flang/runtime/connection.h

   1 //===-- runtime/connection.h ------------------------------------*- 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 // Fortran I/O connection state (abstracted over internal & external units)
  10
  11 #ifndef FORTRAN_RUNTIME_IO_CONNECTION_H_
  12 #define FORTRAN_RUNTIME_IO_CONNECTION_H_
  13
  14 #include "format.h"
  15 #include <cinttypes>
  16 #include <optional>
  17
  18 namespace Fortran::runtime::io {
  19
  20 class IoStatementState;
  21
  22 enum class Direction { Output, Input };
  23 enum class Access { Sequential, Direct, Stream };
  24
  25 // These characteristics of a connection are immutable after being
  26 // established in an OPEN statement.
  27 struct ConnectionAttributes {
  28   Access access{Access::Sequential}; // ACCESS='SEQUENTIAL', 'DIRECT', 'STREAM'
  29   std::optional<bool> isUnformatted; // FORM='UNFORMATTED' if true
  30   bool isUTF8{false}; // ENCODING='UTF-8'
  31   unsigned char internalIoCharKind{0}; // 0->external, 1/2/4->internal
  32   std::optional<std::int64_t> openRecl; // RECL= on OPEN
  33
  34   bool IsRecordFile() const {
  35     // Formatted stream files are viewed as having records, at least on input
  36     return access != Access::Stream || !isUnformatted.value_or(true);
  37   }
  38
  39   template <typename CHAR = char> constexpr bool useUTF8() const {
  40     // For wide CHARACTER kinds, always use UTF-8 for formatted I/O.
  41     // For single-byte CHARACTER, encode characters >= 0x80 with
  42     // UTF-8 iff the mode is set.
  43     return internalIoCharKind == 0 && (sizeof(CHAR) > 1 || isUTF8);
  44   }
  45 };
  46
  47 struct ConnectionState : public ConnectionAttributes {
  48   bool IsAtEOF() const; // true when read has hit EOF or endfile record
  49   bool IsAfterEndfile() const; // true after ENDFILE until repositioned
  50
  51   // All positions and measurements are always in units of bytes,
  52   // not characters.  Multi-byte character encodings are possible in
  53   // both internal I/O (when the character kind of the variable is 2 or 4)
  54   // and external formatted I/O (when the encoding is UTF-8).
  55   std::size_t RemainingSpaceInRecord() const;
  56   bool NeedAdvance(std::size_t) const;
  57   void HandleAbsolutePosition(std::int64_t);
  58   void HandleRelativePosition(std::int64_t);
  59
  60   void BeginRecord() {
  61     positionInRecord = 0;
  62     furthestPositionInRecord = 0;
  63     unterminatedRecord = false;
  64   }
  65
  66   std::optional<std::int64_t> EffectiveRecordLength() const {
  67     // When an input record is longer than an explicit RECL= from OPEN
  68     // it is effectively truncated on input.
  69     return openRecl && recordLength && *openRecl < *recordLength ? openRecl
  70                                                                  : recordLength;
  71   }
  72
  73   std::optional<std::int64_t> recordLength;
  74
  75   std::int64_t currentRecordNumber{1}; // 1 is first
  76
  77   // positionInRecord is the 0-based bytes offset in the current record
  78   // to/from which the next data transfer will occur.  It can be past
  79   // furthestPositionInRecord if moved by an X or T or TR control edit
  80   // descriptor.
  81   std::int64_t positionInRecord{0};
  82
  83   // furthestPositionInRecord is the 0-based byte offset of the greatest
  84   // position in the current record to/from which any data transfer has
  85   // occurred, plus one.  It can be viewed as a count of bytes processed.
  86   std::int64_t furthestPositionInRecord{0}; // max(position+bytes)
  87
  88   // Set at end of non-advancing I/O data transfer
  89   std::optional<std::int64_t> leftTabLimit; // offset in current record
  90
  91   // currentRecordNumber value captured after ENDFILE/REWIND/BACKSPACE statement
  92   // or an end-of-file READ condition on a sequential access file
  93   std::optional<std::int64_t> endfileRecordNumber;
  94
  95   // Mutable modes set at OPEN() that can be overridden in READ/WRITE & FORMAT
  96   MutableModes modes; // BLANK=, DECIMAL=, SIGN=, ROUND=, PAD=, DELIM=, kP
  97
  98   // Set when processing repeated items during list-directed & NAMELIST input
  99   // in order to keep a span of records in frame on a non-positionable file,
 100   // so that backspacing to the beginning of the repeated item doesn't require
 101   // repositioning the external storage medium when that's impossible.
 102   bool pinnedFrame{false};
 103
 104   // Set when the last record of a file is not properly terminated
 105   // so that a non-advancing READ will not signal EOR.
 106   bool unterminatedRecord{false};
 107 };
 108
 109 // Utility class for capturing and restoring a position in an input stream.
 110 class SavedPosition {
 111 public:
 112   explicit SavedPosition(IoStatementState &);
 113   ~SavedPosition();
 114
 115 private:
 116   IoStatementState &io_;
 117   ConnectionState saved_;
 118 };
 119
 120 } // namespace Fortran::runtime::io
 121 #endif // FORTRAN_RUNTIME_IO_CONNECTION_H_