bshell.vim: fix highlighting of nested parentheses in string interpolation

bshell.vim/syntax/bshell.vim

@@ -14,8 +14,29 @@ syn keyword	bshellKeyword	func
 syn keyword	bshellTodo	contained TODO FIXME XXX NOTE HACK TBD
 syn match	bshellLineComment	/#.*$/ contains=bshellTodo
 
-syn region	bshellInterpolatedString	matchgroup=bshellString start=+"+ end=+"+ extend contains=bshellVariable
+syn region	bshellInterpolation	matchgroup=bshellInterpolationDelimiter start=+$(+ end=+)+ keepend contained contains=@bshellAll
+syn region	bshellInterpolation	matchgroup=bshellInterpolationDelimiter start="$(" end=")" contained contains=ALL
+syn region	bshellInterpolatedString	matchgroup=bshellString start=+"+ end=+"+ extend contains=bshellVariable,bshellInterpolation
 syn region	bshellLiteralString	matchgroup=bshellString start=+\'+ end=+\'+ extend contains=bshellSpecialChar,bshellSpecialError,bshellUnicodeNumber,@Spell
+syn region	bshell1NestedParentheses	start="(" skip="\\\\\|\\)" matchgroup=bshellInterpolation end=")" transparent contained
+
+syn case	ignore
+syn match	bshellInteger	"\<0b[01_]*[01]\%([lu]\|lu\|ul\)\=\>" display
+syn match	bshellInteger	"\<\d\+\%(_\+\d\+\)*\%([lu]\|lu\|ul\)\=\>" display
+syn match	bshellInteger	"\<-\d\+\%(_\+\d\+\)*\%([lu]\|lu\|ul\)\=\>" display
+syn match	bshellInteger	"\<0x[[:xdigit:]_]*\x\%([lu]\|lu\|ul\)\=\>" display
+syn match	bshellReal	"\<\d\+\%(_\+\d\+\)*\.\d\+\%(_\+\d\+\)*\%\(e[-+]\=\d\+\%(_\+\d\+\)*\)\=[fdm]\=" display
+syn match	bshellReal	"\.\d\+\%(_\+\d\+\)*\%(e[-+]\=\d\+\%(_\+\d\+\)*\)\=[fdm]\=\>" display
+syn match	bshellReal	"\<\d\+\%(_\+\d\+\)*e[-+]\=\d\+\%(_\+\d\+\)*[fdm]\=\>" display
+syn match	bshellReal	"\<\d\+\%(_\+\d\+\)*[fdm]\>" display
+syn case	match
+syn cluster     bshellNumber	contains=bshellInteger,bshellReal
+
+syn cluster	bshellLiteral	contains=@bshellNumber,@bshellString
+syn cluster	bshellAll	contains=@bshellLiteral,bshellVariable,bshellKeyword,bshellArgFlag,bshellFunctionRef,bshellSymbolOp,bshellKeywordOp
+
+syn match	bshellSymbolOp	"[+]" display
+syn keyword	bshellKeywordOp	is not understands and or
 
 hi def link	bshellKeyword	Statement
 hi def link	bshellArgFlag	Tag
@@ -25,7 +46,12 @@ hi def link	bshellFunctionRef	Function
 hi def link	bshellString	String
 hi def link	bshellInterpolatedString	String
 hi def link	bshellLiteralString	String
+hi def link	bshellInteger	Number
+hi def link	bshellReal	Float
 
+hi def link	bshellInterpolationDelimiter Delimiter
+hi def link	bshellSymbolOp Operator
+hi def link	bshellKeywordOp Operator
 
 " The default highlighting.
 " hi def link	bshellUnspecifiedStatement	Statement

bshell/ast/ast.c

@@ -0,0 +1,166 @@
+#include "ast.h"
+
+#include "../status.h"
+
+#include <assert.h>
+#include <stdlib.h>
+#include <string.h>
+
+extern struct ast_node_definition int_ast_node;
+extern struct ast_node_definition double_ast_node;
+extern struct ast_node_definition word_ast_node;
+extern struct ast_node_definition var_ast_node;
+extern struct ast_node_definition string_ast_node;
+extern struct ast_node_definition fstring_ast_node;
+extern struct ast_node_definition cmdcall_ast_node;
+extern struct ast_node_definition pipeline_ast_node;
+extern struct ast_node_definition redirection_ast_node;
+
+static const struct ast_node_definition *ast_node_defintions[] = {
+	[AST_INT] = &int_ast_node,
+	[AST_DOUBLE] = &double_ast_node,
+	[AST_WORD] = &word_ast_node,
+	[AST_VAR] = &var_ast_node,
+	[AST_STRING] = &string_ast_node,
+	[AST_FSTRING] = &fstring_ast_node,
+	[AST_CMDCALL] = &cmdcall_ast_node,
+	[AST_PIPELINE] = &pipeline_ast_node,
+	[AST_REDIRECTION] = &redirection_ast_node,
+};
+static const size_t nr_ast_node_definitions
+	= sizeof ast_node_defintions / sizeof ast_node_defintions[0];
+
+struct ast_node *ast_node_create(enum ast_node_type type)
+{
+	assert(type < nr_ast_node_definitions);
+
+	const struct ast_node_definition *def = ast_node_defintions[type];
+	struct ast_node *out = malloc(def->def_node_size);
+	if (!out) {
+		return NULL;
+	}
+
+	memset(out, 0x0, def->def_node_size);
+
+	out->n_type = type;
+
+	return out;
+}
+
+void ast_node_destroy(struct ast_node *node)
+{
+	assert(node->n_type < nr_ast_node_definitions);
+
+	struct ast_iterator it = {0};
+	ast_iterator_enqueue(&it, node);
+
+	while (1) {
+		node = ast_iterator_peek(&it);
+		if (!node) {
+			break;
+		}
+
+		const struct ast_node_definition *def
+			= ast_node_defintions[node->n_type];
+
+		if (def->def_cleanup) {
+			def->def_cleanup(node);
+		}
+
+		ast_iterator_dequeue(&it);
+		free(node);
+	}
+}
+
+void ast_node_iterate(struct ast_node *node, struct ast_iterator *it)
+{
+	ast_iterator_enqueue(it, node);
+}
+
+void ast_node_to_string(const struct ast_node *node, fx_bstr *out)
+{
+	const struct ast_node_definition *def
+		= ast_node_defintions[node->n_type];
+	if (def->def_to_string) {
+		def->def_to_string(node, out);
+	}
+}
+
+#define ENUM_STR(x)                                                            \
+	case x:                                                                \
+		return #x
+
+const char *ast_node_type_to_string(enum ast_node_type type)
+{
+	switch (type) {
+		ENUM_STR(AST_NONE);
+		ENUM_STR(AST_INT);
+		ENUM_STR(AST_DOUBLE);
+		ENUM_STR(AST_WORD);
+		ENUM_STR(AST_STRING);
+		ENUM_STR(AST_FSTRING);
+		ENUM_STR(AST_VAR);
+		ENUM_STR(AST_VAR_SPLAT);
+		ENUM_STR(AST_FLAG);
+		ENUM_STR(AST_CMDCALL);
+		ENUM_STR(AST_PIPELINE);
+		ENUM_STR(AST_REDIRECTION);
+	default:
+		return "<unknown>";
+	}
+}
+
+struct ast_node *ast_iterator_peek(struct ast_iterator *it)
+{
+	fx_queue_entry *cur = fx_queue_first(&it->it_queue);
+	if (!cur) {
+		return NULL;
+	}
+
+	return fx_unbox(struct ast_node, cur, n_it.e_entry);
+}
+
+struct ast_node *ast_iterator_dequeue(struct ast_iterator *it)
+{
+	fx_queue_entry *cur = fx_queue_first(&it->it_queue);
+	if (!cur) {
+		return NULL;
+	}
+
+	struct ast_node *node = fx_unbox(struct ast_node, cur, n_it.e_entry);
+	const struct ast_node_definition *def
+		= ast_node_defintions[node->n_type];
+
+	it->it_insert_after = cur;
+	if (def->def_collect_children) {
+		def->def_collect_children(node, it);
+	}
+
+	fx_queue_pop_front(&it->it_queue);
+	return fx_unbox(struct ast_node, cur, n_it.e_entry);
+}
+
+void ast_iterator_enqueue(struct ast_iterator *it, struct ast_node *node)
+{
+	unsigned long new_depth = 0;
+
+	fx_queue_entry *cur = fx_queue_first(&it->it_queue);
+	if (cur) {
+		struct ast_node *cur_node
+			= fx_unbox(struct ast_node, cur, n_it.e_entry);
+		new_depth = cur_node->n_it.e_depth + 1;
+	}
+
+	node->n_it.e_depth = new_depth;
+
+	if (!it->it_insert_after) {
+		fx_queue_push_back(&it->it_queue, &node->n_it.e_entry);
+		return;
+	}
+
+	fx_queue_insert_after(
+		&it->it_queue,
+		&node->n_it.e_entry,
+		it->it_insert_after);
+	it->it_insert_after = &node->n_it.e_entry;
+}

