parse822.c 23.3 KB
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229
/* GNU mailutils - a suite of utilities for electronic mail
   Copyright (C) 1999, 2000, 2001 Free Software Foundation, Inc.

   This program is free software; you can redistribute it and/or modify
   it under the terms of the GNU General Library Public License as published by
   the Free Software Foundation; either version 2, or (at your option)
   any later version.

   This program is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
   GNU Library General Public License for more details.

   You should have received a copy of the GNU Library General Public License
   along with this program; if not, write to the Free Software
   Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.  */

/*
Things to consider:

  - A group should create an address node for a group, accessable
    with address_get_personal(). Perhaps an is_group() would be
    useful? Test that a zero-length phrase is rejected! So these
    are invalid:
      : a@b ;
      "" : ;

  - When parsing phrase, should I ignore non-ascii, or replace with a
    '?' character? Right now parsing fails.

  - Make domain optional in addr-spec, for parsing address lists
    provided to local mail utilities, but NOT in the addr-spec of a
    route-addr.

  - Are comments allowed in domain-literals?

  - Need a way to mark the *end* of a group. Maybe add a field to _address,
    int group_end;, so if you care, you can search for the end of
    a group with address_is_group_end();

  - Need a way to parse "<>", it's a valid SMTP address...

  - Need a way to parse ",,,", it's a valid address-list, it just doesn't
    have any addresses.

  - Functions for forming email addresses, quoting display-name, etc.

  - The personal for ""Sam"" <sam@here> is "Sam", and for "'s@b'" <s@b>
    is 's@b', should I strip those outside parentheses, or is that
    too intrusive? Maybe an apps business if it wants to?

  - Should we do best effort parsing, so parsing "sam@locahost, foo@"
    gets one address, or just say it is or it isn't in RFC format?
    Right now we're strict, we'll see how it goes.

  - parse field names and bodies?
  - parse dates?
  - parse Received: field?

  - test for memory leaks on malloc failure
  - fix the realloc, try a struct _string { char* b, size_t sz };

  - get example mail from drums, and from the perl code.
*/

#ifdef HAVE_CONFIG_H
# include <config.h>
#endif

#include <assert.h>
#include <errno.h>
#include <ctype.h>
#include <string.h>
#include <stdlib.h>

#include "address0.h"

#include <mailutils/parse822.h>

#ifdef EOK
# undef EOK
#endif

#define EOK	0
#define EPARSE	ENOENT

/*
 * Some convenience functions for dealing with dynamically re-sized
 * strings.
 */

static int str_append_n(char** to, const char* from, size_t n)
{
    size_t l = 0;

    /* if not to, then silently discard data */
    if(!to) {
	return EOK;
    }

    if(*to) {
	char* bigger;

	l = strlen(*to);

	bigger = realloc(*to, l + n + 1);

	if(!bigger) {
	    return ENOMEM;
	}

	*to = bigger;
    } else {
	*to = malloc(n + 1);
    }

    strncpy(&to[0][l], from, n);

    /* strncpy is lame, nul terminate our buffer */

    to[0][l + n] = 0;

    return EOK;
}
static int str_append(char** to, const char* from)
{
    return str_append_n(to, from, strlen(from));
}
static int str_append_char(char** to, char c)
{
    return str_append_n(to, &c, 1);
}
static int str_append_range(char** to, const char* b, const char* e)
{
    return str_append_n(to, b, e - b);
}
static void str_free(char** s)
{
    if(s && *s) {
	free(*s);
	*s = 0;
    }
}

