textscan

package
v0.6.13 Latest Latest
Warning

This package is not in the latest version of its module.

 Go to latest Published: Oct 13, 2023 License: MIT Imports: 10 Imported by: 4

README

TextScan

Package textscan implements text scanner for quickly parse text contents. Can use for parse like INI, Properties format contents.

Install

go get github.com/gookit/goutil/strutil/textscan

Examples

package main

import (
	"fmt"

	"github.com/gookit/goutil/dump"
	"github.com/gookit/goutil/strutil/textscan"
	"github.com/gookit/goutil/testutil/assert"
)

func main() {
	ts := textscan.TextScanner{}
	ts.AddMatchers(
		&textscan.CommentsMatcher{},
		&textscan.KeyValueMatcher{},
	)

	ts.SetInput(`
# comments 1
name = inhere

// comments 2
age = 28

/*
multi line
comments 3
*/
desc = '''
a multi
line string
'''
`)

	data := make(map[string]string)
	err := ts.Each(func(t textscan.Token) {
		fmt.Println("====> Token kind:", t.Kind())
		fmt.Println(t.String())

		if t.Kind() == textscan.TokValue {
			v := t.(*textscan.ValueToken)
			data[v.Key()] = v.Value()
		}
	})

	dump.P(data, err)
}

Output:

====> Token kind: Comments
# comments 1
====> Token kind: Value
key: name
value: "inhere"
comments: 
====> Token kind: Comments
// comments 2
====> Token kind: Value
key: age
value: "28"
comments: 
====> Token kind: Comments
/*
multi line
comments 3
*/
====> Token kind: Value
key: desc
value: "\n\na multi\nline string\n"
comments: 

==== Collected data:
map[string]string { #len=3
  "desc": string("
a multi
line string
"), #len=22
  "name": string("inhere"), #len=6
  "age": string("28"), #len=2
},

Projects using textscan

textscan is used in these projects:

Documentation

Overview

Package textscan Implemented a parser that quickly scans and analyzes text content. It can be used to parse INI, Properties and other formats

Index

Examples

Constants

View Source 
const (
	MultiLineValMarkS = "'''"
	MultiLineValMarkD = `"""`
	MultiLineValMarkH = "<<<" // heredoc at start. <<<TXT ... TXT
	MultiLineValMarkQ = "\\"  // at end. eg: properties contents
	MultiLineCmtEnd   = "*/"
)

define special chars consts

Variables

View Source 
var ErrCommentsNotEnd = errors.New("not end of multi-line comments")

ErrCommentsNotEnd error

View Source 
var ErrMLineValueNotEnd = errors.New("not end of multi line value")

ErrMLineValueNotEnd error

Functions

func AddKind 

func AddKind(k Kind, name string)

AddKind add global kind to kinds

func CommentsDetect 

func CommentsDetect(str string, inlineChars []byte) (ok, more bool, err error)

CommentsDetect check.

- inlineChars: #

default match:

  • inline #, //
  • multi line: /*

func CommentsDetectEnd 

func CommentsDetectEnd(line string) bool

CommentsDetectEnd multi line comments end

func HasKind 

func HasKind(k Kind) bool

HasKind check

func IsKindToken 

func IsKindToken(k Kind, tok Token) bool

IsKindToken check

func KindString 

func KindString(k Kind) string

KindString name

Types

type BaseToken 

type BaseToken struct {
	// contains filtered or unexported fields
}

BaseToken struct

func (*BaseToken) IsValid 

func (t *BaseToken) IsValid() bool

IsValid token

func (*BaseToken) Kind 

func (t *BaseToken) Kind() Kind

Kind type

func (*BaseToken) String 

func (t *BaseToken) String() string

String of token

func (*BaseToken) Value 

func (t *BaseToken) Value() string

Value of token

type CommentToken 

type CommentToken struct {
	BaseToken
	// contains filtered or unexported fields
}

CommentToken struct

func NewCommentToken 

func NewCommentToken(val string) *CommentToken

NewCommentToken instance.

