Installation
************

   This document is divided into four parts: the first part (For the
impatient) provides a quick overview of configuration and compilation
instructions.  It also lists those options to the configure shell script
which are unique to gnuplot.  Part two describes the environment variables
used by gnuplot.  Part three explains in detail how --with-PACKAGE=PATH
works, and how this option interacts with --libdir and --includedir.
Part four addresses some platform specific problems and testing issues.
There are also some hints how to build gnuplot with pre-ANSI C compilers.
Part five finally describes a few graphical front ends for gnuplot.

 The description of preprocessor options previously found here has been
moved to a new file called 0PORTING. Generic installation instructions
for `configure', such as distributed with FSF/GNU packages, have been
moved to the file INSTALL.gnu.

For the impatient
=================

   First, tune term.h to choose which terminal drivers you wish to enable.
If you want to support gif output, you need to download, compile and
install the gd library : see term/gif.trm for details.

   If you want to support png output (free gif alternative), you need
png and z libraries : see term/png.trm for details.  Note that the
png library will probably not compile without an ANSI/ISO C compiler.

   A complete overview of configure options is below in the Basic
Installation section. On platforms which do not support GNU
autoconf/configure, ie. most non-Unix platforms, look for a makefile
that looks suitable, (makefile.emx for emx on dos, makefile.nt for
MS VC++ 4.x on NT and probably win95, makefile.win for borland c on
win95/nt, makefile.wc for DOS Watcom C etc.)

File location defaults
----------------------

PREFIX                          /usr/local
gnuplot                         PREFIX/bin
gnuplot_x11                     PREFIX/bin
gnuplot.1                       PREFIX/man/man1
gnuplot.gih                     PREFIX/share

   The only files where the runtime location is defined at compile time are
gnuplot.gih and show.c. This is important if gnuplot is not installed by running
'make install'.

   The default path for the help library, gnuplot.gih, can be controlled in
several different ways:

 o with configure's --prefix= option, eg.
   ./configure --prefix=/gnuplot
   gnuplot will look for the online help file, gnuplot.gih, in /gnuplot/share.
   Attention: This option affects the location of *all* installed files!

 o with configure's --datadir= option, eg.
   ./configure --datadir='/gnuplot/docs'
   gnuplot will look for the online help file, gnuplot.gih, in /gnuplot/docs.

 o at make time, eg.
   make HELPFILE='/gnuplot/docs/gnuplot.gih'
   gnuplot will look for the online help file /gnuplot/docs/gnupot.gih.

 o at execution time by specifying the helpfile with the environment
   variable GNUHELP (see "Environment Variables" below).

   The default location of the executables can be controlled in different
ways:

 o with configure's --prefix= option, eg.
   ./configure --prefix=/gnuplot
   gnuplot will be installed as /gnuplot/bin/gnuplot.
   Attention: This affects the location of *all* installed files!

 o with configure's --bindir= option, eg.
   ./configure --bindir='/gnuplot/bin'
   gnuplot will be installed as /gnuplot/bin/gnuplot.

 o at make time, eg.
   make BINDIR='/gnuplot/bin'
   gnuplot will be installed as /gnuplot/bin/gnuplot.

   If the executables are not installed in the default location, files
written by the `save' command cannot be used as scripts on Un*x platforms.
There is no way to specify the location of the executables during
runtime. Other platforms are not affected by this. 

   The gnuplot demo files are not installed by default. If desired, they
should be copied manually to a location of choice.

Unix, configure
---------------

   On Unix, use
$ ./configure
$ make
[ Optionally run demos--see "How to test gnuplot" below. ]
$ make install

   If gcc is installed, it is used by default. A different compiler can be
used through the CC environment variable:

(Bourne shell)
$ CC=c89 ./configure

(C shell)
$ setenv CC c89
$ ./configure
 
   Any environment settings for CFLAGS are included into the Makefile,
so please make sure that these are really needed.
 
   There are several options available for configure that you may want
to change.  A complete list of options is available through

$ ./configure --help

  --prefix=PREFIX         Install architecture-independent files in PREFIX
                          [/usr/local]. The gnuplot (and gnuplot_x11) binary
                          is installed in PREFIX/bin.
  --bindir=DIR            user executables in DIR [PREFIX/bin]
  --datadir=DIR           Read-only architecture-independent data in DIR
                          [PREFIX/share]. The gnuplot help file is installed
                          in this directory.
  --libdir=DIR            object code libraries in DIR [EPREFIX/lib]
  --includedir=DIR        C header files in DIR [PREFIX/include]
  --mandir=DIR            Man documentation in DIR [PREFIX/man]
  --with-readline (*)     use the builtin minimal readline function
  --without-readline      do not use the builtin readline function
  --with-readline=gnu     use the GNU readline library
  --with-readline=PATH            Specify the location of GNU readline
  --with-plot             use the Unix plot library
  --with-plot=PATH                Specify the location of GNU libplot
  --with-gd (*)           Enable gif terminal with Tom Boutell's gd library
                          (requires GD library)
  --with-gd=PATH                  Specify the location of libgd
  --with-png (*)          Enable png terminal
                          (requires libpng and libz)
  --with-png=PATH                 Specify the location of libpng
  --with-cwdrc            check current directory for .gnuplot file,
                          normally disabled for security reasons
  --with-lasergnu         install lasergnu printer script
  --with-linux-vga        use the Linux SVGA console driver
                          (requires /usr/lib/libvga)
  --enable-apollo         Apollo Graphics Primitive Resource
  --enable-gpr            Apollo Graphics Primitive Resource (fixed-size window)
  --enable-cgi            enable CGI terminal (SCO only)
  --enable-iris           enable IRIS terminal (IRIS4D only)
  --enable-sun            enable sunview terminal (Sun only)
  --enable-unixpc         enable unixpc terminal (ATT 3b1 or ATT 7300)
  --enable-system-time    use the system time routines in time.c
  --with-x (*)            Use the X Window System

 Options marked with (*) are on by default, ie. these features or packages
will be used if configure can detect them, even if the corresponding option
is not specified.  Please note that the `--with-PACKAGE' options can have
additional arguments:

 o `--with-PACKAGE' is equivalent to `--with-PACKAGE=yes'
 o `--with-PACKAGE=no' will disable checking for PACKAGE. It has the same
   effect as `--without-PACKAGE'.
 o `--with-PACKAGE=PATH' will check for PACKAGE in PATH

Unix, no configure
------------------

  The older, no-longer-preferred, way is to copy makefile.unx to Makefile
      cp makefile.unx Makefile
  Look through the Makefile to see if you need to make any changes.
  See especially the HELPDEST and TERMFLAGS variables.  Edit if needed.
  Alternatively, all these variables may be set as command line arguments to
  'make'. For example:

        make <MACHINE> HELPDEST='/usr/um/misc/lib' \
                DEST='/usr/um/misc/bin' READLINE=

  Type
      make
  For further instructions.
  If that works, try
      make install
  For further instructions.

VMS
---

   On VMS, you can use MMS, MMK, or another make utility, or you can
use BUILDVMS.COM.  The supplied files work with Alpha/VMS V6.2 and
DECC V5.7; you may get warnings or more serious errors depending on
the versions of the C compiler, the C run-time libraries, and VMS on
your system.

  To compile using MMK:
     MMK/DESCRIPTION=MAKEFILE.VMS
  To compile using MMS on VAX:
     MMS/DESCRIPTION=MAKEFILE.VMS
  To compile using MMS on Alpha:
     MMS/DESCRIPTION=MAKEFILE.VMS/MACRO=__ALPHA__=1

  Alternatively, you can use MAKE_VMS.COM.  This command file will run
MMK or MMS to build gnuplot.  If you don't have either one installed, it
will use the command file.  The first parameter on the command line is
the compiler you want to use (default DECC).

  It may sometimes be useful to add  `/IGNORE=WARNING' to the `MMS' call.

  To compile using GNUC or VAXC:
     add /MACRO=GNUC or /MACRO=VAXC to the above command
  To compile with another make utility:
     check the documentation of your utility to specify the description
     file and any necessary macros (__ALPHA__, GNUC, or VAXC)
  Or if you don't have a suitable make:
     @BUILDVMS
  To tell gnuplot where to find the help library:
     $ define gnuplot$help disk:[directory]gnuplot.hlb
  Alternatively, put the help in the main system help library.

AmigaOS
-------

Using Aztec C 5.2a
      make -f makefile.ami
Using SAS/C 6.1 or later versions
      smake -f makefile.amg
      You can also copy makefile.amg to smakefile and just type `smake'.

