Thread View: gnu.g++.lib.bug
3 messages
3 total messages
Started by Marc van Woerkom
Mon, 03 Aug 1992 23:29
Plans on iostream-lib
Author: Marc van Woerkom
Date: Mon, 03 Aug 1992 23:29
Date: Mon, 03 Aug 1992 23:29
32 lines
1587 bytes
1587 bytes
Hello!
I have some questions on the future of the I/O part of libg++:
* To me, the iostream concept is much more appealing than the stdio
stuff from C. But the current implementation is not behaving well
enough, to forget stdio. I encounterd problems with manipulators,
random positioning and binary I/O.
-> Is it a future goal, to make the iostream package more compatible
with cfront's one? (e.g. the examples in Stroustrup's C++PL2)
-> Are there plans for a 'binary filemode', to allow more general
I/O?
-> Is it possible to extend the documentation with some examples?
(I'm using the emx port of gcc-2.1 with libg++-2.2, and it's
sometimes hard to figure out the UNIX customs..)
-> Do you think it's realistic to do all I/O with the iostream
package or will there still be need for C way?
Bye, Marc
Marc van Woerkom ****** Aachen Center for Solidification in Space *******
ACCESS e.V., Intzestr. 5 E-Mail: marc@pxcl1.gi.rwth-aachen.de
D-5100 Aachen, Germany Marc_van-Woerkom@ac3.maus.de
Fax: +49 241 38578 "Trust me, I know what I'm doing!" (Sledge Hammer)
Re: Plans on iostream-lib
Author: bothner
Date: Tue, 04 Aug 1992 07:34
Date: Tue, 04 Aug 1992 07:34
32 lines
1585 bytes
1585 bytes
marc@pxcl1.gi.rwth-aachen.de (Marc van Woerkom) writes: > -> Is it a future goal, to make the iostream package more compatible > with cfront's one? (e.g. the examples in Stroustrup's C++PL2) Somewhat. The next release will use multiple inhertiance, and template-based iomanipulators. Which specific example in C++PL2 doesn't work? > -> Are there plans for a 'binary filemode', to allow more general > I/O? I don't know. What is a "binary filemode" (for Unix) and why do you need it? > -> Is it possible to extend the documentation with some examples? Yes, please do so ... > -> Do you think it's realistic to do all I/O with the iostream > package or will there still be need for C way? If your're calling C functions that except a (FILE*) argument, then you will need to do I/O the C way ... Otherwise, note that libg++/iostream/stdio contains a *complete* implementation of stdio built *on top of* the streambuf classes in libg++/iostream. This the the stdio implementation used in Linux, so it is clearly realistic to do all I/O with the iostream package. --Per Bothner Cygnus Support bothner@cygnus.com
Re: Plans on iostream-lib
Author: John Edstrom
Date: Thu, 06 Aug 1992 12:05
Date: Thu, 06 Aug 1992 12:05
32 lines
1507 bytes
1507 bytes
In article <9208042134.AA18130@cygnus.com> bothner@cygnus.com writes: >marc@pxcl1.gi.rwth-aachen.de (Marc van Woerkom) writes: > >> -> Is it a future goal, to make the iostream package more compatible >> with cfront's one? (e.g. the examples in Stroustrup's C++PL2) > >Somewhat. The next release will use multiple inhertiance, >and template-based iomanipulators. Which specific example >in C++PL2 doesn't work? > >> -> Are there plans for a 'binary filemode', to allow more general >> I/O? > >I don't know. What is a "binary filemode" (for Unix) and why do you >need it? > A bit off topic but how about mapping memory to files? That's something I'd like to see. Just a suggestion. > > --Per Bothner >Cygnus Support bothner@cygnus.com > John Edstrom | edstrom @ elmer.hsc.ucalgary.ca -- RM 2104, HSc Building, Div. Neuroscience U. Calgary School of Medicine, 3330 Hospital Drive NW Calgary, Alberta T2N 3Y4 (403) 220 4493 (wk)
Thread Navigation
This is a paginated view of messages in the thread with full content displayed inline.
Messages are displayed in chronological order, with the original post highlighted in green.
Use pagination controls to navigate through all messages in large threads.
Back to All Threads