2019-10-31 09:06:25 +08:00
|
|
|
package org
|
|
|
|
|
|
|
|
import (
|
|
|
|
"regexp"
|
|
|
|
"strings"
|
|
|
|
"unicode"
|
|
|
|
)
|
|
|
|
|
|
|
|
type Block struct {
|
|
|
|
Name string
|
|
|
|
Parameters []string
|
|
|
|
Children []Node
|
2020-09-05 23:45:10 +08:00
|
|
|
Result Node
|
|
|
|
}
|
|
|
|
|
|
|
|
type Result struct {
|
|
|
|
Node Node
|
2019-10-31 09:06:25 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
type Example struct {
|
|
|
|
Children []Node
|
|
|
|
}
|
|
|
|
|
|
|
|
var exampleLineRegexp = regexp.MustCompile(`^(\s*):(\s(.*)|\s*$)`)
|
|
|
|
var beginBlockRegexp = regexp.MustCompile(`(?i)^(\s*)#\+BEGIN_(\w+)(.*)`)
|
|
|
|
var endBlockRegexp = regexp.MustCompile(`(?i)^(\s*)#\+END_(\w+)`)
|
2020-09-05 23:45:10 +08:00
|
|
|
var resultRegexp = regexp.MustCompile(`(?i)^(\s*)#\+RESULTS:`)
|
|
|
|
var exampleBlockEscapeRegexp = regexp.MustCompile(`(^|\n)([ \t]*),([ \t]*)(\*|,\*|#\+|,#\+)`)
|
2019-10-31 09:06:25 +08:00
|
|
|
|
|
|
|
func lexBlock(line string) (token, bool) {
|
|
|
|
if m := beginBlockRegexp.FindStringSubmatch(line); m != nil {
|
|
|
|
return token{"beginBlock", len(m[1]), strings.ToUpper(m[2]), m}, true
|
|
|
|
} else if m := endBlockRegexp.FindStringSubmatch(line); m != nil {
|
|
|
|
return token{"endBlock", len(m[1]), strings.ToUpper(m[2]), m}, true
|
|
|
|
}
|
|
|
|
return nilToken, false
|
|
|
|
}
|
|
|
|
|
2020-09-05 23:45:10 +08:00
|
|
|
func lexResult(line string) (token, bool) {
|
|
|
|
if m := resultRegexp.FindStringSubmatch(line); m != nil {
|
|
|
|
return token{"result", len(m[1]), "", m}, true
|
|
|
|
}
|
|
|
|
return nilToken, false
|
|
|
|
}
|
|
|
|
|
2019-10-31 09:06:25 +08:00
|
|
|
func lexExample(line string) (token, bool) {
|
|
|
|
if m := exampleLineRegexp.FindStringSubmatch(line); m != nil {
|
|
|
|
return token{"example", len(m[1]), m[3], m}, true
|
|
|
|
}
|
|
|
|
return nilToken, false
|
|
|
|
}
|
|
|
|
|
|
|
|
func isRawTextBlock(name string) bool { return name == "SRC" || name == "EXAMPLE" || name == "EXPORT" }
|
|
|
|
|
|
|
|
func (d *Document) parseBlock(i int, parentStop stopFn) (int, Node) {
|
|
|
|
t, start := d.tokens[i], i
|
2021-04-23 08:08:53 +08:00
|
|
|
name, parameters := t.content, splitParameters(t.matches[3])
|
2019-10-31 09:06:25 +08:00
|
|
|
trim := trimIndentUpTo(d.tokens[i].lvl)
|
|
|
|
stop := func(d *Document, i int) bool {
|
|
|
|
return i >= len(d.tokens) || (d.tokens[i].kind == "endBlock" && d.tokens[i].content == name)
|
|
|
|
}
|
2020-09-05 23:45:10 +08:00
|
|
|
block, i := Block{name, parameters, nil, nil}, i+1
|
2019-10-31 09:06:25 +08:00
|
|
|
if isRawTextBlock(name) {
|
|
|
|
rawText := ""
|
|
|
|
for ; !stop(d, i); i++ {
|
|
|
|
rawText += trim(d.tokens[i].matches[0]) + "\n"
|
|
|
|
}
|
2020-09-05 23:45:10 +08:00
|
|
|
if name == "EXAMPLE" || (name == "SRC" && len(parameters) >= 1 && parameters[0] == "org") {
|
|
|
|
rawText = exampleBlockEscapeRegexp.ReplaceAllString(rawText, "$1$2$3$4")
|
|
|
|
}
|
2019-10-31 09:06:25 +08:00
|
|
|
block.Children = d.parseRawInline(rawText)
|
|
|
|
} else {
|
|
|
|
