[libc] implement inet_aton (llvm#162651)

This patch adds the implementation for `inet_aton` function. Since this
function is not explicitly included in POSIX, I have marked it with
`llvm_libc_ext`. It is widely available and commonly used, and can also
be used to implement `inet_addr`, which is included in POSIX.

Author: c8ef

libc/config/linux/aarch64/entrypoints.txt

@@ -945,6 +945,7 @@ if(LLVM_LIBC_FULL_BUILD)
     # arpa/inet.h entrypoints
     libc.src.arpa.inet.htonl
     libc.src.arpa.inet.htons
+    libc.src.arpa.inet.inet_aton
     libc.src.arpa.inet.ntohl
     libc.src.arpa.inet.ntohs
 

libc/config/linux/riscv/entrypoints.txt

@@ -1077,6 +1077,7 @@ if(LLVM_LIBC_FULL_BUILD)
     # arpa/inet.h entrypoints
     libc.src.arpa.inet.htonl
     libc.src.arpa.inet.htons
+    libc.src.arpa.inet.inet_aton
     libc.src.arpa.inet.ntohl
     libc.src.arpa.inet.ntohs
 

libc/config/linux/x86_64/entrypoints.txt

@@ -1113,6 +1113,7 @@ if(LLVM_LIBC_FULL_BUILD)
     # arpa/inet.h entrypoints
     libc.src.arpa.inet.htonl
     libc.src.arpa.inet.htons
+    libc.src.arpa.inet.inet_aton
     libc.src.arpa.inet.ntohl
     libc.src.arpa.inet.ntohs
 

libc/docs/dev/undefined_behavior.rst

@@ -156,3 +156,10 @@ parsed as normal. For l64a it's unspecified what happens if the input value is
 negative. For LLVM-libc, all inputs to l64a are treated as unsigned 32 bit ints.
 Additionally, the return of l64a is in a thread-local buffer that's overwritten
 on each call.
+
+`inet_aton` and Non-Standard Binary Integers
+--------------------------------------------
+The current implementation of the `inet_aton` function utilizes the same code
+as `strtol` to parse IPv4 numbers-and-dots notations. This approach may permit
+the use of binary integers (prefixed with 0b), which is not supported by the
+standard.

libc/include/arpa/inet.yaml

@@ -1,7 +1,8 @@
 header: arpa/inet.h
 header_template: inet.h.def
 macros: []
-types: []
+types:
+  - type_name: in_addr
 enums: []
 objects: []
 functions:
@@ -17,6 +18,13 @@ functions:
     return_type: uint16_t
     arguments:
       - type: uint16_t
+  - name: inet_aton
+    standards:
+      - llvm_libc_ext
+    return_type: int
+    arguments:
+      - type: const char *
+      - type: in_addr *
   - name: ntohl
     standards:
       - POSIX

libc/src/arpa/inet/CMakeLists.txt

@@ -22,6 +22,19 @@ add_entrypoint_object(
     libc.src.__support.common
 )
 
+add_entrypoint_object(
+  inet_aton
+  SRCS
+    inet_aton.cpp
+  HDRS
+    inet_aton.h
+  DEPENDS
+    libc.include.arpa_inet
+    libc.include.llvm-libc-types.in_addr
+    libc.src.__support.common
+    libc.src.__support.str_to_integer
+)
+
 add_entrypoint_object(
   ntohl
   SRCS

libc/src/arpa/inet/inet_aton.cpp

@@ -0,0 +1,57 @@
+//===-- Implementation of inet_aton function ------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "src/arpa/inet/inet_aton.h"
+#include "src/__support/common.h"
+#include "src/__support/endian_internal.h"
+#include "src/__support/str_to_integer.h"
+
+namespace LIBC_NAMESPACE_DECL {
+
+LLVM_LIBC_FUNCTION(int, inet_aton, (const char *cp, in_addr *inp)) {
+  constexpr int IPV4_MAX_DOT_NUM = 3;
+  unsigned long parts[IPV4_MAX_DOT_NUM + 1] = {0};
+  int dot_num = 0;
+
+  for (; dot_num <= IPV4_MAX_DOT_NUM; ++dot_num) {
+    auto result = internal::strtointeger<unsigned long>(cp, 0);
+    parts[dot_num] = result;
+
+    if (result.has_error() || result.parsed_len == 0)
+      return 0;
+    char next_char = *(cp + result.parsed_len);
+    if (next_char != '.' && next_char != '\0')
+      return 0;
+    else if (next_char == '\0')
+      break;
+    else
+      cp += (result.parsed_len + 1);
+  }
+
+  if (dot_num > IPV4_MAX_DOT_NUM)
+    return 0;
+
+  // converts the Internet host address cp from the IPv4 numbers-and-dots
+  // notation (a[.b[.c[.d]]]) into binary form (in network byte order)
+  unsigned long result = 0;
+  for (int i = 0; i <= dot_num; ++i) {
+    unsigned long max_part =
+        i == dot_num ? (0xffffffffUL >> (8 * dot_num)) : 0xffUL;
+    if (parts[i] > max_part)
+      return 0;
+    int shift = i == dot_num ? 0 : 8 * (IPV4_MAX_DOT_NUM - i);
+    result |= parts[i] << shift;
+  }
+
+  if (inp)
+    inp->s_addr = Endian::to_big_endian(static_cast<uint32_t>(result));
+
+  return 1;
+}
+
+} // namespace LIBC_NAMESPACE_DECL

libc/src/arpa/inet/inet_aton.h

@@ -0,0 +1,21 @@
+//===-- Implementation header of inet_aton ----------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_LIBC_SRC_ARPA_INET_INET_ATON_H
+#define LLVM_LIBC_SRC_ARPA_INET_INET_ATON_H
+
+#include "include/llvm-libc-types/in_addr.h"
+#include "src/__support/macros/config.h"
+
+namespace LIBC_NAMESPACE_DECL {
+
+int inet_aton(const char *cp, in_addr *inp);
+
+} // namespace LIBC_NAMESPACE_DECL
+
+#endif // LLVM_LIBC_SRC_ARPA_INET_INET_ATON_H

libc/test/src/arpa/inet/CMakeLists.txt

@@ -22,6 +22,17 @@ add_libc_unittest(
     libc.src.arpa.inet.ntohs
 )
 
+add_libc_unittest(
+  inet_aton
+  SUITE
+    libc_arpa_inet_unittests
+  SRCS
+    inet_aton_test.cpp
+  DEPENDS
+    libc.src.arpa.inet.htonl
+    libc.src.arpa.inet.inet_aton
+)
+
 add_libc_unittest(
   ntohl
   SUITE

libc/test/src/arpa/inet/inet_aton_test.cpp

@@ -0,0 +1,92 @@
+//===-- Unittests for inet_aton -------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "src/arpa/inet/htonl.h"
+#include "src/arpa/inet/inet_aton.h"
+#include "test/UnitTest/Test.h"
+
+namespace LIBC_NAMESPACE_DECL {
+
+TEST(LlvmLibcInetAton, ValidTest) {
+  in_addr a;
+
+  // a.b.c.d
+  a.s_addr = 0;
+  ASSERT_EQ(1, inet_aton("127.1.2.4", &a));
+  ASSERT_EQ(htonl(0x7f010204), a.s_addr);
+
+  // a.b.c
+  a.s_addr = 0;
+  ASSERT_EQ(1, inet_aton("127.1.4", &a));
+  ASSERT_EQ(htonl(0x7f010004), a.s_addr);
+
+  // a.b
+  a.s_addr = 0;
+  ASSERT_EQ(1, inet_aton("127.1", &a));
+  ASSERT_EQ(htonl(0x7f000001), a.s_addr);
+
+  // a
+  a.s_addr = 0;
+  ASSERT_EQ(1, inet_aton("0x7f000001", &a));
+  ASSERT_EQ(htonl(0x7f000001), a.s_addr);
+
+  // Hex (0x) and mixed-case hex digits.
+  a.s_addr = 0;
+  ASSERT_EQ(1, inet_aton("0xFf.0.0.1", &a));
+  ASSERT_EQ(htonl(0xff000001), a.s_addr);
+
+  // Hex (0X) and mixed-case hex digits.
+  a.s_addr = 0;
+  ASSERT_EQ(1, inet_aton("0XfF.0.0.1", &a));
+  ASSERT_EQ(htonl(0xff000001), a.s_addr);
+
+  // Octal.
+  a.s_addr = 0;
+  ASSERT_EQ(1, inet_aton("0177.0.0.1", &a));
+  ASSERT_EQ(htonl(0x7f000001), a.s_addr);
+
+  a.s_addr = 0;
+  ASSERT_EQ(1, inet_aton("036", &a));
+  ASSERT_EQ(htonl(036U), a.s_addr);
+}
+
+TEST(LlvmLibcInetAton, InvalidTest) {
+  ASSERT_EQ(0, inet_aton("", nullptr));           // Empty.
+  ASSERT_EQ(0, inet_aton("x", nullptr));          // Leading junk.
+  ASSERT_EQ(0, inet_aton("127.0.0.1x", nullptr)); // Trailing junk.
+  ASSERT_EQ(0, inet_aton("09.0.0.1", nullptr));   // Invalid octal.
+  ASSERT_EQ(0, inet_aton("0xg.0.0.1", nullptr));  // Invalid hex.
+  ASSERT_EQ(0, inet_aton("1.2.3.4.5", nullptr));  // Too many dots.
+  ASSERT_EQ(0, inet_aton("1.2.3.4.", nullptr));   // Trailing dot.
+
+  // Out of range a.b.c.d form.
+  ASSERT_EQ(0, inet_aton("999.0.0.1", nullptr));
+  ASSERT_EQ(0, inet_aton("0.999.0.1", nullptr));
+  ASSERT_EQ(0, inet_aton("0.0.999.1", nullptr));
+  ASSERT_EQ(0, inet_aton("0.0.0.999", nullptr));
+
+  // Out of range a.b.c form.
+  ASSERT_EQ(0, inet_aton("256.0.0", nullptr));
+  ASSERT_EQ(0, inet_aton("0.256.0", nullptr));
+  ASSERT_EQ(0, inet_aton("0.0.0x10000", nullptr));
+
+  // Out of range a.b form.
+  ASSERT_EQ(0, inet_aton("256.0", nullptr));
+  ASSERT_EQ(0, inet_aton("0.0x1000000", nullptr));
+
+  // Out of range a form.
+  ASSERT_EQ(0, inet_aton("0x100000000", nullptr));
+
+  // 64-bit overflow.
+  ASSERT_EQ(0, inet_aton("0x10000000000000000", nullptr));
+
+  // Out of range octal.
+  ASSERT_EQ(0, inet_aton("0400.0.0.1", nullptr));
+}
+
+} // namespace LIBC_NAMESPACE_DECL