2012-03-27 23:13:14 +00:00
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
2018-12-28 15:30:48 +00:00
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> < html xmlns = "http://www.w3.org/1999/xhtml" > < head > < meta http-equiv = "Content-Type" content = "text/html; charset=UTF-8" / > < title > Backwards Compatibility< / title > < meta name = "generator" content = "DocBook XSL Stylesheets Vsnapshot" / > < meta name = "keywords" content = "ISO C++, backwards" / > < meta name = "keywords" content = "ISO C++, library" / > < meta name = "keywords" content = "ISO C++, runtime, library" / > < link rel = "home" href = "../index.html" title = "The GNU C++ Library" / > < link rel = "up" href = "appendix_porting.html" title = "Appendix / > < link rel = "prev" href = "api.html" title = "API Evolution and Deprecation History" / > < link rel = "next" href = "appendix_free.html" title = "Appendix / > < / head > < body > < div class = "navheader" > < table width = "100%" summary = "Navigation header" > < tr > < th colspan = "3" align = "center" > Backwards Compatibility< / th > < / tr > < tr > < td width = "20%" align = "left" > < a accesskey = "p" href = "api.html" > Prev< / a > < / td > < th width = "60%" align = "center" > Appendix
2012-03-27 23:13:14 +00:00
Porting and Maintenance
2014-09-21 17:33:12 +00:00
< / th > < td width = "20%" align = "right" > < a accesskey = "n" href = "appendix_free.html" > Next< / a > < / td > < / tr > < / table > < hr / > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h2 class = "title" style = "clear: both" > < a id = "manual.appendix.porting.backwards" > < / a > Backwards Compatibility< / h2 > < / div > < / div > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h3 class = "title" > < a id = "backwards.first" > < / a > First< / h3 > < / div > < / div > < / div > < p > The first generation GNU C++ library was called libg++. It was a
2012-03-27 23:13:14 +00:00
separate GNU project, although reliably paired with GCC. Rumors imply
that it had a working relationship with at least two kinds of
dinosaur.
< / p > < p > Some background: libg++ was designed and created when there was no
ISO standard to provide guidance. Classes like linked lists are now
provided for by < code class = "classname" > list< T> < / code > and do not need to be
created by < code class = "function" > genclass< / code > . (For that matter, templates exist
now and are well-supported, whereas genclass (mostly) predates them.)
< / p > < p > There are other classes in libg++ that are not specified in the
ISO Standard (e.g., statistical analysis). While there are a lot of
really useful things that are used by a lot of people, the Standards
Committee couldn't include everything, and so a lot of those
< span class = "quote" > “< span class = "quote" > obvious< / span > ”< / span > classes didn't get included.
2014-09-21 17:33:12 +00:00
< / p > < p > Known Issues include many of the limitations of its immediate ancestor.< / p > < p > Portability notes and known implementation limitations are as follows.< / p > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.first.ios_base" > < / a > No < code class = "code" > ios_base< / code > < / h4 > < / div > < / div > < / div > < p > At least some older implementations don't have < code class = "code" > std::ios_base< / code > , so you should use < code class = "code" > std::ios::badbit< / code > , < code class = "code" > std::ios::failbit< / code > and < code class = "code" > std::ios::eofbit< / code > and < code class = "code" > std::ios::goodbit< / code > .
< / p > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.first.cout_cin" > < / a > No < code class = "code" > cout< / code > in < code class = "filename" > < ostream.h> < / code > , no < code class = "code" > cin< / code > in < code class = "filename" > < istream.h> < / code > < / h4 > < / div > < / div > < / div > < p >
2012-03-27 23:13:14 +00:00
In earlier versions of the standard,
< code class = "filename" > < fstream.h> < / code > ,
< code class = "filename" > < ostream.h> < / code >
and < code class = "filename" > < istream.h> < / code >
used to define
< code class = "code" > cout< / code > , < code class = "code" > cin< / code > and so on. ISO C++ specifies that one needs to include
< code class = "filename" > < iostream> < / code >
explicitly to get the required definitions.
< / p > < p > Some include adjustment may be required.< / p > < p > This project is no longer maintained or supported, and the sources
archived. For the desperate,
2014-09-21 17:33:12 +00:00
the < a class = "link" href = "http://gcc.gnu.org/extensions.html" target = "_top" > GCC extensions
2012-03-27 23:13:14 +00:00
page< / a > describes where to find the last libg++ source. The code is
considered replaced and rewritten.
2014-09-21 17:33:12 +00:00
< / p > < / div > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h3 class = "title" > < a id = "backwards.second" > < / a > Second< / h3 > < / div > < / div > < / div > < p >
2012-03-27 23:13:14 +00:00
The second generation GNU C++ library was called libstdc++, or
libstdc++-v2. It spans the time between libg++ and pre-ISO C++
standardization and is usually associated with the following GCC
releases: egcs 1.x, gcc 2.95, and gcc 2.96.
< / p > < p >
The STL portions of this library are based on SGI/HP STL release 3.11.
< / p > < p >
This project is no longer maintained or supported, and the sources
archived. The code is considered replaced and rewritten.
< / p > < p >
Portability notes and known implementation limitations are as follows.
2014-09-21 17:33:12 +00:00
< / p > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.second.std" > < / a > Namespace < code class = "code" > std::< / code > not supported< / h4 > < / div > < / div > < / div > < p >
2012-03-27 23:13:14 +00:00
Some care is required to support C++ compiler and or library
implementation that do not have the standard library in
< code class = "code" > namespace std< / code > .
< / p > < p >
The following sections list some possible solutions to support compilers
that cannot ignore < code class = "code" > std::< / code > -qualified names.
< / p > < p >
First, see if the compiler has a flag for this. Namespace
back-portability-issues are generally not a problem for g++
compilers that do not have libstdc++ in < code class = "code" > std::< / code > , as the
compilers use < code class = "option" > -fno-honor-std< / code > (ignore
< code class = "code" > std::< / code > , < code class = "code" > :: = std::< / code > ) by default. That is,
the responsibility for enabling or disabling < code class = "code" > std::< / code > is
on the user; the maintainer does not have to care about it. This
probably applies to some other compilers as well.
