bird/conf/cf-lex.l
Ondrej Zajicek (work) 29dda184e5 Conf: Fix parsing full-length IPv6 addresses
Lexer expression for bytestring was too loose, accepting also
full-length IPv6 addresses. It should be restricted such that
colon is used between every byte or never.

Fix the regex and also add some test cases for it.

Thanks to Alexander Zubkov for the bugreport
2022-01-05 16:38:49 +01:00

862 lines
19 KiB
Plaintext

/*
* BIRD -- Configuration Lexer
*
* (c) 1998--2000 Martin Mares <mj@ucw.cz>
*
* Can be freely distributed and used under the terms of the GNU GPL.
*/
/**
* DOC: Lexical analyzer
*
* The lexical analyzer used for configuration files and CLI commands
* is generated using the |flex| tool accompanied by a couple of
* functions maintaining the hash tables containing information about
* symbols and keywords.
*
* Each symbol is represented by a &symbol structure containing name
* of the symbol, its lexical scope, symbol class (%SYM_PROTO for a
* name of a protocol, %SYM_CONSTANT for a constant etc.) and class
* dependent data. When an unknown symbol is encountered, it's
* automatically added to the symbol table with class %SYM_VOID.
*
* The keyword tables are generated from the grammar templates
* using the |gen_keywords.m4| script.
*/
%{
#undef REJECT /* Avoid name clashes */
#include <errno.h>
#include <stdlib.h>
#include <stdarg.h>
#include <stdint.h>
#include <unistd.h>
#include <libgen.h>
#include <glob.h>
#include <fcntl.h>
#include <sys/stat.h>
#include <sys/types.h>
#include <sys/stat.h>
#define PARSER 1
#include "nest/bird.h"
#include "nest/route.h"
#include "nest/protocol.h"
#include "filter/filter.h"
#include "filter/f-inst.h"
#include "conf/conf.h"
#include "conf/cf-parse.tab.h"
#include "lib/string.h"
#include "lib/hash.h"
struct keyword {
byte *name;
int value;
struct keyword *next;
};
#include "conf/keywords.h"
/* Could be defined by Bison in cf-parse.tab.h, inteferes with SYM hash */
#ifdef SYM
#undef SYM
#endif
static uint cf_hash(const byte *c);
#define KW_KEY(n) n->name
#define KW_NEXT(n) n->next
#define KW_EQ(a,b) !strcmp(a,b)
#define KW_FN(k) cf_hash(k)
#define KW_ORDER 8 /* Fixed */
#define SYM_KEY(n) n->name, n->scope->active
#define SYM_NEXT(n) n->next
#define SYM_EQ(a,s1,b,s2) !strcmp(a,b) && s1 == s2
#define SYM_FN(k,s) cf_hash(k)
#define SYM_ORDER 6 /* Initial */
#define SYM_REHASH sym_rehash
#define SYM_PARAMS /8, *1, 2, 2, 6, 20
HASH_DEFINE_REHASH_FN(SYM, struct symbol)
HASH(struct keyword) kw_hash;
struct sym_scope *conf_this_scope;
linpool *cfg_mem;
int (*cf_read_hook)(byte *buf, unsigned int max, int fd);
struct include_file_stack *ifs;
static struct include_file_stack *ifs_head;
#define QUOTED_BUFFER_SIZE 4096
static BUFFER_(char) quoted_buffer;
static char quoted_buffer_data[QUOTED_BUFFER_SIZE];
static inline void quoted_buffer_init(void) {
quoted_buffer.used = 0;
quoted_buffer.size = QUOTED_BUFFER_SIZE;
quoted_buffer.data = quoted_buffer_data;
}
#define MAX_INCLUDE_DEPTH 8
#define YY_INPUT(buf,result,max) result = cf_read_hook(buf, max, ifs->fd);
#define YY_NO_UNPUT
#define YY_FATAL_ERROR(msg) cf_error(msg)
#define YY_USER_ACTION ifs->chno += yyleng; ifs->toklen = yyleng;
static void cf_include(char *arg, int alen);
static int check_eof(void);
static enum yytokentype cf_lex_symbol(const char *data);
%}
%option noyywrap
%option noinput
%option nounput
%option noreject
%x COMMENT CCOMM CLI QUOTED APOSTROPHED INCLUDE
ALPHA [a-zA-Z_]
DIGIT [0-9]
XIGIT [0-9a-fA-F]
ALNUM [a-zA-Z_0-9]
WHITE [ \t]
%%
^{WHITE}*include{WHITE}*\" {
if (!ifs->depth)
cf_error("Include not allowed in CLI");
BEGIN(INCLUDE);
}
<INCLUDE>[^"\n]+["]{WHITE}*; {
char *start, *end;
start = yytext;
end = strchr(start, '"');
*end = 0;
if (start == end)
cf_error("Include with empty argument");
cf_include(start, end-start);
BEGIN(INITIAL);
}
<INCLUDE>["] cf_error("Include with empty argument");
<INCLUDE>. cf_error("Unterminated include");
<INCLUDE>\n cf_error("Unterminated include");
<INCLUDE><<EOF>> cf_error("Unterminated include");
{DIGIT}+:{DIGIT}+ {
uint len1 UNUSED, len2;
u64 l;
char *e;
errno = 0;
l = bstrtoul10(yytext, &e);
if (!e || (*e != ':') || (errno == ERANGE) || (l >> 32))
cf_error("ASN out of range");
if (l >> 16)
{
len1 = 32;
len2 = 16;
cf_lval.i64 = (2ULL << 48) | (((u64) l) << len2);
}
else
{
len1 = 16;
len2 = 32;
cf_lval.i64 = 0 | (((u64) l) << len2);
}
errno = 0;
l = bstrtoul10(e+1, &e);
if (!e || *e || (errno == ERANGE) || (l >> len2))
cf_error("Number out of range");
cf_lval.i64 |= l;
return VPN_RD;
}
[02]:{DIGIT}+:{DIGIT}+ {
uint len1, len2;
u64 l;
char *e;
if (yytext[0] == '0')
{
cf_lval.i64 = 0;
len1 = 16;
len2 = 32;
}
else
{
cf_lval.i64 = 2ULL << 48;
len1 = 32;
len2 = 16;
}
errno = 0;
l = bstrtoul10(yytext+2, &e);
if (!e || (*e != ':') || (errno == ERANGE) || (l >> len1))
cf_error("ASN out of range");
cf_lval.i64 |= ((u64) l) << len2;
errno = 0;
l = bstrtoul10(e+1, &e);
if (!e || *e || (errno == ERANGE) || (l >> len2))
cf_error("Number out of range");
cf_lval.i64 |= l;
return VPN_RD;
}
{DIGIT}+\.{DIGIT}+\.{DIGIT}+\.{DIGIT}+:{DIGIT}+ {
unsigned long int l;
ip4_addr ip4;
char *e;
cf_lval.i64 = 1ULL << 48;
e = strchr(yytext, ':');
*e++ = '\0';
if (!ip4_pton(yytext, &ip4))
cf_error("Invalid IPv4 address %s in Route Distinguisher", yytext);
cf_lval.i64 |= ((u64) ip4_to_u32(ip4)) << 16;
errno = 0;
l = bstrtoul10(e, &e);
if (!e || *e || (errno == ERANGE) || (l >> 16))