bshell/ast/ast.h

@@ -0,0 +1,126 @@
+#ifndef AST_H_
+#define AST_H_
+
+#include "../status.h"
+
+#include <fx/bstr.h>
+#include <fx/queue.h>
+
+struct lex_token;
+
+enum ast_node_type {
+	AST_NONE = 0x00u,
+	AST_INT,
+	AST_DOUBLE,
+	AST_WORD,
+	AST_STRING,
+	AST_FSTRING,
+	AST_VAR,
+	AST_VAR_SPLAT,
+	AST_FLAG,
+	AST_CMDCALL,
+	AST_PIPELINE,
+	AST_REDIRECTION,
+};
+
+struct ast_iterator_entry {
+	fx_queue_entry e_entry;
+	unsigned long e_depth;
+};
+
+struct ast_node {
+	enum ast_node_type n_type;
+	struct ast_node *n_parent;
+	fx_queue_entry n_entry;
+	struct ast_iterator_entry n_it;
+};
+
+struct int_ast_node {
+	struct ast_node n_base;
+	struct lex_token *n_value;
+};
+
+struct double_ast_node {
+	struct ast_node n_base;
+	struct lex_token *n_value;
+};
+
+struct word_ast_node {
+	struct ast_node n_base;
+	struct lex_token *n_value;
+};
+
+struct string_ast_node {
+	struct ast_node n_base;
+	struct lex_token *n_value;
+};
+
+struct fstring_ast_node {
+	struct ast_node n_base;
+	fx_queue n_elements;
+};
+
+struct var_ast_node {
+	struct ast_node n_base;
+	struct lex_token *n_ident;
+};
+
+struct var_splat_ast_node {
+	struct ast_node n_base;
+	struct lex_token *n_ident;
+};
+
+struct cmdcall_ast_node {
+	struct ast_node n_base;
+	fx_queue n_args;
+	fx_queue n_redirect;
+};
+
+struct pipeline_ast_node {
+	struct ast_node n_base;
+	fx_queue n_stages;
+};
+
+struct redirection_ast_node {
+	struct ast_node n_base;
+	bool n_append : 1;
+	bool n_out_is_fd : 1;
+	bool n_out_is_expr : 1;
+
+	unsigned int n_in, n_out;
+	struct ast_node *n_out_path_expr;
+	const char *n_out_path;
+	struct lex_token *n_out_tok;
+};
+
+struct ast_iterator {
+	struct ast_node *it_cur;
+	fx_queue it_queue;
+	unsigned int it_depth;
+	fx_queue_entry *it_insert_after;
+};
+
+struct ast_node_definition {
+	enum ast_node_type def_id;
+	size_t def_node_size;
+	enum bshell_status (*def_collect_children)(
+		struct ast_node *,
+		struct ast_iterator *);
+	enum bshell_status (*def_cleanup)(struct ast_node *);
+	void (*def_to_string)(const struct ast_node *, fx_bstr *);
+};
+
+extern struct ast_node *ast_node_create(enum ast_node_type type);
+extern void ast_node_destroy(struct ast_node *node);
+extern void ast_node_iterate(struct ast_node *node, struct ast_iterator *it);
+extern void ast_node_to_string(const struct ast_node *node, fx_bstr *out);
+
+extern const char *ast_node_type_to_string(enum ast_node_type type);
+
+extern struct ast_node *ast_iterator_peek(struct ast_iterator *it);
+extern struct ast_node *ast_iterator_dequeue(struct ast_iterator *it);
+extern void ast_iterator_enqueue(
+	struct ast_iterator *it,
+	struct ast_node *node);
+
+#endif

bshell/ast/cmdcall.c

@@ -0,0 +1,31 @@
+#include "ast.h"
+
+static enum bshell_status collect_children(
+	struct ast_node *node,
+	struct ast_iterator *it)
+{
+	struct cmdcall_ast_node *cmdcall = (struct cmdcall_ast_node *)node;
+	fx_queue_entry *cur = fx_queue_first(&cmdcall->n_args);
+	while (cur) {
+		struct ast_node *child
+			= fx_unbox(struct ast_node, cur, n_entry);
+		ast_iterator_enqueue(it, child);
+		cur = fx_queue_next(cur);
+	}
+
+	cur = fx_queue_first(&cmdcall->n_redirect);
+	while (cur) {
+		struct ast_node *child
+			= fx_unbox(struct ast_node, cur, n_entry);
+		ast_iterator_enqueue(it, child);
+		cur = fx_queue_next(cur);
+	}
+
+	return BSHELL_SUCCESS;
+}
+
+struct ast_node_definition cmdcall_ast_node = {
+	.def_id = AST_CMDCALL,
+	.def_node_size = sizeof(struct cmdcall_ast_node),
+	.def_collect_children = collect_children,
+};

bshell/ast/double.c

@@ -0,0 +1,6 @@
+#include "ast.h"
+
+struct ast_node_definition double_ast_node = {
+	.def_id = AST_DOUBLE,
+	.def_node_size = sizeof(struct double_ast_node),
+};

bshell/ast/fstring.c

@@ -0,0 +1,6 @@
+#include "ast.h"
+
+struct ast_node_definition fstring_ast_node = {
+	.def_id = AST_FSTRING,
+	.def_node_size = sizeof(struct fstring_ast_node),
+};

bshell/ast/int.c

@@ -0,0 +1,14 @@
+#include "../parse/token.h"
+#include "ast.h"
+
+static void to_string(const struct ast_node *node, fx_bstr *out)
+{
+	struct int_ast_node *i = (struct int_ast_node *)node;
+	fx_bstr_write_fmt(out, NULL, "%lld", i->n_value->tok_int);
+}
+
+struct ast_node_definition int_ast_node = {
+	.def_id = AST_INT,
+	.def_node_size = sizeof(struct int_ast_node),
+	.def_to_string = to_string,
+};

bshell/ast/pipeline.c

@@ -0,0 +1,23 @@
+#include "ast.h"
+
+static enum bshell_status collect_children(
+	struct ast_node *node,
+	struct ast_iterator *it)
+{
+	struct pipeline_ast_node *pipeline = (struct pipeline_ast_node *)node;
+	fx_queue_entry *cur = fx_queue_first(&pipeline->n_stages);
+	while (cur) {
+		struct ast_node *child
+			= fx_unbox(struct ast_node, cur, n_entry);
+		ast_iterator_enqueue(it, child);
+		cur = fx_queue_next(cur);
+	}
+
+	return BSHELL_SUCCESS;
+}
+
+struct ast_node_definition pipeline_ast_node = {
+	.def_id = AST_PIPELINE,
+	.def_node_size = sizeof(struct pipeline_ast_node),
+	.def_collect_children = collect_children,
+};

bshell/ast/redirection.c

@@ -0,0 +1,49 @@
+#include "ast.h"
+
+static enum bshell_status collect_children(
+	struct ast_node *node,
+	struct ast_iterator *it)
+{
+	struct redirection_ast_node *redirection
+		= (struct redirection_ast_node *)node;
+
+	if (redirection->n_out_path_expr) {
+		ast_iterator_enqueue(it, redirection->n_out_path_expr);
+	}
+
+	return BSHELL_SUCCESS;
+}
+
+static void to_string(const struct ast_node *node, fx_bstr *out)
+{
+	struct redirection_ast_node *redirection
+		= (struct redirection_ast_node *)node;
+	fx_bstr_write_fmt(out, NULL, "&%u", redirection->n_in);
+
+	if (redirection->n_append) {
+		fx_bstr_write_fmt(out, NULL, " >>");
+	} else {
+		fx_bstr_write_fmt(out, NULL, " >");
+	}
+
+	if (redirection->n_out_is_fd) {
+		fx_bstr_write_fmt(out, NULL, " &");
+	} else {
+		fx_bstr_write_fmt(out, NULL, " ");
+	}
+
+	if (redirection->n_out_is_expr) {
+		fx_bstr_write_fmt(out, NULL, "<expr>");
+	} else if (redirection->n_out_path) {
+		fx_bstr_write_fmt(out, NULL, "'%s'", redirection->n_out_path);
+	} else {
+		fx_bstr_write_fmt(out, NULL, "%u", redirection->n_out);
+	}
+}
+
+struct ast_node_definition redirection_ast_node = {
+	.def_id = AST_REDIRECTION,
+	.def_node_size = sizeof(struct redirection_ast_node),
+	.def_collect_children = collect_children,
+	.def_to_string = to_string,
+};

bshell/ast/string.c

@@ -0,0 +1,15 @@
+#include "../parse/token.h"
+#include "ast.h"
+
+static void to_string(const struct ast_node *node, fx_bstr *out)
+{
+	const struct string_ast_node *string
+		= (const struct string_ast_node *)node;
+	fx_bstr_write_fmt(out, NULL, "%s", string->n_value->tok_str);
+}
+
+struct ast_node_definition string_ast_node = {
+	.def_id = AST_STRING,
+	.def_node_size = sizeof(struct string_ast_node),
+	.def_to_string = to_string,
+};

bshell/ast/var.c

@@ -0,0 +1,14 @@
+#include "../parse/token.h"
+#include "ast.h"
+
+static void to_string(const struct ast_node *node, fx_bstr *out)
+{
+	const struct var_ast_node *var = (const struct var_ast_node *)node;
+	fx_bstr_write_fmt(out, NULL, "%s", var->n_ident->tok_str);
+}
+
+struct ast_node_definition var_ast_node = {
+	.def_id = AST_VAR,
+	.def_node_size = sizeof(struct var_ast_node),
+	.def_to_string = to_string,
+};

bshell/ast/word.c

@@ -0,0 +1,14 @@
+#include "../parse/token.h"
+#include "ast.h"
+
+static void to_string(const struct ast_node *node, fx_bstr *out)
+{
+	const struct word_ast_node *word = (const struct word_ast_node *)node;
+	fx_bstr_write_fmt(out, NULL, "%s", word->n_value->tok_str);
+}
+
+struct ast_node_definition word_ast_node = {
+	.def_id = AST_WORD,
+	.def_node_size = sizeof(struct word_ast_node),
+	.def_to_string = to_string,
+};

bshell/debug.c

@@ -0,0 +1,135 @@
+#include "debug.h"
+
+#include "ast/ast.h"
+#include "parse/token.h"
+
+#include <fx/string.h>
+#include <fx/term/print.h>
+#include <stdio.h>
+
+extern void print_lex_token(struct lex_token *tok)
+{
+	printf("[%lu:%lu - %lu:%lu] ",
+	       tok->tok_start.c_row,
+	       tok->tok_start.c_col,
+	       tok->tok_end.c_row,
+	       tok->tok_end.c_col);
+
+	switch (tok->tok_type) {
+	case TOK_KEYWORD:
+		fx_puts("[magenta]");
+		break;
+	case TOK_SYMBOL:
+		fx_puts("[blue]");
+		break;
+	case TOK_INT:
+	case TOK_DOUBLE:
+		fx_puts("[yellow]");
+		break;
+	case TOK_FLAG:
+		fx_puts("[red]");
+		break;
+	case TOK_WORD:
+	case TOK_VAR:
+	case TOK_VAR_SPLAT:
+		fx_puts("[cyan]");
+		break;
+	case TOK_STRING:
+		fx_puts("[green]");
+		break;
+	case TOK_STR_START:
+		fx_puts("[green]");
+		break;
+	case TOK_STR_END:
+		fx_puts("[green]");
+		break;
+	case TOK_LINEFEED:
+		fx_puts("[dark_grey]");
+		break;
+	default:
+		break;
+	}
+
+	fx_puts(token_type_to_string(tok->tok_type));
+
+	switch (tok->tok_type) {
+	case TOK_WORD:
+	case TOK_FLAG:
+	case TOK_STRING:
+	case TOK_VAR:
+	case TOK_VAR_SPLAT:
+		printf("(%s)", tok->tok_str);
+		break;
+	case TOK_SYMBOL:
+		printf("(%s)", token_symbol_to_string(tok->tok_symbol));
+		break;
+	case TOK_KEYWORD:
+		printf("(%s)", token_keyword_to_string(tok->tok_keyword));
+		break;
+	case TOK_INT:
+		printf("(%lld)", tok->tok_int);
+		break;
+	case TOK_DOUBLE:
+		printf("(%lf)", tok->tok_double);
+		break;
+	default:
+		break;
+	}
+
+	fx_puts("[reset]\n");
+}
+
+void print_ast_node(struct ast_node *node)
+{
+	struct ast_iterator it = {0};
+	ast_node_iterate(node, &it);
+
+	while (1) {
+		node = ast_iterator_peek(&it);
+		if (!node) {
+			break;
+		}
+
+		for (unsigned long i = 0; i < node->n_it.e_depth; i++) {
+			fx_puts("  ");
+		}
+
+		switch (node->n_type) {
+		case AST_REDIRECTION:
+		case AST_PIPELINE:
+			fx_puts("[blue]");
+			break;
+		case AST_CMDCALL:
+			fx_puts("[red]");
+			break;
+		case AST_INT:
+		case AST_DOUBLE:
+			fx_puts("[yellow]");
+			break;
+		case AST_WORD:
+			fx_puts("[cyan]");
+			break;
+		case AST_STRING:
+		case AST_FSTRING:
+			fx_puts("[green]");
+			break;
+		default:
+			break;
+		}
+
+		fx_printf("%s", ast_node_type_to_string(node->n_type));
+
+		char s[128] = {0};
+		fx_bstr str;
+		fx_bstr_begin(&str, s, sizeof s);
+		ast_node_to_string(node, &str);
+
+		if (fx_bstr_get_size(&str)) {
+			fx_printf("(%s)", fx_bstr_end(&str));
+		}
+
+		fx_printf("[reset]\n");
+
+		ast_iterator_dequeue(&it);
+	}
+}

bshell/debug.h

@@ -0,0 +1,12 @@
+#ifndef DEBUG_H_
+#define DEBUG_H_
+
+#include <stdbool.h>
+
+struct ast_node;
+struct lex_token;
+
+extern void print_lex_token(struct lex_token *tok);
+extern void print_ast_node(struct ast_node *node);
+
+#endif

bshell/parse/lex.h

@@ -0,0 +1,69 @@
+#ifndef LEX_H_
+#define LEX_H_
+
+#include "../status.h"
+
+#include <fx/queue.h>
+#include <fx/string.h>
+#include <fx/stringstream.h>
+
+struct lex_token;
+struct line_source;
+
+enum lex_flags {
+	LEX_PRINT_TOKENS = 0x01u,
+};
+
+enum lex_state_type_id {
+	LEX_STATE_STATEMENT = 0x01u,
+	LEX_STATE_EXPRESSION = 0x02u,
+	LEX_STATE_COMMAND = 0x04u,
+	LEX_STATE_ARITHMETIC = 0x08u,
+	LEX_STATE_STRING = 0x10u,
+};
+
+struct lex_token_def {
+	int id;
+	const char *name;
+	uint64_t name_hash;
+	enum lex_state_type_id enabled_states;
+};
+
+struct lex_symbol_node {
+	char s_char;
+	struct lex_token_def *s_def;
+
+	fx_queue_entry s_entry;
+	fx_queue s_children;
+};
+
+struct lex_state {
+	const struct lex_state_type *s_type;
+	unsigned int s_paren_depth;
+	fx_queue_entry s_entry;
+	fx_string *s_tempstr;
+};
+
+struct lex_ctx {
+	enum lex_flags lex_flags;
+	fx_queue lex_tokens;
+	struct line_source *lex_src;
+	fx_stringstream *lex_buf;
+	fx_string *lex_tmp;
+	fx_wchar lex_ch;
+	fx_queue lex_state;
+	struct lex_symbol_node *lex_sym_tree;
+	enum bshell_status lex_status;
+};
+
+extern enum bshell_status lex_ctx_init(
+	struct lex_ctx *ctx,
+	enum lex_flags flags,
+	struct line_source *src);
+extern enum bshell_status lex_ctx_cleanup(struct lex_ctx *ctx);
+
+extern struct lex_token *lex_ctx_peek(struct lex_ctx *ctx);
+extern struct lex_token *lex_ctx_claim(struct lex_ctx *ctx);
+extern void lex_ctx_discard(struct lex_ctx *ctx);
+
+#endif

bshell/parse/lex/arithmetic.c

@@ -0,0 +1,136 @@
+#include "lex-internal.h"
+
+static enum bshell_status arithmetic_symbol(struct lex_ctx *ctx)
+{
+	const struct lex_token_def *sym = NULL;
+	enum bshell_status status = read_symbol(ctx, &sym);
+
+	if (status != BSHELL_SUCCESS) {
+		return status;
+	}
+
+	struct lex_token *tok = NULL;
+	switch (sym->id) {
+	case SYM_SQUOTE:
+		status = read_literal_string(ctx, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+		enqueue_token(ctx, tok);
+		return BSHELL_SUCCESS;
+
+	case SYM_HASH:
+		return read_line_comment(ctx);
+	case SYM_DQUOTE:
+		if (!lex_state_push(ctx, LEX_STATE_STRING)) {
+			return BSHELL_ERR_NO_MEMORY;
+		}
+
+		return BSHELL_SUCCESS;
+	case SYM_DOLLAR:
+		status = read_var(ctx, TOK_VAR, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		enqueue_token(ctx, tok);
+		return status;
+	case SYM_AT:
+		status = read_var(ctx, TOK_VAR_SPLAT, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		enqueue_token(ctx, tok);
+		return status;
+	case SYM_DOLLAR_LEFT_BRACE:
+		status = read_braced_var(ctx, TOK_VAR, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		enqueue_token(ctx, tok);
+		return status;
+	case SYM_AT_LEFT_BRACE:
+		status = read_braced_var(ctx, TOK_VAR_SPLAT, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		enqueue_token(ctx, tok);
+		return status;
+	default:
+		break;
+	}
+
+	push_symbol(ctx, sym->id);
+
+	switch (sym->id) {
+	case SYM_LEFT_PAREN:
+		lex_state_push(ctx, LEX_STATE_EXPRESSION);
+		return BSHELL_SUCCESS;
+	case SYM_DOLLAR_LEFT_PAREN:
+		lex_state_push(ctx, LEX_STATE_STATEMENT);
+		return BSHELL_SUCCESS;
+	case SYM_RIGHT_PAREN:
+		lex_state_pop(ctx);
+		return BSHELL_SUCCESS;
+	case SYM_SEMICOLON:
+		lex_state_change(ctx, LEX_STATE_STATEMENT);
+		return BSHELL_SUCCESS;
+	default:
+		break;
+	}
+
+	return BSHELL_SUCCESS;
+}
+
+static enum bshell_status arithmetic_word(struct lex_ctx *ctx)
+{
+	struct lex_token *word = NULL;
+	enum bshell_status status = read_word(ctx, &word);
+	if (status != BSHELL_SUCCESS) {
+		return status;
+	}
+
+	bool converted = convert_word_to_keyword(word);
+	if (!converted) {
+		converted = convert_word_to_int(word);
+	}
+
+	enqueue_token(ctx, word);
+	return BSHELL_SUCCESS;
+}
+
+static enum bshell_status arithmetic_pump_token(struct lex_ctx *ctx)
+{
+	fx_wchar c = peek_char(ctx);
+	bool newline = false;
+
+	while (fx_wchar_is_space(c)) {
+		if (c == '\n') {
+			newline = true;
+		}
+
+		advance_char_noread(ctx);
+		c = peek_char_noread(ctx);
+	}
+
+	if (newline) {
+		struct lex_token *tok = lex_token_create(TOK_LINEFEED);
+		enqueue_token(ctx, tok);
+		lex_state_change(ctx, LEX_STATE_STATEMENT);
+		return BSHELL_SUCCESS;
+	}
+
+	if (char_can_begin_symbol(ctx, c)) {
+		return arithmetic_symbol(ctx);
+	}
+
+	return arithmetic_word(ctx);
+}
+
+const struct lex_state_type lex_arithmetic_state = {
+	.s_id = LEX_STATE_ARITHMETIC,
+	.s_pump_token = arithmetic_pump_token,
+};

bshell/parse/lex/command.c

@@ -0,0 +1,131 @@
+#include "lex-internal.h"
+
+static enum bshell_status command_symbol(struct lex_ctx *ctx)
+{
+	const struct lex_token_def *sym = NULL;
+	enum bshell_status status = read_symbol(ctx, &sym);
+
+	if (status != BSHELL_SUCCESS) {
+		return status;
+	}
+
+	struct lex_token *tok = NULL;
+	switch (sym->id) {
+	case SYM_SQUOTE:
+		status = read_literal_string(ctx, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+		enqueue_token(ctx, tok);
+		return BSHELL_SUCCESS;
+
+	case SYM_HASH:
+		return read_line_comment(ctx);
+	case SYM_DQUOTE:
+		if (!lex_state_push(ctx, LEX_STATE_STRING)) {
+			return BSHELL_ERR_NO_MEMORY;
+		}
+
+		return BSHELL_SUCCESS;
+	case SYM_DOLLAR:
+		status = read_var(ctx, TOK_VAR, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		enqueue_token(ctx, tok);
+		return status;
+	case SYM_AT:
+		status = read_var(ctx, TOK_VAR_SPLAT, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		enqueue_token(ctx, tok);
+		return status;
+	case SYM_DOLLAR_LEFT_BRACE:
+		status = read_braced_var(ctx, TOK_VAR, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		enqueue_token(ctx, tok);
+		return status;
+	case SYM_AT_LEFT_BRACE:
+		status = read_braced_var(ctx, TOK_VAR_SPLAT, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		enqueue_token(ctx, tok);
+		return status;
+	default:
+		break;
+	}
+
+	push_symbol(ctx, sym->id);
+
+	switch (sym->id) {
+	case SYM_LEFT_PAREN:
+		lex_state_push(ctx, LEX_STATE_EXPRESSION);
+		return BSHELL_SUCCESS;
+	case SYM_DOLLAR_LEFT_PAREN:
+		lex_state_push(ctx, LEX_STATE_STATEMENT);
+		return BSHELL_SUCCESS;
+	case SYM_RIGHT_PAREN:
+		lex_state_pop(ctx);
+		return BSHELL_SUCCESS;
+	case SYM_SEMICOLON:
+		lex_state_change(ctx, LEX_STATE_STATEMENT);
+		return BSHELL_SUCCESS;
+	default:
+		break;
+	}
+
+	return BSHELL_SUCCESS;
+}
+
+static enum bshell_status command_word(struct lex_ctx *ctx)
+{
+	struct lex_token *word = NULL;
+	enum bshell_status status = read_word(ctx, &word);
+	if (status != BSHELL_SUCCESS) {
+		return status;
+	}
+
+	enqueue_token(ctx, word);
+	return BSHELL_SUCCESS;
+}
+
+enum bshell_status command_pump_token(struct lex_ctx *ctx)
+{
+	fx_wchar c = peek_char(ctx);
+	bool newline = false;
+
+	while (fx_wchar_is_space(c)) {
+		if (c == '\n') {
+			newline = true;
+		}
+
+		advance_char_noread(ctx);
+		c = peek_char_noread(ctx);
+	}
+
+	if (newline) {
+		struct lex_token *tok = lex_token_create(TOK_LINEFEED);
+		enqueue_token(ctx, tok);
+		lex_state_change(ctx, LEX_STATE_STATEMENT);
+		return BSHELL_SUCCESS;
+	}
+
+	if (char_can_begin_symbol(ctx, c)) {
+		return command_symbol(ctx);
+	}
+
+	return command_word(ctx);
+}
+
+const struct lex_state_type lex_command_state = {
+	.s_id = LEX_STATE_COMMAND,
+	.s_pump_token = command_pump_token,
+};