< / p > < p >
Second, experiment with a variety of pre-processor tricks.
< / p > < p >
By defining < code class = "code" > std< / code > as a macro, fully-qualified namespace
calls become global. Volia.
< / p > < pre class = "programlisting" >
#ifdef WICKEDLY_OLD_COMPILER
# define std
#endif
< / pre > < p >
Thanks to Juergen Heinzl who posted this solution on gnu.gcc.help.
< / p > < p >
Another pre-processor based approach is to define a macro
< code class = "code" > NAMESPACE_STD< / code > , which is defined to either
< span class = "quote" > “< span class = "quote" > < / span > ”< / span > or < span class = "quote" > “< span class = "quote" > std< / span > ”< / span > based on a compile-type
test. On GNU systems, this can be done with autotools by means of
an autoconf test (see below) for < code class = "code" > HAVE_NAMESPACE_STD< / code > ,
then using that to set a value for the < code class = "code" > NAMESPACE_STD< / code >
macro. At that point, one is able to use
< code class = "code" > NAMESPACE_STD::string< / code > , which will evaluate to
< code class = "code" > std::string< / code > or < code class = "code" > ::string< / code > (i.e., in the
global namespace on systems that do not put < code class = "code" > string< / code > in
< code class = "code" > std::< / code > ).
< / p > < pre class = "programlisting" >
dnl @synopsis AC_CXX_NAMESPACE_STD
dnl
dnl If the compiler supports namespace std, define
dnl HAVE_NAMESPACE_STD.
dnl
dnl @category Cxx
dnl @author Todd Veldhuizen
dnl @author Luc Maisonobe < luc@spaceroots.org>
dnl @version 2004-02-04
dnl @license AllPermissive
AC_DEFUN([AC_CXX_NAMESPACE_STD], [
AC_CACHE_CHECK(if g++ supports namespace std,
ac_cv_cxx_have_std_namespace,
[AC_LANG_SAVE
AC_LANG_CPLUSPLUS
AC_TRY_COMPILE([#include < iostream>
std::istream& is = std::cin;],,
ac_cv_cxx_have_std_namespace=yes, ac_cv_cxx_have_std_namespace=no)
AC_LANG_RESTORE
])
if test "$ac_cv_cxx_have_std_namespace" = yes; then
AC_DEFINE(HAVE_NAMESPACE_STD,,[Define if g++ supports namespace std. ])
fi
])
2014-09-21 17:33:12 +00:00
< / pre > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.second.iterators" > < / a > Illegal iterator usage< / h4 > < / div > < / div > < / div > < p >
2012-03-27 23:13:14 +00:00
The following illustrate implementation-allowed illegal iterator
use, and then correct use.
2014-09-21 17:33:12 +00:00
< / p > < div class = "itemizedlist" > < ul class = "itemizedlist" style = "list-style-type: disc; " > < li class = "listitem" > < p >
2012-03-27 23:13:14 +00:00
you cannot do < code class = "code" > ostream::operator< < (iterator)< / code >
to print the address of the iterator => use
< code class = "code" > operator< < & *iterator< / code > instead
< / p > < / li > < li class = "listitem" > < p >
you cannot clear an iterator's reference (< code class = "code" > iterator =
0< / code > ) => use < code class = "code" > iterator = iterator_type();< / code >
< / p > < / li > < li class = "listitem" > < p >
< code class = "code" > if (iterator)< / code > won't work any more => use
< code class = "code" > if (iterator != iterator_type())< / code >
2014-09-21 17:33:12 +00:00
< / p > < / li > < / ul > < / div > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.second.isspace" > < / a > < code class = "code" > isspace< / code > from < code class = "filename" > < cctype> < / code > is a macro
2012-03-27 23:13:14 +00:00
< / h4 > < / div > < / div > < / div > < p >
Glibc 2.0.x and 2.1.x define < code class = "filename" > < ctype.h> < / code > functionality as macros
(isspace, isalpha etc.).
< / p > < p >
This implementations of libstdc++, however, keep these functions
as macros, and so it is not back-portable to use fully qualified
names. For example:
< / p > < pre class = "programlisting" >
#include < cctype>
int main() { std::isspace('X'); }
< / pre > < p >
Results in something like this:
< / p > < pre class = "programlisting" >
std:: (__ctype_b[(int) ( ( 'X' ) )] & (unsigned short int) _ISspace ) ;
< / pre > < p >
A solution is to modify a header-file so that the compiler tells
< code class = "filename" > < ctype.h> < / code > to define functions
instead of macros:
< / p > < pre class = "programlisting" >
// This keeps isalnum, et al from being propagated as macros.
#if __linux__
# define __NO_CTYPE 1
#endif
< / pre > < p >
Then, include < code class = "filename" > < ctype.h> < / code >
< / p > < p >
Another problem arises if you put a < code class = "code" > using namespace
std;< / code > declaration at the top, and include
< code class = "filename" > < ctype.h> < / code > . This will
result in ambiguities between the definitions in the global namespace
(< code class = "filename" > < ctype.h> < / code > ) and the
definitions in namespace < code class = "code" > std::< / code >
(< code class = "code" > < cctype> < / code > ).
2014-09-21 17:33:12 +00:00
< / p > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.second.at" > < / a > No < code class = "code" > vector::at< / code > , < code class = "code" > deque::at< / code > , < code class = "code" > string::at< / code > < / h4 > < / div > < / div > < / div > < p >
2012-03-27 23:13:14 +00:00
One solution is to add an autoconf-test for this:
< / p > < pre class = "programlisting" >
AC_MSG_CHECKING(for container::at)
AC_TRY_COMPILE(
[
#include < vector>
#include < deque>
#include < string>
using namespace std;
],
[
deque< int> test_deque(3);
test_deque.at(2);
vector< int> test_vector(2);
test_vector.at(1);
string test_string(< span class = "quote" > “< span class = "quote" > test_string< / span > ”< / span > );
test_string.at(3);
],
[AC_MSG_RESULT(yes)
AC_DEFINE(HAVE_CONTAINER_AT)],
[AC_MSG_RESULT(no)])
< / pre > < p >
If you are using other (non-GNU) compilers it might be a good idea
to check for < code class = "code" > string::at< / code > separately.