      At the beginning of makefile.amg you will find a configuration
      section where you have to adjust all settings which control the build
      process. Most important is probably to select the terminal devices
      which should be supported. You can create a version offering
      png graphics as well as gif support. Use the appropriate settings
      for CPU and MATH to compile for a specific machine type.
      Be sure to enable only those devices for which you have the 
      necessary software already installed. Also ensure that the
      libraries and the sources are all compiled using the same
      settings for CPU and MATH.
Using gcc: see Unix

Atari/TOS
---------

Using gcc 2.x as unix cross- or native compiler
	make -f makefile.st
(Edit top of makefile.st for name of crosscompiler or choose native setting)
Using PureC
	use gnuplot.prj
Using TurboC
	use gnuplot.prj
(Edit gnuplot.prj according to notes at the beginning)

MS-Windows
----------

Using Microsoft C 7.0 and compiling for MS-Windows
      copy makefile.msw makefile
      nmake
  Put wgnuplot.exe, wgnuplot.dll, wgnuplot.hlp and wgnuplot.mnu
  in the windows directory.

Using Borland C++ 3.1 and compiling for MS-Windows
      copy makefile.win makefile
  Edit makefile to change TC.
      make
  Put wgnuplot.exe, wgnuplot.dll, wgnuplot.hlp and wgnuplot.mnu
  in the windows directory.

Using Visual C++ and compiling for Windows NT/95/98
      nmake -f makefile.nt
 or
      copy makefile.nt makefile
      nmake

Using the MinGW32 port of egcs: you need the Micrsoft SDK for the
moment.
      copy makefile.mgw makefile
  Look through the Makefile to see if you need to make any changes.
      make 
      make install

Using the Cygwin port of gcc, which includes MinGW32: you need the
Micrsoft SDK for the moment.
      copy makefile.cyg makefile
  Look through the Makefile to see if you need to make any changes.
      make
  Put wgnuplot.exe, wgnuplot.dll, wgnuplot.hlp and wgnuplot.mnu
  in the windows directory.

MSDOS
-----

Using DJGPP
  Basically the same as Unix+configure. Andris Pavenis <pavenis@lanet.lv>
  has contributed a shell script which makes the whole process even easier:
  just run djconfig.sh, and then make.
  For novices, it is recommended to use makefile.dj2 instead:
      copy makefile.dj2 makefile
      make

Using Microsoft C 7.0.
      copy makefile.msc makefile
      nmake 

Using Borland C++ 3.0
      copy makefile.tc makefile
  Edit makefile to change TC, BIN, BGI, BGIOBJ. You may also want to turn
  off overlays (See manual for more on overlays).
      make

The file gnuplot.gih is needed for help on the PC.
If the file gnuplot.gih is not in the default directory, then use:
    set GNUHELP={full path name of gnuplot.gih}

OS/2
----

To compile under OS/2 (2.x and above) you need the development
suite EMX 0.9 (including gcc). You should also have
GNU Make and IBM's IPFC  (Information Presentation Facility Compiler,
available from the Developer's Toolkit; nowadays it's accessible
through an IBM website for free!).

At the beginning of Makefile.os2 you will find a configuration
section where you have to adjust all settings which control the build
process. Most important is probably to select the terminal devices
which should be supported. You can create a version offering
PM graphics as well as X11 support (to use with XFree86).
Support for these two terminals is provided by additional
executables.
Be sure to enable only those devices for which you have the 
necessary software already installed. Also ensure that the
libraries and the sources are all compiled (not) using
the '-Zmt' flags.

Executing 
  make -f makefile.os2
should create a default build while
  make -f makefile.os2 help
will show you all pre-defined targets.

See other sections of the manuals for more information about
installing/using gnuplot on OS/2.

Environment Variables
=====================

See 'help environment'.

If the environment variable GNUTERM is found, it is used as the terminal
type. Otherwise, in some cases the variable TERM will be used, or the
hardware may be automatically detected.

The PC version looks for the environment variable GNUPLOT to contain
the name of the directory from which to load the initialization file
GNUPLOT.INI.  See the help on 'start_up' for more information.

HOME is examined as a directory where a .gnuplot startup file might be
found. See help on "start-up".

If defined, the environment variable GNUHELP is used for the name
of the .gih help file, otherwise HELPFILE (defined in makefile or
command.c) is used.

The VMS version looks for the logical name GNUPLOT$HELP to locate
the help library.

The CGI drivers need the CGIPATH environment variable to set the path
to the CGI agents, and the CGIDISP and/or CGIPRNT environment variables
to set the output devices.

 If creating dynamically linked executables for the X11 Window System, it
may be necessary to add flags to the LIBS variable in Makefile
to make sure gnuplot finds all required libraries at runtime.  Systems like
SunOS and Solaris use -R to specify the runtime library search path, whereas
OSF/Dec Unix, Irix and Linux use -rpath. If this is not possible, the
LD_LIBRARY_PATH environment variable should be set.

About --with-PACKAGE[=PATH]
***************************

 The following description applies to gnuplot only.

