notepad-plus-plus-legacy/scintilla/lexers/LexA68k.cxx
2019-05-04 21:14:48 +03:00

346 lines
12 KiB
C++

// Scintilla source code edit control
/** @file LexA68k.cxx
** Lexer for Assembler, just for the MASM syntax
** Written by Martial Demolins AKA Folco
**/
// Copyright 2010 Martial Demolins <mdemolins(a)gmail.com>
// The License.txt file describes the conditions under which this software
// may be distributed.
#include <stdlib.h>
#include <string.h>
#include <stdio.h>
#include <stdarg.h>
#include <assert.h>
#include <ctype.h>
#include "ILexer.h"
#include "Scintilla.h"
#include "SciLexer.h"
#include "WordList.h"
#include "LexAccessor.h"
#include "Accessor.h"
#include "StyleContext.h"
#include "CharacterSet.h"
#include "LexerModule.h"
using namespace Scintilla;
// Return values for GetOperatorType
#define NO_OPERATOR 0
#define OPERATOR_1CHAR 1
#define OPERATOR_2CHAR 2
/**
* IsIdentifierStart
*
* Return true if the given char is a valid identifier first char
*/
static inline bool IsIdentifierStart (const int ch)
{
return (isalpha(ch) || (ch == '_') || (ch == '\\'));
}
/**
* IsIdentifierChar
*
* Return true if the given char is a valid identifier char
*/
static inline bool IsIdentifierChar (const int ch)
{
return (isalnum(ch) || (ch == '_') || (ch == '@') || (ch == ':') || (ch == '.'));
}
/**
* GetOperatorType
*
* Return:
* NO_OPERATOR if char is not an operator
* OPERATOR_1CHAR if the operator is one char long
* OPERATOR_2CHAR if the operator is two chars long
*/
static inline int GetOperatorType (const int ch1, const int ch2)
{
int OpType = NO_OPERATOR;
if ((ch1 == '+') || (ch1 == '-') || (ch1 == '*') || (ch1 == '/') || (ch1 == '#') ||
(ch1 == '(') || (ch1 == ')') || (ch1 == '~') || (ch1 == '&') || (ch1 == '|') || (ch1 == ','))
OpType = OPERATOR_1CHAR;
else if ((ch1 == ch2) && (ch1 == '<' || ch1 == '>'))
OpType = OPERATOR_2CHAR;
return OpType;
}
/**
* IsBin
*
* Return true if the given char is 0 or 1
*/
static inline bool IsBin (const int ch)
{
return (ch == '0') || (ch == '1');
}
/**
* IsDoxygenChar
*
* Return true if the char may be part of a Doxygen keyword
*/
static inline bool IsDoxygenChar (const int ch)
{
return isalpha(ch) || (ch == '$') || (ch == '[') || (ch == ']') || (ch == '{') || (ch == '}');
}
/**
* ColouriseA68kDoc
*
* Main function, which colourises a 68k source
*/
static void ColouriseA68kDoc (Sci_PositionU startPos, Sci_Position length, int initStyle, WordList *keywordlists[], Accessor &styler)
{
// Used to buffer a string, to be able to compare it using built-in functions
char Buffer[100];
// Used to know the length of an operator
int OpType;
// Get references to keywords lists
WordList &cpuInstruction = *keywordlists[0];
WordList &registers = *keywordlists[1];
WordList &directive = *keywordlists[2];
WordList &extInstruction = *keywordlists[3];
WordList &alert = *keywordlists[4];
WordList &doxygenKeyword = *keywordlists[5];
// Instanciate a context for our source
StyleContext sc(startPos, length, initStyle, styler);
/************************************************************
*
* Parse the source
*