2014-09-21 17:33:12 +00:00
< / p > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.second.eof" > < / a > No < code class = "code" > std::char_traits< char> ::eof< / code > < / h4 > < / div > < / div > < / div > < p >
2012-03-27 23:13:14 +00:00
Use some kind of autoconf test, plus this:
< / p > < pre class = "programlisting" >
#ifdef HAVE_CHAR_TRAITS
#define CPP_EOF std::char_traits< char> ::eof()
#else
#define CPP_EOF EOF
#endif
2014-09-21 17:33:12 +00:00
< / pre > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.second.stringclear" > < / a > No < code class = "code" > string::clear< / code > < / h4 > < / div > < / div > < / div > < p >
2012-03-27 23:13:14 +00:00
There are two functions for deleting the contents of a string:
< code class = "code" > clear< / code > and < code class = "code" > erase< / code > (the latter returns the
string).
< / p > < pre class = "programlisting" >
void
clear() { _M_mutate(0, this-> size(), 0); }
< / pre > < pre class = "programlisting" >
basic_string&
erase(size_type __pos = 0, size_type __n = npos)
{
return this-> replace(_M_check(__pos), _M_fold(__pos, __n),
_M_data(), _M_data());
}
< / pre > < p >
Unfortunately, < code class = "code" > clear< / code > is not implemented in this
version, so you should use < code class = "code" > erase< / code > (which is probably
faster than < code class = "code" > operator=(charT*)< / code > ).
2014-09-21 17:33:12 +00:00
< / p > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.second.ostreamform_istreamscan" > < / a >
2012-03-27 23:13:14 +00:00
Removal of < code class = "code" > ostream::form< / code > and < code class = "code" > istream::scan< / code >
extensions
< / h4 > < / div > < / div > < / div > < p >
These are no longer supported. Please use stringstreams instead.
2014-09-21 17:33:12 +00:00
< / p > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.second.stringstreams" > < / a > No < code class = "code" > basic_stringbuf< / code > , < code class = "code" > basic_stringstream< / code > < / h4 > < / div > < / div > < / div > < p >
2012-03-27 23:13:14 +00:00
Although the ISO standard < code class = "code" > i/ostringstream< / code > -classes are
provided, (< code class = "filename" > < sstream> < / code > ), for
compatibility with older implementations the pre-ISO
< code class = "code" > i/ostrstream< / code > (< code class = "filename" > < strstream> < / code > ) interface is also provided,
with these caveats:
2014-09-21 17:33:12 +00:00
< / p > < div class = "itemizedlist" > < ul class = "itemizedlist" style = "list-style-type: disc; " > < li class = "listitem" > < p >
2012-03-27 23:13:14 +00:00
< code class = "code" > strstream< / code > is considered to be deprecated
< / p > < / li > < li class = "listitem" > < p >
< code class = "code" > strstream< / code > is limited to < code class = "code" > char< / code >
< / p > < / li > < li class = "listitem" > < p >
with < code class = "code" > ostringstream< / code > you don't have to take care of
terminating the string or freeing its memory
< / p > < / li > < li class = "listitem" > < p >
< code class = "code" > istringstream< / code > can be re-filled (clear();
str(input);)
< / p > < / li > < / ul > < / div > < p >
You can then use output-stringstreams like this:
< / p > < pre class = "programlisting" >
#ifdef HAVE_SSTREAM
# include < sstream>
#else
# include < strstream>
#endif
#ifdef HAVE_SSTREAM
std::ostringstream oss;
#else
std::ostrstream oss;
#endif
2015-08-28 15:33:40 +00:00
oss < < "Name=" < < m_name < < ", number=" < < m_number < < std::endl;
2012-03-27 23:13:14 +00:00
...
#ifndef HAVE_SSTREAM
oss < < std::ends; // terminate the char*-string
#endif
// str() returns char* for ostrstream and a string for ostringstream
// this also causes ostrstream to think that the buffer's memory
// is yours
m_label.set_text(oss.str());
#ifndef HAVE_SSTREAM
// let the ostrstream take care of freeing the memory
oss.freeze(false);
#endif
< / pre > < p >
Input-stringstreams can be used similarly:
< / p > < pre class = "programlisting" >
std::string input;
...
#ifdef HAVE_SSTREAM
std::istringstream iss(input);
#else
std::istrstream iss(input.c_str());
#endif
int i;
iss > > i;
< / pre > < p > One (the only?) restriction is that an istrstream cannot be re-filled:
< / p > < pre class = "programlisting" >
std::istringstream iss(numerator);
iss > > m_num;
// this is not possible with istrstream
iss.clear();
iss.str(denominator);
iss > > m_den;
< / pre > < p >
If you don't care about speed, you can put these conversions in
a template-function:
< / p > < pre class = "programlisting" >
template < class X>
void fromString(const string& input, X& any)
{
#ifdef HAVE_SSTREAM
std::istringstream iss(input);
#else
std::istrstream iss(input.c_str());
#endif
X temp;
iss > > temp;
if (iss.fail())
throw runtime_error(..)
any = temp;
}
< / pre > < p >
Another example of using stringstreams is in < a class = "link" href = "strings.html#strings.string.shrink" title = "Shrink to Fit" > this howto< / a > .
< / p > < p > There is additional information in the libstdc++-v2 info files, in
particular < span class = "quote" > “< span class = "quote" > info iostream< / span > ”< / span > .
2014-09-21 17:33:12 +00:00
< / p > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.second.wchar" > < / a > Little or no wide character support< / h4 > < / div > < / div > < / div > < p >
2012-03-27 23:13:14 +00:00
Classes < code class = "classname" > wstring< / code > and
< code class = "classname" > char_traits< wchar_t> < / code > are
not supported.
