structured logging: support key/value pairs with line breaks

The initial structured logging output (almost) always produced a single line
per log message, with quoting of strings to represent line breaks as \n.  This
made the output hard to read (see
kubernetes/kubernetes#104868).

It was still possible to get line breaks when formatting a value with `%+v` and
that ended up emitting line breaks; this was probably not intended.

Now string values are only quoted if they contain no line break. If they do,
start/end markers delimit the text which appears on its own lines, with
indention to ensure that those additional lines are not accidentally treated as
a new log message when they happen to contain the klog header. It also makes
the output more readable.

Tabs are used as indention for the value because it makes it stand out visually
and because it ensures that inline ">" characters are not mistaken for the
end-of-value delimiter.

The result of `fmt.Sprintf("%+v")` is printed verbatim without quoting when it
contains no line break, otherwise as multi-line value with delimiter and
indention.

Traditional output:

 I1112 14:06:35.783354  328441 structured_logging.go:42] someData printed using InfoF: {hello world 0}
 I1112 14:06:35.783472  328441 structured_logging.go:43] longData printed using InfoF: {long Multiple
 lines
 with quite a bit
 of text. 0}
 I1112 14:06:35.783483  328441 structured_logging.go:44] stringData printed using InfoF,
 with the message across multiple lines:
 long: Multiple
 lines
 with quite a bit
 of text.
 I1112 14:06:35.898176  142908 structured_logging.go:54] logData printed using InfoF:
 {log output from some program I0000 12:00:00.000000  123456 main.go:42] Starting
 E0000 12:00:01.000000  123456 main.go:43] Failed for some reason
  0}

Old InfoS output before this commit:

 I1112 14:06:35.783512  328441 structured_logging.go:50] "using InfoS" someData={Name:hello Data:world internal:0}
 I1112 14:06:35.783529  328441 structured_logging.go:51] "using InfoS" longData={Name:long Data:Multiple
 lines
 with quite a bit
 of text. internal:0}
 I1112 14:06:35.783549  328441 structured_logging.go:52] "using InfoS with\nthe message across multiple lines" int=1 stringData="long: Multiple\nlines\nwith quite a bit\nof text." str="another value"
 I1112 14:06:35.783565  328441 structured_logging.go:61] "Did something" item="foobar"
 I1112 14:06:35.783576  328441 structured_logging.go:63] "This is a full sentence." item="foobar"
 I1112 14:06:35.898278  142908 structured_logging.go:65] "using InfoS" logData={Name:log output from some program Data:I0000 12:00:00.000000  123456 main.go:42] Starting
 E0000 12:00:01.000000  123456 main.go:43] Failed for some reason
  internal:0}

New InfoS output:
I1126 10:31:50.378182  121736 structured_logging.go:58] "using InfoS" someData={Name:hello Data:world internal:0}
I1126 10:31:50.378204  121736 structured_logging.go:59] "using InfoS" longData=<
	{Name:long Data:Multiple
	lines
	with quite a bit
	of text. internal:0}
 >
I1126 10:31:50.378228  121736 structured_logging.go:60] "using InfoS with\nthe message across multiple lines" int=1 stringData=<
	long: Multiple
	lines
	with quite a bit
	of text.
 > str="another value"
I1126 10:31:50.378249  121736 structured_logging.go:65] "using InfoS" logData=<
	{Name:log output from some program Data:I0000 12:00:00.000000  123456 main.go:42] Starting
	E0000 12:00:01.000000  123456 main.go:43] Failed for some reason
	 internal:0}
 >

Performance is the same as before in most cases. Handling of a v1.Container
struct with line breaks in the output gets faster, probably because printing
each line individually is more efficient than quoting.

$ $GOPATH/bin/benchstat /tmp/reorder /tmp/multi-line
name                               old time/op    new time/op    delta
Logging/container/structured-36      39.6µs ± 1%    21.5µs ± 0%  -45.82%  (p=0.008 n=5+5)
Logging/error-value/structured-36    3.14µs ± 4%    3.11µs ± 2%     ~     (p=0.548 n=5+5)
Logging/error/structured-36          3.13µs ± 3%    3.13µs ± 3%     ~     (p=1.000 n=5+5)
Logging/simple/structured-36         2.91µs ± 2%    2.87µs ± 2%     ~     (p=0.310 n=5+5)
Logging/values/structured-36         4.77µs ± 5%    4.75µs ± 4%     ~     (p=1.000 n=5+5)