 Every `--with-PACKAGE' option sets a `with_package' variable in
configure.  Depending on how `--with-PACKAGE' was invoked, there are
only three different possible values for the `with_package' variable:

 Option                   $with_package
----------------------------------------
(not specified)           yes
--with-package            yes
--with-package=yes        yes
--with-package=no         no
--with-package=PATH       PATH
--without-package         no

 This means that configure will always (by default) try to locate PACKAGE
unless `--with-package=no' or `--without-package' was specified. Examples
for with-PACKAGE options used by gnuplot's configure are `--with-gd',
`--with-png'. The corresponding package variables are `with_gd',
and `with_png'.

 For gnuplot, configure uses the macros `gp_SEARCH_LIBDIRS' and
`gp_SEARCH_HEADERDIRS' to search for libraries and includes, resp.  These
macros are different from the standard GNU autoconf macros for this purpose,
`AC_CHECK_LIB' and `AC_CHECK_HEADER/S' (see autoconf manual). The
gp_XXX macros are closely tied to gnuplot, and cannot be used for other
packages without modification.

 The difference between these package specific macros and standard
autoconf macros is that they do search more than one (the default)
directory to determine the location of the specified file.

 Libraries will be searched in at least three, and up to five different
directories, depending on the PATH specified with `--with-PACKAGE=PATH':

 o the default linker path (this is really more than one directory)
 o /usr/local/lib, which is a fairly common place for 3rd party libs
 o the PATH directory, trailing `/lib/libPACKAGE.a' stripped off
 o the PATH directory, trailing `/lib/libPACKAGE.a' stripped off and
   `/lib' added
 o the PATH directory, trailing `/libPACKAGE.a' stripped off.

 Using the default linker path as the first search directory has one
big, but not so obvious advantage: if all libraries are in the same
directory, the path to the first library specified with --with-PACKAGE
is added to the linker path with a `-L' option.  The library searched
for with the next --with-PACKAGE option will now be found faster, because
only one additional directory is checked.  This is also true for header files.

 Example: the user has unpacked the gd library distribution into
/home/user/source/gd:

$ configure --with-gd=/home/user/source/gd

will search the following directories for libgd.a:

 o the default linker path
 o /usr/local/lib
 o /home/user/source/gd
 o /home/user/source/gd/lib
 o /home/user/source/gd

 Include files will be searched in at least two, and up to three different
directories, depending on the PATH specified with `--with-PACKAGE=PATH':

 o the default include path (can be more than one diretcory)
 o /usr/local/include, a fairly common place for 3rd party headers
 o the PATH directory, trailing `/lib/libPACKAGE.a' stripped off
 o the PATH directory, trailing `/lib/libPACKAGE.a' stripped off and
   `/include' added

 Caveat: the `gp_SEARCH_HEADERDIRS' macro should only be used after the
corresponding `gp_SEARCH_LIBDIRS' macro to take full advantage of the
search algorithm.  This is a reasonable assumption nevertheless, because
it allows for structuring configure.in so that the test for a header file
is skipped if the corresponding library is not found.

 Here are some more examples:

 o the gd and png libraries are in /usr/local/gnu/lib, which is
   automatically searched by gcc.  The corresponding header files are
   in /usr/local/gnu/include, which is not in gcc's include search path.
   Solution:

   $ ./configure --with-gd=/usr/local/gnu

 o the gd and png libraries are in /opt/gnu/lib, the header files are
   in /opt/gnu/include.  Solution:
 
   $ ./configure --with-gd=/opt/gnu

 o the gd and png libraries and headers are all in different directories,
   none of which is in the compilers search path:

   $ ./configure --with-gd=/tmp/gd --with-png=/tmp/libpng

[I would like to implement the more generic `--site-includes' and
`--site-libraries' options for additional flexibility, but this is
painfully difficult with the current autoconf.  It also seems that such
a feature would not comply with GNU coding standards.]

Platform problems and testing
*****************************

   This section addresses trouble shooting and testing issues. Userland
questions are answered in the FAQ.

Platform notes
==============

   Generally, if you think that configure has made a mistake in detecting
platform features, there are two ways to switch these off. 

   Example:
configure was for some reason unable to detect the memset() function, but
you are sure it is ok to use on your platform. Now you can either edit
config.cache and change

 ac_cv_func_memset=${ac_cv_func_memset='no'}  to
 ac_cv_func_memset=${ac_cv_func_memset='yes'}  

and rerun configure, or, edit config.h and change

/* #undef HAVE_MEMSET */ to
#define HAVE_MEMSET 1

   Note that changing such defines at compile time, eg. via
'make DEFS=-DHAVE_MEMSET' is wrong, because the DEFS variable in Makefile
may contain other defines (make DEFS='-DHAVE_CONFIG_H -DHAVE_MEMSET'
should work, though).

