| products/Sources/formale Sprachen/C/Lyx/development/lyx3/ |
|
Quellcode-Bibliothek
© Kompilation durch diese Firma
[Weder Korrektheit noch Funktionsfähigkeit der Software werden zugesichert.]
Datei: ReadMe Sprache: Unknown
|
Spracherkennung für: .y vermutete Sprache: Text {Text[109] Latech[171] Haskell[239]} [Methode: Schwerpunktbildung, einfache Gewichte, sechs Dimensionen] Brief discussion on LaTeX grammar --------------------------------- LaTeX syntax is relatively simple. Any command is like this: \command_name [<options>]{<argument_1>}... This can be expressed as a grammar in yacc/bison syntax. Capitalized words are tokens (terminals) returned by lex: command : LT_command options arguments arguments : argument | argument arguments | /* empty: It's possible to not have arguments */ ; argument : '{' anything '}' ; options : '[' word_list ']' | /* empty: It's possible to don't have options */ ; word_list : LT_word ',' word_list | LT_word ; anything : /* Any valid token */ | command | LT_punctuation | LT_accent | LT_etc ; This is quite simple and can be easily parsed by any LALR parser as a yacc or bison generated one. BUT the difficulty comes because the number of arguments depends on the value of LT_command, and that the symbols '{' and '}' have more uses that to separate arguments. This expression: \section{Hello world} {\bf Once} is ambiguos since the parser would interpret the last group as a second argument. Maybe the simpler approach is declaring a grammar for each of the commands that LyX will to understand. Another is that if the parser knows the identity of a command should know also how many arguments should expect. [Aren't these two suggestions almost identical? I mean, if we tell the parser what parameters each command take, we also have a grammar for each of the commands, LyX will understand... Asger] AAS [ Dauer der Verarbeitung: 0.104 Sekunden ] |