name                               old alloc/op   new alloc/op   delta
Logging/container/structured-36      9.35kB ± 0%    9.59kB ± 0%   +2.59%  (p=0.016 n=4+5)
Logging/error-value/structured-36      312B ± 0%      312B ± 0%     ~     (all equal)
Logging/error/structured-36            312B ± 0%      312B ± 0%     ~     (all equal)
Logging/simple/structured-36           288B ± 0%      288B ± 0%     ~     (all equal)
Logging/values/structured-36           464B ± 0%      464B ± 0%     ~     (all equal)

name                               old allocs/op  new allocs/op  delta
Logging/container/structured-36        62.0 ± 0%      63.0 ± 0%   +1.61%  (p=0.008 n=5+5)
Logging/error-value/structured-36      6.00 ± 0%      6.00 ± 0%     ~     (all equal)
Logging/error/structured-36            6.00 ± 0%      6.00 ± 0%     ~     (all equal)
Logging/simple/structured-36           5.00 ± 0%      5.00 ± 0%     ~     (all equal)
Logging/values/structured-36           11.0 ± 0%      11.0 ± 0%     ~     (all equal)

klog.go

@@ -803,10 +803,12 @@ func (l *loggingT) infoS(logger *logr.Logger, filter LogFilter, depth int, msg s
 func (l *loggingT) printS(err error, s severity, depth int, msg string, keysAndValues ...interface{}) {
  // Only create a new buffer if we don't have one cached.
  b := l.getBuffer()
+ // The message is always quoted, even if it contains line breaks.
+ // If developers want multi-line output, they should use a small, fixed
+ // message and put the multi-line output into a value.
  b.WriteString(strconv.Quote(msg))
  if err != nil {
- b.WriteString(" err=")
- b.WriteString(strconv.Quote(err.Error()))
+ kvListFormat(&b.Buffer, "err", err)
  }
  kvListFormat(&b.Buffer, keysAndValues...)
  l.printDepth(s, logging.logr, nil, depth+1, &b.Buffer)
@@ -826,14 +828,17 @@ func kvListFormat(b *bytes.Buffer, keysAndValues ...interface{}) {
  v = missingValue
  }
  b.WriteByte(' ')
+ // Keys are assumed to be well-formed according to
+ // https://github.com/kubernetes/community/blob/master/contributors/devel/sig-instrumentation/migration-to-structured-logging.md#name-arguments
+ // for the sake of performance. Keys with spaces,
+ // special characters, etc. will break parsing.
  if k, ok := k.(string); ok {
  // Avoid one allocation when the key is a string, which
  // normally it should be.
  b.WriteString(k)
  } else {
  b.WriteString(fmt.Sprintf("%s", k))
  }
- b.WriteByte('=')
 
  // The type checks are sorted so that more frequently used ones
  // come first because that is then faster in the common
@@ -842,19 +847,77 @@ func kvListFormat(b *bytes.Buffer, keysAndValues ...interface{}) {
  // (https://github.com/kubernetes/kubernetes/pull/106594#issuecomment-975526235).
  switch v := v.(type) {
  case fmt.Stringer:
- b.WriteString(strconv.Quote(v.String()))
+ writeStringValue(b, true, v.String())
  case string:
- b.WriteString(strconv.Quote(v))
+ writeStringValue(b, true, v)
  case error:
- b.WriteString(strconv.Quote(v.Error()))
+ writeStringValue(b, true, v.Error())
  case []byte:
- // We cannot use the simpler strconv.Quote here
- // because it does not escape unicode characters, which is
- // expected by one test!?
+ // In https://github.com/kubernetes/klog/pull/237 it was decided
+ // to format byte slices with "%+q". The advantages of that are:
+ // - readable output if the bytes happen to be printable
+ // - non-printable bytes get represented as unicode escape
+ // sequences (\uxxxx)
+ //
+ // The downsides are that we cannot use the faster
+ // strconv.Quote here and that multi-line output is not
+ // supported. If developers know that a byte array is
+ // printable and they want multi-line output, they can
+ // convert the value to string before logging it.
+ b.WriteByte('=')
  b.WriteString(fmt.Sprintf("%+q", v))
  default:
- b.WriteString(fmt.Sprintf("%+v", v))
+ writeStringValue(b, false, fmt.Sprintf("%+v", v))
+ }
+ }
+}
+
+func writeStringValue(b *bytes.Buffer, quote bool, v string) {
+ data := []byte(v)
+ index := bytes.IndexByte(data, '\n')
+ if index == -1 {
+ b.WriteByte('=')
+ if quote {
+ // Simple string, quote quotation marks and non-printable characters.
+ b.WriteString(strconv.Quote(v))
+ return
  }
+ // Non-string with no line breaks.
+ b.WriteString(v)
+ return
+ }
+
+ // Complex multi-line string, show as-is with indention like this:
+ // I... "hello world" key=<
+ // <tab>line 1
+ // <tab>line 2
+ // >
+ //
+ // Tabs indent the lines of the value while the end of string delimiter
+ // is indented with a space. That has two purposes:
+ // - visual difference between the two for a human reader because indention
+ // will be different
+ // - no ambiguity when some value line starts with the end delimiter
+ //
+ // One downside is that the output cannot distinguish between strings that
+ // end with a line break and those that don't because the end delimiter
+ // will always be on the next line.
+ b.WriteString("=<\n")
+ for index != -1 {
+ b.WriteByte('\t')
+ b.Write(data[0 : index+1])
+ data = data[index+1:]
+ index = bytes.IndexByte(data, '\n')
+ }
+ if len(data) == 0 {
+ // String ended with line break, don't add another.
+ b.WriteString(" >")
+ } else {
+ // No line break at end of last line, write rest of string and
+ // add one.
+ b.WriteByte('\t')
+ b.Write(data)
+ b.WriteString("\n >")
  }
 }
 

