int rfc822_parse_phrase(struct rfc822_parser_context *ctx, string_t *str)
{
int ret;
/*
phrase = 1*word / obs-phrase
word = atom / quoted-string
obs-phrase = word *(word / "." / CFWS)
*/
if (ctx->data == ctx->end)
return 0;
if (*ctx->data == '.')
return -1;
for (;;) {
if (*ctx->data == '"')
ret = rfc822_parse_quoted_string(ctx, str);
else
ret = rfc822_parse_atom_or_dot(ctx, str);
if (ret <= 0)
return ret;
if (!IS_ATEXT(*ctx->data) && *ctx->data != '"'
&& *ctx->data != '.')
break;
str_append_c(str, ' ');
}
return rfc822_skip_lwsp(ctx);
}
/*
* [RFC2822]
* atext = ALPHA / DIGIT / ; Any character except controls,
* "!" / "#" / ; SP, and specials.
* "$" / "%" / ; Used for atoms
* "&" / "'" /
* "*" / "+" /
* "-" / "/" /
* "=" / "?" /
* "^" / "_" /
* "`" / "{" /
* "|" / "}" /
* "~"
*/
int
XSkip_atextBlock(const char *head, const char *tail, const char **nextp)
{
const char *p;
for (p = head; p < tail && IS_ATEXT(*p); ++p);
*nextp = p;
return *nextp - head;
} // end function : XSkip_atextBlock
int rfc822_parse_dot_atom(struct rfc822_parser_context *ctx, string_t *str)
{
const unsigned char *start;
int ret;
/*
dot-atom = [CFWS] dot-atom-text [CFWS]
dot-atom-text = 1*atext *("." 1*atext)
atext =
; Any character except controls, SP, and specials.
For RFC-822 compatibility allow LWSP around '.'
*/
if (ctx->data == ctx->end || !IS_ATEXT(*ctx->data))
return -1;
for (start = ctx->data++; ctx->data != ctx->end; ) {
if (IS_ATEXT(*ctx->data)) {
ctx->data++;
continue;
}
str_append_n(str, start, ctx->data - start);
if ((ret = rfc822_skip_lwsp(ctx)) <= 0)
return ret;
if (*ctx->data != '.')
return 1;
ctx->data++;
str_append_c(str, '.');
if ((ret = rfc822_skip_lwsp(ctx)) <= 0)
return ret;
start = ctx->data;
}
str_append_n(str, start, ctx->data - start);
return 0;
}
int
valid_localpart(const char *s)
{
#define IS_ATEXT(c) (isalnum((unsigned char)(c)) || strchr(MAILADDR_ALLOWED, (c)))
nextatom:
if (!IS_ATEXT(*s) || *s == '\0')
return 0;
while (*(++s) != '\0') {
if (*s == '.')
break;
if (IS_ATEXT(*s))
continue;
return 0;
}
if (*s == '.') {
s++;
goto nextatom;
}
return 1;
}
int
valid_localpart(char *s)
{
#define IS_ATEXT(c) (isalnum((int)(c)) || strchr("!#$%&'*+-/=?^_`{|}~", (c)))
nextatom:
if (! IS_ATEXT(*s) || *s == '\0')
return 0;
while (*(++s) != '\0') {
if (*s == '.')
break;
if (IS_ATEXT(*s))
continue;
return 0;
}
if (*s == '.') {
s++;
goto nextatom;
}
return 1;
}
int rfc822_parse_atom(struct rfc822_parser_context *ctx, string_t *str)
{
const unsigned char *start;
/*
atom = [CFWS] 1*atext [CFWS]
atext =
; Any character except controls, SP, and specials.
*/
if (ctx->data == ctx->end || !IS_ATEXT(*ctx->data))
return -1;
for (start = ctx->data++; ctx->data != ctx->end; ctx->data++) {
if (IS_ATEXT(*ctx->data))
continue;
str_append_n(str, start, ctx->data - start);
return rfc822_skip_lwsp(ctx);
}
str_append_n(str, start, ctx->data - start);
return 0;
}
int
valid_localpart(const char *s)
{
/*
* RFC 5322 defines theses characters as valid: !#$%&'*+-/=?^_`{|}~
* some of them are potentially dangerous, and not so used after all.
*/
#define IS_ATEXT(c) (isalnum((int)(c)) || strchr("*!%+-/=_", (c)))
nextatom:
if (! IS_ATEXT(*s) || *s == '\0')
return 0;
while (*(++s) != '\0') {
if (*s == '.')
break;
if (IS_ATEXT(*s))
continue;
return 0;
}
if (*s == '.') {
s++;
goto nextatom;
}
return 1;
}
static int
rfc822_parse_atom_or_dot(struct rfc822_parser_context *ctx, string_t *str)
{
const unsigned char *start;
/*
atom = [CFWS] 1*atext [CFWS]
atext =
; Any character except controls, SP, and specials.
The difference between this function and rfc822_parse_dot_atom()
is that this doesn't just silently skip over all the whitespace.
*/
for (start = ctx->data; ctx->data != ctx->end; ctx->data++) {
if (IS_ATEXT(*ctx->data) || *ctx->data == '.')
continue;
str_append_n(str, start, ctx->data - start);
return rfc822_skip_lwsp(ctx);
}
str_append_n(str, start, ctx->data - start);
return 0;
}
static bool
DkimWildcard_matchPubkeyGranularityImpl(const char *pattern_head, const char *pattern_tail,
const char *target_head, const char *target_tail,
bool accept_wildcard)
{
/*
* the ABNF of key-g-tag-lpart says only one wildcard is acceptable.
* But '*' itself is included in dot-atom-text.
* So this function treats the first occurrence of '*' as wildcard,
* and next or later occurrence of '*' as character.
*
* [RFC4871] 3.6.1.
* key-g-tag = %x67 [FWS] "=" [FWS] key-g-tag-lpart
* key-g-tag-lpart = [dot-atom-text] ["*" [dot-atom-text] ]
*/
const char *pattern;
const char *target;
for (pattern = pattern_head, target = target_head; pattern < pattern_tail; ++pattern, ++target) {
if ('*' == *pattern) {
if (accept_wildcard) {
// treat '*' as wildcard
++pattern;
for (const char *bq = target_tail; target <= bq; --bq) {
if (DkimWildcard_matchPubkeyGranularityImpl
(pattern, pattern_tail, bq, target_tail, false)) {
return true;
} // end if
} // end for
return false;
} else {
// treat '*' as (not wildcard but) character
if (target_tail <= target || '*' != *target) {
return false;
} // end if
} // end if
} else if (IS_ATEXT(*pattern) || '.' == *pattern) {
/*
* compare case-sensitively.
*
* [RFC6376] 3.2.
* Values MUST be
* processed as case sensitive unless the specific tag description of
* semantics specifies case insensitivity.
*
* And the local-part of mailbox is essentially treated as case-sensitive.
*
* [RFC5321] 2.4.
* ... The
* local-part of a mailbox MUST BE treated as case sensitive.
*/
if (target_tail <= target || *pattern != *target) {
return false;
} // end if
} else {
// neither atext nor '.' (included in dot-atom-text)
return false;
} // end if
} // end for
return bool_cast(pattern == pattern_tail && target == target_tail);
} // end function: DkimWildcard_matchPubkeyGranularityImpl
