1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670
|
/** @file
Parser: String class part: untaint mechanizm.
Copyright (c) 2001-2017 Art. Lebedev Studio (http://www.artlebedev.com)
Author: Alexandr Petrosian <paf@design.ru> (http://paf.design.ru)
*/
volatile const char * IDENT_UNTAINT_C="$Id: untaint.C,v 1.172 2017/02/07 22:00:45 moko Exp $";
#include "pa_string.h"
#include "pa_hash.h"
#include "pa_exception.h"
#include "pa_table.h"
#include "pa_globals.h"
#include "pa_dictionary.h"
#include "pa_common.h"
#include "pa_charset.h"
#include "pa_request_charsets.h"
#include "pa_sapi.h"
extern "C" { // author forgot to do that
#include "ec.h"
}
#include "pa_sql_connection.h"
// defines
#undef CORD_ec_append
// redefining to intercept flushes and implement whitespace optimization
// of all consequent white space chars leaving only first one
#define CORD_ec_append(x, c) \
{ \
bool skip=false; \
if(optimize) switch(c) { \
case ' ': case '\n': case '\t': \
if(whitespace) \
skip=true; /*skipping subsequent*/ \
else \
whitespace=true; \
break; \
default: \
whitespace=false; \
break; \
} \
if(!skip) { \
if ((x)[0].ec_bufptr == (x)[0].ec_buf + CORD_BUFSZ) { \
CORD_ec_flush_buf(x); \
} \
*((x)[0].ec_bufptr)++ = (c); \
} \
}
#define escape_fragment(action) \
for(; fragment_length--; CORD_next(info->pos)) { \
char c=CORD_pos_fetch(info->pos); \
action \
}
#define encode(need_encode_func, prefix, otherwise) \
if(need_encode_func(c)) { \
CORD_ec_append(info->result, prefix); \
to_hex(c); \
} else \
CORD_ec_append(info->result, otherwise);
#define to_hex(c) \
{ \
CORD_ec_append(info->result, hex_digits[((unsigned char)c) >> 4]); \
CORD_ec_append(info->result, hex_digits[((unsigned char)c) & 0x0F]); \
}
#define to_char(c) { CORD_ec_append(info->result, c); whitespace=false; }
#define to_string(s) { CORD_ec_append_cord(info->result, (CORD)(s)); whitespace=false; }
#define _default CORD_ec_append(info->result, c)
inline bool need_file_encode(unsigned char c){
// russian letters and space ENABLED
// encoding only these...
return strchr(
"*?\"<>|"
#ifndef WIN32
":\\"
#endif
, c)!=0;
}
inline bool need_uri_encode(unsigned char c){
return !(pa_isalnum(c) || strchr("_-./*", c));
}
inline bool need_regex_escape(unsigned char c){
return strchr("\\^$.[]|()?*+{}-", c)!=0;
}
inline bool need_parser_code_escape(unsigned char c){
return strchr("^$;@()[]{}:#\"", c)!=0;
}
// String
/*
HTTP-header = field-name ":" [ field-value ] CRLF
field-name = token
field-value = *( field-content | LWS )
field-content = <the OCTETs making up the field-value
and consisting of either *TEXT or combinations
of token, tspecials, and quoted-string>
token = 1*<any CHAR except CTLs or tspecials>
word = token | quoted-string
quoted-string = ( <"> *(qdtext | quoted-pair ) <"> )
qdtext = <any TEXT except <">>
quoted-pair = "\" CHAR
OCTET = <any 8-bit sequence of data>
CHAR = <any US-ASCII character (octets 0 - 127)>
tspecials = "(" | ")" | "<" | ">" | "@"
| "," | ";" | ":" | "\" | <">
| "/" | "[" | "]" | "?" | "="
| "{" | "}" | SP | HT
SP = <US-ASCII SP, space (32)>
HT = <US-ASCII HT, horizontal-tab (9)>
LWS = [CRLF] 1*( SP | HT )
TEXT = <any OCTET except CTLs, but including LWS>
CTL = <any US-ASCII control character (octets 0 - 31) and DEL (127)>
if(strchr("()<>@,;:\\\"/[]?={} \t", *ptr))
*/
inline bool need_quote_http_header(const char* ptr, size_t size) {
for(; size--; ptr++)
if(strchr(";\\\"= \t" /* excluded ()<>@, :/ ? []{} */, *ptr))
return true;
return false;
}
#ifndef DOXYGEN
struct Append_fragment_info {
String::Language lang;
String::Languages* dest_languages;
size_t dest_body_plan_length;
};
#endif
int append_fragment_optimizing(char alang, size_t asize, Append_fragment_info* info) {
const String::Language lang=(String::Language)(unsigned char)alang;
// main idea here:
// tainted piece would get OPTIMIZED bit from 'lang'
// clean piece would be marked OPTIMIZED manually
// pieces with determined languages [not tainted|clean] would retain theirs langs
info->dest_languages->append(info->dest_body_plan_length,
lang==String::L_TAINTED?