bshell/parse/lex/expression.c

@@ -0,0 +1,134 @@
+#include "lex-internal.h"
+
+static enum bshell_status expression_symbol(struct lex_ctx *ctx)
+{
+	const struct lex_token_def *sym = NULL;
+	enum bshell_status status = read_symbol(ctx, &sym);
+
+	if (status != BSHELL_SUCCESS) {
+		return status;
+	}
+
+	struct lex_token *tok = NULL;
+
+	switch (sym->id) {
+	case SYM_DQUOTE:
+		if (!lex_state_push(ctx, LEX_STATE_STRING)) {
+			return BSHELL_ERR_NO_MEMORY;
+		}
+
+		return BSHELL_SUCCESS;
+	case SYM_DOLLAR:
+		status = read_var(ctx, TOK_VAR, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		enqueue_token(ctx, tok);
+		lex_state_change(ctx, LEX_STATE_ARITHMETIC);
+		return status;
+	case SYM_AT:
+		status = read_var(ctx, TOK_VAR_SPLAT, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		enqueue_token(ctx, tok);
+		lex_state_change(ctx, LEX_STATE_ARITHMETIC);
+		return status;
+	case SYM_DOLLAR_LEFT_BRACE:
+		status = read_braced_var(ctx, TOK_VAR, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		enqueue_token(ctx, tok);
+		lex_state_change(ctx, LEX_STATE_ARITHMETIC);
+		return status;
+	case SYM_AT_LEFT_BRACE:
+		status = read_braced_var(ctx, TOK_VAR_SPLAT, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		enqueue_token(ctx, tok);
+		lex_state_change(ctx, LEX_STATE_ARITHMETIC);
+		return status;
+	default:
+		break;
+	}
+
+	push_symbol(ctx, sym->id);
+
+	switch (sym->id) {
+	case SYM_LEFT_PAREN:
+		lex_state_push(ctx, LEX_STATE_EXPRESSION);
+		return BSHELL_SUCCESS;
+	case SYM_DOLLAR_LEFT_PAREN:
+		lex_state_push(ctx, LEX_STATE_STATEMENT);
+		return BSHELL_SUCCESS;
+	case SYM_RIGHT_PAREN:
+		lex_state_pop(ctx);
+		return BSHELL_SUCCESS;
+	case SYM_SEMICOLON:
+		lex_state_change(ctx, LEX_STATE_STATEMENT);
+		return BSHELL_SUCCESS;
+	default:
+		break;
+	}
+
+	return BSHELL_SUCCESS;
+}
+
+static enum bshell_status expression_word(struct lex_ctx *ctx)
+{
+	struct lex_token *word = NULL;
+	enum bshell_status status = read_word(ctx, &word);
+	if (status != BSHELL_SUCCESS) {
+		return status;
+	}
+
+	bool converted = convert_word_to_int(word);
+
+	if (converted) {
+		lex_state_change(ctx, LEX_STATE_ARITHMETIC);
+	} else {
+		lex_state_change(ctx, LEX_STATE_COMMAND);
+	}
+
+	enqueue_token(ctx, word);
+	return BSHELL_SUCCESS;
+}
+
+static enum bshell_status expression_pump_token(struct lex_ctx *ctx)
+{
+	fx_wchar c = peek_char(ctx);
+	bool newline = false;
+
+	while (fx_wchar_is_space(c)) {
+		if (c == '\n') {
+			newline = true;
+		}
+
+		advance_char_noread(ctx);
+		c = peek_char_noread(ctx);
+	}
+
+	if (newline) {
+		struct lex_token *tok = lex_token_create(TOK_LINEFEED);
+		enqueue_token(ctx, tok);
+		lex_state_change(ctx, LEX_STATE_STATEMENT);
+		return BSHELL_SUCCESS;
+	}
+
+	if (char_can_begin_symbol(ctx, c)) {
+		return expression_symbol(ctx);
+	}
+
+	return expression_word(ctx);
+}
+
+const struct lex_state_type lex_expression_state = {
+	.s_id = LEX_STATE_EXPRESSION,
+	.s_pump_token = expression_pump_token,
+};

bshell/parse/lex/lex-internal.h

@@ -0,0 +1,75 @@
+#ifndef PARSE_LEX_INTERNAL_H_
+#define PARSE_LEX_INTERNAL_H_
+
+#include "../../status.h"
+#include "../lex.h"
+#include "../token.h"
+
+struct lex_ctx;
+
+typedef enum bshell_status (*lex_state_pump_token)(struct lex_ctx *);
+typedef enum bshell_status (*lex_state_begin)(struct lex_ctx *);
+typedef enum bshell_status (*lex_state_end)(struct lex_ctx *);
+
+struct lex_state_type {
+	enum lex_state_type_id s_id;
+	lex_state_pump_token s_pump_token;
+	lex_state_begin s_begin;
+	lex_state_end s_end;
+};
+
+extern enum bshell_status pump_token_statement(struct lex_ctx *ctx);
+extern enum bshell_status pump_token_expression(struct lex_ctx *ctx);
+extern enum bshell_status pump_token_command(struct lex_ctx *ctx);
+extern enum bshell_status pump_token_arithmetic(struct lex_ctx *ctx);
+extern enum bshell_status pump_token_string(struct lex_ctx *ctx);
+
+extern struct lex_state *lex_state_push(
+	struct lex_ctx *ctx,
+	enum lex_state_type_id state_type);
+extern void lex_state_pop(struct lex_ctx *ctx);
+extern struct lex_state *lex_state_get(struct lex_ctx *ctx);
+extern void lex_state_change(struct lex_ctx *ctx, enum lex_state_type_id type);
+extern fx_string *lex_state_get_tempstr(struct lex_ctx *ctx);
+
+extern fx_wchar peek_char(struct lex_ctx *ctx);
+extern fx_wchar peek_char_noread(struct lex_ctx *ctx);
+extern void advance_char(struct lex_ctx *ctx);
+extern void advance_char_noread(struct lex_ctx *ctx);
+
+extern bool string_is_valid_number(const char *s, long long *out);
+extern bool convert_word_to_int(struct lex_token *tok);
+extern bool convert_word_to_keyword(struct lex_token *tok);
+
+extern void enqueue_token(struct lex_ctx *ctx, struct lex_token *tok);
+
+extern enum bshell_status read_word(
+	struct lex_ctx *ctx,
+	struct lex_token **out);
+extern enum bshell_status read_symbol(
+	struct lex_ctx *ctx,
+	const struct lex_token_def **out);
+extern enum bshell_status read_literal_string(
+	struct lex_ctx *ctx,
+	struct lex_token **out);
+extern enum bshell_status read_line_comment(struct lex_ctx *lex);
+extern enum bshell_status read_var(
+	struct lex_ctx *ctx,
+	enum token_type type,
+	struct lex_token **out);
+extern enum bshell_status read_braced_var(
+	struct lex_ctx *ctx,
+	enum token_type type,
+	struct lex_token **out);
+
+extern enum bshell_status push_symbol(
+	struct lex_ctx *ctx,
+	enum token_symbol sym);
+
+extern bool char_can_begin_symbol(struct lex_ctx *ctx, char c);
+extern bool char_can_begin_symbol_in_state(
+	struct lex_ctx *ctx,
+	char c,
+	enum lex_state_type_id state_type);
+
+#endif

bshell/parse/lex/statement.c

@@ -0,0 +1,162 @@
+#include "lex-internal.h"
+
+static enum bshell_status statement_symbol(struct lex_ctx *ctx)
+{
+	const struct lex_token_def *sym = NULL;
+	enum bshell_status status = read_symbol(ctx, &sym);
+
+	if (status != BSHELL_SUCCESS) {
+		return status;
+	}
+
+	struct lex_token *tok = NULL;
+	switch (sym->id) {
+	case SYM_SQUOTE:
+		status = read_literal_string(ctx, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+		enqueue_token(ctx, tok);
+		return BSHELL_SUCCESS;
+
+	case SYM_HASH:
+		return read_line_comment(ctx);
+	case SYM_DQUOTE:
+		if (!lex_state_push(ctx, LEX_STATE_STRING)) {
+			return BSHELL_ERR_NO_MEMORY;
+		}
+
+		return BSHELL_SUCCESS;
+	case SYM_DOLLAR:
+		if (!lex_state_push(ctx, LEX_STATE_ARITHMETIC)) {
+			return BSHELL_ERR_NO_MEMORY;
+		}
+
+		status = read_var(ctx, TOK_VAR, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		enqueue_token(ctx, tok);
+		return status;
+	case SYM_AT:
+		if (!lex_state_push(ctx, LEX_STATE_ARITHMETIC)) {
+			return BSHELL_ERR_NO_MEMORY;
+		}
+
+		status = read_var(ctx, TOK_VAR_SPLAT, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		enqueue_token(ctx, tok);
+		return status;
+	case SYM_DOLLAR_LEFT_BRACE:
+		if (!lex_state_push(ctx, LEX_STATE_ARITHMETIC)) {
+			return BSHELL_ERR_NO_MEMORY;
+		}
+
+		status = read_braced_var(ctx, TOK_VAR, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		enqueue_token(ctx, tok);
+		return status;
+	case SYM_AT_LEFT_BRACE:
+		if (!lex_state_push(ctx, LEX_STATE_ARITHMETIC)) {
+			return BSHELL_ERR_NO_MEMORY;
+		}
+
+		status = read_braced_var(ctx, TOK_VAR_SPLAT, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		enqueue_token(ctx, tok);
+		return status;
+	default:
+		break;
+	}
+
+	push_symbol(ctx, sym->id);
+
+	switch (sym->id) {
+	case SYM_LEFT_PAREN:
+		lex_state_push(ctx, LEX_STATE_EXPRESSION);
+		return BSHELL_SUCCESS;
+	case SYM_LEFT_BRACE:
+	case SYM_DOLLAR_LEFT_PAREN:
+		lex_state_push(ctx, LEX_STATE_STATEMENT);
+		return BSHELL_SUCCESS;
+	case SYM_RIGHT_PAREN:
+	case SYM_RIGHT_BRACE:
+		lex_state_pop(ctx);
+		return BSHELL_SUCCESS;
+	default:
+		break;
+	}
+
+	if (sym->enabled_states & LEX_STATE_COMMAND) {
+		lex_state_change(ctx, LEX_STATE_COMMAND);
+	} else if (sym->enabled_states & LEX_STATE_ARITHMETIC) {
+		lex_state_change(ctx, LEX_STATE_ARITHMETIC);
+	}
+
+	return BSHELL_SUCCESS;
+}
+
+static enum bshell_status statement_word(struct lex_ctx *ctx)
+{
+	struct lex_token *word = NULL;
+	enum bshell_status status = read_word(ctx, &word);
+	if (status != BSHELL_SUCCESS) {
+		return status;
+	}
+
+	bool converted = convert_word_to_keyword(word);
+	if (!converted) {
+		converted = convert_word_to_int(word);
+	}
+
+	if (converted) {
+		lex_state_change(ctx, LEX_STATE_ARITHMETIC);
+	} else {
+		lex_state_change(ctx, LEX_STATE_COMMAND);
+	}
+
+	enqueue_token(ctx, word);
+	return BSHELL_SUCCESS;
+}
+
+static enum bshell_status statement_pump_token(struct lex_ctx *ctx)
+{
+	fx_wchar c = peek_char(ctx);
+	bool newline = false;
+
+	while (fx_wchar_is_space(c)) {
+		if (c == '\n') {
+			newline = true;
+		}
+
+		advance_char_noread(ctx);
+		c = peek_char_noread(ctx);
+	}
+
+	if (newline) {
+		struct lex_token *tok = lex_token_create(TOK_LINEFEED);
+		enqueue_token(ctx, tok);
+		return BSHELL_SUCCESS;
+	}
+
+	if (char_can_begin_symbol(ctx, c)) {
+		return statement_symbol(ctx);
+	}
+
+	return statement_word(ctx);
+}
+
+const struct lex_state_type lex_statement_state = {
+	.s_id = LEX_STATE_STATEMENT,
+	.s_pump_token = statement_pump_token,
+};

