📄 standard_library.html
字号:
<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML//EN"><html><head> <title>C++ I/O library (experimental)</title> <META name="description" content="A freely available implementation of the Standard C++ Library, including the STL and the IOstream/locale library"> <META name="keywords" content="STL, standard library, iostream, locale"></head><BODY TEXT="#000000" LINK="#006600" ALINK="#003300" VLINK="#7C7F87" BGCOLOR="#FFFFFF"><A HREF="/"><IMG SRC="/images/common/sgilogo_small.gif" ALT="SGI Logo" WIDTH="80" HEIGHT="72" BORDER="0"></A><P><Center><h1 Align="Center">C++ I/O library (experimental)</h1></Center> <P>This is an <em>experimental</em> snapshot of SGI's standard C++library. You should <A Href="#snapshot">download</A> it if you are interested in contributing to an ongoing development effort, but not(yet) if you are looking for a library implementation that isguaranteed to work "out of the box".<P>This snapshot includes the entire standard C++ library: STL,iostream/locale, and numerics (<tt>valarray</tt> and<tt>complex</tt>). It does not include a standard C libraryimplementation; it is intended to interoperate with an existing Clibrary.<P>Please note that some aspects of an I/O library are inherentlyplatform-specific. (See below for more details.) This library hasbeen tested on IRIX 6.x using the SGI MIPSpro 7.3 compiler, and onLinux, using glibc 2.0 and the egcs 1.1.2 compilers. It passes simpletests on Microsoft Windows NT using the Microsoft Visual C++ 6.0compiler (with Service Pack 3). Compiling it on other platforms willalmost certainly require at least some work. Porting to anotherversion of Unix, using a relatively recent C++ compiler, should not bedifficult.<P>We welcome suggestions, patches, and bug fixes. Please send comments to <A Href="mailto:stl@sgi.com">stl@sgi.com</A>.<h2>Some notes on the structure of the I/O library</h2><ul><li> It shares some files, such as char_traits.h, with our STL. These snapshots include the full STL distribution as well as the iostreams implementation, since it's hard to draw a line between these two parts of the standard C++ library. <P><li> Certain parts of the STL depend on the I/O library, and they behave differently depending on whether they use "classic" AT&T cfront iostreams or a standard-conforming I/O implementation with templatized iostreams. To use the new iostreams library you must tell the compiler where to find the headers, add the I/O library to the link line, and tell the STL that it should use standard-conforming iostreams. A compilation line may look something like this: <pre> g++ -I. -D__STL_USE_NEW_IOSTREAMS -g hello.cxx libCio.a </pre> Alternatively, you can edit the <tt>stl_config.h</tt> header so that the macro <tt>__STL_USE_NEW_IOSTREAMS</tt> gets defined automatically.<P><li> Locales are a tricky point. It is impossible to layer C++ locales on top of the C library locales. The problem is that the standard C library only has one active locale at a time, as determined by the global <tt>setlocale</tt> function. In C++, though, locales are objects, and there can be any number of active locales at the same time. In a single-threaded world it would be possible (although horribly inefficient) to implement this with <tt>setlocale</tt>, but there's no safe way to do that in a multithreaded environment. <P> This iostream/locale library is intended to coexist with a preexisting C library. It assumes that for any C library implementation there is some way---just not a portable way---to get at the underlying data files so that we can have multiple active locales. We defined an interface, found in in <tt>c_locale.h</tt>, for this functionality. Our C++ I/O library requires some implementation for the <tt>c_locale.h</tt> interface. We believe that implementing the <tt>c_locale.h</tt> interface is the major work involved in porting the I/O library to a new platform. <P> We have implemented the c_locale interface for the IRIX C library (it is not included here because it includes code that, for legal reasons, SGI can't release as open source), and for the GNU glibc 2.x library. <P> Additionally, we provide a stub implementation of the c_locale interface. This interface is used only for named locales that the OS knows about, such as (in IRIX) "de", "es_MX", and "fr_BE". It isn't used for user-defined locales, or for the classic locale. So the stub implementation takes any string, and returns a failure indication when the user attempts to create a named locale. That's perfectly conforming, according to the C++ standard, and it's even useful. (At least for people who don't have a pressing need for internationalization.) This can serve as a stopgap for platforms where we don't yet have a real c_locale implementation.<P><li> There are a couple of smaller places where we have system-dependent code. One, of course, is in <tt>basic_filebuf</tt> (see especially <tt>fstream.cxx</tt>), which makes system calls for reading and writing. (On Unix, it uses such system calls as <tt>open</tt>, <tt>close</tt>, <tt>read</tt>, <tt>write</tt>, <tt>fcntl</tt>, <tt>lseek</tt>, <tt>mmap</tt>, and <tt>munmap</tt>.) <p> We have tried to abstract out as much system-dependent stuff as possible into the <tt>_Filebuf_base</tt> class. At present it supports IRIX and Linux; porting to other versions of Unix should be easy. There is also some incomplete support for NT; issues there include the '\n' <-> CRLF translation in text mode, and the use of native Win32 system calls, especially for memory mapped I/O.