libpsl/src/psl.c

939 lines
24 KiB
C
Raw Normal View History

2014-03-20 22:43:04 +01:00
/*
* Copyright(c) 2014 Tim Ruehsen
*
2014-03-24 17:29:56 +01:00
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
* DEALINGS IN THE SOFTWARE.
*
* This file is part of libpsl.
2014-03-20 22:43:04 +01:00
*
2014-03-24 17:29:56 +01:00
* Public Suffix List routines
2014-03-20 22:43:04 +01:00
*
* Changelog
* 19.03.2014 Tim Ruehsen created from libmget/cookie.c
*
*/
2014-05-12 12:20:59 +02:00
/* need _GNU_SOURCE for qsort_r() */
2014-03-21 11:05:09 +01:00
#ifndef _GNU_SOURCE
# define _GNU_SOURCE
#endif
2014-03-20 22:43:04 +01:00
#if HAVE_CONFIG_H
# include <config.h>
#endif
#if ENABLE_NLS != 0
# include <libintl.h>
# define _(STRING) gettext(STRING)
#else
# define _(STRING) STRING
# define ngettext(STRING1,STRING2,N) STRING2
#endif
2014-03-20 22:43:04 +01:00
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <ctype.h>
2014-06-17 17:14:02 +02:00
#include <alloca.h>
#ifdef WITH_LIBICU
# include <unicode/uversion.h>
# include <unicode/ustring.h>
# include <unicode/uidna.h>
# include <unicode/ucnv.h>
#endif
2014-03-20 22:43:04 +01:00
#include <libpsl.h>
2014-06-17 17:14:02 +02:00
/* number of elements within an array */
#define countof(a) (sizeof(a)/sizeof(*(a)))
/**
* SECTION:libpsl
* @short_description: Public Suffix List library functions
* @title: libpsl
* @stability: unstable
* @include: libpsl.h
*
* [Public Suffix List](http://publicsuffix.org/) library functions.
*
*/
2014-03-20 22:43:04 +01:00
#define countof(a) (sizeof(a)/sizeof(*(a)))
typedef struct {
char
label_buf[48];
2014-03-20 22:43:04 +01:00
const char *
label;
unsigned short
length;
unsigned char
2014-05-12 12:20:59 +02:00
nlabels, /* number of labels */
wildcard; /* this is a wildcard rule (e.g. *.sapporo.jp) */
2014-03-20 22:43:04 +01:00
} _psl_entry_t;
2014-05-12 12:20:59 +02:00
/* stripped down version libmget vector routines */
2014-03-20 22:43:04 +01:00
typedef struct {
int
2014-05-12 12:20:59 +02:00
(*cmp)(const _psl_entry_t *, const _psl_entry_t *); /* comparison function */
2014-03-20 22:43:04 +01:00
_psl_entry_t
2014-05-12 12:20:59 +02:00
**entry; /* pointer to array of pointers to elements */
2014-03-20 22:43:04 +01:00
int
2014-05-12 12:20:59 +02:00
max, /* allocated elements */
cur; /* number of elements in use */
2014-03-20 22:43:04 +01:00
} _psl_vector_t;
struct _psl_ctx_st {
_psl_vector_t
*suffixes,
*suffix_exceptions;
};
2014-05-12 12:20:59 +02:00
/* include the PSL data compiled by 'psl2c' */
2014-06-19 12:06:54 +02:00
#ifndef _LIBPSL_INCLUDED_BY_PSL2C
# include "suffixes.c"
#else
/* if this source file is included by psl2c.c, provide empty builtin data */
2014-06-20 17:04:22 +02:00
static _psl_entry_t suffixes[1];
static _psl_entry_t suffix_exceptions[1];
2014-06-19 12:06:54 +02:00
static time_t _psl_file_time;
static time_t _psl_compile_time;
static const char _psl_sha1_checksum[] = "";
static const char _psl_filename[] = "";
#endif
2014-03-24 17:29:56 +01:00
2014-05-12 12:20:59 +02:00
/* references to this PSL will result in lookups to built-in data */
static const psl_ctx_t
_builtin_psl;
2014-03-24 17:29:56 +01:00
2014-03-20 22:43:04 +01:00
static _psl_vector_t *_vector_alloc(int max, int (*cmp)(const _psl_entry_t *, const _psl_entry_t *))
{
