1
0
Files
org2newsboat/lexer.go

80 lines
1.6 KiB
Go

package main
import (
"regexp"
"strings"
)
// Lexer embbed a temporary buffer to store "content"
// (url, description, tags) and an array of tokens
type Lexer struct {
Buf string
Tokens []Token
}
func (l *Lexer) AddToken(t string, s string) {
l.Tokens = append(l.Tokens, Token{t, s})
}
// Return last tokenized item, useful to determine context for a symbol
func (l Lexer) LastToken() Token {
var lastToken Token
if len(l.Tokens) > 0 {
lastToken = l.Tokens[len(l.Tokens)-1]
}
return lastToken
}
func (l Lexer) CountToken(t string) int {
var counter int
for i := range l.Tokens {
if l.Tokens[i].Value == t {
counter ++
}
}
return counter
}
func (l *Lexer) ProcessSimpleLink(s string) {
ss := strings.Split(s[3:], " ")
for w := range ss {
if w == 0 {
url := strings.TrimSpace(ss[w])
l.AddToken("URL", url)
} else {
if ss[w] != "" && ss[w] != " " {
tag := strings.ReplaceAll(ss[w], ":", "")
l.AddToken("TAG", tag)
}
}
}
}
// Tokenize a given line s from the org file
func (l Lexer) Process(s string) []Token {
if s[3] != '[' {
l.ProcessSimpleLink(s)
return l.Tokens
}
re := regexp.MustCompile(`(?:\[\[)(?P<url>\S+)(?:\]\[)(?P<desc>.+)(?:\]\])(?P<tags>.+)?`)
matches := re.FindStringSubmatch(s)
if len(matches) > 1 {
l.AddToken("URL", strings.TrimSpace(matches[1]))
}
if len(matches) > 2 {
l.AddToken("DESC", strings.TrimSpace(matches[2]))
}
if len(matches) > 3 {
tags := strings.Split(matches[3], " ")
for t := range tags {
if tags[t] != "" && tags[t] != " " {
tag := strings.ReplaceAll(tags[t], ":", "")
l.AddToken("TAG", strings.TrimSpace(tag))
}
}
}
return l.Tokens
}