cf_error("Number out of range");
cf_lval.i64 |= l;
return VPN_RD;
}
{DIGIT}+\.{DIGIT}+\.{DIGIT}+\.{DIGIT}+ {
if (!ip4_pton(yytext, &cf_lval.ip4))
cf_error("Invalid IPv4 address %s", yytext);
return IP4;
}
{XIGIT}{2}((:{XIGIT}{2}){15,}|({XIGIT}{2}){15,}) {
char *s = yytext;
size_t len = 0, i;
struct bytestring *bytes;
byte *b;
while (*s) {
len++;
s += 2;
if (*s == ':')
s++;
}
bytes = cfg_allocz(sizeof(*bytes) + len);
bytes->length = len;
b = &bytes->data[0];
s = yytext;
errno = 0;
for (i = 0; i < len; i++) {
*b = bstrtobyte16(s);
if (errno == ERANGE)
cf_error("Invalid hex string");
b++;
s += 2;
if (*s == ':')
s++;
}
cf_lval.bs = bytes;
return BYTESTRING;
}
({XIGIT}*::|({XIGIT}*:){3,})({XIGIT}*|{DIGIT}+\.{DIGIT}+\.{DIGIT}+\.{DIGIT}+) {
if (!ip6_pton(yytext, &cf_lval.ip6))
cf_error("Invalid IPv6 address %s", yytext);
return IP6;
}
0x{XIGIT}+ {
char *e;
unsigned long int l;
errno = 0;
l = bstrtoul16(yytext+2, &e);
if (!e || *e || errno == ERANGE || (unsigned long int)(unsigned int) l != l)
cf_error("Number out of range");
cf_lval.i = l;
return NUM;
}
{DIGIT}+ {
char *e;
unsigned long int l;
errno = 0;
l = bstrtoul10(yytext, &e);
if (!e || *e || errno == ERANGE || (unsigned long int)(unsigned int) l != l)
cf_error("Number out of range");
cf_lval.i = l;
return NUM;
}
else: {
/* Hack to distinguish if..else from else: in case */
return ELSECOL;
}
['] {
BEGIN(APOSTROPHED);
quoted_buffer_init();
}
<APOSTROPHED>{ALNUM}|[-]|[.:] BUFFER_PUSH(quoted_buffer) = yytext[0];
<APOSTROPHED>\n cf_error("Unterminated symbol");
<APOSTROPHED><<EOF>> cf_error("Unterminated symbol");
<APOSTROPHED>['] {
BEGIN(INITIAL);
BUFFER_PUSH(quoted_buffer) = 0;
return cf_lex_symbol(quoted_buffer_data);
}
<APOSTROPHED>. cf_error("Invalid character in apostrophed symbol");
({ALPHA}{ALNUM}*) {
return cf_lex_symbol(yytext);
}
<CLI>(.|\n) {
BEGIN(INITIAL);
return CLI_MARKER;
}
\.\. {
return DDOT;
}
[={}:;,.()+*/%<>~\[\]?!\|-] {
return yytext[0];
}
["] {
BEGIN(QUOTED);
quoted_buffer_init();
}
<QUOTED>\n cf_error("Unterminated string");
<QUOTED><<EOF>> cf_error("Unterminated string");
<QUOTED>["] {
BEGIN(INITIAL);
BUFFER_PUSH(quoted_buffer) = 0;
cf_lval.t = cfg_strdup(quoted_buffer_data);
return TEXT;
}
<QUOTED>. BUFFER_PUSH(quoted_buffer) = yytext[0];
<INITIAL,COMMENT><<EOF>> { if (check_eof()) return END; }
{WHITE}+
\n ifs->lino++; ifs->chno = 0;
# BEGIN(COMMENT);
\/\* BEGIN(CCOMM);
. cf_error("Unknown character");
<COMMENT>\n {
ifs->lino++;
ifs->chno = 0;
BEGIN(INITIAL);
}
<COMMENT>.
<CCOMM>\*\/ BEGIN(INITIAL);
<CCOMM>\n ifs->lino++; ifs->chno = 0;
<CCOMM>\/\* cf_error("Comment nesting not supported");
<CCOMM><<EOF>> cf_error("Unterminated comment");
<CCOMM>.