bshell/parse/lex/string.c

@@ -0,0 +1,136 @@
+#include "lex-internal.h"
+
+static enum bshell_status string_symbol(struct lex_ctx *ctx)
+{
+	const struct lex_token_def *sym = NULL;
+	enum bshell_status status = read_symbol(ctx, &sym);
+
+	if (status != BSHELL_SUCCESS) {
+		return status;
+	}
+
+	struct lex_token *tok = NULL;
+
+	switch (sym->id) {
+	case SYM_DOLLAR_LEFT_PAREN:
+		status = push_symbol(ctx, sym->id);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		lex_state_push(ctx, LEX_STATE_STATEMENT);
+		return BSHELL_SUCCESS;
+	case SYM_DQUOTE:
+		lex_state_pop(ctx);
+		return BSHELL_SUCCESS;
+	case SYM_DOLLAR:
+		status = read_var(ctx, TOK_VAR, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		enqueue_token(ctx, tok);
+		return status;
+	case SYM_AT:
+		status = read_var(ctx, TOK_VAR_SPLAT, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		enqueue_token(ctx, tok);
+		return status;
+	case SYM_DOLLAR_LEFT_BRACE:
+		status = read_braced_var(ctx, TOK_VAR, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		enqueue_token(ctx, tok);
+		return status;
+	case SYM_AT_LEFT_BRACE:
+		status = read_braced_var(ctx, TOK_VAR_SPLAT, &tok);
+		if (status != BSHELL_SUCCESS) {
+			return status;
+		}
+
+		enqueue_token(ctx, tok);
+		return status;
+	default:
+		break;
+	}
+
+	return BSHELL_ERR_BAD_SYNTAX;
+}
+
+static enum bshell_status string_content(struct lex_ctx *ctx)
+{
+	fx_wchar c = FX_WCHAR_INVALID;
+	fx_string *temp = lex_state_get_tempstr(ctx);
+	fx_string_clear(temp);
+
+	while (1) {
+		c = peek_char(ctx);
+		if (c == FX_WCHAR_INVALID) {
+			/* EOF without end of string */
+			ctx->lex_status = BSHELL_ERR_BAD_SYNTAX;
+		}
+
+		if (char_can_begin_symbol(ctx, c)) {
+			break;
+		}
+
+		fx_string_append_wc(temp, c);
+		advance_char(ctx);
+	}
+
+	if (fx_string_get_size(temp, FX_STRLEN_NORMAL) == 0) {
+		return BSHELL_SUCCESS;
+	}
+
+	struct lex_token *tok = lex_token_create_with_string(
+		TOK_STRING,
+		fx_string_get_cstr(temp));
+	enqueue_token(ctx, tok);
+
+	return BSHELL_SUCCESS;
+}
+
+static enum bshell_status string_begin(struct lex_ctx *ctx)
+{
+	struct lex_token *tok = lex_token_create(TOK_STR_START);
+	if (!tok) {
+		return BSHELL_ERR_NO_MEMORY;
+	}
+
+	enqueue_token(ctx, tok);
+	return BSHELL_SUCCESS;
+}
+
+static enum bshell_status string_end(struct lex_ctx *ctx)
+{
+	struct lex_token *tok = lex_token_create(TOK_STR_END);
+	if (!tok) {
+		return BSHELL_ERR_NO_MEMORY;
+	}
+
+	enqueue_token(ctx, tok);
+	return BSHELL_SUCCESS;
+}
+
+static enum bshell_status string_pump_token(struct lex_ctx *ctx)
+{
+	fx_wchar c = peek_char(ctx);
+
+	if (char_can_begin_symbol(ctx, c)) {
+		return string_symbol(ctx);
+	}
+
+	return string_content(ctx);
+}
+
+const struct lex_state_type lex_string_state = {
+	.s_id = LEX_STATE_STRING,
+	.s_begin = string_begin,
+	.s_end = string_end,
+	.s_pump_token = string_pump_token,
+};

bshell/parse/parse.c

@@ -0,0 +1,30 @@
+#include "parse.h"
+
+#include "../ast/ast.h"
+#include "lex.h"
+#include "syntax.h"
+#include "token.h"
+
+#include <stdio.h>
+#include <string.h>
+
+enum bshell_status parse_ctx_init(struct parse_ctx *ctx, struct lex_ctx *src)
+{
+	memset(ctx, 0x0, sizeof *ctx);
+
+	ctx->p_src = src;
+
+	return BSHELL_SUCCESS;
+}
+
+void parse_ctx_cleanup(struct parse_ctx *ctx)
+{
+}
+
+struct ast_node *parse_ctx_read_node(struct parse_ctx *ctx)
+{
+	struct ast_node *result = NULL;
+	bool ok = parse_statement(ctx, &result);
+
+	return ok ? result : NULL;
+}

bshell/parse/parse.h

@@ -0,0 +1,21 @@
+#ifndef PARSE_H_
+#define PARSE_H_
+
+#include "../status.h"
+
+struct lex_ctx;
+struct ast_node;
+
+struct parse_ctx {
+	struct lex_ctx *p_src;
+	enum bshell_status p_status;
+};
+
+extern enum bshell_status parse_ctx_init(
+	struct parse_ctx *ctx,
+	struct lex_ctx *src);
+extern void parse_ctx_cleanup(struct parse_ctx *ctx);
+
+extern struct ast_node *parse_ctx_read_node(struct parse_ctx *ctx);
+
+#endif

bshell/parse/syntax.h

@@ -0,0 +1,48 @@
+#ifndef PARSE_SYNTAX_H_
+#define PARSE_SYNTAX_H_
+
+#include "../ast/ast.h"
+#include "lex.h"
+#include "parse.h"
+#include "token.h"
+
+#include <stdbool.h>
+
+enum parse_operand_flags {
+	OPERAND_BASIC = 0x01u,
+};
+
+extern struct lex_token *peek_token(struct parse_ctx *ctx);
+extern enum token_type peek_token_type(struct parse_ctx *ctx);
+extern enum token_keyword peek_unknown_keyword(struct parse_ctx *ctx);
+extern enum token_symbol peek_unknown_symbol(struct parse_ctx *ctx);
+
+extern struct lex_token *claim_token(struct parse_ctx *ctx);
+extern void discard_token(struct parse_ctx *ctx);
+
+extern bool peek_linefeed(struct parse_ctx *ctx);
+extern bool peek_symbol(struct parse_ctx *ctx, enum token_symbol sym);
+extern bool peek_word(struct parse_ctx *ctx, struct lex_token **out);
+extern bool peek_int(struct parse_ctx *ctx);
+
+extern bool parse_linefeed(struct parse_ctx *ctx);
+extern bool parse_symbol(struct parse_ctx *ctx, enum token_symbol sym);
+extern bool parse_keyword(struct parse_ctx *ctx, enum token_keyword kw);
+extern bool parse_int(struct parse_ctx *ctx, long long *out);
+extern bool parse_flag(struct parse_ctx *ctx, struct lex_token **out);
+
+extern bool peek_arith_expr(struct parse_ctx *ctx);
+extern bool parse_arith_expr(struct parse_ctx *ctx, struct ast_node **out);
+extern bool parse_operand(
+	struct parse_ctx *ctx,
+	enum parse_operand_flags flags,
+	struct ast_node **out);
+
+extern bool parse_statement(struct parse_ctx *ctx, struct ast_node **out);
+
+extern bool peek_command(struct parse_ctx *ctx);
+extern bool parse_command(struct parse_ctx *ctx, struct ast_node **out);
+extern bool parse_cmdcall(struct parse_ctx *ctx, struct ast_node **out);
+extern bool parse_redirect(struct parse_ctx *ctx, struct ast_node **out);
+
+#endif

bshell/parse/syntax/arith.c

@@ -0,0 +1,27 @@
+#include "../syntax.h"
+
+bool peek_arith_expr(struct parse_ctx *ctx)
+{
+	switch (peek_token_type(ctx)) {
+	case TOK_SYMBOL:
+		switch (peek_unknown_symbol(ctx)) {
+		case SYM_PLUS:
+		case SYM_HYPHEN:
+			return true;
+		default:
+			return false;
+		}
+	case TOK_INT:
+	case TOK_DOUBLE:
+	case TOK_STRING:
+	case TOK_STR_START:
+		return true;
+	default:
+		return false;
+	}
+}
+
+bool parse_arith_expr(struct parse_ctx *ctx, struct ast_node **out)
+{
+	return false;
+}