/*
 * Character Classification - could be rewritten in a C library
 * independent way, my system's C library matches the RFC
 * definitions. I don't know if that's guaranteed.
 *
 * Note that all return values are:
 *   1 -> TRUE
 *   0 -> FALSE
 * This may be appear different than the 0 == success return
 * values of the other functions, but I was getting lost in
 * boolean arithmetic.
 */
int parse822_is_char(char c)
{
    return isascii(c);
}
int parse822_is_digit(char c)
{
    /* digit = <any ASCII decimal digit> */

    return isdigit(c);
}
int parse822_is_ctl(char c)
{
    return iscntrl(c) || c == 127 /* DEL */;
}
int parse822_is_space(char c)
{
    return c == ' ';
}
int parse822_is_htab(char c)
{
    return c == '\t';
}
int parse822_is_lwsp_char(char c)
{
    return parse822_is_space(c) || parse822_is_htab(c);
}
int parse822_is_special(char c)
{
    return strchr("()<>@,;:\\\".[]", c) ? 1 : 0;
}
int parse822_is_atom_char(char c)
{
    return
	parse822_is_char(c) &&
	!parse822_is_special(c) &&
	!parse822_is_space(c) &&
	!parse822_is_ctl(c)
	;
}
int parse822_is_q_text(char c)
{
    return
	parse822_is_char(c)
	&& c != '"'
	&& c != '\\'
	&& c != '\r'
	;
}
int parse822_is_d_text(char c)
{
    return
	parse822_is_char(c)
	&& c != '['
	&& c != ']'
	&& c != '\\'
	&& c != '\r'
	;
}
/*
 * SMTP's version of qtext, called <q> in the RFC 821 syntax,
 * also excludes <LF>.
 */
int parse822_is_smtp_q(char c)
{
    return parse822_is_q_text(c)
	&& c != '\n';
}

/***** From RFC 822, 3.3 Lexical Tokens *****/

int parse822_skip_ws(const char** p, const char* e)
{
    while((*p != e) && parse822_is_lwsp_char(**p)) {
	*p += 1;
    }
    return EOK;
}

int parse822_skip_comments(const char** p, const char* e)
{
    int status;

    while((status = parse822_comment(p, e, 0)) == EOK)
	;

    return EOK;
}

int parse822_digits(const char** p, const char* e,
	int min, int max, int* digits)
{
    const char* save = *p;

    int i = 0;

    assert(digits);

    *digits = 0;

    while(*p < e && parse822_is_digit(**p)) {
	*digits = *digits * 10 + **p - '0';
	*p += 1;
	++i;
	if(max != 0 && i == max) {
	    break;
	}
    }
    if(i < min) {
	*p = save;
	return EPARSE;
    }

    return EOK;
}

int parse822_special(const char** p, const char* e, char c)
{
    parse822_skip_ws(p, e); /* not comments, they start with a special... */

    if((*p != e) && **p == c) {
	*p += 1;
	return EOK;
    }
    return EPARSE;
}

int parse822_comment(const char** p, const char* e, char** comment)
{
    /* comment = "(" *(ctext / quoted-pair / comment) ")"
     * ctext = <any char except "(", ")", "\", & CR, including lwsp>
     */
    const char* save = *p;
    int rc;

    if((rc = parse822_special(p, e, '('))) {
	return rc;
    }

    while(*p != e) {
	char c = **p;

	if(c == ')') {
	    *p += 1;
	    return EOK; /* found end-of-comment */
	} else if(c == '(') {
	    rc = parse822_comment(p, e, comment);
	} else if(c == '\\') {
	    rc = parse822_quoted_pair(p, e, comment);
	} else if(c == '\r') {
	    /* invalid character... */
	    *p += 1;
	} else if(parse822_is_char(c)) {
	    rc = str_append_char(comment, c);
	    *p += 1;
	} else {
	    /* invalid character... */
	    *p += 1;
	}
	if(rc != EOK)
	    break;
    }

    if(*p == e) {
	rc = EPARSE; /* end-of-comment not found */
    }

    *p = save;

    assert(rc != EOK);

    return rc;
}

int parse822_atom(const char** p, const char* e, char** atom)
{
    /* atom = 1*<an atom char> */

    const char* save = *p;
    int rc = EPARSE;

    parse822_skip_comments(p, e);

    save = *p;

    while((*p != e) && parse822_is_atom_char(**p))
    {
	rc = str_append_char(atom, **p);
	*p += 1;
	if(rc != EOK) {
	    *p = save;
	    break;
	}
    }
    return rc;
}

int parse822_quoted_pair(const char** p, const char* e, char** qpair)
{
    /* quoted-pair = "\" char */

    int rc;

