prometheus/pkg/textparse/promlex.l

%{
// Copyright 2017 The Prometheus Authors
// 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 textparse

import (
    "fmt"
)

const (
    sInit = iota
    sComment
    sMeta1
    sMeta2
    sLabels
    sLValue
    sValue
    sTimestamp
)

// Lex is called by the parser generated by "go tool yacc" to obtain each
// token. The method is opened before the matching rules block and closed at
// the end of the file.
func (l *promlexer) Lex() token {
    if l.i >= len(l.b) {
        return tEOF
    }
    c := l.b[l.i]
    l.start = l.i

%}

D     [0-9]
L     [a-zA-Z_]
M     [a-zA-Z_:]
C     [^\n]

%x sComment sMeta1 sMeta2 sLabels sLValue sValue sTimestamp

%yyc c
%yyn c = l.next()
%yyt l.state


%%

\0                                    return tEOF
\n                                    l.state = sInit; return tLinebreak
<*>[ \t]+                             return tWhitespace

#[ \t]+                               l.state = sComment
#                                     return l.consumeComment()
<sComment>HELP[\t ]+                  l.state = sMeta1; return tHelp
<sComment>TYPE[\t ]+                  l.state = sMeta1; return tType
<sMeta1>{M}({M}|{D})*                 l.state = sMeta2; return tMName
<sMeta2>{C}*                          l.state = sInit; return tText

{M}({M}|{D})*                         l.state = sValue; return tMName
<sValue>\{                            l.state = sLabels; return tBraceOpen
<sLabels>{L}({L}|{D})*                return tLName
<sLabels>\}                           l.state = sValue; return tBraceClose
<sLabels>=                            l.state = sLValue; return tEqual
<sLabels>,                            return tComma
<sLValue>\"(\\.|[^\\"])*\"            l.state = sLabels; return tLValue
<sValue>[^{ \t\n]+                    l.state = sTimestamp; return tValue
<sTimestamp>{D}+                      return tTimestamp
<sTimestamp>\n                        l.state = sInit; return tLinebreak

%%
    // Workaround to gobble up comments that started with a HELP or TYPE
    // prefix. We just consume all characters until we reach a newline.
    // This saves us from adding disproportionate complexity to the parser.
    if l.state == sComment {
        return l.consumeComment()
    }
    return tInvalid
}

func (l *promlexer) consumeComment() token {
    for c := l.cur(); ; c = l.next() {
        switch c {
        case 0:
            return tEOF
        case '\n':
            l.state = sInit
            return tComment
        }
    }
}