2014-09-21 17:33:12 +00:00
< / p > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.second.iostream_templates" > < / a > No templatized iostreams< / h4 > < / div > < / div > < / div > < p >
2012-03-27 23:13:14 +00:00
Classes < code class = "classname" > wfilebuf< / code > and
< code class = "classname" > wstringstream< / code > are not supported.
2014-09-21 17:33:12 +00:00
< / p > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.second.thread_safety" > < / a > Thread safety issues< / h4 > < / div > < / div > < / div > < p >
2012-03-27 23:13:14 +00:00
Earlier GCC releases had a somewhat different approach to
threading configuration and proper compilation. Before GCC 3.0,
configuration of the threading model was dictated by compiler
command-line options and macros (both of which were somewhat
thread-implementation and port-specific). There were no
guarantees related to being able to link code compiled with one
set of options and macro setting with another set.
< / p > < p >
For GCC 3.0, configuration of the threading model used with
libraries and user-code is performed when GCC is configured and
built using the --enable-threads and --disable-threads options.
The ABI is stable for symbol name-mangling and limited functional
compatibility exists between code compiled under different
threading models.
< / p > < p >
The libstdc++ library has been designed so that it can be used in
multithreaded applications (with libstdc++-v2 this was only true
of the STL parts.) The first problem is finding a
< span class = "emphasis" > < em > fast< / em > < / span > method of implementation portable to
all platforms. Due to historical reasons, some of the library is
written against per-CPU-architecture spinlocks and other parts
against the gthr.h abstraction layer which is provided by gcc. A
minor problem that pops up every so often is different
interpretations of what "thread-safe" means for a
2018-12-28 15:30:48 +00:00
library (not a general program). We currently use the < a class = "link" href = "https://web.archive.org/web/20171225062613/http://www.sgi.com/tech/stl/thread_safety.html" target = "_top" > same
2012-03-27 23:13:14 +00:00
definition that SGI< / a > uses for their STL subset. However,
the exception for read-only containers only applies to the STL
components. This definition is widely-used and something similar
will be used in the next version of the C++ standard library.
< / p > < p >
Here is a small link farm to threads (no pun) in the mail
archives that discuss the threading problem. Each link is to the
first relevant message in the thread; from there you can use
"Thread Next" to move down the thread. This farm is in
latest-to-oldest order.
2014-09-21 17:33:12 +00:00
< / p > < div class = "itemizedlist" > < ul class = "itemizedlist" style = "list-style-type: disc; " > < li class = "listitem" > < p >
Our threading expert Loren gives a breakdown of < a class = "link" href = "http://gcc.gnu.org/ml/libstdc++/2001-10/msg00024.html" target = "_top" > the
2012-03-27 23:13:14 +00:00
six situations involving threads< / a > for the 3.0
release series.
< / p > < / li > < li class = "listitem" > < p >
2014-09-21 17:33:12 +00:00
< a class = "link" href = "http://gcc.gnu.org/ml/libstdc++/2001-05/msg00384.html" target = "_top" >
2012-03-27 23:13:14 +00:00
This message< / a > inspired a recent updating of issues with
threading and the SGI STL library. It also contains some
example POSIX-multithreaded STL code.
< / p > < / li > < / ul > < / div > < p >
(A large selection of links to older messages has been removed;
many of the messages from 1999 were lost in a disk crash, and the
few people with access to the backup tapes have been too swamped
with work to restore them. Many of the points have been
superseded anyhow.)
2014-09-21 17:33:12 +00:00
< / p > < / div > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h3 class = "title" > < a id = "backwards.third" > < / a > Third< / h3 > < / div > < / div > < / div > < p > The third generation GNU C++ library is called libstdc++, or
2012-03-27 23:13:14 +00:00
libstdc++-v3.
< / p > < p > The subset commonly known as the Standard Template Library
2014-09-21 17:33:12 +00:00
(clauses 23 through 25, mostly) is adapted from the final release
2012-03-27 23:13:14 +00:00
of the SGI STL (version 3.3), with extensive changes.
< / p > < p > A more formal description of the V3 goals can be found in the
official < a class = "link" href = "source_design_notes.html" title = "Design Notes" > design document< / a > .
2017-10-07 00:16:47 +00:00
< / p > < p > Portability notes and known implementation limitations are as follows.< / p > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.third.headers" > < / a > Pre-ISO headers removed< / h4 > < / div > < / div > < / div > < p > The pre-ISO C++ headers
2012-03-27 23:13:14 +00:00
(< code class = "filename" > < iostream.h> < / code > ,
< code class = "filename" > < defalloc.h> < / code > etc.) are
2017-10-07 00:16:47 +00:00
not supported.
< / p > < p > For those of you new to ISO C++ (welcome, time travelers!), the
ancient pre-ISO headers have new names.
The C++ FAQ has a good explanation in < a class = "link" href = "https://isocpp.org/wiki/faq/coding-standards#std-headers" target = "_top" > What's
2014-09-21 17:33:12 +00:00
the difference between < xxx> and < xxx.h> headers?< / a > .
2017-10-07 00:16:47 +00:00
< / p > < p > Porting between pre-ISO headers and ISO headers is simple: headers
2012-03-27 23:13:14 +00:00
like < code class = "filename" > < vector.h> < / code > can be replaced with < code class = "filename" > < vector> < / code > and a using
directive < code class = "code" > using namespace std;< / code > can be put at the global
scope. This should be enough to get this code compiling, assuming the
other usage is correct.
2014-09-21 17:33:12 +00:00
< / p > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.third.hash" > < / a > Extension headers hash_map, hash_set moved to ext or backwards< / h4 > < / div > < / div > < / div > < p > At this time most of the features of the SGI STL extension have been
2012-03-27 23:13:14 +00:00
replaced by standardized libraries.
In particular, the < code class = "classname" > unordered_map< / code > and
< code class = "classname" > unordered_set< / code > containers of TR1 and C++ 2011
are suitable replacements for the non-standard
< code class = "classname" > hash_map< / code > and < code class = "classname" > hash_set< / code >
containers in the SGI STL.