    /* need TWO characters to be available */
    if((e - *p) < 2)
	return EPARSE;

    if(**p != '\\')
	return EPARSE;

    if((rc = str_append_char(qpair, *(*p + 1))))
	return rc;

    *p += 2;

    return EOK;
}

int parse822_quoted_string(const char** p, const char* e, char** qstr)
{
    /* quoted-string = <"> *(qtext/quoted-pair) <">
     * qtext = char except <">, "\", & CR, including lwsp-char
     */

    const char* save = *p;
    int rc;

    parse822_skip_comments(p, e);

    save = *p;

    if((rc = parse822_special(p, e, '"')))
	return rc;

    while(*p != e)
    {
	char c = **p;

	if(c == '"') {
	    *p += 1;
	    return EOK; /* found end-of-qstr */
	} else if(c == '\\') {
	    rc = parse822_quoted_pair(p, e, qstr);
	} else if(c == '\r') {
	    /* invalid character... */
	    *p += 1;
	} else if(parse822_is_char(c)) {
	    rc = str_append_char(qstr, c);
	    *p += 1;
	} else {
	    /* invalid character... */
	    *p += 1;
	}
	if(rc) {
	    *p = save;
	    return rc;
	}
    }
    *p = save;

    return EPARSE; /* end-of-qstr not found */
}

int parse822_word(const char** p, const char* e, char** word)
{
    /* word = atom / quoted-string */

    const char* save = *p;
    int rc = EOK;

    parse822_skip_comments(p, e);

    save = *p;

    {
	char* qstr = 0;
	if((rc = parse822_quoted_string(p, e, &qstr)) == EOK) {
	    rc = str_append(word, qstr);

	    str_free(&qstr);

	    if(rc != EOK)
		*p = save;

	    return rc;
	}
    }

    if(rc != EPARSE) {
	/* it's fatal */
	return rc;
    }

    /* Necessary because the quoted string could have found
     * a partial string (invalid syntax). Thus reset, the atom
     * will fail to if the syntax is invalid.
     */

    {
	char* atom = 0;
	if(parse822_atom(p, e, &atom) == EOK) {
	    rc = str_append(word, atom);

	    str_free(&atom);

	    if(rc != EOK)
		*p = save;

	    return rc;
	}
    }

    return EPARSE;
}

int parse822_phrase(const char** p, const char* e, char** phrase)
{
    /* phrase = 1*word */

    const char* save = *p;
    int rc;

    if((rc = parse822_word(p, e, phrase)))
	return rc;

    /* ok, got the 1 word, now append all the others we can */
    {
	char* word = 0;

	while((rc = parse822_word(p, e, &word)) == EOK) {
	    rc = str_append_char(phrase, ' ');

	    if(rc == EOK)
		rc = str_append(phrase, word);

	    str_free(&word);

	    if(rc != EOK)
		break;
	}
	if(rc == EPARSE)
	    rc = EOK; /* its not an error to find no more words */
    }
    if(rc)
	*p = save;

    return rc;
}

/***** From RFC 822, 6.1 Address Specification Syntax *****/

static address_t new_mb(void) {
    return calloc(1, sizeof(struct _address));
}

static int fill_mb(
	address_t* a,
	char* comments, char* personal, char* local, char* domain)
{
    int rc = EOK;

