From 5174de7cd33c3d4fa591c9c93859ff9989b07e8c Mon Sep 17 00:00:00 2001 From: "Christoph M. Becker" Date: Wed, 13 May 2020 09:36:52 +0200 Subject: [PATCH] Fix #77423: parse_url() will deliver a wrong host to user To avoid that `parse_url()` returns an erroneous host, which would be valid for `FILTER_VALIDATE_URL`, we make sure that only userinfo which is valid according to RFC 3986 is treated as such. For consistency with the existing url parsing code, we use ctype functions, although that is not necessarily correct. --- ext/standard/tests/strings/url_t.phpt | 6 ++-- ext/standard/tests/url/bug77423.phpt | 30 +++++++++++++++++++ .../tests/url/parse_url_basic_001.phpt | 6 ++-- .../tests/url/parse_url_basic_003.phpt | 2 +- .../tests/url/parse_url_basic_005.phpt | 2 +- .../tests/url/parse_url_unterminated.phpt | 6 ++-- ext/standard/url.c | 24 +++++++++++++-- 7 files changed, 60 insertions(+), 16 deletions(-) create mode 100644 ext/standard/tests/url/bug77423.phpt diff --git a/ext/standard/tests/strings/url_t.phpt b/ext/standard/tests/strings/url_t.phpt index caa93cb9cf01..dc13018b14a0 100644 --- a/ext/standard/tests/strings/url_t.phpt +++ b/ext/standard/tests/strings/url_t.phpt @@ -589,15 +589,13 @@ $sample_urls = array ( string(16) "some_page_ref123" } ---> http://secret@hideout@www.php.net:80/index.php?test=1&test2=char&test3=mixesCI#some_page_ref123: array(7) { +--> http://secret@hideout@www.php.net:80/index.php?test=1&test2=char&test3=mixesCI#some_page_ref123: array(6) { ["scheme"]=> string(4) "http" ["host"]=> - string(11) "www.php.net" + string(26) "secret@hideout@www.php.net" ["port"]=> int(80) - ["user"]=> - string(14) "secret@hideout" ["path"]=> string(10) "/index.php" ["query"]=> diff --git a/ext/standard/tests/url/bug77423.phpt b/ext/standard/tests/url/bug77423.phpt new file mode 100644 index 000000000000..be03fe95e24e --- /dev/null +++ b/ext/standard/tests/url/bug77423.phpt @@ -0,0 +1,30 @@ +--TEST-- +Bug #77423 (parse_url() will deliver a wrong host to user) +--FILE-- + +--EXPECT-- +bool(false) +array(3) { + ["scheme"]=> + string(4) "http" + ["host"]=> + string(19) "php.net\@aliyun.com" + ["path"]=> + string(7) "/aaa.do" +} +bool(false) +array(2) { + ["scheme"]=> + string(5) "https" + ["host"]=> + string(26) "example.com\uFF03@bing.com" +} diff --git a/ext/standard/tests/url/parse_url_basic_001.phpt b/ext/standard/tests/url/parse_url_basic_001.phpt index f3abd703b263..89b1b7b3d683 100644 --- a/ext/standard/tests/url/parse_url_basic_001.phpt +++ b/ext/standard/tests/url/parse_url_basic_001.phpt @@ -514,15 +514,13 @@ echo "Done"; string(16) "some_page_ref123" } ---> http://secret@hideout@www.php.net:80/index.php?test=1&test2=char&test3=mixesCI#some_page_ref123: array(7) { +--> http://secret@hideout@www.php.net:80/index.php?test=1&test2=char&test3=mixesCI#some_page_ref123: array(6) { ["scheme"]=> string(4) "http" ["host"]=> - string(11) "www.php.net" + string(26) "secret@hideout@www.php.net" ["port"]=> int(80) - ["user"]=> - string(14) "secret@hideout" ["path"]=> string(10) "/index.php" ["query"]=> diff --git a/ext/standard/tests/url/parse_url_basic_003.phpt b/ext/standard/tests/url/parse_url_basic_003.phpt index 1eb64d6a1b1c..da35b927821a 100644 --- a/ext/standard/tests/url/parse_url_basic_003.phpt +++ b/ext/standard/tests/url/parse_url_basic_003.phpt @@ -62,7 +62,7 @@ echo "Done"; --> http://secret:@www.php.net/index.php?test=1&test2=char&test3=mixesCI#some_page_ref123 : string(11) "www.php.net" --> http://:hideout@www.php.net:80/index.php?test=1&test2=char&test3=mixesCI#some_page_ref123 : string(11) "www.php.net" --> http://secret:hideout@www.php.net/index.php?test=1&test2=char&test3=mixesCI#some_page_ref123 : string(11) "www.php.net" ---> http://secret@hideout@www.php.net:80/index.php?test=1&test2=char&test3=mixesCI#some_page_ref123 : string(11) "www.php.net" +--> http://secret@hideout@www.php.net:80/index.php?test=1&test2=char&test3=mixesCI#some_page_ref123 : string(26) "secret@hideout@www.php.net" --> http://secret:hid:out@www.php.net:80/index.php?test=1&test2=char&test3=mixesCI#some_page_ref123 : string(11) "www.php.net" --> nntp://news.php.net : string(12) "news.php.net" --> ftp://ftp.gnu.org/gnu/glic/glibc.tar.gz : string(11) "ftp.gnu.org" diff --git a/ext/standard/tests/url/parse_url_basic_005.phpt b/ext/standard/tests/url/parse_url_basic_005.phpt index dfbe7e7971e4..731a4adb3149 100644 --- a/ext/standard/tests/url/parse_url_basic_005.phpt +++ b/ext/standard/tests/url/parse_url_basic_005.phpt @@ -62,7 +62,7 @@ echo "Done"; --> http://secret:@www.php.net/index.php?test=1&test2=char&test3=mixesCI#some_page_ref123 : string(6) "secret" --> http://:hideout@www.php.net:80/index.php?test=1&test2=char&test3=mixesCI#some_page_ref123 : string(0) "" --> http://secret:hideout@www.php.net/index.php?test=1&test2=char&test3=mixesCI#some_page_ref123 : string(6) "secret" ---> http://secret@hideout@www.php.net:80/index.php?test=1&test2=char&test3=mixesCI#some_page_ref123 : string(14) "secret@hideout" +--> http://secret@hideout@www.php.net:80/index.php?test=1&test2=char&test3=mixesCI#some_page_ref123 : NULL --> http://secret:hid:out@www.php.net:80/index.php?test=1&test2=char&test3=mixesCI#some_page_ref123 : string(6) "secret" --> nntp://news.php.net : NULL --> ftp://ftp.gnu.org/gnu/glic/glibc.tar.gz : NULL diff --git a/ext/standard/tests/url/parse_url_unterminated.phpt b/ext/standard/tests/url/parse_url_unterminated.phpt index 7c9150a513d9..f859fc7d407f 100644 --- a/ext/standard/tests/url/parse_url_unterminated.phpt +++ b/ext/standard/tests/url/parse_url_unterminated.phpt @@ -522,15 +522,13 @@ echo "Done"; string(16) "some_page_ref123" } ---> http://secret@hideout@www.php.net:80/index.php?test=1&test2=char&test3=mixesCI#some_page_ref123: array(7) { +--> http://secret@hideout@www.php.net:80/index.php?test=1&test2=char&test3=mixesCI#some_page_ref123: array(6) { ["scheme"]=> string(4) "http" ["host"]=> - string(11) "www.php.net" + string(26) "secret@hideout@www.php.net" ["port"]=> int(80) - ["user"]=> - string(14) "secret@hideout" ["path"]=> string(10) "/index.php" ["query"]=> diff --git a/ext/standard/url.c b/ext/standard/url.c index a33091a86b75..113e0100243a 100644 --- a/ext/standard/url.c +++ b/ext/standard/url.c @@ -92,6 +92,22 @@ static const char *binary_strcspn(const char *s, const char *e, const char *char return e; } +static int is_userinfo_valid(const char *str, size_t len) +{ + const char *valid = "-._~!$&'()*+,;=:"; + const char *p = str; + while (p - str < len) { + if (isalpha(*p) || isdigit(*p) || strchr(valid, *p)) { + p++; + } else if (*p == '%' && p - str <= len - 3 && isdigit(*(p+1)) && isxdigit(*(p+2))) { + p += 3; + } else { + return 0; + } + } + return 1; +} + /* {{{ php_url_parse */ PHPAPI php_url *php_url_parse_ex(char const *str, size_t length) { @@ -233,13 +249,17 @@ PHPAPI php_url *php_url_parse_ex2(char const *str, size_t length, zend_bool *has ret->pass = zend_string_init(pp, (p-pp), 0); php_replace_controlchars_ex(ZSTR_VAL(ret->pass), ZSTR_LEN(ret->pass)); } else { - ret->user = zend_string_init(s, (p-s), 0); - php_replace_controlchars_ex(ZSTR_VAL(ret->user), ZSTR_LEN(ret->user)); + if (!is_userinfo_valid(s, p-s)) { + goto check_port; + } + ret->user = zend_string_init(s, (p-s), 0); + php_replace_controlchars_ex(ZSTR_VAL(ret->user), ZSTR_LEN(ret->user)); } s = p + 1; } +check_port: /* check for port */ if (s < ue && *s == '[' && *(e-1) == ']') { /* Short circuit portscan,