< / p > < p > Header files < code class = "filename" > < hash_map> < / code > and < code class = "filename" > < hash_set> < / code > moved
to < code class = "filename" > < ext/hash_map> < / code > and < code class = "filename" > < ext/hash_set> < / code > ,
respectively. At the same time, all types in these files are enclosed
in < code class = "code" > namespace __gnu_cxx< / code > . Later versions deprecate
these files, and suggest using TR1's < code class = "filename" > < unordered_map> < / code >
and < code class = "filename" > < unordered_set> < / code > instead.
< / p > < p > The extensions are no longer in the global or < code class = "code" > std< / code >
namespaces, instead they are declared in the < code class = "code" > __gnu_cxx< / code >
namespace. For maximum portability, consider defining a namespace
alias to use to talk about extensions, e.g.:
< / p > < pre class = "programlisting" >
#ifdef __GNUC__
#if __GNUC__ < 3
#include < hash_map.h>
namespace extension { using ::hash_map; }; // inherit globals
#else
#include < backward/hash_map>
#if __GNUC__ == 3 & & __GNUC_MINOR__ == 0
namespace extension = std; // GCC 3.0
#else
namespace extension = ::__gnu_cxx; // GCC 3.1 and later
#endif
#endif
#else // ... there are other compilers, right?
namespace extension = std;
#endif
extension::hash_map< int,int> my_map;
< / pre > < p > This is a bit cleaner than defining typedefs for all the
instantiations you might need.
< / p > < p > The following autoconf tests check for working HP/SGI hash containers.
< / p > < pre class = "programlisting" >
# AC_HEADER_EXT_HASH_MAP
AC_DEFUN([AC_HEADER_EXT_HASH_MAP], [
AC_CACHE_CHECK(for ext/hash_map,
ac_cv_cxx_ext_hash_map,
[AC_LANG_SAVE
AC_LANG_CPLUSPLUS
ac_save_CXXFLAGS="$CXXFLAGS"
CXXFLAGS="$CXXFLAGS -Werror"
AC_TRY_COMPILE([#include < ext/hash_map> ], [using __gnu_cxx::hash_map;],
ac_cv_cxx_ext_hash_map=yes, ac_cv_cxx_ext_hash_map=no)
CXXFLAGS="$ac_save_CXXFLAGS"
AC_LANG_RESTORE
])
if test "$ac_cv_cxx_ext_hash_map" = yes; then
AC_DEFINE(HAVE_EXT_HASH_MAP,,[Define if ext/hash_map is present. ])
fi
])
< / pre > < pre class = "programlisting" >
# AC_HEADER_EXT_HASH_SET
AC_DEFUN([AC_HEADER_EXT_HASH_SET], [
AC_CACHE_CHECK(for ext/hash_set,
ac_cv_cxx_ext_hash_set,
[AC_LANG_SAVE
AC_LANG_CPLUSPLUS
ac_save_CXXFLAGS="$CXXFLAGS"
CXXFLAGS="$CXXFLAGS -Werror"
AC_TRY_COMPILE([#include < ext/hash_set> ], [using __gnu_cxx::hash_set;],
ac_cv_cxx_ext_hash_set=yes, ac_cv_cxx_ext_hash_set=no)
CXXFLAGS="$ac_save_CXXFLAGS"
AC_LANG_RESTORE
])
if test "$ac_cv_cxx_ext_hash_set" = yes; then
AC_DEFINE(HAVE_EXT_HASH_SET,,[Define if ext/hash_set is present. ])
fi
])
2014-09-21 17:33:12 +00:00
< / pre > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.third.nocreate_noreplace" > < / a > No < code class = "code" > ios::nocreate/ios::noreplace< / code > .
< / h4 > < / div > < / div > < / div > < p > Historically these flags were used with iostreams to control whether
new files are created or not when opening a file stream, similar to the
< code class = "code" > O_CREAT< / code > and < code class = "code" > O_EXCL< / code > flags for the
< code class = "function" > open(2)< / code > system call. Because iostream modes correspond
to < code class = "function" > fopen(3)< / code > modes these flags are not supported.
For input streams a new file will not be created anyway, so
< code class = "code" > ios::nocreate< / code > is not needed.
For output streams, a new file will be created if it does not exist, which is
consistent with the behaviour of < code class = "function" > fopen< / code > .
< / p > < p > When one of these flags is needed a possible alternative is to attempt
to open the file using < span class = "type" > std::ifstream< / span > first to determine whether
the file already exists or not. This may not be reliable however, because
whether the file exists or not could change between opening the
< span class = "type" > std::istream< / span > and re-opening with an output stream. If you need
to check for existence and open a file as a single operation then you will
need to use OS-specific facilities outside the C++ standard library, such
as < code class = "function" > open(2)< / code > .
< / p > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.third.streamattach" > < / a >
2012-03-27 23:13:14 +00:00
No < code class = "code" > stream::attach(int fd)< / code >
< / h4 > < / div > < / div > < / div > < p >
Phil Edwards writes: It was considered and rejected for the ISO
standard. Not all environments use file descriptors. Of those
that do, not all of them use integers to represent them.
< / p > < p >
For a portable solution (among systems which use
file descriptors), you need to implement a subclass of
< code class = "code" > std::streambuf< / code > (or
< code class = "code" > std::basic_streambuf< ..> < / code > ) which opens a file
given a descriptor, and then pass an instance of this to the
stream-constructor.
< / p > < p >
An extension is available that implements this.
2019-06-02 15:48:37 +00:00
< code class = "filename" > < ext/stdio_filebuf.h> < / code >
contains a derived class called
< code class = "classname" > __gnu_cxx::stdio_filebuf< / code > .
2012-03-27 23:13:14 +00:00
This class can be constructed from a C < code class = "code" > FILE*< / code > or a file
descriptor, and provides the < code class = "code" > fd()< / code > function.
< / p > < p >
For another example of this, refer to
2014-09-21 17:33:12 +00:00
< a class = "link" href = "http://www.josuttis.com/cppcode/fdstream.html" target = "_top" > fdstream example< / a >
2012-03-27 23:13:14 +00:00
by Nicolai Josuttis.