    *a = new_mb();

    if(!*a) {
	return ENOMEM;
    }

    (*a)->comments = comments;
    (*a)->personal = personal;

    do {
	/* loop exists only to break out of */
        if(!local || !domain) {
	  /* no email to construct */
	  break;
	}
	if((rc = parse822_quote_local_part(&(*a)->email, local)))
	    break;
	if((rc = str_append(&(*a)->email, "@")))
	    break;
	if((rc = str_append(&(*a)->email, domain)))
	    break;
    } while(0);

    (*a)->local_part = local;
    (*a)->domain = domain;

    if(rc != EOK) {
	/* note that the arguments have NOT been freed, we only own
	 * them on success. */
	free(*a);
    }

    return rc;
}

int parse822_address_list(address_t* a, const char* s)
{
    /* address-list = #(address) */

    const char** p = &s;
    const char*  e = &s[strlen(s)];
    int rc = EOK;
    address_t* n = a; /* the next address we'll be creating */

    rc = parse822_address(p, e, n);

    /* A list may start with a leading <,>, we'll find out if
     * that's not the case at the top of the while, but give
     * this a conditional OK unless there was some other kind
     * of error.
     */
    if(rc != EOK && rc != EPARSE) {
	return rc;
    }
    while(*p < e)
    {
	parse822_skip_comments(p, e);

	/* An address can contain a group, so an entire
	 * list of addresses may have been appended, or no
	 * addresses at all. Walk to the end.
	 */
	while(*n) {
	    n = &(*n)->next;
	}

	/* Remember that ',,a@b' is a valid list! So, we must find
	 * the <,>, but the address after it can be empty.
	 */
	if((rc = parse822_special(p, e, ','))) {
	     break;
	}
	parse822_skip_comments(p, e);

	rc = parse822_address(p, e, n);

	if(rc == EOK || rc == EPARSE) {
	    /* that's cool, it may be a <,>, we'll find out if it isn't
	     * at the top of the loop
	     */
	    rc = EOK;
	} else {
	    /* anything else is a fatal error, break out */
	    break;
	}
    }

    if(rc) {
	address_destroy(a);
    }

    return rc;
}

int parse822_address(const char** p, const char* e, address_t* a)
{
    /* address = mailbox / group */

    int rc;

    if((rc = parse822_mail_box(p, e, a)) == EPARSE)
    	rc = parse822_group(p, e, a);

    return rc;
}

int parse822_group(const char** p, const char* e, address_t* a)
{
    /* group = phrase ":" [#mailbox] ";" */

    const char* save = *p;
    address_t* asave = a; /* so we can destroy these if parsing fails */
    int rc;
    char* phrase = 0;

    parse822_skip_comments(p, e);

    *p = save;

    if((rc = parse822_phrase(p, e, &phrase))) {
	return rc;
    }

    parse822_skip_comments(p, e);

    if((rc = parse822_special(p, e, ':'))) {
	*p = save;
	return rc;
    }

    /* fake up an address node for the group's descriptive phrase, if
     * it fails, clean-up will happen after the loop
     */
    if((rc = fill_mb(a, 0, phrase, 0, 0)) == EOK) {
      a = &(*a)->next;
    } else {
      str_free(&phrase);
    }

    /* Basically, on each loop, we may find a mailbox, but we must find
     * a comma after the mailbox, otherwise we've popped off the end
     * of the list.
     */
    while(!rc) {
	parse822_skip_comments(p, e);

	/* it's ok not be a mailbox, but other errors are fatal */
	rc = parse822_mail_box(p, e, a);
	if(rc == EOK) {
	    a = &(*a)->next;

	    parse822_skip_comments(p, e);
	} else if(rc != EPARSE) {
	    break;
	}

	if((rc = parse822_special(p, e, ','))) {
	    /* the commas aren't optional */
	    break;
	}
    }
    if(rc == EPARSE) {
	rc = EOK; /* ok, as long as we find the ";" next */
    }

    if(rc || (rc = parse822_special(p, e, ';'))) {
	*p = save;

	address_destroy(asave);
    }

    return rc;
}

int parse822_mail_box(const char** p, const char* e, address_t* a)
{
    /* mailbox = addr-spec [ "(" comment ")" ] / [phrase] route-addr
     *
     *	Note: we parse the ancient comment on the right since
     *	  it's such "common practice". :-(
     *	Note: phrase is called display-name in drums.
     *	Note: phrase is optional in drums, though not in RFC 822.
     */
    const char* save = *p;
    int rc;

