SED navigation bar go to SED home page go to Dataplot home page go to NIST home page SED Home Page SED Staff SED Projects SED Products and Publications Search Dataplot Pages

Downloading DATAPLOT for Unix Hosts

Contact Information May 2020

Alan Heckert
Statistical Engineering Division
National Institute Of Standards and Technology
Gaithersburg, Md. 20899-8980
(301) 975-2899 or FTS 879-2899
alan.heckert@nist.gov
Jim Filliben
Statistical Engineering Division
National Institute Of Standards and Technology
Gaithersburg, Md. 20899-8980
(301) 975-2855 or FTS 879-2855
james.filliben@nist.gov

Introduction Dataplot is a command driven program. You can optionally run Dataplot using a graphical user interface (GUI) that is implemented using the Tcl/Tk scripting language. Reference will be made below to the "command line" version and the "GUI" version.

The Tcl/Tk software is typically installed on most Linux platforms. Dataplot should work with version 8.4 or higher of Tcl/Tk.

Binary Installations
Package Managers for Installing Software There are a large number of Linux distros and each of these distros typically provide a package manager for binary installations of software. This section provides links to package files.

Some of the advantages of using the package manager files are fast installation (no compilation is required) and any needed dependencies are automatically installed. The disadvantages are that super user permission is required and packages are specific to a given distro and to the version of the distro.

Red Hat Package Manager Red Hat systems (Red Hat Enterprise, CentOS, Fedora) use the Red Hat Package Manager (RPM).

The following RPM files are currently available

After you have downloaded the appropriate "rpm" file, you can install Dataplot by entering the command (this is demonstrated for the Fedora 30 rpm file)

    sudo yum install ./dataplot-20200523-1.fc30.x86_64.rpm

Thanks to Chris Schanzle for putting the RPM package together.

If your system is not supported, you can create an RPM file. Download the source from github (see the next section). Then see the INSTALL file for instructions on creating the RPM file.

Other Distros Currently, we only provide binary installations for the Red Hat Package Manager. We anticipate providing packages for a few additional distros.

We are aware that others have created packages for other distros. If you would like to contribute a package, contact alan.heckert@nist.gov.

Installation from Source
Building from Source Dataplot can also be installed from source. Source builds require that the gfortran/gcc compilers are installed on your local platform.

We have tested the source build on CentOS, Fedora, Ubuntu (a Debian derivative), Manjaro (an arch derivative), cygwin, MSYS2, and MinGW64.

Install any Needed Dependencies
Step 1: Install Any Desired Prerequisites For a source build, the gfortran/gcc compilers need to be installed on your local platform.

Dataplot can optionally utilize several external libraries that are available for Linux systems.

NOTE:

    Package managers can simplify the process of ensuring that needed dependencies are installed. The dependencies below are used by a number of popular open source programs, so they can typically be installed using the package manager for your variant of Linux. Although most or all of these dependencies may already be installed, to build Dataplot from source you also need the developer libraries installed (these are less likely to be installed). We provide the needed commands for the developer libraries for Red Hat and Ubuntu systems. However, the process should be similar for other Linux distros and the documentation for your platform should provide the needed commands. Also, only the X11 dependency should be considered essential. The remaining dependencies provide useful capabilities, but you can skip installing them if you do not anticipate needing them. We provide a brief description of their purpose to help you make the appropriate choices for which dependencies to install.

The following dependencies are supported.

  • The X11 library is used to generate screen graphics. X11 will be available on almost all Linux systems. If you need to install the developer libraries, do the following

      Red Hat: sudo dnf install libx11-devel
      Ubuntu: sudo apt install libx11-dev

  • The GD library is used to generate bit-map graphics (PNG, JPEG, GIF, and TIFF) and to read images in these formats. GD is used by many programs, so it is likely to be installed on your system. If you need to install the developer libraries, do the following