_psl_vector_t *v;
if (!(v = calloc(1, sizeof(_psl_vector_t))))
return NULL;
if (!(v->entry = malloc(max * sizeof(_psl_entry_t *)))) {
free(v);
return NULL;
}
v->max = max;
v->cmp = cmp;
return v;
}
static void _vector_free(_psl_vector_t **v)
{
if (v && *v) {
if ((*v)->entry) {
int it;
for (it = 0; it < (*v)->cur; it++)
free((*v)->entry[it]);
free((*v)->entry);
}
free(*v);
}
}
static _psl_entry_t *_vector_get(const _psl_vector_t *v, int pos)
{
if (pos < 0 || !v || pos >= v->cur) return NULL;
return v->entry[pos];
}
2014-05-12 12:20:59 +02:00
/* the entries must be sorted by */
2014-03-20 22:43:04 +01:00
static int _vector_find(const _psl_vector_t *v, const _psl_entry_t *elem)
{
if (v) {
int l, r, m;
int res;
2014-05-12 12:20:59 +02:00
/* binary search for element (exact match) */
2014-03-20 22:43:04 +01:00
for (l = 0, r = v->cur - 1; l <= r;) {
m = (l + r) / 2;
if ((res = v->cmp(elem, v->entry[m])) > 0) l = m + 1;
else if (res < 0) r = m - 1;
else return m;
}
}
2014-05-12 12:20:59 +02:00
return -1; /* not found */
2014-03-20 22:43:04 +01:00
}
static int _vector_add(_psl_vector_t *v, const _psl_entry_t *elem)
{
if (v) {
void *elemp;
elemp = malloc(sizeof(_psl_entry_t));
memcpy(elemp, elem, sizeof(_psl_entry_t));
if (v->max == v->cur)
v->entry = realloc(v->entry, (v->max *= 2) * sizeof(_psl_entry_t *));
v->entry[v->cur++] = elemp;
return v->cur - 1;
}
return -1;
}
static int _compare(const void *p1, const void *p2, void *v)
{
return ((_psl_vector_t *)v)->cmp(*((_psl_entry_t **)p1), *((_psl_entry_t **)p2));
}
static void _vector_sort(_psl_vector_t *v)
{
if (v && v->cmp)
qsort_r(v->entry, v->cur, sizeof(_psl_vector_t *), _compare, v);
}
2014-05-12 12:20:59 +02:00
static int _vector_size(_psl_vector_t *v)
2014-03-22 20:35:56 +01:00
{
return v ? v->cur : 0;
}
2014-05-12 12:20:59 +02:00
/* by this kind of sorting, we can easily see if a domain matches or not */
2014-03-20 22:43:04 +01:00
static int _suffix_compare(const _psl_entry_t *s1, const _psl_entry_t *s2)
{
int n;
if ((n = s2->nlabels - s1->nlabels))
2014-05-12 12:20:59 +02:00
return n; /* most labels first */
2014-03-20 22:43:04 +01:00
2014-03-22 20:35:56 +01:00
if ((n = s1->length - s2->length))
2014-05-12 12:20:59 +02:00
return n; /* shorter rules first */
2014-03-20 22:43:04 +01:00
return strcmp(s1->label, s2->label ? s2->label : s2->label_buf);
2014-03-20 22:43:04 +01:00
}
2014-03-29 18:12:45 +01:00
static int _suffix_init(_psl_entry_t *suffix, const char *rule, size_t length)
2014-03-20 22:43:04 +01:00
{
const char *src;
char *dst;
suffix->label = suffix->label_buf;
if (length >= sizeof(suffix->label_buf) - 1) {
suffix->nlabels = 0;
2014-05-12 12:20:59 +02:00
/* fprintf(stderr, _("Suffix rule too long (%zd, ignored): %s\n"), length, rule); */
2014-03-29 18:12:45 +01:00
return -1;
2014-03-20 22:43:04 +01:00
}
if (*rule == '*') {
if (*++rule != '.') {
suffix->nlabels = 0;
2014-05-12 12:20:59 +02:00
/* fprintf(stderr, _("Unsupported kind of rule (ignored): %s\n"), rule); */
2014-03-29 18:12:45 +01:00
return -2;
2014-03-20 22:43:04 +01:00
}
rule++;
suffix->wildcard = 1;
suffix->length = (unsigned char)length - 2;
} else {
suffix->wildcard = 0;
suffix->length = (unsigned char)length;
}
suffix->nlabels = 1;
for (dst = suffix->label_buf, src = rule; *src;) {
if (*src == '.')
