653 lines
22 KiB
C
653 lines
22 KiB
C
#undef I3__FILE__
|
||
#define I3__FILE__ "config_parser.c"
|
||
/*
|
||
* vim:ts=4:sw=4:expandtab
|
||
*
|
||
* i3 - an improved dynamic tiling window manager
|
||
* © 2009-2012 Michael Stapelberg and contributors (see also: LICENSE)
|
||
*
|
||
* config_parser.c: hand-written parser to parse configuration directives.
|
||
*
|
||
* See also src/commands_parser.c for rationale on why we use a custom parser.
|
||
*
|
||
* This parser works VERY MUCH like src/commands_parser.c, so read that first.
|
||
* The differences are:
|
||
*
|
||
* 1. config_parser supports the 'number' token type (in addition to 'word' and
|
||
* 'string'). Numbers are referred to using &num (like $str).
|
||
*
|
||
* 2. Criteria are not executed immediately, they are just stored.
|
||
*
|
||
* 3. config_parser recognizes \n and \r as 'end' token, while commands_parser
|
||
* ignores them.
|
||
*
|
||
* 4. config_parser skips the current line on invalid inputs and follows the
|
||
* nearest <error> token.
|
||
*
|
||
*/
|
||
#include <stdio.h>
|
||
#include <stdlib.h>
|
||
#include <string.h>
|
||
#include <unistd.h>
|
||
#include <stdbool.h>
|
||
#include <stdint.h>
|
||
|
||
#include "all.h"
|
||
|
||
// Macros to make the YAJL API a bit easier to use.
|
||
#define y(x, ...) yajl_gen_ ## x (command_output.json_gen, ##__VA_ARGS__)
|
||
#define ystr(str) yajl_gen_string(command_output.json_gen, (unsigned char*)str, strlen(str))
|
||
|
||
/*******************************************************************************
|
||
* The data structures used for parsing. Essentially the current state and a
|
||
* list of tokens for that state.
|
||
*
|
||
* The GENERATED_* files are generated by generate-commands-parser.pl with the
|
||
* input parser-specs/configs.spec.
|
||
******************************************************************************/
|
||
|
||
#include "GENERATED_config_enums.h"
|
||
|
||
typedef struct token {
|
||
char *name;
|
||
char *identifier;
|
||
/* This might be __CALL */
|
||
cmdp_state next_state;
|
||
union {
|
||
uint16_t call_identifier;
|
||
} extra;
|
||
} cmdp_token;
|
||
|
||
typedef struct tokenptr {
|
||
cmdp_token *array;
|
||
int n;
|
||
} cmdp_token_ptr;
|
||
|
||
#include "GENERATED_config_tokens.h"
|
||
|
||
/*******************************************************************************
|
||
* The (small) stack where identified literals are stored during the parsing
|
||
* of a single command (like $workspace).
|
||
******************************************************************************/
|
||
|
||
struct stack_entry {
|
||
/* Just a pointer, not dynamically allocated. */
|
||
const char *identifier;
|
||
enum {
|
||
STACK_STR = 0,
|
||
STACK_LONG = 1,
|
||
} type;
|
||
union {
|
||
char *str;
|
||
long num;
|
||
} val;
|
||
};
|
||
|
||
/* 10 entries should be enough for everybody. */
|
||
static struct stack_entry stack[10];
|
||
|
||
/*
|
||
* Pushes a string (identified by 'identifier') on the stack. We simply use a
|
||
* single array, since the number of entries we have to store is very small.