func (*CommentToken) HasMore 

func (t *CommentToken) HasMore() bool

HasMore is multi line values

func (*CommentToken) MergeSame 

func (t *CommentToken) MergeSame(tok Token) error

MergeSame comments token

func (*CommentToken) ScanMore 

func (t *CommentToken) ScanMore(ts *TextScanner) error

ScanMore scan multi line values

func (*CommentToken) String 

func (t *CommentToken) String() string

String for token

func (*CommentToken) Value 

func (t *CommentToken) Value() string

Value fo token

type CommentsMatcher 

type CommentsMatcher struct {
	// InlineChars for match inline comments. default is: #
	InlineChars []byte
	// MatchFn for comments line
	// - mark 	useful on multi line comments
	MatchFn func(text string) (ok, more bool, err error)
	// DetectEnd for multi line comments
	DetectEnd func(text string) bool
}

CommentsMatcher match comments lines. will auto merge prev comments token

func (*CommentsMatcher) Match 

func (m *CommentsMatcher) Match(text string, prev Token) (Token, error)

Match comments token

func (*CommentsMatcher) MatchEnd 

func (m *CommentsMatcher) MatchEnd(text string) bool

MatchEnd for multi line comments

type ErrScan 

type ErrScan struct {
	Msg  string // error message
	Line int    // error line number, start 1
	Text string // text contents on error
}

ErrScan error on scan or parse contents

func (ErrScan) Error 

func (e ErrScan) Error() string

Error string

type HandleFn 

type HandleFn func(t Token)

HandleFn for token

type KeyValueMatcher 

type KeyValueMatcher struct {
	// Separator string for split key and value, default is "="
	Separator string
	// MergeComments collect previous comments token to value token.
	// If set as True, on each s.Scan() please notice skip TokComments
	MergeComments bool
	// InlineComment parse and split inline comment
	InlineComment bool
	// DisableMultiLine value parse
	DisableMultiLine bool
	// KeyCheckFn set func check key string is valid
	KeyCheckFn func(key string) error
}

KeyValueMatcher match key-value token. Support parse `KEY=VALUE` line text contents.

func (*KeyValueMatcher) DetectEnd 

func (m *KeyValueMatcher) DetectEnd(mark, text string) (ok bool, val string)

DetectEnd for multi line value

func (*KeyValueMatcher) Match 

func (m *KeyValueMatcher) Match(text string, prev Token) (Token, error)

Match text line.

type Kind 

type Kind uint8

Kind type 

const (
	TokInvalid Kind = iota
	TokKey
	TokValue
	TokComments
)

builtin defined kinds

func (Kind) String 

func (k Kind) String() string

String name for kind

type LiteToken 

type LiteToken interface {
	Kind() Kind
	Value() string
	IsValid() bool
}

LiteToken interface

type Matcher 

type Matcher interface {
	// Match text line by kind, if success returns a new Token
	Match(line string, prev Token) (tok Token, err error)
}

Matcher interface

type Parser 

type Parser struct {

	// Func for handle tokens
	Func HandleFn
	// contains filtered or unexported fields
}

Parser struct

func NewParser 

func NewParser(fn HandleFn) *Parser

NewParser instance

func (*Parser) AddMatchers 

func (p *Parser) AddMatchers(ms ...Matcher)

AddMatchers register token matchers

func (*Parser) Parse 

func (p *Parser) Parse(bs []byte) error

Parse input bytes

func (*Parser) ParseFrom 

func (p *Parser) ParseFrom(r io.Reader) error

ParseFrom input reader

func (*Parser) ParseText 

func (p *Parser) ParseText(text string) error

ParseText input string

type StringToken 

type StringToken struct {
	BaseToken
}

StringToken struct

func NewEmptyToken 

func NewEmptyToken() *StringToken

NewEmptyToken instance. Can use for want skip parse some contents

func NewStringToken 

func NewStringToken(k Kind, val string) *StringToken

NewStringToken instance.

func (*StringToken) HasMore 

func (t *StringToken) HasMore() bool

HasMore is multi line values

func (*StringToken) MergeSame 

func (t *StringToken) MergeSame(_ Token) error

MergeSame implements

func (*StringToken) ScanMore 

func (t *StringToken) ScanMore(_ *TextScanner) error

ScanMore implements

type TextScanner 

type TextScanner struct {
	// contains filtered or unexported fields
}

TextScanner struct.