suffix->nlabels++;
2014-06-17 17:14:02 +02:00
*dst++ = *src++;
2014-03-20 22:43:04 +01:00
}
*dst = 0;
2014-03-29 18:12:45 +01:00
return 0;
2014-03-20 22:43:04 +01:00
}
2014-06-17 17:14:02 +02:00
static int _psl_is_public_suffix(const psl_ctx_t *psl, const char *domain)
2014-03-20 22:43:04 +01:00
{
_psl_entry_t suffix, *rule;
const char *p, *label_bak;
unsigned short length_bak;
2014-05-12 12:20:59 +02:00
/* this function should be called without leading dots, just make sure */
2014-03-20 22:43:04 +01:00
suffix.label = domain + (*domain == '.');
suffix.length = strlen(suffix.label);
suffix.wildcard = 0;
suffix.nlabels = 1;
for (p = suffix.label; *p; p++)
if (*p == '.')
suffix.nlabels++;
2014-05-12 12:20:59 +02:00
/* if domain has enough labels, it is public */
2014-03-24 17:29:56 +01:00
if (psl == &_builtin_psl)
rule = &suffixes[0];
else
rule = _vector_get(psl->suffixes, 0);
2014-03-20 22:43:04 +01:00
if (!rule || rule->nlabels < suffix.nlabels - 1)
return 0;
2014-03-20 22:43:04 +01:00
2014-03-24 17:29:56 +01:00
if (psl == &_builtin_psl)
rule = bsearch(&suffix, suffixes, countof(suffixes), sizeof(suffixes[0]), (int(*)(const void *, const void *))_suffix_compare);
else
rule = _vector_get(psl->suffixes, _vector_find(psl->suffixes, &suffix));
2014-03-20 22:43:04 +01:00
if (rule) {
2014-05-12 12:20:59 +02:00
/* definitely a match, no matter if the found rule is a wildcard or not */
return 1;
2014-03-28 22:02:42 +01:00
} else if (suffix.nlabels == 1) {
2014-05-12 12:20:59 +02:00
/* unknown TLD, this is the prevailing '*' match */
return 1;
2014-03-20 22:43:04 +01:00
}
label_bak = suffix.label;
length_bak = suffix.length;
if ((suffix.label = strchr(suffix.label, '.'))) {
suffix.label++;
suffix.length = strlen(suffix.label);
suffix.nlabels--;
2014-03-24 17:29:56 +01:00
if (psl == &_builtin_psl)
rule = bsearch(&suffix, suffixes, countof(suffixes), sizeof(suffixes[0]), (int(*)(const void *, const void *))_suffix_compare);
else
rule = _vector_get(psl->suffixes, _vector_find(psl->suffixes, &suffix));
2014-03-20 22:43:04 +01:00
if (rule) {
if (rule->wildcard) {
2014-05-12 12:20:59 +02:00
/* now that we matched a wildcard, we have to check for an exception */
2014-03-20 22:43:04 +01:00
suffix.label = label_bak;
suffix.length = length_bak;
suffix.nlabels++;
2014-03-24 17:29:56 +01:00
if (psl == &_builtin_psl) {
if (bsearch(&suffix, suffix_exceptions, countof(suffix_exceptions), sizeof(suffix_exceptions[0]), (int(*)(const void *, const void *))_suffix_compare))
2014-05-12 12:20:59 +02:00
return 0; /* found an exception, so 'domain' is not a public suffix */
2014-03-24 17:29:56 +01:00
} else {
if (_vector_get(psl->suffix_exceptions, _vector_find(psl->suffix_exceptions, &suffix)) != 0)
2014-05-12 12:20:59 +02:00
return 0; /* found an exception, so 'domain' is not a public suffix */
2014-03-24 17:29:56 +01:00
}
2014-03-20 22:43:04 +01:00
return 1;
2014-03-20 22:43:04 +01:00
}
}
}
return 0;
2014-03-20 22:43:04 +01:00
}
2014-06-17 17:14:02 +02:00
/**
* psl_is_public_suffix:
* @psl: PSL context
* @domain: Domain string
*
* This function checks if @domain is a public suffix by the means of the
* [Mozilla Public Suffix List](http://publicsuffix.org).
*
* For cookie domain checking see psl_is_cookie_domain_acceptable().
*
* International @domain names have to be either in lowercase UTF-8 or in ASCII form (punycode).
* Other encodings result in unexpected behavior.
*
2014-06-17 17:14:02 +02:00
* @psl is a context returned by either psl_load_file(), psl_load_fp() or
* psl_builtin().
*
* Returns: 1 if domain is a public suffix, 0 if not.
