Bug 4393: compile fails on OS X

Also fixes related issues with -latomic detection and linking
to binaries that do not use C++11 atomics.

diff --git a/configure.ac b/configure.ac
index 99dcd9f4b15710567902de30c2ff18940a30cd45..671ad9c3decfcd313dee04e1b60abb86a3796460 100644 (file)
--- a/configure.ac
+++ b/configure.ac
@@ -458,10 +458,11 @@ if test "x$with_dl" = "xyes"; then
 fi
 
 ## check for atomics library before anything that might need it
-AC_SEARCH_LIBS([__atomic_load_8],[atomic])
-if test "x$ac_cv_search___atomic_load_8" = "-latomic"; then
-  ATOMICLIB="-latomic"
-fi
+# AC_SEARCH_LIBS pollutes LIBS
+SQUID_STATE_SAVE(LIBATOMIC)
+AC_SEARCH_LIBS([__atomic_load_8],[atomic],[
+  test "$ac_res" = "none required" || ATOMICLIB=$ac_res],[])
+SQUID_STATE_ROLLBACK(LIBATOMIC)
 AC_SUBST(ATOMICLIB)
 
 AC_SEARCH_LIBS([shm_open], [rt])

diff --git a/include/rfc3986.h b/include/rfc3986.h
new file mode 100644 (file)
index 0000000..43f6103
--- /dev/null
+++ b/include/rfc3986.h
@@ -0,0 +1,121 @@
+/*
+ * Copyright (C) 1996-2015 The Squid Software Foundation and contributors
+ *
+ * Squid software is distributed under GPLv2+ license and includes
+ * contributions from numerous individuals and organizations.
+ * Please see the COPYING and CONTRIBUTORS files for details.
+ */
+
+#ifndef SQUID_INCLUDE_RFC3986_H
+#define SQUID_INCLUDE_RFC3986_H
+
+#include "src/base/CharacterSet.h"
+
+// RFC 1738 URL-encoding character sets
+namespace Rfc1738
+{
+extern const CharacterSet
+        Unsafe,  // RFC 1738 unsafe set
+        Reserved, // RFC 1738 Reserved set
+        Unescaped; // RFC 1738 Unsafe and RFC 5234 CTL
+}
+
+// RFC 3986 URL-encoding
+namespace Rfc3986
+{
+
+extern const CharacterSet
+        GenDelims,// RFC 3986 gen-delims set
+        SubDelims,// RFC 3986 sub-delims set
+        Reserved, // RFC 3986 reserved characters set
+        Unreserved, // RFC 3986 unreserved characters set
+        Unescaped, // CTL and unsafe except for percent symbol
+        All;
+
+// integer representation of hex numeric characters
+extern int fromhex[256];
+
+// hex representation of each UTF-8 character
+extern const char * const tohex[256];
+
+/** unescape a percent-encoded string
+ *
+ * API-compatible with std::string and SBuf
+ */
+template <class String>
+String unescape(const String &s)
+{
+    typename String::size_type pos=s.find('%');
+    if (pos == String::npos)
+        return s;
+    String rv;
+    rv.reserve(s.length());
+    const auto e=s.end();
+    for (auto in = s.begin(); in != e; ++in) {
+        if (*in != '%') { // normal case, copy and continue
+            rv.push_back(*in);
+            continue;
+        }
+        auto ti = in;
+        ++ti;
+        if (ti == e) { // String ends in %
+            rv.push_back(*in);
+            break;
+        }
+        if (*ti == '%') { //double '%' escaping
+            rv.push_back(*in);
+            ++in;
+            continue;
+        }
+        const int v1 = fromhex[*ti];
+        if (v1 < 0) { // decoding failed at first hextdigit
+            rv.push_back(*in);
+            continue;
+        }
+        ++ti;
+        if (ti == e) { // String ends in '%[[:hexdigit:]]'
+            rv.push_back(*in);
+            continue;
+        }
+        const int v2 = fromhex[*ti];
+        if (v2 < 0) { // decoding failed at second hextdigit
+            rv.push_back(*in);
+            continue;
+        }
+        const int x = v1 << 4 | v2;
+        if (x > 0 && x <= 255) {
+            rv.push_back(static_cast<char>(x));
+            ++in; ++in;
+            continue;
+        }
+        rv.push_back(*in);
+    }
+    return rv;
+}
+
+template <class String>
+String escape(const String &s, const CharacterSet &escapeChars = Rfc1738::Unescaped)
+{
+    String rv;
+    bool didEscape = false;
+    rv.reserve(s.length()*2); //TODO: optimize arbitrary constant
+    for (auto c : s) {
+        if (escapeChars[c]) {
+            rv.push_back('%');
+            const char *hex=tohex[c];
+            rv.push_back(hex[0]);
+            rv.push_back(hex[1]);
+            didEscape = true;
+        } else {
+            rv.push_back(c);
+        }
+    }
+    if (didEscape)
+        return rv;
+    else
+        return s;
+}
+
+} // namespace Rfc3986
+
+#endif /* SQUID_INCLUDE_RFC3986_H */

