GNU `recode' converts files between character sets and usages. When exact transliterations are not possible, it may get rid of the offending characters or fall back on approximations. This program recognizes or produces nearly 150 different character sets and is able to transliterate files between almost any pair. Most RFC 1345 character sets are supported. GNU `recode' has been written by Franc,ois Pinard. It is an evolving product, and specifications might change in future releases. Note that `-f' is not completely implemented. If you modify or add an `.l' source file, you need `mawk', GNU `awk' or `nawk', and Flex 2.4.6 or better for remaking `merged.c'. Finally, see further down, below, for various installation hints on special systems. See file `BACKLOG' for a summary of pending mail and articles. See file `COPYING' for copying conditions. See file `INSTALL' for compilation and installation instructions. See file `NEWS' for a list of major changes in the current release. See file `README.DOS' for notes about the MSDOS port. See file `THANKS' for a list of contributors. Use file `File-Latin1' to play with recode, if you feel like it. Your feedback will help us to make a better and more portable product. Mail suggestions and bug reports (including documentation errors) for this program to `bug-gnu-utils@prep.ai.mit.edu'. If you develop new charsets, new methods, or anything along `recode', let me know and share your findings, at `pinard@iro.umontreal.ca'. .-------. | Hints | `-------' Here are a few hints which might help installing `recode' on some systems. Most may be applied by temporary presetting environment variables while calling `./configure'. File `INSTALL' explains this. * You may override DEFAULT_CHARSET while configuring. For example, on an AtariST using `bash', one could do: DEFAULT_CHARSET=AtariST ./configure * Some C compilers, like Apollo's, have a real hard time compiling merged.c. If this is your case, first avoid compiler optimization. From within the Bourne shell, you may use: CPPFLAGS=-DUSE_FPUTC CFLAGS= ./configure * For 80286 based systems, it has been reported that some compilers generate wrong code while optimizing for *small* models. So, from within the Bourne shell, do: CFLAGS=-Ml LDFLAGS=-Ml ./configure to force large memory model. For 80286 Xenix compiler, the last time it was tried a while ago, one ought to use: CFLAGS='-Ml -F2000' LDFLAGS=-Ml ./configure * Some smallish systems have poor `popen' support or trash heavily when processes do fork. In this case, just before doing `make', edit `config.h' and ensure HAVE_POPEN and HAVE_PIPE are *not* defined.
GNU recode has been ported to smallish MSDOS systems, using Borland's
C compiler huge memory model. At site ftp.iro.umontreal.ca in
directory pub/gnuish, fetch reco34ax.zoo, or such, for an all-ready
executable, and reco34as.zoo for sources and documentation.
Here is my own installation recipe. On the UNIX side, I got Larry
Wall's Perl (I'm running 4.019) and Rahul Dhesi's Zoo 2.1. On the
MSDOS side, I got Rahul Dhesi's Zoo 2.1, Thorsten Ohl's port of GNU
make and Borland's Turbo C++ 1.0 system.
1. On UNIX, configure recode right in the distribution directory.
2. Execute `make distdos'. This should create reco34as.zoo, or such.
3. Move the zoo archive to the MSDOS side.
4. On MSDOS, create a directory, cd to it and unpack the zoo file.
5. Execute `make'. This should produce recode.exe.
6. Make reco34ax.zoo out of readme{,.dos}, copying, recode.{inf,exe}.
Even if GNU recode is free software, it is copyrighted. See COPYING.
In particular, make the full source available where the binaries are.
Francois Pinard <pinard@iro.umontreal.ca>
These are the contents of the former NiCE NeXT User Group NeXTSTEP/OpenStep software archive, currently hosted by Netfuture.ch.