*
* Since: 0.1
*/
int psl_is_public_suffix(const psl_ctx_t *psl, const char *domain)
{
if (!psl || !domain)
return 1;
return _psl_is_public_suffix(psl, domain);
}
/**
* psl_unregistrable_domain:
* @psl: PSL context
* @domain: Domain string
*
* This function finds the longest publix suffix part of @domain by the means
* of the [Mozilla Public Suffix List](http://publicsuffix.org).
*
* International @domain names have to be either in lowercase UTF-8 or in ASCII form (punycode).
* Other encodings result in unexpected behavior.
*
* @psl is a context returned by either psl_load_file(), psl_load_fp() or
* psl_builtin().
*
* Returns: Pointer to longest public suffix part of @domain or %NULL if @domain
* does not contain a public suffix (or if @psl is %NULL).
*
* Since: 0.1
*/
const char *psl_unregistrable_domain(const psl_ctx_t *psl, const char *domain)
2014-03-26 17:14:25 +01:00
{
if (!psl || !domain)
return NULL;
2014-05-12 12:20:59 +02:00
/*
* We check from left to right to catch special PSL entries like 'forgot.his.name':
* 'forgot.his.name' and 'name' are in the PSL while 'his.name' is not.
*/
2014-03-26 17:14:25 +01:00
2014-06-17 17:14:02 +02:00
while (!_psl_is_public_suffix(psl, domain)) {
if ((domain = strchr(domain, '.')))
domain++;
else
2014-05-12 12:20:59 +02:00
break; /* prevent endless loop if psl_is_public_suffix() is broken. */
2014-03-26 17:14:25 +01:00
}
return domain;
2014-03-26 17:14:25 +01:00
}
/**
* psl_registrable_domain:
* @psl: PSL context
* @domain: Domain string
*
* This function finds the shortest private suffix part of @domain by the means
* of the [Mozilla Public Suffix List](http://publicsuffix.org).
*
* International @domain names have to be either in lowercase UTF-8 or in ASCII form (punycode).
* Other encodings result in unexpected behavior.
*
* @psl is a context returned by either psl_load_file(), psl_load_fp() or
* psl_builtin().
*
* Returns: Pointer to shortest private suffix part of @domain or %NULL if @domain
* does not contain a private suffix (or if @psl is %NULL).
*
* Since: 0.1
*/
const char *psl_registrable_domain(const psl_ctx_t *psl, const char *domain)
{
const char *p, *regdom = NULL;
if (!psl || !domain || *domain == '.')
return NULL;
2014-05-12 12:20:59 +02:00
/*
* We check from left to right to catch special PSL entries like 'forgot.his.name':
* 'forgot.his.name' and 'name' are in the PSL while 'his.name' is not.
*/
2014-06-17 17:14:02 +02:00
while (!_psl_is_public_suffix(psl, domain)) {
if ((p = strchr(domain, '.'))) {
regdom = domain;
domain = p + 1;
} else
2014-05-12 12:20:59 +02:00
break; /* prevent endless loop if psl_is_public_suffix() is broken. */
}
return regdom;
}
2014-06-17 17:14:02 +02:00
static int _str_is_ascii(const char *s)
{
while (*s > 0) s++;
return !*s;
}
2014-06-18 15:21:22 +02:00
#ifdef WITH_LIBICU
2014-06-17 17:14:02 +02:00
static void _add_punycode_if_needed(UIDNA *idna, _psl_vector_t *v, _psl_entry_t *e)
{
if (_str_is_ascii(e->label_buf))
return;
/* IDNA2008 UTS#46 punycode conversion */
if (idna) {
_psl_entry_t suffix, *suffixp;
char lookupname[128] = "";
UErrorCode status = 0;
UIDNAInfo info = UIDNA_INFO_INITIALIZER;
UChar utf16_dst[128], utf16_src[128];
int32_t utf16_src_length;
u_strFromUTF8(utf16_src, sizeof(utf16_src)/sizeof(utf16_src[0]), &utf16_src_length, e->label_buf, -1, &status);
if (U_SUCCESS(status)) {
int32_t dst_length = uidna_nameToASCII(idna, utf16_src, utf16_src_length, utf16_dst, sizeof(utf16_dst)/sizeof(utf16_dst[0]), &info, &status);
if (U_SUCCESS(status)) {
u_strToUTF8(lookupname, sizeof(lookupname), NULL, utf16_dst, dst_length, &status);
if (U_SUCCESS(status)) {
if (strcmp(e->label_buf, lookupname)) {
/* fprintf(stderr, "libicu '%s' -> '%s'\n", e->label_buf, lookupname); */
_suffix_init(&suffix, lookupname, strlen(lookupname));
suffix.wildcard = e->wildcard;
suffixp = _vector_get(v, _vector_add(v, &suffix));
suffixp->label = suffixp->label_buf; /* set label to changed address */
} /* else ignore */
} /* else
fprintf(stderr, "Failed to convert UTF-16 to UTF-8 (status %d)\n", status); */
} /* else
fprintf(stderr, "Failed to convert to ASCII (status %d)\n", status); */
} /* else
fprintf(stderr, "Failed to convert UTF-8 to UTF-16 (status %d)\n", status); */
}
}
2014-06-18 15:21:22 +02:00
#endif
2014-06-17 17:14:02 +02:00
/**
* psl_load_file:
* @fname: Name of PSL file
*
* This function loads the public suffixes file named @fname.