diff --git a/lib/Makefile.am b/lib/Makefile.am
index bc46a4aa66608627d52a2d7e09032089a76827dd..1bff9457f80a63a8d94de3d478f6e71007bc803f 100644 (file)
--- a/lib/Makefile.am
+++ b/lib/Makefile.am
@@ -59,6 +59,7 @@ libmiscencoding_la_SOURCES = \
        md5.c \
        rfc1738.c \
        rfc2617.c \
+       rfc3986.cc \
        uudecode.c
 
 libmisccontainers_la_SOURCES = \

diff --git a/lib/rfc3986.cc b/lib/rfc3986.cc
new file mode 100644 (file)
index 0000000..d1ab41a
--- /dev/null
+++ b/lib/rfc3986.cc
@@ -0,0 +1,103 @@
+/*
+ * Copyright (C) 1996-2015 The Squid Software Foundation and contributors
+ *
+ * Squid software is distributed under GPLv2+ license and includes
+ * contributions from numerous individuals and organizations.
+ * Please see the COPYING and CONTRIBUTORS files for details.
+ */
+
+#include "squid.h"
+#include "rfc3986.h"
+#include "base/CharacterSet.h"
+
+// these should be moved to rfc1738.cc when it exists
+namespace Rfc1738
+{
+
+/* RFC 1738 section 5:
+
+  safe           = "$" | "-" | "_" | "." | "+"
+  extra          = "!" | "*" | "'" | "(" | ")" | ","
+  national       = "{" | "}" | "|" | "\" | "^" | "~" | "[" | "]" | "`"
+  punctuation    = "<" | ">" | "#" | "%" | <">
+
+  reserved       = ";" | "/" | "?" | ":" | "@" | "&" | "="
+  hex            = digit | "A" | "B" | "C" | "D" | "E" | "F" |
+                   "a" | "b" | "c" | "d" | "e" | "f"
+  escape         = "%" hex hex
+
+  unreserved     = alpha | digit | safe | extra
+  uchar          = unreserved | escape
+  xchar          = unreserved | reserved | escape
+  digits         = 1*digit
+
+*/
+
+const CharacterSet
+    Unsafe("rfc1738:unsafe", " <>\"#%{}|\\^~[]`"),
+    Reserved("rfc1738:reserved", ";/?:@&="),
+    // ? why called unescaped ? its the set which must never be used unescaped
+    Unescaped = (Unsafe + CharacterSet::CTL + CharacterSet::OBSTEXT).rename("rfc1738:unescaped")
+    ;
+
+} // namespace Rfc1738
+
+namespace Rfc3986
+{
+
+const CharacterSet
+    GenDelims("rfc3986:gen-delims",":/?#[]@"),
+    SubDelims("rfc3986:sub-delims","!$&'()*+,;="),
+    Reserved = (GenDelims + SubDelims).rename("rfc3986:reserved"),
+    Unreserved = CharacterSet("rfc3986:unreserved","-._~") +
+        CharacterSet::ALPHA + CharacterSet::DIGIT,
+    // ?
+    All = (Rfc1738::Unsafe + Reserved + CharacterSet::CTL).rename("rfc3986:all")
+    ;
+
+// integer representation of hex numeric characters,
+// or -1 for characters invalid in hex representation
+int fromhex[256] = {
+  // 0-127 (7-bit ASCII)
+  -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, // 0-15
+  -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, // 16-31
+  -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, // 32-47
+   0,  1,  2,  3,  4,  5,  6,  7,  8,  9, -1, -1, -1, -1, -1, -1, // 48-63
+  -1, 10, 11, 12, 13, 14, 15, -1, -1, -1, -1, -1, -1, -1, -1, -1, // 64-79
+  -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, // 80-95
+  -1, 10, 11, 12, 13, 14, 15, -1, -1, -1, -1, -1, -1, -1, -1, -1, // 96-111
+  -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, // 112-127
+
+  // 128-255 (8-bit UTF-8)
+  -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+  -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+  -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+  -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+  -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+  -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+  -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+  -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
+};
+
+// a hex string representation of each UTF-8 character
+const char * const tohex[256] = {
+ "00","01","02","03","04","05","06","07","08","09","0A","0B","0C","0D","0E","0F",
+ "10","11","12","13","14","15","16","17","18","19","1A","1B","1C","1D","1E","1F",
+ "20","21","22","23","24","25","26","27","28","29","2A","2B","2C","2D","2E","2F",
+ "30","31","32","33","34","35","36","37","38","39","3A","3B","3C","3D","3E","3F",
+ "40","41","42","43","44","45","46","47","48","49","4A","4B","4C","4D","4E","4F",
+ "50","51","52","53","54","55","56","57","58","59","5A","5B","5C","5D","5E","5F",
+ "60","61","62","63","64","65","66","67","68","69","6A","6B","6C","6D","6E","6F",
+ "70","71","72","73","74","75","76","77","78","79","7A","7B","7C","7D","7E","7F",
+ "80","81","82","83","84","85","86","87","88","89","8A","8B","8C","8D","8E","8F",
+ "90","91","92","93","94","95","96","97","98","99","9A","9B","9C","9D","9E","9F",
+ "A0","A1","A2","A3","A4","A5","A6","A7","A8","A9","AA","AB","AC","AD","AE","AF",
+ "B0","B1","B2","B3","B4","B5","B6","B7","B8","B9","BA","BB","BC","BD","BE","BF",
+ "C0","C1","C2","C3","C4","C5","C6","C7","C8","C9","CA","CB","CC","CD","CE","CF",
+ "D0","D1","D2","D3","D4","D5","D6","D7","D8","D9","DA","DB","DC","DD","DE","DF",
+ "E0","E1","E2","E3","E4","E5","E6","E7","E8","E9","EA","EB","EC","ED","EE","EF",
+ "F0","F1","F2","F3","F4","F5","F6","F7","F8","F9","FA","FB","FC","FD","FE","FF"
+};
+
+} // namespace Rfc3986
+