info->lang
:lang==String::L_CLEAN?
(String::Language)(String::L_CLEAN|String::L_OPTIMIZE_BIT) // ORing with OPTIMIZED flag
:lang,
asize);
info->dest_body_plan_length+=asize;
return 0; // 0=continue
}
int append_fragment_nonoptimizing(char alang, size_t asize, Append_fragment_info* info) {
const String::Language lang=(String::Language)(unsigned char)alang;
// The core idea: tainted pieces got marked with context's lang
info->dest_languages->append(info->dest_body_plan_length,
lang==String::L_TAINTED?
info->lang
:lang,
asize);
info->dest_body_plan_length+=asize;
return 0; // 0=continue
}
/**
appends to other String without language change
*/
String& String::append_to(String& dest) const {
if(is_empty())
return dest;
// first: fragment infos
dest.langs.appendHelper(dest.body, langs, body);
// next: letters
dest.body<<body;
ASSERT_STRING_INVARIANT(dest);
return dest;
}
/**
appends to other String,
marking all tainted pieces of it with @a lang.
or marking ALL pieces of it with a @a lang when @a forced to,
and propagating OPTIMIZE language bit.
*/
String& String::append_to(String& dest, Language ilang, bool forced) const {
if(is_empty())
return dest;
// first: fragment infos
if(forced) //forcing passed lang?
dest.langs.appendHelper(dest.body, ilang, body);
else {
if(langs.opt.is_not_just_lang){
Append_fragment_info info={ilang, &dest.langs, dest.body.length()};
langs.for_each(body, ilang&L_OPTIMIZE_BIT?
append_fragment_optimizing
:append_fragment_nonoptimizing, &info);
} else {
Language lang=langs.opt.lang;
// see append_fragment_* for explanation
if(ilang&L_OPTIMIZE_BIT){
dest.langs.appendHelper(dest.body,
lang==String::L_TAINTED?
ilang
:lang==String::L_CLEAN?
(String::Language)(String::L_CLEAN|String::L_OPTIMIZE_BIT)
:lang,
body);
} else {
dest.langs.appendHelper(dest.body, lang==String::L_TAINTED ? ilang:lang, body);
}
}
}
// next: letters
dest.body<<body;
ASSERT_STRING_INVARIANT(dest);
return dest;
}
/** http://www.ietf.org/rfc/rfc2047.txt
RFC
(3) As a replacement for a 'word' entity within a 'phrase', for example,
one that precedes an address in a From, To, or Cc header. The ABNF
definition for 'phrase' from RFC 822 thus becomes:
phrase = 1*( encoded-word / word )
In this case the set of characters that may be used in a "Q"-encoded
'encoded-word' is restricted to: <upper and lower case ASCII
letters, decimal digits, "!", "*", "+", "-", "/", "=", and "_"
(underscore, ASCII 95.)>. An 'encoded-word' that appears within a
'phrase' MUST be separated from any adjacent 'word', 'text' or
'special' by 'linear-white-space'.
...