\!\= return NEQ;
\!\~ return NMA;
\<\= return LEQ;
\>\= return GEQ;
\&\& return AND;
\|\| return OR;
\[\= return PO;
\=\] return PC;
%%
static uint
cf_hash(const byte *c)
{
uint h = 13 << 24;
while (*c)
h = h + (h >> 2) + (h >> 5) + ((uint) *c++ << 24);
return h;
}
/*
* IFS stack - it contains structures needed for recursive processing
* of include in config files. On the top of the stack is a structure
* for currently processed file. Other structures are either for
* active files interrupted because of include directive (these have
* fd and flex buffer) or for inactive files scheduled to be processed
* later (when parent requested including of several files by wildcard
* match - these do not have fd and flex buffer yet).
*
* FIXME: Most of these ifs and include functions are really sysdep/unix.
*/
static struct include_file_stack *
push_ifs(struct include_file_stack *old)
{
struct include_file_stack *ret;
ret = cfg_allocz(sizeof(struct include_file_stack));
ret->lino = 1;
ret->prev = old;
return ret;
}
static struct include_file_stack *
pop_ifs(struct include_file_stack *old)
{
yy_delete_buffer(old->buffer);
close(old->fd);
return old->prev;
}
static void
enter_ifs(struct include_file_stack *new)
{
if (!new->buffer)
{
new->fd = open(new->file_name, O_RDONLY);
if (new->fd < 0)
{
ifs = ifs->up;
cf_error("Unable to open included file %s: %m", new->file_name);
}
new->buffer = yy_create_buffer(NULL, YY_BUF_SIZE);
}
yy_switch_to_buffer(new->buffer);
}
/**
* cf_lex_unwind - unwind lexer state during error
*
* cf_lex_unwind() frees the internal state on IFS stack when the lexical
* analyzer is terminated by cf_error().
*/
void
cf_lex_unwind(void)
{
struct include_file_stack *n;
for (n = ifs; n != ifs_head; n = n->prev)
{
/* Memory is freed automatically */
if (n->buffer)
yy_delete_buffer(n->buffer);
if (n->fd)
close(n->fd);
}
ifs = ifs_head;
}
static void
cf_include(char *arg, int alen)
{
struct include_file_stack *base_ifs = ifs;
int new_depth, rv, i;
char *patt;
glob_t g = {};
new_depth = ifs->depth + 1;
if (new_depth > MAX_INCLUDE_DEPTH)
cf_error("Max include depth reached");
/* expand arg to properly handle relative filenames */
if (*arg != '/')
{
int dlen = strlen(ifs->file_name);
char *dir = alloca(dlen + 1);
patt = alloca(dlen + alen + 2);
memcpy(dir, ifs->file_name, dlen + 1);
sprintf(patt, "%s/%s", dirname(dir), arg);
}
else
patt = arg;
/* Skip globbing if there are no wildcards, mainly to get proper
response when the included config file is missing */
if (!strpbrk(arg, "?*["))
{
ifs = push_ifs(ifs);
ifs->file_name = cfg_strdup(patt);
ifs->depth = new_depth;
ifs->up = base_ifs;
enter_ifs(ifs);
return;
}
/* Expand the pattern */
rv = glob(patt, GLOB_ERR | GLOB_NOESCAPE, NULL, &g);
if (rv == GLOB_ABORTED)
cf_error("Unable to match pattern %s: %m", patt);
if ((rv != 0) || (g.gl_pathc <= 0))
return;
/*
* Now we put all found files to ifs stack in reverse order, they
* will be activated and processed in order as ifs stack is popped
* by pop_ifs() and enter_ifs() in check_eof().