klog_test.go

@@ -943,6 +943,12 @@ func TestVInfoS(t *testing.T) {
  return time.Date(2006, 1, 2, 15, 4, 5, .067890e9, time.Local)
  }
  pid = 1234
+ myData := struct {
+ Data string
+ }{
+ Data: `This is some long text
+with a line break.`,
+ }
  var testDataInfo = []struct {
  msg string
  format string
@@ -963,6 +969,26 @@ func TestVInfoS(t *testing.T) {
  format: "I0102 15:04:05.067890 1234 klog_test.go:%d] \"test\" err=\"test error\"\n",
  keysValues: []interface{}{"err", errors.New("test error")},
  },
+ {
+ msg: `first message line
+second message line`,
+ format: `I0102 15:04:05.067890 1234 klog_test.go:%d] "first message line\nsecond message line" multiLine=<
+ first value line
+ second value line
+ >
+`,
+ keysValues: []interface{}{"multiLine", `first value line
+second value line`},
+ },
+ {
+ msg: `message`,
+ format: `I0102 15:04:05.067890 1234 klog_test.go:%d] "message" myData=<
+ {Data:This is some long text
+ with a line break.}
+ >
+`,
+ keysValues: []interface{}{"myData", myData},
+ },
  }
 
  logging.verbosity.Set("2")
@@ -987,7 +1013,7 @@ func TestVInfoS(t *testing.T) {
  want = ""
  }
  if contents(infoLog) != want {
- t.Errorf("V(%d).InfoS has unexpected output: \n got:\t%s\nwant:\t%s", l, contents(infoLog), want)
+ t.Errorf("V(%d).InfoS has unexpected output:\ngot:\n%s\nwant:\n%s\n", l, contents(infoLog), want)
  }
  }
  }
@@ -1031,7 +1057,7 @@ func TestErrorS(t *testing.T) {
  }
  want := fmt.Sprintf(e.format, line)
  if contents(errorLog) != want {
- t.Errorf("ErrorS has wrong format: \n got:\t%s\nwant:\t%s", contents(errorLog), want)
+ t.Errorf("ErrorS has wrong format:\ngot:\n%s\nwant:\n%s\n", contents(errorLog), want)
  }
  }
  }
@@ -1075,6 +1101,20 @@ func TestKvListFormat(t *testing.T) {
  keysValues: []interface{}{"pod", "kubedns", "bytes", []byte("��=� ⌘")},
  want: " pod=\"kubedns\" bytes=\"\\ufffd\\ufffd=\\ufffd \\u2318\"",
  },
+ {
+ keysValues: []interface{}{"multiLineString", `Hello world!
+ Starts with tab.
+ Starts with spaces.
+No whitespace.`,
+ "pod", "kubedns",
+ },
+ want: ` multiLineString=<
+ Hello world!
+ Starts with tab.
+ Starts with spaces.
+ No whitespace.
+ > pod="kubedns"`,
+ },
  {
  keysValues: []interface{}{"pod", "kubedns", "maps", map[string]int{"three": 4}},
  want: " pod=\"kubedns\" maps=map[three:4]",
@@ -1113,7 +1153,7 @@ func TestKvListFormat(t *testing.T) {
  b := &bytes.Buffer{}
  kvListFormat(b, d.keysValues...)
  if b.String() != d.want {
- t.Errorf("kvlist format error:\n got:\n\t%s\nwant:\t%s", b.String(), d.want)
+ t.Errorf("kvlist format error:\ngot:\n%s\nwant:\n%s\n", b.String(), d.want)
  }
  }
 }