|
||
*
|
||
*/
|
||
static void push_string(const char *identifier, const char *str) {
|
||
for (int c = 0; c < 10; c++) {
|
||
if (stack[c].identifier != NULL &&
|
||
strcmp(stack[c].identifier, identifier) != 0)
|
||
continue;
|
||
if (stack[c].identifier == NULL) {
|
||
/* Found a free slot, let’s store it here. */
|
||
stack[c].identifier = identifier;
|
||
stack[c].val.str = sstrdup(str);
|
||
stack[c].type = STACK_STR;
|
||
} else {
|
||
/* Append the value. */
|
||
char *prev = stack[c].val.str;
|
||
sasprintf(&(stack[c].val.str), "%s,%s", prev, str);
|
||
free(prev);
|
||
}
|
||
return;
|
||
}
|
||
|
||
/* When we arrive here, the stack is full. This should not happen and
|
||
* means there’s either a bug in this parser or the specification
|
||
* contains a command with more than 10 identified tokens. */
|
||
fprintf(stderr, "BUG: commands_parser stack full. This means either a bug "
|
||
"in the code, or a new command which contains more than "
|
||
"10 identified tokens.\n");
|
||
exit(1);
|
||
}
|
||
|
||
static void push_long(const char *identifier, long num) {
|
||
for (int c = 0; c < 10; c++) {
|
||
if (stack[c].identifier != NULL)
|
||
continue;
|
||
/* Found a free slot, let’s store it here. */
|
||
stack[c].identifier = identifier;
|
||
stack[c].val.num = num;
|
||
stack[c].type = STACK_LONG;
|
||
return;
|
||
}
|
||
|
||
/* When we arrive here, the stack is full. This should not happen and
|
||
* means there’s either a bug in this parser or the specification
|
||
* contains a command with more than 10 identified tokens. */
|
||
fprintf(stderr, "BUG: commands_parser stack full. This means either a bug "
|
||
"in the code, or a new command which contains more than "
|
||
"10 identified tokens.\n");
|
||
exit(1);
|
||
|
||
}
|
||
|
||
static const char *get_string(const char *identifier) {
|
||
for (int c = 0; c < 10; c++) {
|
||
if (stack[c].identifier == NULL)
|
||
break;
|
||
if (strcmp(identifier, stack[c].identifier) == 0)
|
||
return stack[c].val.str;
|
||
}
|
||
return NULL;
|
||
}
|
||
|
||
static const long get_long(const char *identifier) {
|
||
for (int c = 0; c < 10; c++) {
|
||
if (stack[c].identifier == NULL)
|
||
break;
|
||
if (strcmp(identifier, stack[c].identifier) == 0)
|
||
return stack[c].val.num;
|
||
}
|
||
return 0;
|
||
}
|
||
|
||
static void clear_stack(void) {
|
||
for (int c = 0; c < 10; c++) {
|
||
if (stack[c].type == STACK_STR && stack[c].val.str != NULL)
|
||
free(stack[c].val.str);
|
||
stack[c].identifier = NULL;
|
||
stack[c].val.str = NULL;
|
||
stack[c].val.num = 0;
|
||
}
|
||
}
|
||
|
||
// TODO: remove this if it turns out we don’t need it for testing.
|
||
#if 0
|
||
/*******************************************************************************
|
||
* A dynamically growing linked list which holds the criteria for the current
|
||
* command.
|
||
******************************************************************************/
|
||
|
||
typedef struct criterion {
|
||
char *type;
|
||
char *value;
|
||
|
||
TAILQ_ENTRY(criterion) criteria;
|
||
} criterion;
|
||
|
||
static TAILQ_HEAD(criteria_head, criterion) criteria =
|
||
TAILQ_HEAD_INITIALIZER(criteria);
|
||
|
||
/*
|
||
* Stores the given type/value in the list of criteria.
|
||
* Accepts a pointer as first argument, since it is 'call'ed by the parser.
|
||
*
|
||
*/
|
||
static void push_criterion(void *unused_criteria, const char *type,
|
||
const char *value) {
|
||
struct criterion *criterion = malloc(sizeof(struct criterion));
|
||
criterion->type = strdup(type);
|
||
criterion->value = strdup(value);
|
||
TAILQ_INSERT_TAIL(&criteria, criterion, criteria);
|
||
}
|
||
|
||
/*
|
||
* Clears the criteria linked list.
|
||
* Accepts a pointer as first argument, since it is 'call'ed by the parser.
|
||
*
|
||
*/
|
||
static void clear_criteria(void *unused_criteria) {
|
||
struct criterion *criterion;
|
||
while (!TAILQ_EMPTY(&criteria)) {
|
||
criterion = TAILQ_FIRST(&criteria);
|
||
free(criterion->type);
|
||
free(criterion->value);
|
||
TAILQ_REMOVE(&criteria, criterion, criteria);
|
||
free(criterion);
|
||
}
|
||
}
|
||
#endif
|
||
|
||
/*******************************************************************************
|
||
* The parser itself.