* To free the allocated resources, call psl_free().
*
* The suffixes are expected to be lowercase UTF-8 encoded if they are international.
*
* Returns: Pointer to a PSL context or %NULL on failure.
*
* Since: 0.1
*/
2014-03-20 22:43:04 +01:00
psl_ctx_t *psl_load_file(const char *fname)
2014-03-22 14:28:55 +01:00
{
FILE *fp;
psl_ctx_t *psl = NULL;
if (!fname)
return NULL;
2014-03-22 14:28:55 +01:00
if ((fp = fopen(fname, "r"))) {
psl = psl_load_fp(fp);
fclose(fp);
}
return psl;
}
/**
* psl_load_fp:
* @fp: FILE pointer
*
* This function loads the public suffixes from a FILE pointer.
* To free the allocated resources, call psl_free().
*
* The suffixes are expected to be lowercase UTF-8 encoded if they are international.
*
* Returns: Pointer to a PSL context or %NULL on failure.
*
* Since: 0.1
*/
2014-03-22 14:28:55 +01:00
psl_ctx_t *psl_load_fp(FILE *fp)
2014-03-20 22:43:04 +01:00
{
psl_ctx_t *psl;
_psl_entry_t suffix, *suffixp;
char buf[256], *linep, *p;
2014-06-17 17:14:02 +02:00
#ifdef WITH_LIBICU
UIDNA *idna;
UErrorCode status = 0;
#endif
2014-03-20 22:43:04 +01:00
2014-03-22 14:28:55 +01:00
if (!fp)
2014-03-20 22:43:04 +01:00
return NULL;
2014-03-22 14:28:55 +01:00
if (!(psl = calloc(1, sizeof(psl_ctx_t))))
return NULL;
2014-03-20 22:43:04 +01:00
2014-06-17 17:14:02 +02:00
#ifdef WITH_LIBICU
idna = uidna_openUTS46(UIDNA_USE_STD3_RULES, &status);
#endif
2014-05-12 12:20:59 +02:00
/*
* as of 02.11.2012, the list at http://publicsuffix.org/list/ contains ~6000 rules and 40 exceptions.
* as of 19.02.2014, the list at http://publicsuffix.org/list/ contains ~6500 rules and 19 exceptions.
*/
2014-03-22 14:28:55 +01:00
psl->suffixes = _vector_alloc(8*1024, _suffix_compare);
psl->suffix_exceptions = _vector_alloc(64, _suffix_compare);
while ((linep = fgets(buf, sizeof(buf), fp))) {
2014-05-12 12:20:59 +02:00
while (isspace(*linep)) linep++; /* ignore leading whitespace */
if (!*linep) continue; /* skip empty lines */
2014-03-22 14:28:55 +01:00
if (*linep == '/' && linep[1] == '/')
2014-05-12 12:20:59 +02:00
continue; /* skip comments */
2014-03-22 14:28:55 +01:00
2014-05-12 12:20:59 +02:00
/* parse suffix rule */
2014-03-22 14:28:55 +01:00
for (p = linep; *linep && !isspace(*linep);) linep++;
*linep = 0;
if (*p == '!') {
2014-05-12 12:20:59 +02:00
/* add to exceptions */
2014-06-17 17:14:02 +02:00
if (_suffix_init(&suffix, p + 1, linep - p - 1) == 0) {
2014-03-29 18:12:45 +01:00
suffixp = _vector_get(psl->suffix_exceptions, _vector_add(psl->suffix_exceptions, &suffix));
2014-06-17 17:14:02 +02:00
suffixp->label = suffixp->label_buf; /* set label to changed address */
2014-06-18 15:21:22 +02:00
#ifdef WITH_LIBICU
2014-06-17 17:14:02 +02:00
_add_punycode_if_needed(idna, psl->suffix_exceptions, suffixp);
2014-06-18 15:21:22 +02:00
#endif
2014-06-17 17:14:02 +02:00
}
2014-03-22 14:28:55 +01:00
} else {
2014-06-17 17:14:02 +02:00
/* add to suffixes */
if (_suffix_init(&suffix, p, linep - p) == 0) {
2014-03-29 18:12:45 +01:00
suffixp = _vector_get(psl->suffixes, _vector_add(psl->suffixes, &suffix));
2014-06-17 17:14:02 +02:00
suffixp->label = suffixp->label_buf; /* set label to changed address */
2014-06-18 15:21:22 +02:00
#ifdef WITH_LIBICU
2014-06-17 17:14:02 +02:00
_add_punycode_if_needed(idna, psl->suffixes, suffixp);
2014-06-18 15:21:22 +02:00
