = regexp.MustCompile(".*```.*")
 )
 
 // Splits a document up into a slice of lines.
@@ -120,35 +122,58 @@ func endMungeTag(desc string) string {
 // Calls 'replace' for all sections of the document not in ``` / ``` blocks. So
 // that you don't have false positives inside those blocks.
 func replaceNonPreformatted(input []byte, replace func([]byte) []byte) []byte {
+	f := splitByPreformatted(input)
 	output := []byte(nil)
+	for _, block := range f {
+		if block.preformatted {
+			output = append(output, block.data...)
+		} else {
+			output = append(output, replace(block.data)...)
+		}
+	}
+	return output
+}
+
+type fileBlock struct {
+	preformatted bool
+	data         []byte
+}
+
+type fileBlocks []fileBlock
+
+func splitByPreformatted(input []byte) fileBlocks {
+	f := fileBlocks{}
+
 	cur := []byte(nil)
-	keepBlock := true
+	preformatted := false
 	// SplitAfter keeps the newline, so you don't have to worry about
 	// omitting it on the last line or anything. Also, the documentation
 	// claims it's unicode safe.
 	for _, line := range bytes.SplitAfter(input, []byte("\n")) {
-		if keepBlock {
-			if preformatRE.Match(line) {
-				cur = replace(cur)
-				output = append(output, cur...)
+		if !preformatted {
+			if preformatRE.Match(line) && !notPreformatRE.Match(line) {
+				if len(cur) > 0 {
+					f = append(f, fileBlock{false, cur})
+				}
 				cur = []byte{}
-				keepBlock = false
+				preformatted = true
 			}
 			cur = append(cur, line...)
 		} else {
 			cur = append(cur, line...)
-			if preformatRE.Match(line) {
-				output = append(output, cur...)
+			if preformatEndRE.Match(line) {
+				if len(cur) > 0 {
+					f = append(f, fileBlock{true, cur})
+				}
 				cur = []byte{}
-				keepBlock = true
+				preformatted = false
 			}
 		}
 	}
-	if keepBlock {
-		cur = replace(cur)
+	if len(cur) > 0 {
+		f = append(f, fileBlock{preformatted, cur})
 	}
-	output = append(output, cur...)
-	return output
+	return f
 }
 
 // As above, but further uses exp to parse the non-preformatted sections.