************************************************************/
for ( ; sc.More(); sc.Forward())
{
/************************************************************
*
* A style always terminates at the end of a line, even for
* comments (no multi-lines comments)
*
************************************************************/
if (sc.atLineStart) {
sc.SetState(SCE_A68K_DEFAULT);
}
/************************************************************
*
* If we are not in "default style", check if the style continues
* In this case, we just have to loop
*
************************************************************/
if (sc.state != SCE_A68K_DEFAULT)
{
if ( ((sc.state == SCE_A68K_NUMBER_DEC) && isdigit(sc.ch)) // Decimal number
|| ((sc.state == SCE_A68K_NUMBER_BIN) && IsBin(sc.ch)) // Binary number
|| ((sc.state == SCE_A68K_NUMBER_HEX) && isxdigit(sc.ch)) // Hexa number
|| ((sc.state == SCE_A68K_MACRO_ARG) && isdigit(sc.ch)) // Macro argument
|| ((sc.state == SCE_A68K_STRING1) && (sc.ch != '\'')) // String single-quoted
|| ((sc.state == SCE_A68K_STRING2) && (sc.ch != '\"')) // String double-quoted
|| ((sc.state == SCE_A68K_MACRO_DECLARATION) && IsIdentifierChar(sc.ch)) // Macro declaration (or global label, we don't know at this point)
|| ((sc.state == SCE_A68K_IDENTIFIER) && IsIdentifierChar(sc.ch)) // Identifier
|| ((sc.state == SCE_A68K_LABEL) && IsIdentifierChar(sc.ch)) // Label (local)
|| ((sc.state == SCE_A68K_COMMENT_DOXYGEN) && IsDoxygenChar(sc.ch)) // Doxygen keyword
|| ((sc.state == SCE_A68K_COMMENT_SPECIAL) && isalpha(sc.ch)) // Alert
|| ((sc.state == SCE_A68K_COMMENT) && !isalpha(sc.ch) && (sc.ch != '\\'))) // Normal comment
{
continue;
}
/************************************************************
*
* Check if current state terminates
*
************************************************************/
// Strings: include terminal ' or " in the current string by skipping it
if ((sc.state == SCE_A68K_STRING1) || (sc.state == SCE_A68K_STRING2)) {
sc.Forward();
}
// If a macro declaration was terminated with ':', it was a label
else if ((sc.state == SCE_A68K_MACRO_DECLARATION) && (sc.chPrev == ':')) {
sc.ChangeState(SCE_A68K_LABEL);
}
// If it wasn't a Doxygen keyword, change it to normal comment
else if (sc.state == SCE_A68K_COMMENT_DOXYGEN) {
sc.GetCurrent(Buffer, sizeof(Buffer));
if (!doxygenKeyword.InList(Buffer)) {
sc.ChangeState(SCE_A68K_COMMENT);
}
sc.SetState(SCE_A68K_COMMENT);
continue;
}
// If it wasn't an Alert, change it to normal comment
else if (sc.state == SCE_A68K_COMMENT_SPECIAL) {
sc.GetCurrent(Buffer, sizeof(Buffer));
if (!alert.InList(Buffer)) {
sc.ChangeState(SCE_A68K_COMMENT);
}
// Reset style to normal comment, or to Doxygen keyword if it begins with '\'
if (sc.ch == '\\') {
sc.SetState(SCE_A68K_COMMENT_DOXYGEN);
}
else {
sc.SetState(SCE_A68K_COMMENT);
}
continue;
}
// If we are in a comment, it's a Doxygen keyword or an Alert
else if (sc.state == SCE_A68K_COMMENT) {
if (sc.ch == '\\') {
sc.SetState(SCE_A68K_COMMENT_DOXYGEN);
}
else {
sc.SetState(SCE_A68K_COMMENT_SPECIAL);
}
continue;
}
// Check if we are at the end of an identifier
// In this case, colourise it if was a keyword.