|
||
******************************************************************************/
|
||
|
||
static cmdp_state state;
|
||
static Match current_match;
|
||
static struct ConfigResult subcommand_output;
|
||
static struct ConfigResult command_output;
|
||
|
||
/* A list which contains the states that lead to the current state, e.g.
|
||
* INITIAL, WORKSPACE_LAYOUT.
|
||
* When jumping back to INITIAL, statelist_idx will simply be set to 1
|
||
* (likewise for other states, e.g. MODE or BAR).
|
||
* This list is used to process the nearest error token. */
|
||
static cmdp_state statelist[10] = { INITIAL };
|
||
/* NB: statelist_idx points to where the next entry will be inserted */
|
||
static int statelist_idx = 1;
|
||
|
||
#include "GENERATED_config_call.h"
|
||
|
||
|
||
static void next_state(const cmdp_token *token) {
|
||
cmdp_state _next_state = token->next_state;
|
||
|
||
//printf("token = name %s identifier %s\n", token->name, token->identifier);
|
||
//printf("next_state = %d\n", token->next_state);
|
||
if (token->next_state == __CALL) {
|
||
subcommand_output.json_gen = command_output.json_gen;
|
||
GENERATED_call(token->extra.call_identifier, &subcommand_output);
|
||
_next_state = subcommand_output.next_state;
|
||
clear_stack();
|
||
}
|
||
|
||
state = _next_state;
|
||
if (state == INITIAL) {
|
||
clear_stack();
|
||
}
|
||
|
||
/* See if we are jumping back to a state in which we were in previously
|
||
* (statelist contains INITIAL) and just move statelist_idx accordingly. */
|
||
for (int i = 0; i < statelist_idx; i++) {
|
||
if (statelist[i] != _next_state)
|
||
continue;
|
||
statelist_idx = i+1;
|
||
return;
|
||
}
|
||
|
||
/* Otherwise, the state is new and we add it to the list */
|
||
statelist[statelist_idx++] = _next_state;
|
||
}
|
||
|
||
/*
|
||
* Returns a pointer to the start of the line (one byte after the previous \r,
|
||
* \n) or the start of the input, if this is the first line.
|
||
*
|
||
*/
|
||
static const char *start_of_line(const char *walk, const char *beginning) {
|
||
while (*walk != '\n' && *walk != '\r' && walk >= beginning) {
|
||
walk--;
|
||
}
|
||
|
||
return walk + 1;
|
||
}
|
||
|
||
/*
|
||
* Copies the line and terminates it at the next \n, if any.
|
||
*
|
||
* The caller has to free() the result.
|
||
*
|
||
*/
|
||
static char *single_line(const char *start) {
|
||
char *result = sstrdup(start);
|
||
char *end = strchr(result, '\n');
|
||
if (end != NULL)
|
||
*end = '\0';
|
||
return result;
|
||
}
|
||
|
||
struct ConfigResult *parse_config(const char *input, struct context *context) {
|
||
/* Dump the entire config file into the debug log. We cannot just use
|
||
* DLOG("%s", input); because one log message must not exceed 4 KiB. */
|
||
const char *dumpwalk = input;
|
||
int linecnt = 1;
|
||
while (*dumpwalk != '\0') {
|
||
char *next_nl = strchr(dumpwalk, '\n');
|
||
if (next_nl != NULL) {
|
||
DLOG("CONFIG(line %3d): %.*s\n", linecnt, (int)(next_nl - dumpwalk), dumpwalk);
|
||
dumpwalk = next_nl + 1;
|
||
} else {
|
||
DLOG("CONFIG(line %3d): %s\n", linecnt, dumpwalk);
|
||
break;
|
||
}
|
||
linecnt++;
|
||
}
|
||
state = INITIAL;
|
||
statelist_idx = 1;
|
||
|
||
/* A YAJL JSON generator used for formatting replies. */
|
||
#if YAJL_MAJOR >= 2
|
||
command_output.json_gen = yajl_gen_alloc(NULL);
|
||
#else
|
||
command_output.json_gen = yajl_gen_alloc(NULL, NULL);
|
||
#endif
|
||
|