(2) The 8-bit hexadecimal value 20 (e.g., ISO-8859-1 SPACE) may be
represented as "_" (underscore, ASCII 95.). (This character may
not pass through some internetwork mail gateways, but its use
will greatly enhance readability of "Q" encoded data with mail
readers that do not support this encoding.) Note that the "_"
always represents hexadecimal 20, even if the SPACE character
occupies a different code position in the character set in use.
paf: obviously,
without "=", or one could not differ "=E0" and "russian letter a"
and without "_", or in would mean 0x20
*/
inline bool mail_header_char_valid_within_Qencoded(char c) {
return (pa_isalnum((unsigned char)c) || strchr("!*+-/", c));
}
inline bool addr_spec_soon(const char *src) {
for(char c; (c=*src); src++)
if(c=='<')
return true;
else if(!(c==' ' || c=='\t'))
return false;
return false;
}
/**
RFC
Upper case should be used for hexadecimal digits "A" through "F"
The 8-bit hexadecimal value 20 (e.g., ISO-8859-1 SPACE)
may be represented as "_"
*/
inline bool mail_header_nonspace_char(char c) {
return c != 0x20;
}
inline void ec_append(CORD_ec& result, bool& optimize, bool& whitespace, CORD_pos pos, size_t size) {
while(size--) {
CORD_ec_append(result, CORD_pos_fetch(pos));
CORD_next(pos);
}
}
inline void pa_CORD_pos_advance(CORD_pos pos, size_t n) {
while(true) {
long avail=CORD_pos_chars_left(pos);
if(avail<=0) {
CORD_next(pos);
if(!--n)
break;
} else if((size_t)avail<n) {
CORD_pos_advance(pos, avail);
n-=avail;
} else { // avail>=n
CORD_pos_advance(pos, n);
break;
}
}
}
#ifndef DOXYGEN
struct Cstr_to_string_body_block_info {
// input
String::Language lang;
SQL_Connection* connection;
const Request_charsets* charsets;
const String::Body* body;
// output
CORD_ec result;
// private
CORD_pos pos;
size_t fragment_begin;
bool whitespace;
const char* exception;
};
#endif
// @todo: replace info->body->mid with something that uses info->pos
int cstr_to_string_body_block(String::Language to_lang, size_t fragment_length, Cstr_to_string_body_block_info* info) {
bool& whitespace=info->whitespace;
size_t fragment_end=info->fragment_begin+fragment_length;
//fprintf(stderr, "%d, %d =%s=\n", to_lang, fragment_length, info->body->cstr());
bool optimize=(to_lang & String::L_OPTIMIZE_BIT)!=0;
if(!optimize)
whitespace=false;
switch(to_lang & ~String::L_OPTIMIZE_BIT) {
case String::L_CLEAN:
case String::L_TAINTED:
case String::L_AS_IS:
// clean piece
// tainted piece, but undefined untaint language
// for VString.as_double of tainted values
// for ^process{body} evaluation
// tainted, untaint language: as-is
ec_append(info->result, optimize, whitespace, info->pos, fragment_length);
break;
case String::L_FILE_SPEC:
// tainted, untaint language: file [name]
{
escape_fragment(
encode(need_file_encode, '_', c);
);
}
break;
case String::L_URI:
// tainted, untaint language: uri
{
const char *fragment_str=info->body->mid(info->fragment_begin, fragment_length).cstr();
// skip source [we use recoded version]
pa_CORD_pos_advance(info->pos, fragment_length);
String::C output(fragment_str, fragment_length);
if(info->charsets)
output=Charset::transcode(output, info->charsets->source(), info->charsets->client());
char c;
for(const char* src=output.str; (c=*src++); )
encode(need_uri_encode, '%', c);
}
break;
case String::L_HTTP_HEADER:
// tainted, untaint language: http-field-content-text
escape_fragment(switch(c) {
case '\n':
case '\r': to_string(" "); break;
default: _default; break;
});
break;
case String::L_MAIL_HEADER:
// tainted, untaint language: mail-header
// http://www.ietf.org/rfc/rfc2047.txt
if(info->charsets) {
size_t mail_size;
const char *mail_ptr=
info->body->mid(info->fragment_begin, mail_size=fragment_length).cstr();
// skip source [we use recoded version]
pa_CORD_pos_advance(info->pos, mail_size);
const char* charset_name=info->charsets->mail().NAME().cstr();
// Subject: Re: parser3: =?koi8-r?Q?=D3=C5=CD=C9=CE=C1=D2?=
bool to_quoted_printable=false;
bool email=false;
uchar c;
for(const char* src=mail_ptr; (c=(uchar)*src++); ) {
if(c=='\r' || c=='\n')
c=' ';
if(to_quoted_printable && (c==',' || c == '"' || addr_spec_soon(src-1/*position to 'c'*/))) {
email=c=='<';
to_string("?=");
to_quoted_printable=false;
}
//RFC + An 'encoded-word' MUST NOT appear in any portion of an 'addr-spec'.