    /* -> addr-spec */
    if((rc = parse822_addr_spec(p, e, a)) == EOK) {
	parse822_skip_ws(p, e);

	/* yuck. */
	if((rc = parse822_comment(p, e, &(*a)->personal)) == EPARSE) {
	    rc = EOK;
	    /* cool if there's no comment, */
	}
	/* but if something else is wrong, destroy the address */
	if(rc) {
	    address_destroy(a);
	    *p = save;
	}

	return rc;
    }
    if(rc != EPARSE) {
	*p = save;
	return rc;
    }

    /* -> phrase route-addr */
    {
	char* phrase = 0;

	rc = parse822_phrase(p, e, &phrase);

	if(rc != EPARSE && rc != EOK) {
	    return rc;
	}

	if((rc = parse822_route_addr(p, e, a))) {
	    *p = save;
	    str_free(&phrase);
	    return rc;
	}

	/* add the phrase */
	(*a)->personal = phrase;
    }

    return EOK;
}

int parse822_route_addr(const char** p, const char* e, address_t* a)
{
    /* route-addr = "<" [route] addr-spec ">" */

    const char* save = *p;
    char* route = 0;
    int rc;

    parse822_skip_comments(p, e);

    if((rc = parse822_special(p, e, '<'))) {
	*p = save;

	return rc;
    }

    parse822_route(p, e, &route);

    if((rc = parse822_addr_spec(p, e, a))) {
	*p = save;

	str_free(&route);

	return rc;
    }

    (*a)->route = route; /* now we don't have to free our local */

    parse822_skip_comments(p, e);

    if((rc = parse822_special(p, e, '>'))) {
	*p = save;

	address_destroy(a);

	return rc;
    }

    return EOK;
}

int parse822_route(const char** p, const char* e, char** route)
{
    /* route = 1#("@" domain ) ":" */

    const char* save = *p;
    char* accumulator = 0;
    int rc = EOK;

    for(;;) {
	parse822_skip_comments(p, e);

	if((rc = parse822_special(p, e, '@'))) {
	    break;
	}

	if((rc = str_append(&accumulator, "@"))) {
	    break;
	}

	parse822_skip_comments(p, e);

	if((rc = parse822_domain(p, e, &accumulator))) {
	    /* it looked like a route, but there's no domain! */
	    break;
	}

	parse822_skip_comments(p, e);

	if((rc = parse822_special(p, e, ',')) == EPARSE) {
	    /* no more routes, but we got one so its ok */
	    rc = EOK;
	    break;
	}
	if((rc = str_append(&accumulator, ", "))) {
	    break;
	}
    }

    parse822_skip_comments(p, e);

    if(!rc) {
	rc = parse822_special(p, e, ':');
    }

    if(!rc) {
	rc = str_append(route, accumulator);
    }
    if(rc) {
	str_free(&accumulator);
	*p = save;
    }

    return rc;
}

int parse822_addr_spec(const char** p, const char* e, address_t* a)
{
    /* addr-spec = local-part "@" domain */

    const char* save = *p;
    char* local_part = 0;
    char* domain = 0;
    int rc;

    rc = parse822_local_part(p, e, &local_part);

    parse822_skip_comments(p, e);

    if(!rc) {
	rc = parse822_special(p, e, '@');
    }

    if(!rc) {
	rc = parse822_domain(p, e, &domain);
    }

    if(!rc) {
	rc = fill_mb(a, 0, 0, local_part, domain);
    }

    if(rc) {
	*p = save;
	str_free(&local_part);
	str_free(&domain);
    }
    return rc;
}

int parse822_local_part(const char** p, const char* e, char** local_part)
{
    /* local-part = word *("." word)
     *
     * Note: rewrite as ->  word ["." local-part]
     */

    const char* save = *p;
    const char* save2 = *p;
    int rc;

    parse822_skip_comments(p, e);

    if((rc = parse822_word(p, e, local_part))) {
	*p = save;
	return rc;
    }
    /* We've got a local-part, but keep looking for more. */

    parse822_skip_comments(p, e);