||
y(array_open);
|
||
|
||
const char *walk = input;
|
||
const size_t len = strlen(input);
|
||
int c;
|
||
const cmdp_token *token;
|
||
bool token_handled;
|
||
linecnt = 1;
|
||
|
||
// TODO: make this testable
|
||
#ifndef TEST_PARSER
|
||
cfg_criteria_init(¤t_match, &subcommand_output, INITIAL);
|
||
#endif
|
||
|
||
/* The "<=" operator is intentional: We also handle the terminating 0-byte
|
||
* explicitly by looking for an 'end' token. */
|
||
while ((walk - input) <= len) {
|
||
/* Skip whitespace before every token, newlines are relevant since they
|
||
* separate configuration directives. */
|
||
while ((*walk == ' ' || *walk == '\t') && *walk != '\0')
|
||
walk++;
|
||
|
||
//printf("remaining input: %s\n", walk);
|
||
|
||
cmdp_token_ptr *ptr = &(tokens[state]);
|
||
token_handled = false;
|
||
for (c = 0; c < ptr->n; c++) {
|
||
token = &(ptr->array[c]);
|
||
|
||
/* A literal. */
|
||
if (token->name[0] == '\'') {
|
||
if (strncasecmp(walk, token->name + 1, strlen(token->name) - 1) == 0) {
|
||
if (token->identifier != NULL)
|
||
push_string(token->identifier, token->name + 1);
|
||
walk += strlen(token->name) - 1;
|
||
next_state(token);
|
||
token_handled = true;
|
||
break;
|
||
}
|
||
continue;
|
||
}
|
||
|
||
if (strcmp(token->name, "number") == 0) {
|
||
/* Handle numbers. We only accept decimal numbers for now. */
|
||
char *end = NULL;
|
||
errno = 0;
|
||
long int num = strtol(walk, &end, 10);
|
||
if ((errno == ERANGE && (num == LONG_MIN || num == LONG_MAX)) ||
|
||
(errno != 0 && num == 0))
|
||
continue;
|
||
|
||
/* No valid numbers found */
|
||
if (end == walk)
|
||
continue;
|
||
|
||
if (token->identifier != NULL)
|
||
push_long(token->identifier, num);
|
||
|
||
/* Set walk to the first non-number character */
|
||
walk = end;
|
||
next_state(token);
|
||
token_handled = true;
|
||
break;
|
||
}
|
||
|
||
if (strcmp(token->name, "string") == 0 ||
|
||
strcmp(token->name, "word") == 0) {
|
||
const char *beginning = walk;
|
||
/* Handle quoted strings (or words). */
|
||
if (*walk == '"') {
|
||
beginning++;
|
||
walk++;
|
||
while (*walk != '\0' && (*walk != '"' || *(walk-1) == '\\'))
|
||
walk++;
|
||
} else {
|
||
if (token->name[0] == 's') {
|
||
while (*walk != '\0' && *walk != '\r' && *walk != '\n')
|
||
walk++;
|
||
} else {
|
||
/* For a word, the delimiters are white space (' ' or
|
||
* '\t'), closing square bracket (]), comma (,) and
|
||
* semicolon (;). */
|
||
while (*walk != ' ' && *walk != '\t' &&
|
||
*walk != ']' && *walk != ',' &&
|
||
*walk != ';' && *walk != '\r' &&
|
||
*walk != '\n' && *walk != '\0')
|
||
walk++;
|
||
}
|
||
}
|
||
if (walk != beginning) {
|
||
char *str = scalloc(walk-beginning + 1);
|
||
/* We copy manually to handle escaping of characters. */
|
||
int inpos, outpos;
|
||
for (inpos = 0, outpos = 0;
|
||
inpos < (walk-beginning);
|
||
inpos++, outpos++) {
|
||
/* We only handle escaped double quotes to not break
|
||
* backwards compatibility with people using \w in
|
||
* regular expressions etc. */
|
||
if (beginning[inpos] == '\\' && beginning[inpos+1] == '"')
|
||
inpos++;
|
||
str[outpos] = beginning[inpos];
|
||
}
|
||
if (token->identifier)
|
||
push_string(token->identifier, str);
|
||
free(str);
|
||
/* If we are at the end of a quoted string, skip the ending
|
||
* double quote. */
|
||
if (*walk == '"')
|
||
walk++;
|
||
next_state(token);
|
||
token_handled = true;
|
||