2014-09-21 17:33:12 +00:00
< / p > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.third.support_cxx98" > < / a >
2012-03-27 23:13:14 +00:00
Support for C++98 dialect.
< / h4 > < / div > < / div > < / div > < p > Check for complete library coverage of the C++1998/2003 standard.
< / p > < pre class = "programlisting" >
# AC_HEADER_STDCXX_98
AC_DEFUN([AC_HEADER_STDCXX_98], [
AC_CACHE_CHECK(for ISO C++ 98 include files,
ac_cv_cxx_stdcxx_98,
[AC_LANG_SAVE
AC_LANG_CPLUSPLUS
AC_TRY_COMPILE([
#include < cassert>
#include < cctype>
#include < cerrno>
#include < cfloat>
#include < ciso646>
#include < climits>
#include < clocale>
#include < cmath>
#include < csetjmp>
#include < csignal>
#include < cstdarg>
#include < cstddef>
#include < cstdio>
#include < cstdlib>
#include < cstring>
#include < ctime>
#include < algorithm>
#include < bitset>
#include < complex>
#include < deque>
#include < exception>
#include < fstream>
#include < functional>
#include < iomanip>
#include < ios>
#include < iosfwd>
#include < iostream>
#include < istream>
#include < iterator>
#include < limits>
#include < list>
#include < locale>
#include < map>
#include < memory>
#include < new>
#include < numeric>
#include < ostream>
#include < queue>
#include < set>
#include < sstream>
#include < stack>
#include < stdexcept>
#include < streambuf>
#include < string>
#include < typeinfo>
#include < utility>
#include < valarray>
#include < vector>
],,
ac_cv_cxx_stdcxx_98=yes, ac_cv_cxx_stdcxx_98=no)
AC_LANG_RESTORE
])
if test "$ac_cv_cxx_stdcxx_98" = yes; then
AC_DEFINE(STDCXX_98_HEADERS,,[Define if ISO C++ 1998 header files are present. ])
fi
])
2014-09-21 17:33:12 +00:00
< / pre > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.third.support_tr1" > < / a >
2012-03-27 23:13:14 +00:00
Support for C++TR1 dialect.
< / h4 > < / div > < / div > < / div > < p > Check for library coverage of the TR1 standard.
< / p > < pre class = "programlisting" >
# AC_HEADER_STDCXX_TR1
AC_DEFUN([AC_HEADER_STDCXX_TR1], [
AC_CACHE_CHECK(for ISO C++ TR1 include files,
ac_cv_cxx_stdcxx_tr1,
[AC_LANG_SAVE
AC_LANG_CPLUSPLUS
AC_TRY_COMPILE([
#include < tr1/array>
#include < tr1/ccomplex>
#include < tr1/cctype>
#include < tr1/cfenv>
#include < tr1/cfloat>
#include < tr1/cinttypes>
#include < tr1/climits>
#include < tr1/cmath>
#include < tr1/complex>
#include < tr1/cstdarg>
#include < tr1/cstdbool>
#include < tr1/cstdint>
#include < tr1/cstdio>
#include < tr1/cstdlib>
#include < tr1/ctgmath>
#include < tr1/ctime>
#include < tr1/cwchar>
#include < tr1/cwctype>
#include < tr1/functional>
#include < tr1/memory>
#include < tr1/random>
#include < tr1/regex>
#include < tr1/tuple>
#include < tr1/type_traits>
#include < tr1/unordered_set>
#include < tr1/unordered_map>
#include < tr1/utility>
],,
ac_cv_cxx_stdcxx_tr1=yes, ac_cv_cxx_stdcxx_tr1=no)
AC_LANG_RESTORE
])
if test "$ac_cv_cxx_stdcxx_tr1" = yes; then
AC_DEFINE(STDCXX_TR1_HEADERS,,[Define if ISO C++ TR1 header files are present. ])
fi
])
< / pre > < p > An alternative is to check just for specific TR1 includes, such as < unordered_map> and < unordered_set> .
< / p > < pre class = "programlisting" >
# AC_HEADER_TR1_UNORDERED_MAP
AC_DEFUN([AC_HEADER_TR1_UNORDERED_MAP], [
AC_CACHE_CHECK(for tr1/unordered_map,
ac_cv_cxx_tr1_unordered_map,
[AC_LANG_SAVE
AC_LANG_CPLUSPLUS
AC_TRY_COMPILE([#include < tr1/unordered_map> ], [using std::tr1::unordered_map;],
ac_cv_cxx_tr1_unordered_map=yes, ac_cv_cxx_tr1_unordered_map=no)
AC_LANG_RESTORE
])
if test "$ac_cv_cxx_tr1_unordered_map" = yes; then
AC_DEFINE(HAVE_TR1_UNORDERED_MAP,,[Define if tr1/unordered_map is present. ])
fi
])
< / pre > < pre class = "programlisting" >
# AC_HEADER_TR1_UNORDERED_SET
AC_DEFUN([AC_HEADER_TR1_UNORDERED_SET], [
AC_CACHE_CHECK(for tr1/unordered_set,
ac_cv_cxx_tr1_unordered_set,
[AC_LANG_SAVE
AC_LANG_CPLUSPLUS
AC_TRY_COMPILE([#include < tr1/unordered_set> ], [using std::tr1::unordered_set;],
ac_cv_cxx_tr1_unordered_set=yes, ac_cv_cxx_tr1_unordered_set=no)
AC_LANG_RESTORE
])
if test "$ac_cv_cxx_tr1_unordered_set" = yes; then
AC_DEFINE(HAVE_TR1_UNORDERED_SET,,[Define if tr1/unordered_set is present. ])
fi
])
2014-09-21 17:33:12 +00:00
< / pre > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.third.support_cxx11" > < / a >
2012-03-27 23:13:14 +00:00
Support for C++11 dialect.
< / h4 > < / div > < / div > < / div > < p > Check for baseline language coverage in the compiler for the C++11 standard.