diff --git a/src/Makefile.am b/src/Makefile.am
index a9205726a5ccdf04af661df2e20060f72db96443..565fdced3a6be4e64acad04ac980dafedac45773 100644 (file)
--- a/src/Makefile.am
+++ b/src/Makefile.am
@@ -3825,6 +3825,27 @@ tests_testEnumIterator_LDADD = \
        $(XTRA_LIBS)
 tests_testEnumIterator_DEPENDENCIES =
 
+check_PROGRAMS += tests/testRFC3986
+tests_testRFC3986_SOURCES = \
+       $(SBUF_SOURCE) \
+       String.cc \
+       tests/stub_debug.cc \
+       tests/stub_libmem.cc \
+       tests/stub_SBufDetailedStats.cc \
+       tests/testRFC3986.h \
+       tests/testRFC3986.cc
+nodist_tests_testRFC3986_SOURCES = \
+       $(TESTSOURCES)
+tests_testRFC3986_LDFLAGS = $(LIBADD_DL)
+tests_testRFC3986_LDADD = \
+       base/libbase.la \
+       $(top_builddir)/lib/libmiscencoding.la \
+       $(SQUID_CPPUNIT_LIBS) \
+       $(COMPAT_LIB) \
+       $(SQUID_CPPUNIT_LA) \
+       $(XTRA_LIBS)
+tests_testRFC3986_DEPENDENCIES =
+
 TESTS += testHeaders
 