break;
|
||
}
|
||
}
|
||
|
||
if (strcmp(token->name, "end") == 0) {
|
||
//printf("checking for end: *%s*\n", walk);
|
||
if (*walk == '\0' || *walk == '\n' || *walk == '\r') {
|
||
next_state(token);
|
||
token_handled = true;
|
||
/* To make sure we start with an appropriate matching
|
||
* datastructure for commands which do *not* specify any
|
||
* criteria, we re-initialize the criteria system after
|
||
* every command. */
|
||
// TODO: make this testable
|
||
#ifndef TEST_PARSER
|
||
cfg_criteria_init(¤t_match, &subcommand_output, INITIAL);
|
||
#endif
|
||
linecnt++;
|
||
walk++;
|
||
break;
|
||
}
|
||
}
|
||
}
|
||
|
||
if (!token_handled) {
|
||
/* Figure out how much memory we will need to fill in the names of
|
||
* all tokens afterwards. */
|
||
int tokenlen = 0;
|
||
for (c = 0; c < ptr->n; c++)
|
||
tokenlen += strlen(ptr->array[c].name) + strlen("'', ");
|
||
|
||
/* Build up a decent error message. We include the problem, the
|
||
* full input, and underline the position where the parser
|
||
* currently is. */
|
||
char *errormessage;
|
||
char *possible_tokens = smalloc(tokenlen + 1);
|
||
char *tokenwalk = possible_tokens;
|
||
for (c = 0; c < ptr->n; c++) {
|
||
token = &(ptr->array[c]);
|
||
if (token->name[0] == '\'') {
|
||
/* A literal is copied to the error message enclosed with
|
||
* single quotes. */
|
||
*tokenwalk++ = '\'';
|
||
strcpy(tokenwalk, token->name + 1);
|
||
tokenwalk += strlen(token->name + 1);
|
||
*tokenwalk++ = '\'';
|
||
} else {
|
||
/* Skip error tokens in error messages, they are used
|
||
* internally only and might confuse users. */
|
||
if (strcmp(token->name, "error") == 0)
|
||
continue;
|
||
/* Any other token is copied to the error message enclosed
|
||
* with angle brackets. */
|
||
*tokenwalk++ = '<';
|
||
strcpy(tokenwalk, token->name);
|
||
tokenwalk += strlen(token->name);
|
||
*tokenwalk++ = '>';
|
||
}
|
||
if (c < (ptr->n - 1)) {
|
||
*tokenwalk++ = ',';
|
||
*tokenwalk++ = ' ';
|
||
}
|
||
}
|
||
*tokenwalk = '\0';
|
||
sasprintf(&errormessage, "Expected one of these tokens: %s",
|
||
possible_tokens);
|
||
free(possible_tokens);
|
||
|
||
|
||
/* Go back to the beginning of the line */
|
||
const char *error_line = start_of_line(walk, input);
|
||
|
||
/* Contains the same amount of characters as 'input' has, but with
|
||
* the unparseable part highlighted using ^ characters. */
|
||
char *position = scalloc(strlen(error_line) + 1);
|
||
const char *copywalk;
|
||
for (copywalk = error_line;
|
||
*copywalk != '\n' && *copywalk != '\r' && *copywalk != '\0';
|
||
copywalk++)
|
||
position[(copywalk - error_line)] = (copywalk >= walk ? '^' : (*copywalk == '\t' ? '\t' : ' '));
|
||
position[(copywalk - error_line)] = '\0';
|
||
|
||
ELOG("CONFIG: %s\n", errormessage);
|
||
ELOG("CONFIG: (in file %s)\n", context->filename);
|
||
char *error_copy = single_line(error_line);
|
||
|
||
/* Print context lines *before* the error, if any. */
|
||
if (linecnt > 1) {
|
||
const char *context_p1_start = start_of_line(error_line-2, input);
|
||
char *context_p1_line = single_line(context_p1_start);
|
||
if (linecnt > 2) {
|
||
const char *context_p2_start = start_of_line(context_p1_start-2, input);
|
||
char *context_p2_line = single_line(context_p2_start);
|
||
ELOG("CONFIG: Line %3d: %s\n", linecnt - 2, context_p2_line);
|
||
free(context_p2_line);
|
||
}
|
||
ELOG("CONFIG: Line %3d: %s\n", linecnt - 1, context_p1_line);