#endif
2014-06-17 17:14:02 +02:00
}
2014-03-20 22:43:04 +01:00
}
}
2014-03-20 22:43:04 +01:00
2014-03-22 14:28:55 +01:00
_vector_sort(psl->suffix_exceptions);
_vector_sort(psl->suffixes);
2014-06-17 17:14:02 +02:00
#ifdef WITH_LIBICU
if (idna)
uidna_close(idna);
#endif
2014-03-20 22:43:04 +01:00
return psl;
}
/**
* psl_load_free:
* @psl: PSL context pointer
*
* This function frees the the PSL context that has been retrieved via
* psl_load_fp() or psl_load_file().
*
* Returns: Pointer to a PSL context private or %NULL on failure.
*
* Since: 0.1
*/
void psl_free(psl_ctx_t *psl)
2014-03-24 17:29:56 +01:00
{
if (psl && psl != &_builtin_psl) {
_vector_free(&psl->suffixes);
_vector_free(&psl->suffix_exceptions);
free(psl);
2014-03-24 17:29:56 +01:00
}
}
/**
* psl_builtin:
*
* This function returns the PSL context that has been generated and built in at compile-time.
* You don't have to free the returned context explicitely.
*
* The builtin data also contains punycode entries, one for each international domain name.
*
* If the generation of built-in data has been disabled during compilation, %NULL will be returned.
* So if using the builtin psl context, you can provide UTF-8 or punycode representations of domains to
* functions like psl_is_public_suffix().
*
* Returns: Pointer to the built in PSL data or NULL if this data is not available.
*
* Since: 0.1
*/
const psl_ctx_t *psl_builtin(void)
{
#ifdef WITH_BUILTIN
return &_builtin_psl;
#else
return NULL;
#endif
}
/**
* psl_suffix_count:
* @psl: PSL context pointer
*
* This function returns number of public suffixes maintained by @psl.
* The number of exceptions within the Public Suffix List are not included.
*
* If the generation of built-in data has been disabled during compilation, 0 will be returned.
*
* Returns: Number of public suffixes entries in PSL context.
*
* Since: 0.1
*/
int psl_suffix_count(const psl_ctx_t *psl)
{
2014-03-24 17:29:56 +01:00
if (psl == &_builtin_psl)
return countof(suffixes);
else if (psl)
2014-03-24 17:29:56 +01:00
return _vector_size(psl->suffixes);
2014-03-29 18:58:24 +01:00
else
return 0;
}
2014-03-22 20:35:56 +01:00
/**
* psl_suffix_exception_count:
* @psl: PSL context pointer
*
* This function returns number of public suffix exceptions maintained by @psl.
*
* If the generation of built-in data has been disabled during compilation, 0 will be returned.
*
* Returns: Number of public suffix exceptions in PSL context.
*
* Since: 0.1
*/
int psl_suffix_exception_count(const psl_ctx_t *psl)
{
2014-03-24 17:29:56 +01:00
if (psl == &_builtin_psl)
return countof(suffix_exceptions);
else if (psl)
2014-03-24 17:29:56 +01:00
return _vector_size(psl->suffix_exceptions);
2014-03-29 18:58:24 +01:00
else
return 0;
}
/**
* psl_builtin_compile_time:
*
* This function returns the time when the Publix Suffix List has been compiled into C code (by psl2c).
*
* If the generation of built-in data has been disabled during compilation, 0 will be returned.
*
* Returns: time_t value or 0.
*
* Since: 0.1
*/
2014-03-24 17:29:56 +01:00
time_t psl_builtin_compile_time(void)
2014-03-20 22:43:04 +01:00
{
2014-03-24 17:29:56 +01:00
return _psl_compile_time;
}
/**
* psl_builtin_file_time:
*
* This function returns the mtime of the Publix Suffix List file that has been built in.