bshell/parse/syntax/command.c

@@ -0,0 +1,419 @@
+#include "../syntax.h"
+
+#include <fx/encoding.h>
+
+static bool parse_cmdcall_arg(struct parse_ctx *ctx, struct ast_node **out)
+{
+	if (ctx->p_status != BSHELL_SUCCESS) {
+		return false;
+	}
+
+	struct lex_token *tok = peek_token(ctx);
+	if (!tok) {
+		return false;
+	}
+
+	struct ast_node *arg = NULL;
+
+	switch (tok->tok_type) {
+	case TOK_WORD: {
+		struct word_ast_node *n
+			= (struct word_ast_node *)ast_node_create(AST_WORD);
+		if (!n) {
+			ctx->p_status = BSHELL_ERR_NO_MEMORY;
+			return false;
+		}
+
+		n->n_value = claim_token(ctx);
+		*out = (struct ast_node *)n;
+		return true;
+	}
+
+#if 0
+	case TOK_FLAG: {
+		struct word_ast_node *n
+			= (struct word_ast_node *)ast_node_create(AST_WORD);
+		if (!n) {
+			ctx->p_status = BSHELL_ERR_NO_MEMORY;
+			return false;
+		}
+
+		n->n_value = claim_token(ctx);
+		*out = (struct ast_node *)n;
+		return true;
+	}
+#endif
+
+	case TOK_VAR: {
+		struct var_ast_node *n
+			= (struct var_ast_node *)ast_node_create(AST_VAR);
+		if (!n) {
+			ctx->p_status = BSHELL_ERR_NO_MEMORY;
+			return false;
+		}
+
+		n->n_ident = claim_token(ctx);
+		*out = (struct ast_node *)n;
+		return true;
+	}
+
+	case TOK_VAR_SPLAT: {
+		struct var_splat_ast_node *n
+			= (struct var_splat_ast_node *)ast_node_create(
+				AST_VAR_SPLAT);
+		if (!n) {
+			ctx->p_status = BSHELL_ERR_NO_MEMORY;
+			return false;
+		}
+
+		n->n_ident = claim_token(ctx);
+		*out = (struct ast_node *)n;
+		return true;
+	}
+
+	case TOK_STRING: {
+		struct string_ast_node *n
+			= (struct string_ast_node *)ast_node_create(AST_STRING);
+		if (!n) {
+			ctx->p_status = BSHELL_ERR_NO_MEMORY;
+			return false;
+		}
+
+		n->n_value = claim_token(ctx);
+		*out = (struct ast_node *)n;
+		return true;
+	}
+
+	default:
+		return false;
+	}
+
+	return true;
+}
+
+static bool parse_redirect_to_fd(
+	struct parse_ctx *ctx,
+	unsigned int in_fd,
+	bool append,
+	struct ast_node **out)
+{
+	if (ctx->p_status != BSHELL_SUCCESS) {
+		return false;
+	}
+
+	struct redirection_ast_node *redirect
+		= (struct redirection_ast_node *)ast_node_create(
+			AST_REDIRECTION);
+
+	redirect->n_in = in_fd;
+	redirect->n_append = append;
+
+	if (!parse_symbol(ctx, SYM_AMPERSAND)) {
+		ast_node_destroy((struct ast_node *)redirect);
+		return false;
+	}
+
+	struct lex_token *out_tok = NULL;
+	struct ast_node *out_expr = NULL;
+	long long out_fd = -1;
+
+	if (peek_word(ctx, &out_tok)) {
+		const char *s = out_tok->tok_str;
+		char *ep;
+		out_fd = strtoll(s, &ep, 10);
+		if (*ep == '\0') {
+			discard_token(ctx);
+			out_tok = NULL;
+		} else {
+			out_fd = -1;
+		}
+	} else if (!parse_cmdcall_arg(ctx, &out_expr)) {
+		return false;
+	}
+
+	redirect->n_out_is_fd = (out_fd >= 0) || out_expr;
+	redirect->n_out_is_expr = out_expr != NULL;
+	redirect->n_out = (unsigned int)out_fd;
+	redirect->n_out_path_expr = out_expr;
+	if (out_tok) {
+		redirect->n_out_tok = claim_token(ctx);
+		redirect->n_out_path = out_tok->tok_str;
+	}
+
+	*out = (struct ast_node *)redirect;
+	return true;
+}
+
+static bool parse_redirect_to_file_squashed(
+	struct parse_ctx *ctx,
+	unsigned int in_fd,
+	bool append,
+	const char *str,
+	struct ast_node **out)
+{
+	if (ctx->p_status != BSHELL_SUCCESS) {
+		return false;
+	}
+
+	struct lex_token *tok = peek_token(ctx);
+	if (*str == '\0') {
+		return false;
+	}
+
+	struct redirection_ast_node *redirect
+		= (struct redirection_ast_node *)ast_node_create(
+			AST_REDIRECTION);
+
+	redirect->n_in = in_fd;
+	redirect->n_append = append;
+	redirect->n_out_is_fd = false;
+	redirect->n_out_is_expr = false;
+	redirect->n_out_path = str;
+
+	redirect->n_out_tok = claim_token(ctx);
+
+	*out = (struct ast_node *)redirect;
+	return true;
+}
+
+static bool parse_redirect_to_file_separate(
+	struct parse_ctx *ctx,
+	unsigned int in_fd,
+	bool append,
+	struct ast_node **out)
+{
+	if (ctx->p_status != BSHELL_SUCCESS) {
+		return false;
+	}
+
+	struct ast_node *out_path = NULL;
+	if (!parse_cmdcall_arg(ctx, &out_path)) {
+		ctx->p_status = BSHELL_ERR_BAD_SYNTAX;
+		return false;
+	}
+
+	struct redirection_ast_node *redirect
+		= (struct redirection_ast_node *)ast_node_create(
+			AST_REDIRECTION);
+
+	redirect->n_in = in_fd;
+	redirect->n_append = append;
+	redirect->n_out_is_fd = false;
+	redirect->n_out_is_expr = true;
+	redirect->n_out_path_expr = out_path;
+
+	*out = (struct ast_node *)redirect;
+	return true;
+}
+
+bool parse_redirect(struct parse_ctx *ctx, struct ast_node **out)
+{
+	struct lex_token *tok = peek_token(ctx);
+	if (!tok || tok->tok_type != TOK_WORD) {
+		return false;
+	}
+
+	unsigned int in_fd = 1;
+	const char *str = tok->tok_str;
+	bool append = false;
+
+	if (fx_wchar_is_number(*str)) {
+		in_fd = *str - '0';
+		str++;
+	}
+
+	if (*str != '>') {
+		return false;
+	}
+
+	str++;
+	if (*str == '>') {
+		append = true;
+		str++;
+	}
+
+	if (*str != '\0') {
+		return parse_redirect_to_file_squashed(
+			ctx,
+			in_fd,
+			append,
+			str,
+			out);
+	}
+
+	discard_token(ctx);
+
+	if (parse_redirect_to_fd(ctx, in_fd, append, out)) {
+		return true;
+	}
+
+	if (parse_redirect_to_file_separate(ctx, in_fd, append, out)) {
+		return true;
+	}
+
+	return false;
+}
+
+static bool peek_cmdcall_item(struct parse_ctx *ctx, bool unrestricted)
+{
+	/* each token type falls into one of three categories:
+	 * - cmdcall item: the token can be used as part of a command call. the
+	 *   token indicates the start of a command call.
+	 * - NOT a cmdcall item: the token cannot be used as part of a command
+	 *   call, usually because it as a cmdcall operator like | or &.
+	 *   encountering one of these tokens ends the cmdcall currently being
+	 *   parsed.
+	 * - RESTRICTED cmdcall item: the token can be used as part of a
+	 *   command, but will not be considered the start of a cmdcall. to run
+	 *   a command with this token as its name, the call operator must be
+	 *   used.
+	 */
+	switch (peek_token_type(ctx)) {
+	case TOK_KEYWORD:
+	case TOK_INT:
+	case TOK_DOUBLE:
+	case TOK_VAR:
+	case TOK_VAR_SPLAT:
+	case TOK_STRING:
+	case TOK_STR_START:
+		return unrestricted;
+	case TOK_SYMBOL:
+		switch (peek_unknown_symbol(ctx)) {
+		case SYM_PLUS:
+		case SYM_HYPHEN:
+			return unrestricted;
+		case SYM_PIPE:
+		case SYM_AMPERSAND:
+		case SYM_SEMICOLON:
+			return false;
+		default:
+			return true;
+		}
+	case TOK_NONE:
+	case TOK_LINEFEED:
+		return false;
+	default:
+		return true;
+	}
+}
+
+bool parse_cmdcall(struct parse_ctx *ctx, struct ast_node **out)
+{
+	struct cmdcall_ast_node *node
+		= (struct cmdcall_ast_node *)ast_node_create(AST_CMDCALL);
+	if (!node) {
+		ctx->p_status = BSHELL_ERR_NO_MEMORY;
+		return false;
+	}
+
+	struct ast_node *child = NULL;
+	bool unrestricted = false;
+	bool ok = true;
+	bool stop = false;
+
+	if (parse_symbol(ctx, SYM_AMPERSAND)) {
+		unrestricted = true;
+	}
+
+	if (!peek_cmdcall_item(ctx, unrestricted)) {
+		return false;
+	}
+
+	struct lex_token *tok = peek_token(ctx);
+	if (!tok) {
+		return false;
+	}
+
+	if (!parse_cmdcall_arg(ctx, &child)) {
+		return false;
+	}
+
+	fx_queue_push_back(&node->n_args, &child->n_entry);
+
+	while (ok && !stop) {
+		if (!peek_cmdcall_item(ctx, true)) {
+			break;
+		}
+
+		struct lex_token *tok = peek_token(ctx);
+		if (!tok) {
+			break;
+		}
+
+		if (parse_redirect(ctx, &child)) {
+			fx_queue_push_back(&node->n_redirect, &child->n_entry);
+		} else if (parse_cmdcall_arg(ctx, &child)) {
+			fx_queue_push_back(&node->n_args, &child->n_entry);
+		} else {
+			ctx->p_status = BSHELL_ERR_BAD_SYNTAX;
+			ok = false;
+			break;
+		}
+	}
+
+	if (!ok) {
+		ast_node_destroy((struct ast_node *)node);
+		node = NULL;
+	}
+
+	*out = (struct ast_node *)node;
+	return ok;
+}
+
+bool peek_command(struct parse_ctx *ctx)
+{
+	if (peek_symbol(ctx, SYM_AMPERSAND)) {
+		return true;
+	}
+
+	return peek_cmdcall_item(ctx, false);
+}
+
+bool parse_command(struct parse_ctx *ctx, struct ast_node **out)
+{
+	struct ast_node *cmdcall = NULL;
+	if (!parse_cmdcall(ctx, &cmdcall)) {
+		return false;
+	}
+
+	struct pipeline_ast_node *pipeline = NULL;
+
+	while (1) {
+		if (parse_symbol(ctx, SYM_SEMICOLON) || parse_linefeed(ctx)) {
+			break;
+		}
+
+		if (!parse_symbol(ctx, SYM_PIPE)) {
+			break;
+		}
+
+		if (!pipeline) {
+			pipeline = (struct pipeline_ast_node *)ast_node_create(
+				AST_PIPELINE);
+			if (!pipeline) {
+				ctx->p_status = BSHELL_ERR_NO_MEMORY;
+				ast_node_destroy(cmdcall);
+				return false;
+			}
+
+			fx_queue_push_back(
+				&pipeline->n_stages,
+				&cmdcall->n_entry);
+		}
+
+		if (!parse_cmdcall(ctx, &cmdcall)) {
+			ctx->p_status = BSHELL_ERR_BAD_SYNTAX;
+			return false;
+		}
+
+		fx_queue_push_back(&pipeline->n_stages, &cmdcall->n_entry);
+	}
+
+	if (pipeline) {
+		*out = (struct ast_node *)pipeline;
+	} else {
+		*out = cmdcall;
+	}
+
+	return true;
+}