consumed, nodes := d.parseMany(i, stop)
|
|
|
|
block.Children = nodes
|
|
|
|
i += consumed
|
|
|
|
}
|
2020-09-05 23:45:10 +08:00
|
|
|
if i >= len(d.tokens) || d.tokens[i].kind != "endBlock" || d.tokens[i].content != name {
|
|
|
|
return 0, nil
|
|
|
|
}
|
|
|
|
if name == "SRC" {
|
|
|
|
consumed, result := d.parseSrcBlockResult(i+1, parentStop)
|
|
|
|
block.Result = result
|
|
|
|
i += consumed
|
|
|
|
}
|
|
|
|
return i + 1 - start, block
|
|
|
|
}
|
|
|
|
|
|
|
|
func (d *Document) parseSrcBlockResult(i int, parentStop stopFn) (int, Node) {
|
|
|
|
start := i
|
|
|
|
for ; !parentStop(d, i) && d.tokens[i].kind == "text" && d.tokens[i].content == ""; i++ {
|
|
|
|
}
|
|
|
|
if parentStop(d, i) || d.tokens[i].kind != "result" {
|
|
|
|
return 0, nil
|
2019-10-31 09:06:25 +08:00
|
|
|
}
|
2020-09-05 23:45:10 +08:00
|
|
|
consumed, result := d.parseResult(i, parentStop)
|
|
|
|
return (i - start) + consumed, result
|
2019-10-31 09:06:25 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
func (d *Document) parseExample(i int, parentStop stopFn) (int, Node) {
|
|
|
|
example, start := Example{}, i
|
|
|
|
for ; !parentStop(d, i) && d.tokens[i].kind == "example"; i++ {
|
|
|
|
example.Children = append(example.Children, Text{d.tokens[i].content, true})
|
|
|
|
}
|
|
|
|
return i - start, example
|
|
|
|
}
|
|
|
|
|
2020-09-05 23:45:10 +08:00
|
|
|
func (d *Document) parseResult(i int, parentStop stopFn) (int, Node) {
|
|
|
|
if i+1 >= len(d.tokens) {
|
|
|
|
return 0, nil
|
|
|
|
}
|
|
|
|
consumed, node := d.parseOne(i+1, parentStop)
|
|
|
|
return consumed + 1, Result{node}
|
|
|
|
}
|
|
|
|
|
2019-10-31 09:06:25 +08:00
|
|
|
func trimIndentUpTo(max int) func(string) string {
|
|
|
|
return func(line string) string {
|
|
|
|
i := 0
|
|
|
|
for ; i < len(line) && i < max && unicode.IsSpace(rune(line[i])); i++ {
|
|
|
|
}
|
|
|
|
return line[i:]
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2021-04-23 08:08:53 +08:00
|
|
|
func splitParameters(s string) []string {
|
|
|
|
parameters, parts := []string{}, strings.Split(s, " :")
|
|
|
|
lang, rest := strings.TrimSpace(parts[0]), parts[1:]
|
|
|
|
if lang != "" {
|
|
|
|
parameters = append(parameters, lang)
|
|
|
|
}
|
|
|
|
for _, p := range rest {
|
|
|
|
kv := strings.SplitN(p+" ", " ", 2)
|
|
|
|
parameters = append(parameters, ":"+kv[0], strings.TrimSpace(kv[1]))
|
|
|
|
}
|
|
|
|
return parameters
|
|
|
|
}
|
|
|
|
|
2020-09-05 23:45:10 +08:00
|
|
|
func (b Block) ParameterMap() map[string]string {
|
|
|
|
if len(b.Parameters) == 0 {
|
|
|
|
return nil
|
|
|
|
}
|
|
|
|
m := map[string]string{":lang": b.Parameters[0]}
|
|
|
|
for i := 1; i+1 < len(b.Parameters); i += 2 {
|
|
|
|
m[b.Parameters[i]] = b.Parameters[i+1]
|
|
|
|
}
|
|
|
|
return m
|
|
|
|
}
|
|
|
|
|
2019-11-05 16:39:03 +08:00
|
|
|
func (n Example) String() string { return orgWriter.WriteNodesAsString(n) }
|
|
|
|
func (n Block) String() string { return orgWriter.WriteNodesAsString(n) }
|
2020-09-05 23:45:10 +08:00
|
|
|
func (n Result) String() string { return orgWriter.WriteNodesAsString(n) }
|