*
* If the generation of built-in data has been disabled during compilation, 0 will be returned.
*
* Returns: time_t value or 0.
*
* Since: 0.1
*/
2014-03-24 17:29:56 +01:00
time_t psl_builtin_file_time(void)
{
return _psl_file_time;
}
/**
* psl_builtin_sha1sum:
*
* This function returns the SHA1 checksum of the Publix Suffix List file that has been built in.
* The returned string is in lowercase hex encoding, e.g. "2af1e9e3044eda0678bb05949d7cca2f769901d8".
*
* If the generation of built-in data has been disabled during compilation, an empty string will be returned.
*
* Returns: String containing SHA1 checksum or an empty string.
*
* Since: 0.1
*/
2014-03-24 17:29:56 +01:00
const char *psl_builtin_sha1sum(void)
{
return _psl_sha1_checksum;
2014-03-20 22:43:04 +01:00
}
2014-04-17 12:31:06 +02:00
/**
* psl_builtin_filename:
*
* This function returns the file name of the Publix Suffix List file that has been built in.
*
* If the generation of built-in data has been disabled during compilation, an empty string will be returned.
*
* Returns: String containing the PSL file name or an empty string.
*
* Since: 0.1
*/
const char *psl_builtin_filename(void)
{
return _psl_filename;
}
/**
* psl_get_version:
*
* Get libpsl version.
*
* Returns: String containing version of libpsl.
*
* Since: 0.2.5
**/
const char *psl_get_version (void)
{
2014-06-18 15:21:22 +02:00
return PACKAGE_VERSION
2014-06-17 17:14:02 +02:00
#ifdef WITH_LIBICU
2014-06-18 15:21:22 +02:00
" (+libicu/" U_ICU_VERSION ")"
2014-06-17 17:14:02 +02:00
#else
2014-06-18 15:21:22 +02:00
" (limited IDNA support)"
2014-06-17 17:14:02 +02:00
#endif
2014-06-18 15:21:22 +02:00
;
}
/**
* psl_is_cookie_domain_acceptable:
* @psl: PSL context pointer
* @hostname: The request hostname.
* @cookie_domain: The domain value from a cookie
*
* This helper function checks whether @cookie_domain is an acceptable cookie domain value for the request
* @hostname.
*
* For international domain names both, @hostname and @cookie_domain, have to be either in lowercase UTF-8
* or in ASCII form (punycode). Other encodings or mixing UTF-8 and punycode result in unexpected behavior.
*
2014-04-16 11:36:37 +02:00
* Examples:
* 1. Cookie domain 'example.com' would be acceptable for hostname 'www.example.com',
* but '.com' or 'com' would NOT be acceptable since 'com' is a public suffix.
*
* 2. Cookie domain 'his.name' would be acceptable for hostname 'remember.his.name',
* but NOT for 'forgot.his.name' since 'forgot.his.name' is a public suffix.
*
* Returns: 1 if acceptable, 0 if not acceptable.
*
* Since: 0.1
*/
int psl_is_cookie_domain_acceptable(const psl_ctx_t *psl, const char *hostname, const char *cookie_domain)
{
2014-04-16 11:36:37 +02:00
const char *p;
size_t hostname_length, cookie_domain_length;
if (!psl || !hostname || !cookie_domain)
return 0;
while (*cookie_domain == '.')
cookie_domain++;
if (!strcmp(hostname, cookie_domain))
2014-05-12 12:20:59 +02:00
return 1; /* an exact match is acceptable (and pretty common) */
cookie_domain_length = strlen(cookie_domain);
hostname_length = strlen(hostname);
if (cookie_domain_length >= hostname_length)
2014-05-12 12:20:59 +02:00
return 0; /* cookie_domain is too long */
p = hostname + hostname_length - cookie_domain_length;
if (!strcmp(p, cookie_domain) && p[-1] == '.') {
2014-05-12 12:20:59 +02:00
/* OK, cookie_domain matches, but it must be longer than the longest public suffix in 'hostname' */
if (!(p = psl_unregistrable_domain(psl, hostname)))
return 1;
if (cookie_domain_length > strlen(p))
return 1;
}
return 0;
}
2014-06-17 17:14:02 +02:00
/**
* psl_str_to_utf8lower:
* @str: string to convert
* @encoding: charset encoding of @str, e.g. 'iso-8859-1' or %NULL
* @locale: locale of @str for to lowercase conversion, e.g. 'de' or %NULL
* @lower: return value containing the converted string
*
* This helper function converts a string to lowercase UTF-8 representation.