bshell/parse/syntax/generic.c

@@ -0,0 +1,143 @@
+#include "../lex.h"
+#include "../parse.h"
+#include "../syntax.h"
+#include "../token.h"
+
+struct lex_token *claim_token(struct parse_ctx *ctx)
+{
+	return lex_ctx_claim(ctx->p_src);
+}
+
+void discard_token(struct parse_ctx *ctx)
+{
+	return lex_ctx_discard(ctx->p_src);
+}
+
+struct lex_token *peek_token(struct parse_ctx *ctx)
+{
+	return lex_ctx_peek(ctx->p_src);
+}
+
+enum token_type peek_token_type(struct parse_ctx *ctx)
+{
+	struct lex_token *tok = peek_token(ctx);
+	return tok ? tok->tok_type : TOK_NONE;
+}
+
+enum token_symbol peek_unknown_symbol(struct parse_ctx *ctx)
+{
+	struct lex_token *tok = peek_token(ctx);
+	return (tok && tok->tok_type == TOK_SYMBOL) ? tok->tok_symbol
+	                                            : SYM_NONE;
+}
+
+enum token_keyword peek_unknown_keyword(struct parse_ctx *ctx)
+{
+	struct lex_token *tok = peek_token(ctx);
+	return (tok && tok->tok_type == TOK_KEYWORD) ? tok->tok_keyword
+	                                             : KW_NONE;
+}
+
+bool peek_word(struct parse_ctx *ctx, struct lex_token **out)
+{
+	struct lex_token *tok = peek_token(ctx);
+	if (tok && tok->tok_type == TOK_WORD) {
+		*out = tok;
+		return true;
+	}
+
+	return false;
+}
+
+bool peek_linefeed(struct parse_ctx *ctx)
+{
+	struct lex_token *tok = peek_token(ctx);
+	if (tok && tok->tok_type == TOK_LINEFEED) {
+		return true;
+	}
+
+	return false;
+}
+
+bool peek_symbol(struct parse_ctx *ctx, enum token_symbol sym)
+{
+	struct lex_token *tok = peek_token(ctx);
+	if (!tok) {
+		return false;
+	}
+
+	if (tok->tok_type != TOK_SYMBOL) {
+		return false;
+	}
+
+	if (tok->tok_symbol != sym) {
+		return false;
+	}
+
+	return true;
+}
+
+bool parse_linefeed(struct parse_ctx *ctx)
+{
+	struct lex_token *tok = peek_token(ctx);
+	if (tok && tok->tok_type == TOK_LINEFEED) {
+		discard_token(ctx);
+		return true;
+	}
+
+	return false;
+}
+
+bool parse_symbol(struct parse_ctx *ctx, enum token_symbol sym)
+{
+	struct lex_token *tok = peek_token(ctx);
+	if (!tok) {
+		return false;
+	}
+
+	if (tok->tok_type != TOK_SYMBOL) {
+		return false;
+	}
+
+	if (tok->tok_symbol != sym) {
+		return false;
+	}
+
+	discard_token(ctx);
+	return true;
+}
+
+bool parse_keyword(struct parse_ctx *ctx, enum token_keyword kw)
+{
+	struct lex_token *tok = peek_token(ctx);
+	if (!tok) {
+		return false;
+	}
+
+	if (tok->tok_type != TOK_KEYWORD) {
+		return false;
+	}
+
+	if (tok->tok_keyword != kw) {
+		return false;
+	}
+
+	discard_token(ctx);
+	return true;
+}
+
+bool parse_int(struct parse_ctx *ctx, long long *out)
+{
+	struct lex_token *tok = peek_token(ctx);
+	if (!tok) {
+		return false;
+	}
+
+	if (tok->tok_type != TOK_INT) {
+		return false;
+	}
+
+	*out = tok->tok_int;
+	discard_token(ctx);
+	return true;
+}

bshell/parse/syntax/statement.c

@@ -0,0 +1,15 @@
+#include "../syntax.h"
+
+bool parse_statement(struct parse_ctx *ctx, struct ast_node **out)
+{
+	bool ok = false;
+	if (peek_arith_expr(ctx)) {
+		ok = parse_arith_expr(ctx, out);
+	}
+
+	if (!ok && peek_command(ctx)) {
+		ok = parse_command(ctx, out);
+	}
+
+	return ok;
+}

bshell/parse/token.c

@@ -0,0 +1,150 @@
+#include "token.h"
+
+#include <fx/string.h>
+#include <stdlib.h>
+#include <string.h>
+
+struct lex_token *lex_token_create(enum token_type type)
+{
+	struct lex_token *out = malloc(sizeof *out);
+	if (!out) {
+		return NULL;
+	}
+
+	memset(out, 0x0, sizeof *out);
+
+	out->tok_type = type;
+
+	return out;
+}
+
+struct lex_token *lex_token_create_with_string(
+	enum token_type type,
+	const char *s)
+{
+	struct lex_token *tok = lex_token_create(type);
+	if (!tok) {
+		return NULL;
+	}
+
+	tok->tok_str = fx_strdup(s);
+	if (!tok->tok_str) {
+		free(tok);
+		return NULL;
+	}
+
+	return tok;
+}
+
+void lex_token_destroy(struct lex_token *tok)
+{
+	switch (tok->tok_type) {
+	case TOK_WORD:
+	case TOK_FLAG:
+	case TOK_STRING:
+		if (tok->tok_str) {
+			free(tok->tok_str);
+		}
+
+		break;
+	default:
+		break;
+	}
+
+	free(tok);
+}
+
+struct lex_token *lex_token_change_type(
+	struct lex_token *tok,
+	enum token_type new_type)
+{
+	switch (tok->tok_type) {
+	case TOK_WORD:
+	case TOK_FLAG:
+	case TOK_STRING:
+		if (tok->tok_str) {
+			free(tok->tok_str);
+			tok->tok_str = NULL;
+		}
+		break;
+	default:
+		break;
+	}
+
+	tok->tok_type = new_type;
+	return tok;
+}
+
+#define ENUM_STR(x)                                                            \
+	case x:                                                                \
+		return #x
+
+const char *token_type_to_string(enum token_type type)
+{
+	switch (type) {
+		ENUM_STR(TOK_NONE);
+		ENUM_STR(TOK_KEYWORD);
+		ENUM_STR(TOK_SYMBOL);
+		ENUM_STR(TOK_INT);
+		ENUM_STR(TOK_DOUBLE);
+		ENUM_STR(TOK_WORD);
+		ENUM_STR(TOK_VAR);
+		ENUM_STR(TOK_VAR_SPLAT);
+		ENUM_STR(TOK_FLAG);
+		ENUM_STR(TOK_STRING);
+		ENUM_STR(TOK_STR_START);
+		ENUM_STR(TOK_STR_END);
+		ENUM_STR(TOK_LINEFEED);
+	default:
+		return "<unknown>";
+	}
+}
+
+const char *token_keyword_to_string(enum token_keyword keyword)
+{
+	switch (keyword) {
+		ENUM_STR(KW_NONE);
+		ENUM_STR(KW_FUNC);
+		ENUM_STR(KW_IF);
+		ENUM_STR(KW_ELSE);
+	default:
+		return "<unknown>";
+	}
+}
+
+const char *token_symbol_to_string(enum token_symbol sym)
+{
+	switch (sym) {
+		ENUM_STR(SYM_NONE);
+		ENUM_STR(SYM_PLUS);
+		ENUM_STR(SYM_HYPHEN);
+		ENUM_STR(SYM_FORWARD_SLASH);
+		ENUM_STR(SYM_ASTERISK);
+		ENUM_STR(SYM_AMPERSAND);
+		ENUM_STR(SYM_PERCENT);
+		ENUM_STR(SYM_SQUOTE);
+		ENUM_STR(SYM_DQUOTE);
+		ENUM_STR(SYM_HASH);
+		ENUM_STR(SYM_SEMICOLON);
+		ENUM_STR(SYM_COMMA);
+		ENUM_STR(SYM_DOLLAR);
+		ENUM_STR(SYM_DOLLAR_LEFT_PAREN);
+		ENUM_STR(SYM_PIPE);
+		ENUM_STR(SYM_AT);
+		ENUM_STR(SYM_AT_LEFT_BRACE);
+		ENUM_STR(SYM_LEFT_BRACE);
+		ENUM_STR(SYM_RIGHT_BRACE);
+		ENUM_STR(SYM_LEFT_BRACKET);
+		ENUM_STR(SYM_RIGHT_BRACKET);
+		ENUM_STR(SYM_LEFT_PAREN);
+		ENUM_STR(SYM_RIGHT_PAREN);
+		ENUM_STR(SYM_EQUAL);
+		ENUM_STR(SYM_PLUS_EQUAL);
+		ENUM_STR(SYM_HYPHEN_EQUAL);
+		ENUM_STR(SYM_FORWARD_SLASH_EQUAL);
+		ENUM_STR(SYM_ASTERISK_EQUAL);
+		ENUM_STR(SYM_PERCENT_EQUAL);
+	default:
+		return "<unknown>";
+	}
+}

