At this stage this is just a comment for users. The directive will have
effect at go tip when go CL125580044 is submitted. When that happens, at
tip parser code generation can be acheived in the default case with:
go generate
in the relevant package directory.
138 lines
3 KiB
Go
138 lines
3 KiB
Go
// Copyright 2014 The Cayley 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.
|
|
|
|
//go:generate ragel -Z -G2 parse.rl
|
|
|
|
// Package cquads implements parsing N-Quads like line-based syntax
|
|
// for RDF datasets.
|
|
//
|
|
// N-Quad parsing is performed as based on a simplified grammar derived from
|
|
// the N-Quads grammar defined by http://www.w3.org/TR/n-quads/.
|
|
//
|
|
// For a complete definition of the grammar, see cquads.rl.
|
|
package cquads
|
|
|
|
import (
|
|
"bufio"
|
|
"bytes"
|
|
"fmt"
|
|
"io"
|
|
"strconv"
|
|
|
|
"github.com/google/cayley/quad"
|
|
)
|
|
|
|
// Decoder implements simplified N-Quad document parsing.
|
|
type Decoder struct {
|
|
r *bufio.Reader
|
|
line []byte
|
|
}
|
|
|
|
// NewDecoder returns an N-Quad decoder that takes its input from the
|
|
// provided io.Reader.
|
|
func NewDecoder(r io.Reader) *Decoder {
|
|
return &Decoder{r: bufio.NewReader(r)}
|
|
}
|
|
|
|
// Unmarshal returns the next valid N-Quad as a quad.Quad, or an error.
|
|
func (dec *Decoder) Unmarshal() (quad.Quad, error) {
|
|
dec.line = dec.line[:0]
|
|
var line []byte
|
|
for {
|
|
for {
|
|
l, pre, err := dec.r.ReadLine()
|
|
if err != nil {
|
|
return quad.Quad{}, err
|
|
}
|
|
dec.line = append(dec.line, l...)
|
|
if !pre {
|
|
break
|
|
}
|
|
}
|
|
if line = bytes.TrimSpace(dec.line); len(line) != 0 && line[0] != '#' {
|
|
break
|
|
}
|
|
dec.line = dec.line[:0]
|
|
}
|
|
triple, err := Parse(string(line))
|
|
if err != nil {
|
|
return quad.Quad{}, fmt.Errorf("failed to parse %q: %v", dec.line, err)
|
|
}
|
|
if !triple.IsValid() {
|
|
return dec.Unmarshal()
|
|
}
|
|
return triple, nil
|
|
}
|
|
|
|
func unEscape(r []rune, isQuoted, isEscaped bool) string {
|
|
if isQuoted {
|
|
r = r[1 : len(r)-1]
|
|
}
|
|
if len(r) >= 2 && r[0] == '<' && r[len(r)-1] == '>' {
|
|
return string(r[1 : len(r)-1])
|
|
}
|
|
if !isEscaped {
|
|
return string(r)
|
|
}
|
|
|
|
buf := bytes.NewBuffer(make([]byte, 0, len(r)))
|
|
|
|
for i := 0; i < len(r); {
|
|
switch r[i] {
|
|
case '\\':
|
|
i++
|
|
var c byte
|
|
switch r[i] {
|
|
case 't':
|
|
c = '\t'
|
|
case 'b':
|
|
c = '\b'
|
|
case 'n':
|
|
c = '\n'
|
|
case 'r':
|
|
c = '\r'
|
|
case 'f':
|
|
c = '\f'
|
|
case '"':
|
|
c = '"'
|
|
case '\'':
|
|
c = '\''
|
|
case '\\':
|
|
c = '\\'
|
|
case 'u':
|
|
rc, err := strconv.ParseInt(string(r[i+1:i+5]), 16, 32)
|
|
if err != nil {
|
|
panic(fmt.Errorf("internal parser error: %v", err))
|
|
}
|
|
buf.WriteRune(rune(rc))
|
|
i += 5
|
|
continue
|
|
case 'U':
|
|
rc, err := strconv.ParseInt(string(r[i+1:i+9]), 16, 32)
|
|
if err != nil {
|
|
panic(fmt.Errorf("internal parser error: %v", err))
|
|
}
|
|
buf.WriteRune(rune(rc))
|
|
i += 9
|
|
continue
|
|
}
|
|
buf.WriteByte(c)
|
|
default:
|
|
buf.WriteRune(r[i])
|
|
}
|
|
i++
|
|
}
|
|
|
|
return buf.String()
|
|
}
|