< / p > < pre class = "programlisting" >
# AC_COMPILE_STDCXX_11
AC_DEFUN([AC_COMPILE_STDCXX_11], [
AC_CACHE_CHECK(if g++ supports C++11 features without additional flags,
ac_cv_cxx_compile_cxx11_native,
[AC_LANG_SAVE
AC_LANG_CPLUSPLUS
AC_TRY_COMPILE([
template < typename T>
struct check final
{
static constexpr T value{ __cplusplus };
};
typedef check< check< bool> > right_angle_brackets;
int a;
decltype(a) b;
typedef check< int> check_type;
check_type c{};
check_type& & cr = static_cast< check_type& & > (c);
static_assert(check_type::value == 201103L, "C++11 compiler");],,
ac_cv_cxx_compile_cxx11_native=yes, ac_cv_cxx_compile_cxx11_native=no)
AC_LANG_RESTORE
])
AC_CACHE_CHECK(if g++ supports C++11 features with -std=c++11,
ac_cv_cxx_compile_cxx11_cxx,
[AC_LANG_SAVE
AC_LANG_CPLUSPLUS
ac_save_CXXFLAGS="$CXXFLAGS"
CXXFLAGS="$CXXFLAGS -std=c++11"
AC_TRY_COMPILE([
template < typename T>
struct check final
{
static constexpr T value{ __cplusplus };
};
typedef check< check< bool> > right_angle_brackets;
int a;
decltype(a) b;
typedef check< int> check_type;
check_type c{};
check_type& & cr = static_cast< check_type& & > (c);
static_assert(check_type::value == 201103L, "C++11 compiler");],,
ac_cv_cxx_compile_cxx11_cxx=yes, ac_cv_cxx_compile_cxx11_cxx=no)
CXXFLAGS="$ac_save_CXXFLAGS"
AC_LANG_RESTORE
])
AC_CACHE_CHECK(if g++ supports C++11 features with -std=gnu++11,
ac_cv_cxx_compile_cxx11_gxx,
[AC_LANG_SAVE
AC_LANG_CPLUSPLUS
ac_save_CXXFLAGS="$CXXFLAGS"
CXXFLAGS="$CXXFLAGS -std=gnu++11"
AC_TRY_COMPILE([
template < typename T>
struct check final
{
static constexpr T value{ __cplusplus };
};
typedef check< check< bool> > right_angle_brackets;
int a;
decltype(a) b;
typedef check< int> check_type;
check_type c{};
check_type& & cr = static_cast< check_type& & > (c);
static_assert(check_type::value == 201103L, "C++11 compiler");],,
ac_cv_cxx_compile_cxx11_gxx=yes, ac_cv_cxx_compile_cxx11_gxx=no)
CXXFLAGS="$ac_save_CXXFLAGS"
AC_LANG_RESTORE
])
if test "$ac_cv_cxx_compile_cxx11_native" = yes ||
test "$ac_cv_cxx_compile_cxx11_cxx" = yes ||
test "$ac_cv_cxx_compile_cxx11_gxx" = yes; then
AC_DEFINE(HAVE_STDCXX_11,,[Define if g++ supports C++11 features. ])
fi
])
< / pre > < p > Check for library coverage of the C++2011 standard.
(Some library headers are commented out in this check, they are
not currently provided by libstdc++).
< / p > < pre class = "programlisting" >
# AC_HEADER_STDCXX_11
AC_DEFUN([AC_HEADER_STDCXX_11], [
AC_CACHE_CHECK(for ISO C++11 include files,
ac_cv_cxx_stdcxx_11,
[AC_REQUIRE([AC_COMPILE_STDCXX_11])
AC_LANG_SAVE
AC_LANG_CPLUSPLUS
ac_save_CXXFLAGS="$CXXFLAGS"
CXXFLAGS="$CXXFLAGS -std=gnu++11"
AC_TRY_COMPILE([
#include < cassert>
#include < ccomplex>
#include < cctype>
#include < cerrno>
#include < cfenv>
#include < cfloat>
#include < cinttypes>
#include < ciso646>
#include < climits>
#include < clocale>
#include < cmath>
#include < csetjmp>
#include < csignal>
#include < cstdalign>
#include < cstdarg>
#include < cstdbool>
#include < cstddef>
#include < cstdint>
#include < cstdio>
#include < cstdlib>
#include < cstring>
#include < ctgmath>
#include < ctime>
// #include < cuchar>
#include < cwchar>
#include < cwctype>
#include < algorithm>
#include < array>
#include < atomic>
#include < bitset>
#include < chrono>
// #include < codecvt>
#include < complex>
#include < condition_variable>
#include < deque>
#include < exception>
#include < forward_list>
#include < fstream>
#include < functional>
#include < future>
#include < initializer_list>
#include < iomanip>
#include < ios>
#include < iosfwd>
#include < iostream>
#include < istream>
#include < iterator>
#include < limits>
#include < list>
#include < locale>
#include < map>
#include < memory>
#include < mutex>
#include < new>
#include < numeric>
#include < ostream>
#include < queue>
#include < random>
#include < ratio>
#include < regex>
#include < scoped_allocator>
#include < set>
#include < sstream>
#include < stack>
#include < stdexcept>
#include < streambuf>
#include < string>
#include < system_error>
#include < thread>
#include < tuple>
#include < typeindex>
#include < typeinfo>
#include < type_traits>
#include < unordered_map>
#include < unordered_set>
#include < utility>
#include < valarray>
#include < vector>
],,
ac_cv_cxx_stdcxx_11=yes, ac_cv_cxx_stdcxx_11=no)
AC_LANG_RESTORE
CXXFLAGS="$ac_save_CXXFLAGS"
])
if test "$ac_cv_cxx_stdcxx_11" = yes; then
AC_DEFINE(STDCXX_11_HEADERS,,[Define if ISO C++11 header files are present. ])
fi
])
< / pre > < p > As is the case for TR1 support, these autoconf macros can be made for a finer-grained, per-header-file check. For
< code class = "filename" > < unordered_map> < / code >
< / p > < pre class = "programlisting" >