* Lowercase UTF-8 is needed as input to the domain checking functions.
*
2014-06-23 12:56:13 +02:00
* @lower is set to %NULL on error.
*
2014-06-17 17:14:02 +02:00
* The return value 'lower' must be freed after usage.
*
2014-06-20 12:36:51 +02:00
* Returns: psl_error_t value.
* PSL_SUCCESS: Success
* PSL_ERR_INVALID_ARG: @str is a %NULL value.
* PSL_ERR_CONVERTER: Failed to open the unicode converter with name @encoding
* PSL_ERR_TO_UTF16: Failed to convert @str to unicode
* PSL_ERR_TO_LOWER: Failed to convert unicode to lowercase
* PSL_ERR_TO_UTF8: Failed to convert unicode to UTF-8
2014-06-17 17:14:02 +02:00
*
* Since: 0.4
*/
2014-06-20 12:36:51 +02:00
psl_error_t psl_str_to_utf8lower(const char *str, const char *encoding, const char *locale, char **lower)
2014-06-17 17:14:02 +02:00
{
2014-06-20 12:36:51 +02:00
int ret = PSL_ERR_INVALID_ARG;
2014-06-17 17:14:02 +02:00
if (lower)
*lower = NULL;
if (!str)
2014-06-20 12:36:51 +02:00
return PSL_ERR_INVALID_ARG;
/* shortcut to avoid costly conversion */
if (_str_is_ascii(str)) {
if (lower) {
char *p;
*lower = strdup(str);
/* convert ASCII string to lowercase */
for (p = *lower; *p; p++)
if (isupper(*p))
*p = tolower(*p);
}
2014-06-20 12:36:51 +02:00
return PSL_SUCCESS;
}
2014-06-17 17:14:02 +02:00
#ifdef WITH_LIBICU
2014-06-18 16:27:29 +02:00
do {
2014-06-17 17:14:02 +02:00
size_t str_length = strlen(str);
UErrorCode status = 0;
UChar *utf16_dst, *utf16_lower;
int32_t utf16_dst_length;
char *utf8_lower;
UConverter *uconv;
/* C89 allocation */
utf16_dst = alloca(sizeof(UChar) * (str_length * 2 + 1));
utf16_lower = alloca(sizeof(UChar) * (str_length * 2 + 1));
utf8_lower = alloca(str_length * 2 + 1);
uconv = ucnv_open(encoding, &status);
if (U_SUCCESS(status)) {
utf16_dst_length = ucnv_toUChars(uconv, utf16_dst, str_length * 2 + 1, str, str_length, &status);
ucnv_close(uconv);
if (U_SUCCESS(status)) {
int32_t utf16_lower_length = u_strToLower(utf16_lower, str_length * 2 + 1, utf16_dst, utf16_dst_length, locale, &status);
if (U_SUCCESS(status)) {
u_strToUTF8(utf8_lower, str_length * 8 + 1, NULL, utf16_lower, utf16_lower_length, &status);
if (U_SUCCESS(status)) {
if (lower)
*lower = strdup(utf8_lower);
2014-06-20 12:36:51 +02:00
ret = PSL_SUCCESS;
2014-06-17 17:14:02 +02:00
} else {
2014-06-20 12:36:51 +02:00
ret = PSL_ERR_TO_UTF8;
/* fprintf(stderr, "Failed to convert UTF-16 to UTF-8 (status %d)\n", status); */
2014-06-17 17:14:02 +02:00
}
} else {
2014-06-20 12:36:51 +02:00
ret = PSL_ERR_TO_LOWER;
/* fprintf(stderr, "Failed to convert UTF-16 to lowercase (status %d)\n", status); */
2014-06-17 17:14:02 +02:00
}
} else {
2014-06-20 12:36:51 +02:00
ret = PSL_ERR_TO_UTF16;
/* fprintf(stderr, "Failed to convert string to UTF-16 (status %d)\n", status); */
2014-06-17 17:14:02 +02:00
}
} else {
2014-06-20 12:36:51 +02:00
ret = PSL_ERR_CONVERTER;
/* fprintf(stderr, "Failed to open converter for '%s' (status %d)\n", encoding, status); */
2014-06-17 17:14:02 +02:00
}
2014-06-18 16:27:29 +02:00
} while (0);
2014-06-17 17:14:02 +02:00
#endif
return ret;
}