 - HP-UX 9.x
   It is recommended to use gcc, although the native compiler cc may
  work with warnings.

 - HP-UX 10.x
   It is recommended to use the native compiler cc, as problems
  have been reported when using gcc. In the cases reported, we have
  yet to establish whether the compiler is at fault, or the compiler
  setup/configuration is broken.

 - IRIX 6.x
   If you want to use the png terminal, you must install your own
  versions of libpng and zlib. The versions supplied with the OS
  are too old. Note that you need be very careful to select the
  correct compiler options/ABI's.

 - MS-DOS
   If ports of common Unix utilities (bash, sed etc) are available,
  gnuplot can be built with DJGPP. Install instructions are the same
  as for Unix (with configure).

 - SunOS 4.x
   An ANSI/ISO C compiler should be used to compile gnuplot. It is
  recommended to install gcc. If this is not an option, the system
  compiler cc can be made to work with Wietse Venema's unproto tool.
  See entry below for instructions.

  As of 1998-11-18, gnuplot compiles again with the native compiler cc.

 - System V.2/Ultrix 4.x/M88 SysV.3
   An ANSI/ISO C compiler should be used to compile gnuplot. It is
  recommended to install gcc. If this is not an option, the system
  compiler cc can be made to work with Wietse Venema's unproto tool.

  unproto is available from
  ftp://ftp.win.tue.nl/pub/unix/unproto5.shar.Z
  ftp://ftp.porcupine.org/pub/lang/unproto5.shar.Z

  After installing unproto, configure gnuplot with

  (Bourne shell syntax)

  $ cd gnuplot
  $ CC='cc -B/full/path/to/unproto/dir/ -tp' ./configure <options>

  or (C shell syntax)

  prompt (41) cd gnuplot
  prompt (42) setenv CC 'cc -B/full/path/to/unproto/dir/ -tp'
  prompt (43) ./configure <options>

  The required compiler options shown here are for System V.2.  For other
  platforms, consult the unproto documentation. `/full/path/to/unproto/dir'
  is the full path name of the directory where unproto is installed. It is
  not necessary to "install" unproto, one can just create a subdirectory
  in gnuplot's source directory and use it from there.

How to test gnuplot
===================

   No comprehensive test suite for gnuplot's features has been written
to date. However, the supplied demo files provide a good method of
testing commonly used features. All command line examples below assume
Unix Bourne shell syntax.

   The demo files can be run interactively by eg.

$ cd gnuplot/demo
$ PATH=..:$PATH ../gnuplot simple.dem

and gnuplot prompts the user to "Hit return to continue" to cycle
through all the plots.  (The PATH= statement is only required if you are 
using the X11 plotting device and the gnuplot_x11 binary has not been
installed yet.)

   To run the demos in a specified file without interaction, one
can use

$ PATH=..:$PATH ../gnuplot simple.dem </dev/null

   To run all demos non-interactively, use

$ PATH=..:$PATH ../gnuplot all.dem </dev/null

   To use a different plotting device than the default (usually X11
under Unix), use eg.

$ GNUTERM=dumb ../gnuplot all.dem </dev/null

   To test the capabilities of the terminal you are using, there is
the 'test' command:

$ gnuplot

        G N U P L O T
        Unix version 3.7
        patchlevel 1
        last modified Tue Jan 12 13:48:06 BST 1999

        Copyright(C) 1986 - 1993, 1998
        Thomas Williams, Colin Kelley and many others

        Send comments and requests for help to info-gnuplot@dartmouth.edu
        Send bugs, suggestions and mods to bug-gnuplot@dartmouth.edu

Terminal type set to 'x11'
gnuplot> test

   `test` creates a display of line and point styles and other useful things
appropriate for the terminal you are using.

Front Ends for gnuplot
======================

 o Xgfe by David Ishee <dmishee@jasper.he.net>, available from
   http://von-mises.home.ml.org/xgfe/xgfe.html
   If you download the source, you need Qt library from
   http://www.troll.no/

 o Bruce Ravel <ravel@phys.washington.edu> has written a new version of
   gnuplot-mode for GNU emacs and XEmacs. This version is based on
   the gnuplot.el file by Gershon Elber which is included with this
   distribution. The package is available from
   http://feff.phys.washington.edu/~ravel/gnuplot/