func NewScanner 

func NewScanner(in any) *TextScanner

NewScanner instance

Example 
package main

import (
	"fmt"

	"github.com/gookit/goutil/strutil/textscan"
)

func main() {
	ts := textscan.NewScanner(`source code`)
	// add token matcher, can add your custom matcher
	ts.AddMatchers(
		&textscan.CommentsMatcher{
			InlineChars: []byte{'#'},
		},
		&textscan.KeyValueMatcher{
			MergeComments: true,
		},
	)

	// scan and parsing
	for ts.Scan() {
		tok := ts.Token()

		if !tok.IsValid() {
			continue
		}

		// Custom handle the parsed token
		if tok.Kind() == textscan.TokValue {
			vt := tok.(*textscan.ValueToken)
			fmt.Println(vt)
		}
	}

	if ts.Err() != nil {
		fmt.Println("ERROR:", ts.Err())
	}
}

Output:

func (*TextScanner) AddKind 

func (s *TextScanner) AddKind(k Kind, name string)

AddKind register new kind

func (*TextScanner) AddMatchers 

func (s *TextScanner) AddMatchers(ms ...Matcher)

AddMatchers register token matchers

func (*TextScanner) Each 

func (s *TextScanner) Each(fn func(t Token)) error

Each every token by given func

func (*TextScanner) Err 

func (s *TextScanner) Err() error

Err get

func (*TextScanner) Line 

func (s *TextScanner) Line() int

Line on current

func (*TextScanner) PrevToken 

func (s *TextScanner) PrevToken() Token

PrevToken get of previous scan.

func (*TextScanner) Scan 

func (s *TextScanner) Scan() bool

Scan source input and parsing. Can use Token() get current parsed token value

Usage: 

ts := textscan.NewScanner(`source ...`)
for ts.Scan() {
	tok := ts.Token()
	// do something...
}
fmt.Println(ts.Err())

func (*TextScanner) ScanNext 

func (s *TextScanner) ScanNext() (ok bool, text string)

ScanNext advance and fetch next line text

func (*TextScanner) SetInput 

func (s *TextScanner) SetInput(in any)

SetInput for scan and parse

func (*TextScanner) SetNext 

func (s *TextScanner) SetNext(text string)

SetNext text for scan and parse

func (*TextScanner) SetSplit 

func (s *TextScanner) SetSplit(fn bufio.SplitFunc)

SetSplit set split func on scan

func (*TextScanner) Token 

func (s *TextScanner) Token() Token

Token get of current scan.

type Token 

type Token interface {
	LiteToken
	String() string
	// HasMore is multi line values
	HasMore() bool
	// ScanMore scan multi line values
	ScanMore(ts *TextScanner) error
	MergeSame(tok Token) error
}

Token parser

type ValueToken 

type ValueToken struct {
	BaseToken
	// contains filtered or unexported fields
}

ValueToken contains key and value contents

func (*ValueToken) Comment 

func (t *ValueToken) Comment() string

Comment lines string

func (*ValueToken) HasComment 

func (t *ValueToken) HasComment() bool

HasComment for the value

func (*ValueToken) HasMore 

func (t *ValueToken) HasMore() bool

HasMore is multi line values

func (*ValueToken) Key 

func (t *ValueToken) Key() string

Key name

func (*ValueToken) Mark added in v0.5.15 

func (t *ValueToken) Mark() string

Mark for multi line values

func (*ValueToken) MergeSame 

func (t *ValueToken) MergeSame(_ Token) error

MergeSame comments token

func (*ValueToken) ScanMore 

func (t *ValueToken) ScanMore(ts *TextScanner) error

ScanMore scan multi line values

func (*ValueToken) String 

func (t *ValueToken) String() string

String of token

func (*ValueToken) Value 

func (t *ValueToken) Value() string

Value text string.

func (*ValueToken) Values added in v0.5.15 

func (t *ValueToken) Values() []string

Values for multi line values

Source Files

View all Source files

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL
go.dev uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn more.