else if ((sc.state == SCE_A68K_IDENTIFIER) && !IsIdentifierChar(sc.ch)) {
sc.GetCurrentLowered(Buffer, sizeof(Buffer)); // Buffer the string of the current context
if (cpuInstruction.InList(Buffer)) { // And check if it belongs to a keyword list
sc.ChangeState(SCE_A68K_CPUINSTRUCTION);
}
else if (extInstruction.InList(Buffer)) {
sc.ChangeState(SCE_A68K_EXTINSTRUCTION);
}
else if (registers.InList(Buffer)) {
sc.ChangeState(SCE_A68K_REGISTER);
}
else if (directive.InList(Buffer)) {
sc.ChangeState(SCE_A68K_DIRECTIVE);
}
}
// All special contexts are now handled.Come back to default style
sc.SetState(SCE_A68K_DEFAULT);
}
/************************************************************
*
* Check if we must enter a new state
*
************************************************************/
// Something which begins at the beginning of a line, and with
// - '\' + an identifier start char, or
// - '\\@' + an identifier start char
// is a local label (second case is used for macro local labels). We set it already as a label, it can't be a macro/equ declaration
if (sc.atLineStart && (sc.ch < 0x80) && IsIdentifierStart(sc.chNext) && (sc.ch == '\\')) {
sc.SetState(SCE_A68K_LABEL);
}
if (sc.atLineStart && (sc.ch < 0x80) && (sc.ch == '\\') && (sc.chNext == '\\')) {
sc.Forward(2);
if ((sc.ch == '@') && IsIdentifierStart(sc.chNext)) {
sc.ChangeState(SCE_A68K_LABEL);
sc.SetState(SCE_A68K_LABEL);
}
}
// Label and macro identifiers start at the beginning of a line
// We set both as a macro id, but if it wasn't one (':' at the end),
// it will be changed as a label.
if (sc.atLineStart && (sc.ch < 0x80) && IsIdentifierStart(sc.ch)) {
sc.SetState(SCE_A68K_MACRO_DECLARATION);
}
else if ((sc.ch < 0x80) && (sc.ch == ';')) { // Default: alert in a comment. If it doesn't match
sc.SetState(SCE_A68K_COMMENT); // with an alert, it will be toggle to a normal comment
}
else if ((sc.ch < 0x80) && isdigit(sc.ch)) { // Decimal numbers haven't prefix
sc.SetState(SCE_A68K_NUMBER_DEC);
}
else if ((sc.ch < 0x80) && (sc.ch == '%')) { // Binary numbers are prefixed with '%'
sc.SetState(SCE_A68K_NUMBER_BIN);
}
else if ((sc.ch < 0x80) && (sc.ch == '$')) { // Hexadecimal numbers are prefixed with '$'
sc.SetState(SCE_A68K_NUMBER_HEX);
}
else if ((sc.ch < 0x80) && (sc.ch == '\'')) { // String (single-quoted)
sc.SetState(SCE_A68K_STRING1);
}
else if ((sc.ch < 0x80) && (sc.ch == '\"')) { // String (double-quoted)
sc.SetState(SCE_A68K_STRING2);
}
else if ((sc.ch < 0x80) && (sc.ch == '\\') && (isdigit(sc.chNext))) { // Replacement symbols in macro are prefixed with '\'
sc.SetState(SCE_A68K_MACRO_ARG);
}
else if ((sc.ch < 0x80) && IsIdentifierStart(sc.ch)) { // An identifier: constant, label, etc...
sc.SetState(SCE_A68K_IDENTIFIER);
}
else {
if (sc.ch < 0x80) {
OpType = GetOperatorType(sc.ch, sc.chNext); // Check if current char is an operator
if (OpType != NO_OPERATOR) {
sc.SetState(SCE_A68K_OPERATOR);
if (OpType == OPERATOR_2CHAR) { // Check if the operator is 2 bytes long
sc.ForwardSetState(SCE_A68K_OPERATOR); // (>> or <<)
}
}
}
}
} // End of for()
sc.Complete();
}
// Names of the keyword lists
static const char * const a68kWordListDesc[] =
{
"CPU instructions",
"Registers",
"Directives",
"Extended instructions",
"Comment special words",
"Doxygen keywords",
0
};
LexerModule lmA68k(SCLEX_A68K, ColouriseA68kDoc, "a68k", 0, a68kWordListDesc);