bshell/parse/token.h

@@ -0,0 +1,133 @@
+#ifndef IVY_LANG_LEX_H_
+#define IVY_LANG_LEX_H_
+
+#include <fx/queue.h>
+#include <stdbool.h>
+
+struct char_cell {
+	unsigned long c_row, c_col;
+};
+
+enum token_type {
+	TOK_NONE = 0,
+	__TOK_INDEX_BASE = 100,
+	TOK_KEYWORD,
+	TOK_SYMBOL,
+	TOK_INT,
+	TOK_DOUBLE,
+	TOK_WORD,
+	TOK_FLAG,
+	TOK_VAR,
+	TOK_VAR_SPLAT,
+	TOK_STRING,
+	TOK_STR_START,
+	TOK_STR_END,
+	TOK_LINEFEED,
+	__TOK_INDEX_LIMIT,
+};
+
+enum token_keyword {
+	KW_NONE = 0,
+	__KW_INDEX_BASE = 200,
+	KW_FUNC,
+	KW_IF,
+	KW_ELSE,
+	__KW_INDEX_LIMIT,
+};
+
+enum token_symbol {
+	SYM_NONE = 0,
+	__SYM_INDEX_BASE = 300,
+	SYM_PLUS,
+	SYM_HYPHEN,
+	SYM_FORWARD_SLASH,
+	SYM_ASTERISK,
+	SYM_AMPERSAND,
+	SYM_PERCENT,
+	SYM_SQUOTE,
+	SYM_DQUOTE,
+	SYM_HASH,
+	SYM_SEMICOLON,
+	SYM_COMMA,
+	SYM_DOLLAR,
+	SYM_DOLLAR_LEFT_PAREN,
+	SYM_DOLLAR_LEFT_BRACE,
+	SYM_PIPE,
+	SYM_AT,
+	SYM_AT_LEFT_BRACE,
+	SYM_LEFT_BRACE,
+	SYM_RIGHT_BRACE,
+	SYM_LEFT_BRACKET,
+	SYM_RIGHT_BRACKET,
+	SYM_LEFT_PAREN,
+	SYM_RIGHT_PAREN,
+	SYM_EQUAL,
+	SYM_PLUS_EQUAL,
+	SYM_HYPHEN_EQUAL,
+	SYM_ASTERISK_EQUAL,
+	SYM_FORWARD_SLASH_EQUAL,
+	SYM_PERCENT_EQUAL,
+	__SYM_INDEX_LIMIT,
+};
+
+struct lex_token {
+	enum token_type tok_type;
+
+	struct char_cell tok_start, tok_end;
+
+	fx_queue_entry tok_entry;
+
+	union {
+		enum token_keyword tok_keyword;
+		enum token_symbol tok_symbol;
+		long long tok_int;
+		double tok_double;
+		char *tok_str;
+	};
+};
+
+extern struct lex_token *lex_token_create(enum token_type type);
+extern struct lex_token *lex_token_create_with_string(
+	enum token_type type,
+	const char *s);
+extern void lex_token_destroy(struct lex_token *tok);
+
+extern struct lex_token *lex_token_change_type(
+	struct lex_token *tok,
+	enum token_type new_type);
+
+static inline bool lex_token_is_symbol(
+	struct lex_token *tok,
+	enum token_symbol sym)
+{
+	return (tok->tok_type == TOK_SYMBOL && tok->tok_symbol == sym);
+}
+static inline bool lex_token_is_keyword(
+	struct lex_token *tok,
+	enum token_keyword kw)
+{
+	return (tok->tok_type == TOK_KEYWORD && tok->tok_keyword == kw);
+}
+static inline bool lex_token_type_has_string_value(enum token_type type)
+{
+	switch (type) {
+	case TOK_WORD:
+	case TOK_STRING:
+	case TOK_FLAG:
+	case TOK_VAR:
+	case TOK_VAR_SPLAT:
+		return true;
+	default:
+		return false;
+	}
+}
+static inline bool lex_token_has_string_value(const struct lex_token *tok)
+{
+	return lex_token_type_has_string_value(tok->tok_type);
+}
+
+extern const char *token_type_to_string(enum token_type type);
+extern const char *token_keyword_to_string(enum token_keyword keyword);
+extern const char *token_symbol_to_string(enum token_symbol sym);
+
+#endif

doc/sample/redirection.bshell

@@ -0,0 +1 @@
+echo hello 2> error.txt | ls -la | echo done; exit -1

doc/sample/simple.bshell

@@ -1,8 +1,14 @@
 func test-function($name) {
-	echo "Hello, $name!"
+	echo "Hello, $name! $(2 + 4 + 2) wow"
 }
 
 # Example of instantiating an FX runtime object.
 $obj = new-object -type-name fx.string -arguments "John Doe"
 
+$hash = @{
+    1       = 'one'
+    2       = 'two'
+    'three' = 3
+}
+
 test-function -name $obj

doc/sample/syntax.bshell

@@ -0,0 +1,173 @@
+# The lexer has three modes: ARITHMETIC, COMMAND, and STRING
+# ARITHMETIC mode is operand-based, all symbols, keywords, and constant parsing
+# is enabled.
+# COMMAND mode is word-based, only a subset of symbols are enabled, no keyword
+# or constant parsing is performed, and more liberal word formations and
+# substitutions are allowed
+# STRING mode is used to read string literals (i.e. those strings that DON'T
+# support variable substitutions). All chars read are appended to the resulting
+# string, with no further parsing performed.
+
+# Initially, the lexer mode is unspecified, until:
+#  a) The lexer reads a character, from which the correct mode is deduced.
+#  b) The parser manually switches the lexer's mode
+# Lexer state supports nesting.
+
+# ARITHMETIC
+# both of these are equivalant
+$a = 2
+# VAR(a)
+# SYMBOL(=)
+# INT(2)
+
+$b=4
+# VAR(b)
+# SYMBOL(=)
+# INT(4)
+
+# ARITHMETIC
+# this is a syntax error (there should be an operator between the two vars)
+$a$b
+# VAR(a)
+# VAR(b)
+
+# When the parser encounters SYMBOL(%) it should switch the lexer to COMMAND
+# mode, which will allow the following word construction to be used.
+# this executes the command whose name is equal to concatenating the values
+# of $a and $b (in this case, '24')
+% $a$b
+# SYMBOL(%)
+# WORD_START
+# VAR(a)
+# VAR(b)
+# WORD_END
+
+# executes the command with the name 'a+2b'. because the first char encountered
+# by the lexer is alphabetic, it reads a regular word in COMMAND mode.
+a+2b
+# WORD(a+2b)
+
+# executes the command with the name '-no$a' ($a is not substituted).
+# the first char encountered is a symbol, which is read as a word in COMMAND
+# mode
+-no$a
+# WORD(-no)
+
+# returns the result of applying the NOT operator to the value of $a.
+# the first char encountered is a symbol, which is read as a word in COMMAND
+# mode. as characters are read, they are compared against registered operators.
+# if a match is found, the operator is emitted, and the parser will switch
+# the lexer to ARITHMETIC mode
+-not$a
+# OP(not)
+# VAR(a)
+
+# executes the command with the name '-not$a' ($a is NOT substituted)
+# because of the preceding hyphen, variable substitution is not performed.
+% -not$a
+# SYMBOL(%)
+# WORD(-not$a)
+
+# executes the command with the name '-not2' ($a IS substituted)
+# variable substitution IS performed in dquote strings regardless of the hyphen.
+% "-not$a"
+# SYMBOL(%)
+# STR_START
+# STRING(-not)
+# VAR(a)
+# STR_END
+
+# interpreted as a command with args ['a', '+b', '/c']
+# the first char encountered is alpbabetic, so the expression is parsed in
+# COMMAND mode
+a +b /c
+# WORD(a)
+# WORD(+b)
+# WORD(/c)
+
+# interpreted as an arithmetic expression (but not a well-formed one)
++b /c
+# SYM(+)
+# WORD(b)
+# SYM(/)
+# WORD(c)
+
+# interpreted as a command with name '%+'
+%+
+# WORD(%+)
+
+# interpreted as a command with args ['%', '+']
+% +
+# WORD(%)
+# WORD(+)
+
+# interpreted as a command with name '%'
+%;
+# WORD(%)
+# SYMBOL(;)
+
+# interpreted as a command with name '+'
+&+
+# SYMBOL(&)
+# WORD(+)
+
+# interpreted as a string, which triggers the parser to enter ARITHMETIC mode
+'hello world'
+# STRING(hello world)
+
+# interpreted as a command with args ['echo', 'hello world']
+echo 'hello world'
+# WORD(echo)
+# STRING(hello world)
+
+# interpreted as an interpolated string
+"Hello $(if ($x -lt 5) { echo 'yes' } else {echo 'no'})"
+
+
+###############################################################################
+# The lexer operates as a state machine, moving between different states as
+# different characters are encountered
+# The states are stored in a stack, to allow recursive parsing.
+# The lexer has the following states:
+# STATEMENT: A generic statement, could be a command, keyword, arithmetic
+#       expression, etc. The next char or symbol encountered will cause the
+#       lexer to switch to the appropriate state type:
+#           letters, word-symbols -> COMMAND
+#           squote -> ARITHMETIC
+#           dquote -> ARITHMETIC, FSTRING
+#           Digits, vars, var-splats, keywords, all other symbols -> ARITHMETIC
+# EXPRESSION: Similar to STATEMENT, but only allows a single command or
+#       arithmetic expression. CANNOT use keywords or statement terminators.
+#           Letters, word-symbols -> COMMAND
+#           squote -> ARITHMETIC
+#           dquote -> ARITHMETIC, FSTRING
+#           Digits, vars, var-splats, keywords, all other symbols -> ARITHMETIC
+# COMMAND: Only words, (f)strings, vars, var-splats, and a subset of symbols are
+#       parsed.
+# ARITHMETIC: Words, strings, vars, var-splats, all symbols, keywords are parsed.
+# STRING: Only a subset of symbols are parsed, all other characters are appended
+#       to the resulting string.
+#
+# Once a state has changed from EXPRESSION to one of the other three state
+# types, certain characters will result in the current state either changing
+# type or being popped from the stack:
+#   STATEMENT: semicolon -> STATEMENT
+#            left-paren, left-brace -> POP
+#   EXPRESSION: semicolon -> POP
+#            left-paren, left-brace -> POP
+#   COMMAND: semicolon -> STATEMENT
+#            left-paren, left-brace -> POP
+#   ARITHMETIC: semicolon -> STATEMENT
+#            left-paren, left-brace -> POP
+#
+# Certain symbols require recursive parsing:
+# - dquote strings allow string interpolation, so expressions withing the string
+#   may be parsed in a different state. Once the expression is complete, the
+#   lexer returns to the previous state.
+# - in most cases, $(...) can be used to delimit sub-expressions (including in
+#   strings. When '$(' is encountered, a new state entry of type EXPRESSION is
+#   pushed onto the stack. When the corresponding ')' is encountered, that state
+#   entry is popped from the stack.
+# - similarly to $(...), (...) can be used to group expressions, just like in
+#   mathematical expressions.
+