if(!email && (
( !to_quoted_printable && (c & 0x80) ) // starting quote-printable-encoding on first 8bit char
|| ( to_quoted_printable && !mail_header_char_valid_within_Qencoded(c) )
)) {
if(!to_quoted_printable) {
to_string("=?");
to_string(charset_name);
to_string("?Q?");
to_quoted_printable=true;
}
encode(mail_header_nonspace_char, '=', '_');
} else
to_char(c);
if(c=='>')
email=false;
}
if(to_quoted_printable) // close
to_string("?=");
} else
ec_append(info->result, optimize, whitespace, info->pos, fragment_length);
break;
case String::L_SQL:
// tainted, untaint language: sql
if(info->connection) {
const char *fragment_str=info->body->mid(info->fragment_begin, fragment_length).cstr();
// skip source [we use recoded version]
pa_CORD_pos_advance(info->pos, fragment_length);
to_string(info->connection->quote(fragment_str, fragment_length));
} else {
info->exception="untaint in SQL language failed - no connection specified";
info->fragment_begin=fragment_end;
return 1; // stop processing. can't throw exception here
}
break;
case String::L_JS:
escape_fragment(switch(c) {
case '\n': to_string("\\n"); break;
case '"': to_string("\\\""); break;
case '\'': to_string("\\'"); break;
case '\\': to_string("\\\\"); break;
case '\xFF': to_string("\\\xFF"); break;
case '\r': to_string("\\r"); break;
default: _default; break;
});
break;
case String::L_XML:
// [2] Char ::= #x9 | #xA | #xD | [#x20-#xD7FF] | [#xE000-#xFFFD] | [#x10000-#x10FFFF]
escape_fragment(switch(c) {
case '\x20':
case '\x9':
case '\xA':
case '\xD': // this is usually removed on input
_default;
break;
case '&': to_string("&"); break;
case '>': to_string(">"); break;
case '<': to_string("<"); break;
case '"': to_string("""); break;
case '\'': to_string("'"); break;
default:
if(((unsigned char)c)<0x20) {
// fixing it, so that libxml would not result
// in fatal error parsing text
// though it really violates standard.
// to indicate there were an error
// replace bad char not to it's code,
// which we can do,
// but rather to '!' to show that input were actually
// invalid.
// life: shows that MSIE can somehow garble form values
// so that they contain these chars.
to_char('!');
} else {
_default;
}
break;
});
break;
case String::L_HTML:
escape_fragment(switch(c) {
case '&': to_string("&"); break;
case '>': to_string(">"); break;
case '<': to_string("<"); break;
case '"': to_string("""); break;
default: _default; break;
});
break;
case String::L_REGEX:
// tainted, untaint language: regex
escape_fragment(
if(need_regex_escape(c))
to_char('\\')
_default;
);
break;
case String::L_JSON:
// tainted, untaint language: json <http://json.org/>
// escape '"' '\' '/' '\n' '\t' '\r' '\b' '\f' chars and escape chars as \uXXXX if output charset != UTF-8
{
if(info->charsets==NULL || info->charsets->client().isUTF8()){
// escaping to \uXXXX is not needed
escape_fragment(switch((unsigned char)c) {
case '\n': to_string("\\n"); break;
case '"' : to_string("\\\""); break;
case '\\': to_string("\\\\"); break;
case '/' : to_string("\\/"); break;
case '\t': to_string("\\t"); break;
case '\r': to_string("\\r"); break;
case '\b': to_string("\\b"); break;
case '\f': to_string("\\f"); break;
case 0xE2: // \u2028 and \u2029 (line/paragraph separators), check bug #1023
if(info->charsets && info->charsets->source().isUTF8() && fragment_length>=2){
CORD_next(info->pos);
char c1=CORD_pos_fetch(info->pos);
CORD_next(info->pos);
char c2=CORD_pos_fetch(info->pos);
if((unsigned char)c1 == 0x80 && ((unsigned char)c2 >= 0xA8 && (unsigned char)c2 <= 0xAF)){
to_string("\\u20");
to_hex(((unsigned char)c2-0x80));