<P><li> Another place where we've got system dependent code is syncing with stdio. The C++ standard requires that there be a mode in which iostreams and stdio are synchronized (the C++ standard is imprecise about just what synchronization means, but, after talking to the people who wrote that part of the standard, it's clear that it means a very close coupling), and it requires that synchronized mode be the default. We could have satisfied this requirement by implementing things in terms of <tt>putc</tt> and <tt>getc</tt>, but that would have been horribly inefficient. Instead we did something uglier, but more efficient: we directly manipulate the pointers inside a <tt>FILE</tt> object. We've encapsulated knowledge of struct <tt>FILE</tt> into the header <tt><stl_stdio_file.h></tt>, which defines a few inline functions for this manipulation. Again, this header has to be modified for every OS that we target. This has not proven to be a serious problem in practice, since most stdio implementations appear to be very similar to each other.<P> (An aside: the C++ standard does not say what type of streambuf is used by the standard streams, <tt>cin</tt>, <tt>cout</tt>, <tt>cerr</tt>, etc. Our choice: the wide streams, and in unsynchronized mode the narrow streams as well, use <tt>basic_filebuf</tt>. In synchronized mode the narrow streams instead use <tt>SGI::stdio_istreambuf</tt> or <tt>SGI::stdio_ostreambuf</tt>. Those classes are defined in the header <tt><stdio_streambuf></tt>, and, of course, they are extensions.)<P><li> Initialization of iostreams is tricky. The standard specifies how it's supposed to work in some detail, and it doesn't seem to be possible to do it portably. Our solution, which is admittedly a ghastly hack, is in <tt>iostream.cxx</tt>. It's nonportable, but it will work on many platforms. On some platforms there are better ways of doing this.</ul><h2>Areas that need more work</h2><ul><li> Portability to other platforms <ul> <li> Workarounds for compilers that don't support the full C++ language <li> Writing a low-level locale implementation for other C libraries. <li> Use native file I/O system calls for non-Unix operating systems. <li> Testing on other configurations </ul><li> The <tt>basic_fstream</tt> class sometimes uses memory-mapped I/O for file input. It should also use memory-mapped I/O for output when possible.<li> Testing of nontrivial code conversions from wide to narrow characters, especially with variable-width and state-dependent encodings, has been sporadic; it's likely that there are still bugs in this area. Most of the relevant code is in <tt><fstream></tt>.<li> The numeric input facets handle <tt>istreambuf_iterator</tt> naively. They should look directly at the buffer inside the streambuf, rather than using <tt>sgetc</tt>, <tt>sbumpc</tt>, and <tt>snextc</tt> to get one character at a time. (See the header <tt><istream></tt> for an example of how this can be done.) The same optimization applies to the time and monetary facets, but they're less important.<li> There has been some performance tuning for better execution speed, but there has been very little performance tuning for code size or compilation speed.</ul><hr><A Name="snapshot"><h1>Snapshots</h1></A> <Table Border Width="100%"><TR><TD> June 8, 2000</TD><TD><A Href="stdlib_20000608.tar.gz">stdlib_20000608.tar.gz</A> <br> <A Href="stdlib_20000608.zip">stdlib_20000608.zip</A></TD><TD>Many small bug fixes and compatibility patches. Change to thelow-level locale API to improve handling of locale "", which indicates that we are to use a locale that corresponds to the user'spreference. (The major changes are in c_locale.h and locale.cxx.)This change is probably a performance improvement in the common case.</TD></TR><TR><TD> May 5, 2000</TD><TD><A Href="stdlib_20000505.tar.gz">stdlib_20000505.tar.gz</A> <br> <A Href="stdlib_20000505.zip">stdlib_20000505.zip</A></TD><TD>Changed <tt>deque</tt> to remove the extension of a third, defaulttemplate parameter by which the user can control the node size. Theextension does not appear to be standard-conforming.</TD></TR><TR><TD> April 18, 2000</TD><TD><A Href="stdlib_20000418.tar.gz">stdlib_20000418.tar.gz</A> <br> <A Href="stdlib_20000418.zip">stdlib_20000418.zip</A></TD><TD>Changed <tt>basic_istream::seekg</tt> and <tt>basic_ostream::seekp</tt>so that they conform to the resolution to library issue 136 adopted in Tokyo. (<tt>seekg</tt> should set only the get pointer,and <tt>seekp</tt> should set only get put pointer.) Fixed a bug thataffects string output when the width is nonzero; thanks to VolkerH. Simonis for a bug report and patch. Changed <tt>bitset</tt> toremove the extension of a second template parameter. The extensiondoes not appear to be standard-conforming (see library issue 94), and it isn't very useful anyway.</TD>⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -