libc/src/stdio/scanf_core/string_converter.cpp

   1 //===-- String type specifier converters for scanf --------------*- C++ -*-===//
   2 //
   3 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
   4 // See https://llvm.org/LICENSE.txt for license information.
   5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
   6 //
   7 //===----------------------------------------------------------------------===//
   8
   9 #include "src/stdio/scanf_core/string_converter.h"
  10
  11 #include "src/__support/CPP/limits.h"
  12 #include "src/__support/ctype_utils.h"
  13 #include "src/__support/macros/config.h"
  14 #include "src/stdio/scanf_core/core_structs.h"
  15 #include "src/stdio/scanf_core/reader.h"
  16
  17 #include <stddef.h>
  18
  19 namespace LIBC_NAMESPACE_DECL {
  20 namespace scanf_core {
  21
  22 int convert_string(Reader *reader, const FormatSection &to_conv) {
  23   // %s "Matches a sequence of non-white-space characters"
  24
  25   // %c "Matches a sequence of characters of exactly the number specified by the
  26   // field width (1 if no field width is present in the directive)"
  27
  28   // %[ "Matches a nonempty sequence of characters from a set of expected
  29   // characters (the scanset)."
  30   size_t max_width = 0;
  31   if (to_conv.max_width > 0) {
  32     max_width = to_conv.max_width;
  33   } else {
  34     if (to_conv.conv_name == 'c') {
  35       max_width = 1;
  36     } else {
  37       max_width = cpp::numeric_limits<size_t>::max();
  38     }
  39   }
  40
  41   char *output = reinterpret_cast<char *>(to_conv.output_ptr);
  42
  43   char cur_char = reader->getc();
  44   size_t i = 0;
  45   for (; i < max_width && cur_char != '\0'; ++i) {
  46     // If this is %s and we've hit a space, or if this is %[] and we've found
  47     // something not in the scanset.
  48     if ((to_conv.conv_name == 's' && internal::isspace(cur_char)) ||
  49         (to_conv.conv_name == '[' && !to_conv.scan_set.test(cur_char))) {
  50       break;
  51     }
  52     // if the NO_WRITE flag is not set, write to the output.
  53     if ((to_conv.flags & NO_WRITE) == 0)
  54       output[i] = cur_char;
  55     cur_char = reader->getc();
  56   }
  57
  58   // We always read one more character than will be used, so we have to put the
  59   // last one back.
  60   reader->ungetc(cur_char);
  61
  62   // If this is %s or %[]
  63   if (to_conv.conv_name != 'c' && (to_conv.flags & NO_WRITE) == 0) {
  64     // Always null terminate the string. This may cause a write to the
  65     // (max_width + 1) byte, which is correct. The max width describes the max
  66     // number of characters read from the input string, and doesn't necessarily
  67     // correspond to the output.
  68     output[i] = '\0';
  69   }
  70
  71   if (i == 0)
  72     return MATCHING_FAILURE;
  73   return READ_OK;
  74 }
  75
  76 } // namespace scanf_core
  77 } // namespace LIBC_NAMESPACE_DECL