 ## Special Universal .h dependency test script

diff --git a/src/tests/testRFC3986.cc b/src/tests/testRFC3986.cc
new file mode 100644 (file)
index 0000000..8638837
--- /dev/null
+++ b/src/tests/testRFC3986.cc
@@ -0,0 +1,132 @@
+/*
+ * Copyright (C) 1996-2015 The Squid Software Foundation and contributors
+ *
+ * Squid software is distributed under GPLv2+ license and includes
+ * contributions from numerous individuals and organizations.
+ * Please see the COPYING and CONTRIBUTORS files for details.
+ */
+
+#include "squid.h"
+#include "SBuf.h"
+#include "testRFC3986.h"
+#include "rfc1738.h"
+#include "unitTestMain.h"
+
+#include <cassert>
+
+/* Being a C library code it is best bodily included and tested with C++ type-safe techniques. */
+#include "lib/rfc3986.cc"
+
+CPPUNIT_TEST_SUITE_REGISTRATION( testRFC3986 );
+
+static void
+performDecodingTest(const std::string &encoded_str, const std::string &plaintext_str)
+{
+    std::string decoded_str = rfc3986_unescape(encoded_str);
+    CPPUNIT_ASSERT_EQUAL(plaintext_str, decoded_str);
+
+    SBuf encoded_sbuf(encoded_str);
+    SBuf plaintext_sbuf(plaintext_str);
+    SBuf decoded_sbuf = rfc3986_unescape(encoded_sbuf);
+    CPPUNIT_ASSERT_EQUAL(plaintext_sbuf, decoded_sbuf);
+}
+
+/* Regular Format de-coding tests */
+void testRFC3986::testUrlDecode()
+{
+    performDecodingTest("%2Fdata%2Fsource%2Fpath","/data/source/path");
+    performDecodingTest("http://foo.invalid%2Fdata%2Fsource%2Fpath",
+                    "http://foo.invalid/data/source/path");
+    // TODO query string
+
+    performDecodingTest("1 w%0Ard","1 w\nrd"); // Newline %0A encoded
+    performDecodingTest("2 w%rd","2 w%rd"); // Un-encoded %
+    performDecodingTest("3 w%%rd","3 w%rd"); // encoded %
+    performDecodingTest("5 Bad String %1","5 Bad String %1"); // corrupt string
+    performDecodingTest("6 Bad String %1A%3","6 Bad String \032%3"); //partly corrupt string
+    performDecodingTest("7 Good String %1A","7 Good String \032"); // non corrupt string
+    //test various endings
+    performDecodingTest("8 word%","8 word%");
+    performDecodingTest("9 word%z","9 word%z");
+    performDecodingTest("10 word%1","10 word%1");
+    performDecodingTest("11 word%1q","11 word%1q");
+    performDecodingTest("12 word%1a","12 word\032");
+ }
+
+// perform a test for std::string, SBuf and if rfc1738flag is != 0 compare
+//  against rfc1738 implementation
+static void
+performEncodingTest(const char *plaintext_str, const char *encoded_str, int rfc1738flag, const CharacterSet  &rfc3986CSet)
+{
+    CPPUNIT_ASSERT_EQUAL(std::string(encoded_str), rfc3986_escape(std::string(plaintext_str), rfc3986CSet));
+    CPPUNIT_ASSERT_EQUAL(SBuf(encoded_str), rfc3986_escape(SBuf(plaintext_str), rfc3986CSet));
+    if (!rfc1738flag)
+        return;
+    char *result = rfc1738_do_escape(plaintext_str, rfc1738flag);
+    CPPUNIT_ASSERT_EQUAL(std::string(encoded_str), std::string(result));
+}
+void testRFC3986::testUrlEncode()
+{
+    /* TEST: Escaping only unsafe characters */
+    performEncodingTest("http://foo.invalid/data/source/path",
+                        "http://foo.invalid/data/source/path",
+                        RFC1738_ESCAPE_UNSAFE, RFC3986::Unsafe);
+
+    /* regular URL (no encoding needed) */
+    performEncodingTest("http://foo.invalid/data/source/path",
+                        "http://foo.invalid/data/source/path",
+                        RFC1738_ESCAPE_UNSAFE, RFC3986::Unsafe);
+
+    /* long string of unsafe # characters */
+    performEncodingTest("################ ################ ################ ################ ################ ################ ################ ################",
+                        "%23%23%23%23%23%23%23%23%23%23%23%23%23%23%23%23%20%23%23%23%23%23%23%23%23%23%23%23%23%23%23%23%23%20%23%23%23%23%23%23%23%23%23%23%23%23%23%23%23%23%20%23%23%23%23%23%23%23%23%23%23%23%23%23%23%23%23%20%23%23%23%23%23%23%23%23%23%23%23%23%23%23%23%23%20%23%23%23%23%23%23%23%23%23%23%23%23%23%23%23%23%20%23%23%23%23%23%23%23%23%23%23%23%23%23%23%23%23%20%23%23%23%23%23%23%23%23%23%23%23%23%23%23%23%23",
+                        RFC1738_ESCAPE_UNSAFE, RFC3986::Unsafe);
+
+    /* TEST: escaping only reserved characters */
+
+    /* regular URL (full encoding requested) */
+    performEncodingTest("http://foo.invalid/data/source/path",
+                        "http%3A%2F%2Ffoo.invalid%2Fdata%2Fsource%2Fpath",
+                        RFC1738_ESCAPE_RESERVED, RFC3986::Reserved);
+
+    /* regular path (encoding wanted for ALL special chars) */
+    performEncodingTest("/data/source/path",
+                        "%2Fdata%2Fsource%2Fpath",
+                        RFC1738_ESCAPE_RESERVED, RFC3986::Reserved);
+
+    /* TEST: safety-escaping a string already partially escaped */
+
+    /* escaping of dangerous characters in a partially escaped string */
+    performEncodingTest("http://foo.invalid/data%2Fsource[]",
+                        "http://foo.invalid/data%2Fsource%5B%5D",
+                        RFC1738_ESCAPE_UNESCAPED, RFC3986::Unescaped);
+
+    /* escaping of hexadecimal 0xFF characters in a partially escaped string */
+    performEncodingTest("http://foo.invalid/data%2Fsource\xFF\xFF",
+                        "http://foo.invalid/data%2Fsource%FF%FF",
+                        RFC1738_ESCAPE_UNESCAPED, RFC3986::Unescaped);
+}
+
+/** SECURITY BUG TESTS: avoid null truncation attacks by skipping %00 bytes */
+void testRFC3986::PercentZeroNullDecoding()
+{
+    /* Attack with %00 encoded NULL */
+    performDecodingTest("w%00rd", "w%00rd");
+
+    /* Attack with %0 encoded NULL */
+    performDecodingTest("w%0rd", "w%0rd");
+
+    /* Handle '0' bytes embeded in encoded % */
+    performDecodingTest("w%%00%rd", "w%00%rd");
+
+    /* Handle NULL bytes with encoded % */
+    performDecodingTest("w%%%00%rd", "w%%00%rd");
+}
+
+void
+testRFC3986::testPerformance()
+{
+
+}
+
+