    /* If we get a parse error, we roll back to save2, but if
     * something else failed, we have to roll back to save.
     */
    save2 = *p;

    rc = parse822_special(p, e, '.');

    if(!rc) {
	char* more = 0;
	if((rc = parse822_local_part(p, e, &more)) == EOK) {
	    /* append more */
	    if((rc = str_append(local_part, ".")) == EOK) {
		rc = str_append(local_part, more);
	    }
	    str_free(&more);
	}
    }

    if(rc == EPARSE) {
	/* if we didn't get more ("." word) pairs, that's ok */
	*p = save2;
	rc = EOK;
    }
    if(rc) {
	/* if anything else failed, that's real */
	*p = save;
	str_free(local_part);
    }
    return rc;
}

int parse822_domain(const char** p, const char* e, char** domain)
{
    /* domain = sub-domain *("." sub-domain)
     *
     * Note: rewrite as -> sub-domain ("." domain)
     */

    const char* save = *p;
    const char* save2 = 0;
    int rc;

    parse822_skip_comments(p, e);

    if((rc = parse822_sub_domain(p, e, domain))) {
	*p = save;
	return rc;
    }


    /* We save before skipping comments to preserve the comment
     * at the end of a domain, the addr-spec may want to abuse it
     * for a personal name.
     */
    save2 = *p;

    /* we've got the 1, keep looking for more */

    parse822_skip_comments(p, e);

    rc = parse822_special(p, e, '.');

    if(!rc) {
	char* more = 0;
	if((rc = parse822_domain(p, e, &more)) == EOK) {
	    if((rc = str_append(domain, ".")) == EOK) {
		rc = str_append(domain, more);
	    }
	    str_free(&more);
	}
    }
    if(rc == EPARSE) {
	/* we didn't parse more ("." sub-domain) pairs, that's ok */
	*p = save2;
	rc = EOK;
    }

    if(rc) {
	/* something else failed, roll it all back */
	*p = save;
	str_free(domain);
    }
    return rc;
}

int parse822_sub_domain(const char** p, const char* e, char** sub_domain)
{
    /* sub-domain = domain-ref / domain-literal
     */

    int rc;

    if((rc = parse822_domain_ref(p, e, sub_domain)) == EPARSE)
    	rc = parse822_domain_literal(p, e, sub_domain);

    return rc;
}

int parse822_domain_ref(const char** p, const char* e, char** domain_ref)
{
    /* domain-ref = atom */

    return parse822_atom(p, e, domain_ref);
}

int parse822_d_text(const char** p, const char* e, char** dtext)
{
    /* d-text = 1*dtext
     *
     * Note: dtext is only defined as a character class in
     *	RFC822, but this definition is more useful for
     *	slurping domain literals.
     */

    const char* start = *p;
    int rc = EOK;

    while(*p < e && parse822_is_d_text(**p)) {
	*p += 1;
    }

    if(start == *p) {
	return EPARSE;
    }

    if((rc = str_append_range(dtext, start, *p))) {
	*p = start;
    }

    return rc;
}

int parse822_domain_literal(const char** p, const char* e, char** domain_literal)
{
    /* domain-literal = "[" *(dtext / quoted-pair) "]" */

    const char* save = *p;
    char* literal = 0;
    int rc;