|
||
free(context_p1_line);
|
||
}
|
||
ELOG("CONFIG: Line %3d: %s\n", linecnt, error_copy);
|
||
ELOG("CONFIG: %s\n", position);
|
||
free(error_copy);
|
||
/* Print context lines *after* the error, if any. */
|
||
for (int i = 0; i < 2; i++) {
|
||
char *error_line_end = strchr(error_line, '\n');
|
||
if (error_line_end != NULL && *(error_line_end + 1) != '\0') {
|
||
error_line = error_line_end + 1;
|
||
error_copy = single_line(error_line);
|
||
ELOG("CONFIG: Line %3d: %s\n", linecnt + i + 1, error_copy);
|
||
free(error_copy);
|
||
}
|
||
}
|
||
|
||
context->has_errors = true;
|
||
|
||
/* Format this error message as a JSON reply. */
|
||
y(map_open);
|
||
ystr("success");
|
||
y(bool, false);
|
||
/* We set parse_error to true to distinguish this from other
|
||
* errors. i3-nagbar is spawned upon keypresses only for parser
|
||
* errors. */
|
||
ystr("parse_error");
|
||
y(bool, true);
|
||
ystr("error");
|
||
ystr(errormessage);
|
||
ystr("input");
|
||
ystr(input);
|
||
ystr("errorposition");
|
||
ystr(position);
|
||
y(map_close);
|
||
|
||
/* Skip the rest of this line, but continue parsing. */
|
||
while ((walk - input) <= len && *walk != '\n')
|
||
walk++;
|
||
|
||
free(position);
|
||
free(errormessage);
|
||
clear_stack();
|
||
|
||
/* To figure out in which state to go (e.g. MODE or INITIAL),
|
||
* we find the nearest state which contains an <error> token
|
||
* and follow that one. */
|
||
bool error_token_found = false;
|
||
for (int i = statelist_idx-1; (i >= 0) && !error_token_found; i--) {
|
||
cmdp_token_ptr *errptr = &(tokens[statelist[i]]);
|
||
for (int j = 0; j < errptr->n; j++) {
|
||
if (strcmp(errptr->array[j].name, "error") != 0)
|
||
continue;
|
||
next_state(&(errptr->array[j]));
|
||
error_token_found = true;
|
||
break;
|
||
}
|
||
}
|
||
|
||
assert(error_token_found);
|
||
}
|
||
}
|
||
|
||
y(array_close);
|
||
|
||
return &command_output;
|
||
}
|
||
|
||
/*******************************************************************************
|
||
* Code for building the stand-alone binary test.commands_parser which is used
|
||
* by t/187-commands-parser.t.
|
||
******************************************************************************/
|
||
|
||
#ifdef TEST_PARSER
|
||
|
||
/*
|
||
* Logs the given message to stdout while prefixing the current time to it,
|
||
* but only if debug logging was activated.
|
||
* This is to be called by DLOG() which includes filename/linenumber
|
||
*
|
||
*/
|
||
void debuglog(char *fmt, ...) {
|
||
va_list args;
|
||
|
||
va_start(args, fmt);
|
||
fprintf(stdout, "# ");
|
||
vfprintf(stdout, fmt, args);
|
||
va_end(args);
|
||
}
|
||
|
||
void errorlog(char *fmt, ...) {
|
||
va_list args;
|
||
|
||
va_start(args, fmt);
|
||
vfprintf(stderr, fmt, args);
|
||
va_end(args);
|
||
}
|
||
|
||
static int criteria_next_state;
|
||
|
||
void cfg_criteria_init(I3_CFG, int _state) {
|
||
criteria_next_state = _state;
|
||
}
|
||
|
||
void cfg_criteria_add(I3_CFG, const char *ctype, const char *cvalue) {
|
||
}
|
||
|
||
void cfg_criteria_pop_state(I3_CFG) {
|
||
result->next_state = criteria_next_state;
|
||
}
|
||
|
||
int main(int argc, char *argv[]) {
|
||
if (argc < 2) {
|
||
fprintf(stderr, "Syntax: %s <command>\n", argv[0]);
|
||
return 1;
|
||
}
|
||
struct context context;
|
||
context.filename = "<stdin>";
|
||
parse_config(argv[1], &context);
|
||
}
|
||
#endif
|