diff --git a/src/tests/testRFC3986.h b/src/tests/testRFC3986.h
new file mode 100644 (file)
index 0000000..2679411
--- /dev/null
+++ b/src/tests/testRFC3986.h
@@ -0,0 +1,38 @@
+/*
+ * Copyright (C) 1996-2015 The Squid Software Foundation and contributors
+ *
+ * Squid software is distributed under GPLv2+ license and includes
+ * contributions from numerous individuals and organizations.
+ * Please see the COPYING and CONTRIBUTORS files for details.
+ */
+
+#ifndef SQUID_LIB_TEST_RFC3986_H
+#define SQUID_LIB_TEST_RFC3986_H
+
+#include <cppunit/extensions/HelperMacros.h>
+
+/**
+ * Test the URL coder RFC 3986 Engine
+ */
+class testRFC3986 : public CPPUNIT_NS::TestFixture
+{
+    CPPUNIT_TEST_SUITE( testRFC3986 );
+    CPPUNIT_TEST( testUrlDecode );
+    CPPUNIT_TEST( testUrlEncode );
+    CPPUNIT_TEST( PercentZeroNullDecoding );
+    CPPUNIT_TEST( testPerformance );
+    CPPUNIT_TEST_SUITE_END();
+
+public:
+
+protected:
+    void testUrlDecode();
+    void testUrlEncode();
+
+    // bugs.
+    void PercentZeroNullDecoding();
+    void testPerformance();
+};
+
+#endif /* SQUID_LIB_TEST_RFC3986_H */
+