    if((rc = parse822_special(p, e, '['))) {
	return rc;
    }
    if((rc = str_append_char(&literal, '['))) {
	*p = save;
	return rc;
    }

    while(
	    (rc = parse822_d_text(p, e, &literal)) == EOK ||
	    (rc = parse822_quoted_pair(p, e, &literal)) == EOK
	) {
	/* Eat all of this we can get! */
    }
    if(rc == EPARSE) {
	rc = EOK;
    }
    if(!rc) {
	rc = parse822_special(p, e, ']');
    }
    if(!rc) {
	rc = str_append_char(&literal, ']');
    }
    if(!rc) {
	rc = str_append(domain_literal, literal);
    }

    str_free(&literal);

    if(rc) {
	*p = save;
    }
    return rc;
}

#if 0

/***** From RFC 822, 3.2 Header Field Definitions *****/

int parse822_field_name(const char** p, const char* e, char** fieldname)
{
    /* field-name = 1*<any char, excluding ctlS, space, and ":"> ":" */

    Ptr save = p;

    Rope fn;

    while(*p != e) {
	char c = *p;

	if(!parse822_is_char(c))
	    break;

	if(parse822_is_ctl(c))
	    break;
	if(parse822_is_space(c))
	    break;
	if(c == ':')
	    break;

	fn.append(c);
	*p += 1;
    }
    /* must be at least one char in the field name */
    if(fn.empty()) {
	p = save;
	return 0;
    }
    parse822_skip_comments(p, e);

    if(!parse822_special(p, e, ':')) {
	p = save;
	return 0;
    }

    fieldname = fn;

    return 1;
}

int parse822_field_body(const char** p, const char* e, Rope& fieldbody)
{
    /* field-body = *text [CRLF lwsp-char field-body] */

    Ptr save = p;

    Rope fb;

    for(;;)
    {
	Ptr eol = p;
	while(eol != e) {
	    char c = *eol;
	    if(eol[0] == '\r' && (eol+1) != e && eol[1] == '\n')
		break;
	    ++eol;
	}
	fb.append(p, eol);
	p = eol;
	if(eol == e)
	    break; /* no more, so we're done */

	assert(p[0] == '\r');
	assert(p[1] == '\n');

	p += 2;

	if(*p == e)
	    break; /* no more, so we're done */

	/* check if next line is a continuation line */
	if(*p != ' ' && *p != '\t')
	    break;
    }

    fieldbody = fb;

    return 1;
}
#endif

/***** RFC 822 Quoting Functions *****/

int parse822_quote_string(char** quoted, const char* raw)
{
  /* quoted-string = <"> *(qtext/quoted-pair) <">
   *
   * So double quote the string, and back quote anything that
   * isn't qtext.
   */

  int rc = EOK;
  const char* s;

  if(!raw || !quoted || *quoted) {
    return EINVAL;
  }

  s = raw;

  rc = str_append_char(quoted, '"');

  while(!rc && *s) {
    if(!parse822_is_q_text(*s)) {
      rc = str_append_char(quoted, '\\');
    }

    if(!rc) {
      rc = str_append_char(quoted, *s);
    }
    ++s;
  }

  if(!rc) {
    rc = str_append_char(quoted, '"');
  }

  if(rc) {
    str_free(quoted);
  }
  return rc;
}

int parse822_quote_local_part(char** quoted, const char* raw)
{
  /* local-part = word * ("." word)
   * word = atom / quoted-string
   *
   * So, if any character isn't a "." or an atom character, we quote
   * the whole thing as a string, for simplicity, otherwise just
   * copy it.
   */

  const char* s = 0;
 
  if(!raw || !quoted || *quoted) {
    return EINVAL;
  }
  s = raw;

  while(*s) {
    if(*s != '.' && !parse822_is_atom_char(*s)) {
	return parse822_quote_string(quoted, raw);
    }
    ++s;
  }

  /* if we don't have to quote it, just copy it over */

  return str_append(quoted, raw);
}