|
 | | | | Building Xerces-C++ with ICU using bundled Perl scripts on Windows | | | | |
| | |
As mentioned earlier, Xerces-C++ may be built in stand-alone mode using
native encoding support and also using ICU where you get support over 180
different encodings. ICU stands for International Components for Unicode and is an
open source distribution from IBM. You can get
ICU libraries from
IBM's developerWorks site
or go to the ICU
download page
directly.
 | Important: Please remember that ICU and
Xerces-C++ must be built with the same compiler,
preferably with the same version. You cannot for example,
build ICU with a threaded version of the xlC compiler and
build Xerces-C++ with a non-threaded one. |
There are two options to build Xerces-C++ with ICU. One is to use the
MSDEV GUI environment, and the other is to invoke the compiler from the
command line.
Using, the GUI environment, requires one to edit the project files.
Here, we will describe only the second option. It involves using the
perl script 'packageBinaries.pl'.
Prerequisites:
- Perl 5.004 or higher
- Cygwin tools or MKS Toolkit
- zip.exe
Extract Xerces-C++ source files from the .zip archive using WinZip, say
in the root directory (an arbitrary drive x:). It should create a directory like
'x:\xerces-c-src1_6_0'.
Extract the ICU files, using WinZip, in root directory of the disk
where you have installed Xerces-C++, sources. After extraction, there
should be a new directory 'x:\icu' which contains all the ICU
source files.
Start a command prompt to get a new shell window. Make sure you have
perl, cygwin tools (uname, rm,
cp, ...), and zip.exe somewhere in the
path. Next setup the environment for MSVC using
'VCVARS32.BAT' or a similar file. Then at the prompt
enter:
| | | | set XERCESCROOT=x:\xerces-c-src1_6_0
set ICUROOT=x:\icu
cd x:\xerces-c-src1_6_0\scripts
perl packageBinaries.pl -s x:\xerces-c-src1_6_0 -o x:\temp\xerces-c1_6_0-win32 -t icu | | | | |
(Match the source directory to your system; the target directory can be
anything you want.)
If everything is setup right and works right, then you should see a
binary drop created in the target directory specified above. This script
will build both ICU and Xerces-C++, copy the files (relevant to the binary
drop) to the target directory.
For a description of options available, you can enter:
|
| | | | I wish to port Xerces to my favourite platform. Do you have any suggestions? | | | | |
| | |
All platform dependent code in Xerces has been
isolated to a couple of files, which should ease the porting
effort. Here are the basic steps that should be followed to
port Xerces.
- The directory
src/util/Platforms contains the
platform sensitive files while src/util/Compilers contains
all development environment sensitive files. Each operating
system has a file of its own and each development environment
has another one of its own too.
As an example, the Win32 platform as a Win32Defs.hpp file
and the Visual C++ environment has a VCPPDefs.hpp file.
These files set up certain define tokens, typedefs,
constants, etc... that will drive the rest of the code to
do the right thing for that platform and development
environment. AIX/CSet have their own AIXDefs.hpp and
CSetDefs.hpp files, and so on. You should create new
versions of these files for your platform and environment
and follow the comments in them to set up your own.
Probably the comments in the Win32 and Visual C++ will be
the best to follow, since that is where the main
development is done.
- Next, edit the file
XercesDefs.hpp, which is where all
of the fundamental stuff comes into the system. You will
see conditional sections in there where the above
per-platform and per-environment headers are brought in.
Add the new ones for your platform under the appropriate
conditionals.
- Now edit
AutoSense.hpp. Here we set canonical Xerces
internal #define tokens which indicate the platform and
compiler. These definitions are based on known platform
and compiler defines.
AutoSense.hpp is included in XercesDefs.hpp and the
canonical platform and compiler settings thus defined will
make the particular platform and compiler headers to be
the included at compilation.
It might be a little tricky to decipher this file so be
careful. If you are using say another compiler on Win32,
probably it will use similar tokens so that the platform
will get picked up already using what is already there.
- Once this is done, you will then need to implement a
version of the platform utilities for your platform.
Each operating system has a file which implements some
methods of the XMLPlatformUtils class, specific to that
operating system. These are not terribly complex, so it
should not be a lot of work. The Win32 verions is called
Win32PlatformUtils.cpp, the AIX version is
AIXPlatformUtils.cpp and so on. Create one for your
platform, with the correct name, and empty out all of the
implementation so that just the empty shells of the
methods are there (with dummy returns where needed to make
the compiler happy.) Once you've done that, you can start
to get it to build without any real implementation.
- Once you have the system building, then start
implementing your own platform utilties methods. Follow
the comments in the Win32 version as to what they do, the
comments will be improved in subsequent versions, but they
should be fairly obvious now. Once you have these
implementations done, you should be able to start
debugging the system using the demo programs.
Other concerns are:
- Does ICU compile on your platform? If not, then you'll need to
create a transcoder implementation that uses your local transcoding
services. The Iconv transcoder should work for you, though perhaps
with some modifications.
- What message loader will you use? To get started, you can use the
"in memory" one, which is very simple and easy. Then, once you get
going, you may want to adapt the message catalog message loader, or
write one of your own that uses local services.
That is the work required in a nutshell!
|
| | | | What should I define XMLCh to be? | | | | |
| | |
XMLCh should be defined to be a type suitable for holding a
utf-16 encoded (16 bit) value, usually an unsigned short.
All XML data is handled within Xerces-C++ as strings of
XMLCh characters. Regardless of the size of the
type chosen, the data stored in variables of type XMLCh
will always be utf-16 encoded values.
Unlike XMLCh, the encoding
of wchar_t is platform dependent. Sometimes it is utf-16
(AIX, Windows), sometimes ucs-4 (Solaris,
Linux), sometimes it is not based on Unicode at all
(HP/UX, AS/400, system 390).
Some earlier releases of xerce-c defined XMLCh to be the
same type as wchar_t on most platforms, with the goal of making
it possible to pass XMLCh strings to library or system functions
that were expecting wchar_t paramters. This approach has
been abandonded because of
-
Portability problems with any code that assumes that
the types of XMLCh and wchar_t are compatible
- Excessive memory usage, especially in the DOM, on
platforms with 32 bit wchar_t.
- utf-16 encoded XMLCh is not always compatible with
ucs-4 encoded wchar_t on Solaris and Linux. The
problem occurs with Unicode characters with values
greater than 64k; in ucs-4 the value is stored as
a single 32 bit quatity. With utf-16, the value
will be stored as a "surrogate pair" of two 16 bit
values. Even with XMLCh equated to wchar_t, xerces will
still create the utf-16 encoded surrogate pairs, which
are illegal in ucs-4 encoded wchar_t strings.
|
|
|
