miki

commit e0ab93a77f243af664423974da884ec30a4876c5
parent acf3ea53f035cf711f77cc403fe9c24391e152cb
Author: weetat <wtenison@protonmail.com>
Date:   Fri, 28 Feb 2025 18:25:28 +0000

Add EBNF grammar for fmd

Also rename tokenLiteral to tokenWord, add ( and ) to the lexer symbols,
and comment what each keyword does.

Diffstat:
A
src/fmd/fmd.ebnf
 | 
16
++++++++++++++++
M
src/fmd/lexer.go
 | 
18
++++++++++++++----
M
src/fmd/parser.go
 | 
5
+++--

3 files changed, 33 insertions(+), 6 deletions(-)
diff --git a/src/fmd/fmd.ebnf b/src/fmd/fmd.ebnf
@@ -0,0 +1,15 @@
+<fmd> ::= (<command> | <block>)*
+
+<command> ::=   <keyword> ":" <word>| 
+                <keyword> ":" "{" <block> "}" |
+                <keyword> "(" <block> ")" ":" <word> |
+                <keyword> "(" <block> ")" ":" "{" <block> "}" |
+
+<block> ::= (<word> | <space>)+
+<word> ::= (<letter> | <number> | <symbol>)+
+<keyword> ::= "b" | "i" | "h" [1-6] | "@" | "^" | "#"
+
+<space>  ::= " "
+<letter> ::= [a-z] | [A-Z]
+<number> ::= [0-9]+ | (<number> "." <number>)
+<symbol> ::= "." | "," | ";" | "-" | "!" | "#" | "$" | "%" | "^" | "&" | "*" | "(" | ")"
+\ No newline at end of file
diff --git a/src/fmd/lexer.go b/src/fmd/lexer.go
@@ -11,7 +11,7 @@ type tokenType int
 
 const (
 	// WORDS
-	tokenLiteral tokenType = iota // 0. (letters, numbers, symbols)+
+	tokenWord    tokenType = iota // 0. (letters, numbers, symbols)+
 	tokenKeyword                  // 1. (valid keywords)
 
 	// SPACING
@@ -24,6 +24,8 @@ const (
 	tokenColon  // 6. :
 	tokenLBrace // 7. {
 	tokenRBrace // 8. }
+	tokenLParen // 9. (
+	tokenRParen // 10. )
 )
 
 type token struct {
@@ -31,9 +33,17 @@ type token struct {
 	Value string
 }
 
-var symbols = []string{" ", "\n", "\t", "\\", ":", "{", "}"}
+var symbols = []string{" ", "\n", "\t", "\\", ":", "{", "}", "(", ")"}
 var keywords = []string{
-	"b", "i", "h1", "h2", "h3", "h4", "h5", "h6", "@", "#", "col", "img", "_"}
+	"b",                                // bold
+	"i",                                // italic
+	"h1", "h2", "h3", "h4", "h5", "h6", // header
+	"@",   // link
+	"#",   // tag
+	"col", // column container
+	"img", // image
+	"_",   // empty container (<div>)
+}
 
 type lexer struct {
 	input  []byte // text input to scan
@@ -85,7 +95,7 @@ func (l *lexer) Exec() []token {
 				if slices.Contains(keywords, _tokenString) {
 					l.tokens = append(l.tokens, token{tokenKeyword, _tokenString})
 				} else if _tokenString != "" {
-					l.tokens = append(l.tokens, token{tokenLiteral, _tokenString})
+					l.tokens = append(l.tokens, token{tokenWord, _tokenString})
 				}
 			}
 		}
diff --git a/src/fmd/parser.go b/src/fmd/parser.go
@@ -91,14 +91,14 @@ func (p *parser) parseBlock() node {
 }
 
 func (p *parser) parseLiteral() node {
-	_literal := node{nType: nodeLiteral, value: p.consume(tokenLiteral).Value}
+	_literal := node{nType: nodeLiteral, value: p.consume(tokenWord).Value}
 	return _literal
 }
 
 func (p *parser) isNextCommand() bool {
 	return p.peekNextN(tokenKeyword, 0) &&
 		p.peekNextN(tokenColon, 1) &&
-		p.peekNextN(tokenLiteral, 2) ||
+		p.peekNextN(tokenWord, 2) ||
 		p.peekNextN(tokenLBrace, 2)
 }
 
@@ -113,6 +113,7 @@ func (p *parser) consume(t tokenType) token {
 		p.pos++
 		return token
 	}
+	// TODO: Error out gracefully and send empty AST.
 	panic(fmt.Sprintf("__PANIC\tfmd:parser:Invalid token:%v", t))
 }