} else {
CORD_ec_append(info->result, c);
CORD_ec_append(info->result, c1);
CORD_ec_append(info->result, c2);
}
fragment_length-=2;
} else {
_default;
}
break;
default:
if((unsigned char)c < 0x20){
to_string("\\u00");
to_hex(c);
} else {
_default;
}
break;
});
} else {
const char *fragment_str=info->body->mid(info->fragment_begin, fragment_length).cstr();
// skip source [we use recoded version]
pa_CORD_pos_advance(info->pos, fragment_length);
to_string(Charset::escape_JSON(String::C(fragment_str, fragment_length), info->charsets->source()).str);
}
}
break;
case String::L_HTTP_COOKIE:
// tainted, untaint language: cookie (3.3.0 and higher: %uXXXX in UTF-8)
if(info->charsets) {
const char *fragment_str=info->body->mid(info->fragment_begin, fragment_length).cstr();
// skip source [we use recoded version]
pa_CORD_pos_advance(info->pos, fragment_length);
to_string(Charset::escape(String::C(fragment_str, fragment_length), info->charsets->source()).str);
} else
ec_append(info->result, optimize, whitespace, info->pos, fragment_length);
break;
case String::L_PARSER_CODE:
// for auto-untaint in process
escape_fragment(
if(need_parser_code_escape(c))
to_char('^');
_default;
);
break;
default:
SAPI::abort("unknown untaint language #%d",
static_cast<int>(to_lang)); // should never
break; // never
}
info->fragment_begin=fragment_end;
return 0; // 0=continue
}
String::Body String::cstr_to_string_body_taint(Language lang, SQL_Connection* connection, const Request_charsets *charsets) const {
if(is_empty())
return String::Body();
Cstr_to_string_body_block_info info;
// input
info.lang=lang;
info.connection=connection;
info.charsets=charsets;
info.body=&body;
// output
CORD_ec_init(info.result);
// private
body.set_pos(info.pos, 0);
info.fragment_begin=0;
info.exception=0;
info.whitespace=true;
cstr_to_string_body_block(lang, length(), &info);
if(info.exception)
throw Exception(0,
0,
info.exception);
return String::Body(CORD_ec_to_cord(info.result));
}
int cstr_to_string_body_block_untaint(char alang, size_t fragment_length, Cstr_to_string_body_block_info* info){
const String::Language lang=(String::Language)(unsigned char)alang;
// see append_fragment_* for explanation
if(info->lang&String::L_OPTIMIZE_BIT)
return cstr_to_string_body_block(
lang==String::L_TAINTED?
info->lang
:lang==String::L_CLEAN?
(String::Language)(String::L_CLEAN|String::L_OPTIMIZE_BIT)
:lang,
fragment_length, info);
else
return cstr_to_string_body_block(lang==String::L_TAINTED ? info->lang:lang, fragment_length, info);
}
String::Body String::cstr_to_string_body_untaint(Language lang, SQL_Connection* connection, const Request_charsets *charsets) const {
if(is_empty())
return String::Body();
Cstr_to_string_body_block_info info;
// input
info.lang=lang;
info.connection=connection;
info.charsets=charsets;
info.body=&body;
// output
CORD_ec_init(info.result);
// private
body.set_pos(info.pos, 0);
info.fragment_begin=0;
info.exception=0;
info.whitespace=true;
langs.for_each(body, cstr_to_string_body_block_untaint, &info);
if(info.exception)
throw Exception(0,
0,
info.exception);
return String::Body(CORD_ec_to_cord(info.result));
}
const char* String::untaint_and_transcode_cstr(Language lang, const Request_charsets *charsets) const {
if(charsets && &charsets->source() != &charsets->client()){
// Note: L_URI is allready transcoded during untaint, but transcode does not affect %XX
return Charset::transcode(cstr_to_string_body_untaint(lang, 0, charsets), charsets->source(), charsets->client()).cstr();
} else
return cstr_to_string_body_untaint(lang, 0, charsets).cstr();
}
|