quip: Add initial lexer sources + example ref

Signed-off-by: Ian Moffett <ian@mirocom.org>
This commit is contained in:
2026-03-21 07:15:54 -04:00
parent 0fa8ad12ca
commit 6f7b4b5e31
6 changed files with 214 additions and 2 deletions

19
example/ref.quip Normal file
View File

@@ -0,0 +1,19 @@
.cc clang
.ld lld-link
CFLAGS ::
-Iinclude/
-mgeneral-regs-only
-ffreestanding
$CONF # expands to -D chain
~
.obiter echo $OBJ
.obiter objcopy -O binary $OBJ $OBJ.bin
.oblink
# configuration section
# these get
conf FOO : true
conf BAR : 123

127
frontend/lexer.c Normal file
View File

@@ -0,0 +1,127 @@
/*
* Copyright (c) 2026, Mirocom Laboratories
* Provided under the BSD-3 clause
*
* Abstract:
* This file implements the lexer.
* Author:
* Ian M. Moffett <ian@mirocom.org>
*/
#include <sys/types.h>
#include <stddef.h>
#include <stdbool.h>
#include <unistd.h>
#include <errno.h>
#include <ctype.h>
#include "frontend/token.h"
#include "frontend/lexer.h"
/*
* Test if a given character counts as a whitespace character
*
* @c: Character to test
*/
static inline bool
lexer_is_ws(char c)
{
switch (c) {
case '\t':
case '\f':
case '\a':
case ' ':
return true;
}
return false;
}
/*
* Consume a single character in a buffered manner from the
* build source file.
*
* @state: Quip state machine
*
* Returns the fetched character on success, otherwise a value
* of '\0' on failure.
*/
static char
lexer_buffer_consume(struct quip_state *state)
{
ssize_t n;
if (state == NULL) {
return '\0';
}
/*
* If there is nothing in the lexer-side buffer, fill it
* with what we can.
*/
if (state->lex_buf_cap == 0) {
n = read(state->in_fd, state->lex_buf, LEX_FILEBUF_LEN);
if (n <= 0)
return '\0';
state->lex_buf_cap = n;
}
/* Grab a single character if not empty */
if (state->lex_buf_i < state->lex_buf_cap) {
return state->lex_buf[state->lex_buf_i++];
}
/* Empty, reset everything and try again */
state->lex_buf_cap = 0;
state->lex_buf_i = 0;
return lexer_buffer_consume(state);
}
/*
* Consume a single character optionally skipping whitespace
*
* @state: Quip state machine
* @skip_ws: If true, skip whitespace
*/
static char
lexer_consume(struct quip_state *state, bool skip_ws)
{
char c;
while ((c = lexer_buffer_consume(state)) != '\0') {
if (skip_ws && lexer_is_ws(c)) {
continue;
}
return c;
}
return '\0';
}
int
lexer_scan(struct quip_state *state, struct token *tokres)
{
char c;
if (state == NULL || tokres == NULL) {
return -1;
}
if ((c = lexer_consume(state, true)) == '\0') {
return -1;
}
switch (c) {
case '\n':
tokres->type = TT_NEWLINE;
tokres->c = c;
return 0;
case ':':
tokres->type = TT_COLON;
tokres->c = c;
return 0;
}
return -1;
}

View File

@@ -11,7 +11,11 @@
#ifndef COMMON_KNOBS_H
#define COMMON_KNOBS_H 1
/* Core settings */
#define QUIP_VERSION "0.0.1"
#define QUIP_FILEPATH "build.quip"
/* Buffering settings */
#define LEX_FILEBUF_LEN 16
#endif /* !_COMMON_KNOBS_H_ */

View File

@@ -13,16 +13,23 @@
#include <stdint.h>
#include <stddef.h>
#include "common/knobs.h"
/*
* Represents the build state machine
*
* @in_fd: Input file descriptor of build file
* @line_num: Current line number
* @in_fd: Input file descriptor of build file
* @line_num: Current line number
* @lex_buf: Used to reduce system call frequency
* @lex_buf_cap: Lexer buffer capacity
* @lex_buf_i: Lexer buffer index
*/
struct quip_state {
int in_fd;
size_t line_num;
char lex_buf[LEX_FILEBUF_LEN];
size_t lex_buf_cap;
size_t lex_buf_i;
};
/*

19
include/frontend/lexer.h Normal file
View File

@@ -0,0 +1,19 @@
#ifndef FRONTEND_LEXER_H
#define FRONTEND_LEXER_H 1
#include <stdint.h>
#include <stddef.h>
#include "frontend/token.h"
#include "common/state.h"
/*
* Scan for a single token within the source file
*
* @state: Quip state
* @tokres: Token result is written here
*
* Returns zero on success
*/
int lexer_scan(struct quip_state *state, struct token *tokres);
#endif /* !FRONTEND_LEXER_H */

36
include/frontend/token.h Normal file
View File

@@ -0,0 +1,36 @@
/*
* Copyright (c) 2026, Mirocom Laboratories
* Provided under the BSD-3 clause
*
* Abstract:
* This file defines the list of valid tokens.
* Author:
* Ian M. Moffett <ian@mirocom.org>
*/
#ifndef FRONTEND_TOKEN_H
#define FRONTEND_TOKEN_H 1
/*
* Represents valid token types
*/
typedef enum {
TT_NONE, /* [none] */
TT_NAME, /* [name] */
TT_NEWLINE, /* [newline] */
TT_COLON, /* ':' */
} tt_t;
/*
* Represents a lexical token
*
* @type: Token type
*/
struct token {
tt_t type;
union {
char c;
};
};
#endif /* !FRONTEND_TOKEN_H */