# AC_HEADER_UNORDERED_MAP
AC_DEFUN([AC_HEADER_UNORDERED_MAP], [
AC_CACHE_CHECK(for unordered_map,
ac_cv_cxx_unordered_map,
[AC_REQUIRE([AC_COMPILE_STDCXX_11])
AC_LANG_SAVE
AC_LANG_CPLUSPLUS
ac_save_CXXFLAGS="$CXXFLAGS"
CXXFLAGS="$CXXFLAGS -std=gnu++11"
AC_TRY_COMPILE([#include < unordered_map> ], [using std::unordered_map;],
ac_cv_cxx_unordered_map=yes, ac_cv_cxx_unordered_map=no)
CXXFLAGS="$ac_save_CXXFLAGS"
AC_LANG_RESTORE
])
if test "$ac_cv_cxx_unordered_map" = yes; then
AC_DEFINE(HAVE_UNORDERED_MAP,,[Define if unordered_map is present. ])
fi
])
< / pre > < pre class = "programlisting" >
# AC_HEADER_UNORDERED_SET
AC_DEFUN([AC_HEADER_UNORDERED_SET], [
AC_CACHE_CHECK(for unordered_set,
ac_cv_cxx_unordered_set,
[AC_REQUIRE([AC_COMPILE_STDCXX_11])
AC_LANG_SAVE
AC_LANG_CPLUSPLUS
ac_save_CXXFLAGS="$CXXFLAGS"
CXXFLAGS="$CXXFLAGS -std=gnu++11"
AC_TRY_COMPILE([#include < unordered_set> ], [using std::unordered_set;],
ac_cv_cxx_unordered_set=yes, ac_cv_cxx_unordered_set=no)
CXXFLAGS="$ac_save_CXXFLAGS"
AC_LANG_RESTORE
])
if test "$ac_cv_cxx_unordered_set" = yes; then
AC_DEFINE(HAVE_UNORDERED_SET,,[Define if unordered_set is present. ])
fi
])
< / pre > < p >
Some C++11 features first appeared in GCC 4.3 and could be enabled by
< code class = "option" > -std=c++0x< / code > and < code class = "option" > -std=gnu++0x< / code > for GCC
releases which pre-date the 2011 standard. Those C++11 features and GCC's
support for them were still changing until the 2011 standard was finished,
but the autoconf checks above could be extended to test for incomplete
C++11 support with < code class = "option" > -std=c++0x< / code > and
< code class = "option" > -std=gnu++0x< / code > .
2014-09-21 17:33:12 +00:00
< / p > < / div > < div class = "section" > < div class = "titlepage" > < div > < div > < h4 class = "title" > < a id = "backwards.third.iterator_type" > < / a >
2012-03-27 23:13:14 +00:00
< code class = "code" > Container::iterator_type< / code > is not necessarily < code class = "code" > Container::value_type*< / code >
< / h4 > < / div > < / div > < / div > < p >
This is a change in behavior from older versions. Now, most
< span class = "type" > iterator_type< / span > typedefs in container classes are POD
objects, not < span class = "type" > value_type< / span > pointers.
2015-08-28 15:33:40 +00:00
< / p > < / div > < / div > < div class = "bibliography" > < div class = "titlepage" > < div > < div > < h3 class = "title" > < a id = "backwards.biblio" > < / a > Bibliography< / h3 > < / div > < / div > < / div > < div class = "biblioentry" > < a id = "id-1.3.6.3.8.5.2" > < / a > < p > < span class = "title" > < em >
2014-09-21 17:33:12 +00:00
< a class = "link" href = "http://www.kegel.com/gcc/gcc4.html" target = "_top" >
2012-03-27 23:13:14 +00:00
Migrating to GCC 4.1
< / a >
2015-08-28 15:33:40 +00:00
< / em > . < / span > < span class = "author" > < span class = "firstname" > Dan< / span > < span class = "surname" > Kegel< / span > . < / span > < / p > < / div > < div class = "biblioentry" > < a id = "id-1.3.6.3.8.5.3" > < / a > < p > < span class = "title" > < em >
2017-10-07 00:16:47 +00:00
< a class = "link" href = "https://lists.debian.org/debian-gcc/2006/03/msg00405.html" target = "_top" >
2012-03-27 23:13:14 +00:00
Building the Whole Debian Archive with GCC 4.1: A Summary
< / a >
2015-08-28 15:33:40 +00:00
< / em > . < / span > < span class = "author" > < span class = "firstname" > Martin< / span > < span class = "surname" > Michlmayr< / span > . < / span > < / p > < / div > < div class = "biblioentry" > < a id = "id-1.3.6.3.8.5.4" > < / a > < p > < span class = "title" > < em >
2014-09-21 17:33:12 +00:00
< a class = "link" href = "http://annwm.lbl.gov/~leggett/Atlas/gcc-3.2.html" target = "_top" >
2012-03-27 23:13:14 +00:00
Migration guide for GCC-3.2
< / a >
2014-09-21 17:33:12 +00:00
< / em > . < / span > < / p > < / div > < / div > < / div > < div class = "navfooter" > < hr / > < table width = "100%" summary = "Navigation footer" > < tr > < td width = "40%" align = "left" > < a accesskey = "p" href = "api.html" > Prev< / a > < / td > < td width = "20%" align = "center" > < a accesskey = "u" href = "appendix_porting.html" > Up< / a > < / td > < td width = "40%" align = "right" > < a accesskey = "n" href = "appendix_free.html" > Next< / a > < / td > < / tr > < tr > < td width = "40%" align = "left" valign = "top" > API Evolution and Deprecation History < / td > < td width = "20%" align = "center" > < a accesskey = "h" href = "../index.html" > Home< / a > < / td > < td width = "40%" align = "right" valign = "top" >
2012-03-27 23:13:14 +00:00
Free Software Needs Free Documentation
2015-08-28 15:33:40 +00:00
< / td > < / tr > < / table > < / div > < / body > < / html >