*/
for(i = g.gl_pathc - 1; i >= 0; i--)
{
char *fname = g.gl_pathv[i];
struct stat fs;
if (stat(fname, &fs) < 0)
{
globfree(&g);
cf_error("Unable to stat included file %s: %m", fname);
}
if (fs.st_mode & S_IFDIR)
continue;
/* Prepare new stack item */
ifs = push_ifs(ifs);
ifs->file_name = cfg_strdup(fname);
ifs->depth = new_depth;
ifs->up = base_ifs;
}
globfree(&g);
enter_ifs(ifs);
}
static int
check_eof(void)
{
if (ifs == ifs_head)
{
/* EOF in main config file */
ifs->lino = 1; /* Why this? */
return 1;
}
ifs = pop_ifs(ifs);
enter_ifs(ifs);
return 0;
}
static struct symbol *
cf_new_symbol(const byte *c)
{
struct symbol *s;
uint l = strlen(c);
if (l > SYM_MAX_LEN)
cf_error("Symbol too long");
s = cfg_allocz(sizeof(struct symbol) + l + 1);
*s = (struct symbol) { .scope = conf_this_scope, .class = SYM_VOID, };
strcpy(s->name, c);
if (!new_config->sym_hash.data)
HASH_INIT(new_config->sym_hash, new_config->pool, SYM_ORDER);
HASH_INSERT2(new_config->sym_hash, SYM, new_config->pool, s);
add_tail(&(new_config->symbols), &(s->n));
return s;
}
/**
* cf_find_symbol - find a symbol by name
* @cfg: specificed config
* @c: symbol name
*
* This functions searches the symbol table in the config @cfg for a symbol of
* given name. First it examines the current scope, then the second recent one
* and so on until it either finds the symbol and returns a pointer to its
* &symbol structure or reaches the end of the scope chain and returns %NULL to
* signify no match.
*/
struct symbol *
cf_find_symbol(const struct config *cfg, const byte *c)
{
struct symbol *s;
if (cfg->sym_hash.data &&
(s = HASH_FIND(cfg->sym_hash, SYM, c, 1)))
return s;
/* In CLI command parsing, fallback points to the current config, otherwise it is NULL. */
if (cfg->fallback &&
cfg->fallback->sym_hash.data &&
(s = HASH_FIND(cfg->fallback->sym_hash, SYM, c, 1)))
return s;
return NULL;
}
/**
* cf_get_symbol - get a symbol by name
* @c: symbol name
*
* This functions searches the symbol table of the currently parsed config
* (@new_config) for a symbol of given name. It returns either the already
* existing symbol or a newly allocated undefined (%SYM_VOID) symbol if no
* existing symbol is found.
*/
struct symbol *
cf_get_symbol(const byte *c)
{
return cf_find_symbol(new_config, c) ?: cf_new_symbol(c);
}
/**
* cf_localize_symbol - get the local instance of given symbol
* @sym: the symbol to localize
*
* This functions finds the symbol that is local to current scope
* for purposes of cf_define_symbol().
*/
struct symbol *
cf_localize_symbol(struct symbol *sym)
{
/* If the symbol type is void, it has been recently allocated just in this scope. */
if (!sym->class)
return sym;
/* If the scope is the current, it is already defined in this scope. */
if (sym->scope == conf_this_scope)
cf_error("Symbol already defined");
/* Not allocated here yet, doing it now. */
return cf_new_symbol(sym->name);
}
struct symbol *
cf_default_name(char *template, int *counter)
{
char buf[SYM_MAX_LEN];
struct symbol *s;
char *perc = strchr(template, '%');
for(;;)
{
bsprintf(buf, template, ++(*counter));
s = cf_get_symbol(buf);
if (s->class == SYM_VOID)
return s;
if (!perc)
break;
}
cf_error("Unable to generate default name");
}
static enum yytokentype
cf_lex_symbol(const char *data)
{
/* Have we defined such a symbol? */
struct symbol *sym = cf_get_symbol(data);
cf_lval.s = sym;
if (sym->class != SYM_VOID)
return CF_SYM_KNOWN;
/* Is it a keyword? */
struct keyword *k = HASH_FIND(kw_hash, KW, data);
if (k)
{
if (k->value > 0)
return k->value;
else
{
cf_lval.i = -k->value;
return ENUM;
}
}
/* OK, undefined symbol */
cf_lval.s = sym;
return CF_SYM_UNDEFINED;
}
static void
cf_lex_init_kh(void)
{
HASH_INIT(kw_hash, &root_pool, KW_ORDER);
struct keyword *k;
for (k=keyword_list; k->name; k++)
HASH_INSERT(kw_hash, KW, k);
}
/**
* cf_lex_init - initialize the lexer
* @is_cli: true if we're going to parse CLI command, false for configuration
* @c: configuration structure
*
* cf_lex_init() initializes the lexical analyzer and prepares it for
* parsing of a new input.
*/
void
cf_lex_init(int is_cli, struct config *c)
{
if (!kw_hash.data)
cf_lex_init_kh();
ifs_head = ifs = push_ifs(NULL);
if (!is_cli)
{
ifs->file_name = c->file_name;
ifs->fd = c->file_fd;
ifs->depth = 1;
}
yyrestart(NULL);
ifs->buffer = YY_CURRENT_BUFFER;
if (is_cli)
BEGIN(CLI);
else
BEGIN(INITIAL);
c->root_scope = cfg_allocz(sizeof(struct sym_scope));
conf_this_scope = c->root_scope;
conf_this_scope->active = 1;
}
/**
* cf_push_scope - enter new scope
* @sym: symbol representing scope name
*
* If we want to enter a new scope to process declarations inside
* a nested block, we can just call cf_push_scope() to push a new
* scope onto the scope stack which will cause all new symbols to be
* defined in this scope and all existing symbols to be sought for
* in all scopes stored on the stack.
*/
void
cf_push_scope(struct symbol *sym)
{
struct sym_scope *s = cfg_allocz(sizeof(struct sym_scope));
s->next = conf_this_scope;
conf_this_scope = s;
s->active = 1;
s->name = sym;
s->slots = 0;
}
/**
* cf_pop_scope - leave a scope
*
* cf_pop_scope() pops the topmost scope from the scope stack,
* leaving all its symbols in the symbol table, but making them
* invisible to the rest of the config.
*/
void
cf_pop_scope(void)
{
conf_this_scope->active = 0;
conf_this_scope = conf_this_scope->next;
ASSERT(conf_this_scope);
}
/**
* cf_symbol_class_name - get name of a symbol class
* @sym: symbol
*
* This function returns a string representing the class
* of the given symbol.
*/
char *
cf_symbol_class_name(struct symbol *sym)
{
switch (sym->class)
{
case SYM_VOID:
return "undefined";
case SYM_PROTO:
return "protocol";
case SYM_TEMPLATE:
return "protocol template";
case SYM_FUNCTION:
return "function";
case SYM_FILTER:
return "filter";
case SYM_TABLE:
return "routing table";
case SYM_ATTRIBUTE:
return "custom attribute";
case SYM_CONSTANT_RANGE:
return "constant";
case SYM_VARIABLE_RANGE:
return "variable";
default:
return "unknown type";
}
}
/**
* DOC: Parser
*
* Both the configuration and CLI commands are analyzed using a syntax
* driven parser generated by the |bison| tool from a grammar which
* is constructed from information gathered from grammar snippets by
* the |gen_parser.m4| script.
*
* Grammar snippets are files (usually with extension |.Y|) contributed
* by various BIRD modules in order to provide information about syntax of their
* configuration and their CLI commands. Each snipped consists of several
* sections, each of them starting with a special keyword: |CF_HDR| for
* a list of |#include| directives needed by the C code, |CF_DEFINES|
* for a list of C declarations, |CF_DECLS| for |bison| declarations
* including keyword definitions specified as |CF_KEYWORDS|, |CF_GRAMMAR|
* for the grammar rules, |CF_CODE| for auxiliary C code and finally
* |CF_END| at the end of the snippet.
*
* To create references between the snippets, it's possible to define
* multi-part rules by utilizing the |CF_ADDTO| macro which adds a new
* alternative to a multi-part rule.
*
* CLI commands are defined using a |CF_CLI| macro. Its parameters are:
* the list of keywords determining the command, the list of parameters,
* help text for the parameters and help text for the command.
*
* Values of |enum| filter types can be defined using |CF_ENUM| with
* the following parameters: name of filter type, prefix common for all
* literals of this type and names of all the possible values.
*/