0001040: open_wmemstream() and mbstate_t - Austin Group Defect Tracker

Austin Group Defect Tracker

Aardvark Mark IV

Viewing Issue Simple Details [ Jump to Notes ]

[ Issue History ] [ Print ]

Category

Severity

Type

Date Submitted

Last Update

0001040

[1003.1(2013)/Issue7+TC1] System Interfaces

Objection

Error

2016-03-31 09:14

2024-06-11 08:57

Reporter

geoffclare

View Status

public

Assigned To

Priority

normal

Resolution

Accepted

Status

Closed

Name

Geoff Clare

Organization

The Open Group

User Reference

Section

2.5.2 Stream Orientation and Encoding Rules

Page Number

498

Line Number

17062, 17057, 17049, 17038

Interp Status

---

Final Accepted Text

Summary

0001040: open_wmemstream() and mbstate_t

Description

Since streams opened with open_wmemstream() do not involve any
wide-character to multi-byte character conversion, the requirements
in the standard relating to the mbstate_t object associated with
wide-oriented streams should not apply to streams opened with
open_wmemstream().

Desired Action

On page 498 line 17038 section 2.5.2

Change:

Each wide-oriented stream has an associated mbstate_t object ...

to:

Each wide-oriented stream [CX]that was not opened with open_wmemstream()[/CX] has an associated mbstate_t object ...

On page 498 line 17049 section 2.5.2

Change:

... the external file associated with a wide-oriented stream is a sequence of (possibly multi-byte) characters ...

to:

... the external file associated with a wide-oriented stream [CX]that was not opened with open_wmemstream()[/CX] is a sequence of (possibly multi-byte) characters ...

On page 498 line 17057 section 2.5.2

Change:

The wide-character input functions read characters from the stream and convert them ...

to:

[CX]On streams that were not opened with open_wmemstream()[/CX], the wide-character input functions read characters from the stream and convert them ...

On page 498 line 17062 section 2.5.2

Change:

The wide-character output functions convert wide characters to (possibly multi-byte) characters ...

to:

[CX]On streams that were not opened with open_wmemstream()[/CX], the wide-character output functions convert wide characters to (possibly multi-byte) characters ...

Tags

tc3-2008

Attached Files

Relationships

Notes
(0003112) shware_systems (reporter) 2016-03-31 14:29	I agree the use of mbstate_t to always down convert, width wise, wchars to a multibyte encoding on output and up convert on input not necessary, but they are still required to handle the %c and %s format specifiers, without the l modifier, for wscan() and wprint(). Also, to properly handle Unicode in Issue 8, the mbstate_t needs fields to process multiword encodings in char16_t and char32_t arrays used as arguments to those, in the absence of a mwstate_t type. I think the change proposed doesn't cover that sufficiently. At present, "possibly multibyte" covers "possibly multiword" implicitly.

Issue History
Date Modified	Username	Field	Change
2016-03-31 09:14	geoffclare	New Issue
2016-03-31 09:14	geoffclare	Name	=> Geoff Clare
2016-03-31 09:14	geoffclare	Organization	=> The Open Group
2016-03-31 09:14	geoffclare	Section	=> 2.5.2 Stream Orientation and Encoding Rules
2016-03-31 09:14	geoffclare	Page Number	=> 498
2016-03-31 09:14	geoffclare	Line Number	=> 17062, 17057, 17049, 17038
2016-03-31 09:14	geoffclare	Interp Status	=> ---
2016-03-31 14:29	shware_systems	Note Added: 0003112
2017-03-16 16:31	Don Cragun	Status	New => Resolved
2017-03-16 16:31	Don Cragun	Resolution	Open => Accepted
2017-03-16 16:32	Don Cragun	Tag Attached: tc3-2008
2019-10-23 10:20	geoffclare	Status	Resolved => Applied
2024-06-11 08:57	agadmin	Status	Applied => Closed

Mantis 1.1.6[^]