      Red Hat: sudo dnf install gd-devel
      Ubuntu: sudo apt install libgd2-dev

  • The Cairo library provides several graphics output devices. Although the output devices added by Cairo are already supported by Dataplot, an advantage of using the Cairo versions is that the appearance across different devices is more consistent. The Cairo driver in Dataplot should be considered "beta". Cairo is used by a number of popular programs, so it is likely to be installed on your system. If you need to install the developer libraries, do the following

      Red Hat: sudo dnf install cairo-devel
      Ubuntu: sudo apt install libcairo2-dev

  • The readline library provides command history and editing. The readline library is used by many programs, so it is likely to be installed on your system. If you need to install the developer libraries, do the following

      Red Hat: sudo dnf install readline-devel
      Ubuntu: sudo apt install readline-dev

  • The LIBPLOT library (plotutils) provides several graphics output devices that are not otherwise available in Dataplot (Adobe Illustrator, binary CGM, netPBM image format). Although LIBPLOT may already be installed on your local platform, it is less likely than the other libraries to already be installed. To install the plotutils library and the associated developer files, do the following for Red Hat

      Red Hat: sudo dnf install plotutils
      Red Hat: sudo dnf install plotutils-devel
      Ubuntu: sudo apt install plotutils
      Ubuntu: sudo apt install libplot-dev

    The default Makefile does not activate the LIBPLOT feature.

NOTE: For some Red Hat systems, you may need to use "yum" instead of "dnf".

Download and Unpack the Dataplot Files
Step 2: Download and Unpack the Dataplot Files The Dataplot files are available on the following github site

Select the "Clone or Download" menu and then select the "Download ZIP" text. This will save the file

    dataplot-master.zip

Move this file to a convenient location on your system and then enter

    unzip dataplot-master.zip

Alternatively, you can perform the download and unpack with the following command

curl -L https://github.com/usnistgov/dataplot/tarball/master | tar -xz

Build Dataplot
Step 3: Build Dataplot After the unzip command, enter the command

    cd dataplot/src

If you have installed the above developer libraries and you have root permissions, then you should not need to edit the Makefile. Otherwise, check the following two items in the Makefile.

  1. The "PREFIX=$(DESTDIR)/usr/local" line. If you have root permissions, you can leave this line as is or change it to "PREFIX=$(DESTDIR)/usr" if you prefer. If you do not have root permissions, we recommend changing this line to either "PREFIX=$(HOME)" or "PREFIX=/var/tmp".

  2. The line "FEATURES = GD CAIRO X11 READLINE NCURSES" specifies what dependencies you have installed. If you did not install any of the developer libraries, you may want to change this to "FEATURES = X11". The PLOT feature (LIBPLOT) is omitted by default. If you have the LIBPLOT libraries installed, you can add the PLOT feature.

The INSTALL file provides additional detail (e.g., if you want to build a version that can handle larger data set sizes). However, in most cases you should not need to modify anything other than the PREFIX and FEATURES lines.

After making any needed modifications to the Makefile, enter

    make

If the Makefile does not show any errors, then do

    sudo make install

If you did not set the PREFIX line to a location that requires root permissions, then you can omit the "sudo".

Problems with Downloading or Installation If you have problems with the download or installation, contact alan.heckert@nist.gov for assistance.
Test Dataplot Installation
Step 4a: Test Command Line Version of Dataplot You can test the installation of the command line version of Dataplot by entering the following commands.
    dataplot
    device 1 x11
    call minitest.dp
    exit
Even if you only intend to use the GUI version, I recommend testing the command line version first since the GUI version is in fact running the command line version.

Be sure that the directory containing the Dataplot executable is in your path. For example, if you have the executable in /usr/local/bin, you can add the following line to your .cshrc file (if you use the c-shell)

    set path=(/usr/local/bin $path)
For the Bourne shell, the comparable line for the .bashrc file is
    PATH=/usr/local/bin:$PATH; export PATH
Step 4b: Test the GUI version of Dataplot To test the Dataplot GUI, enter

    xdataplot

Dataplot Configuration Files There are two configuration files that you may want to edit at some point.

  • The "/usr/local/lib/dplogf.tex" file contains a startup file that will be executed when you initiate Dataplot. This is an ASCII file containing Dataplot commands that is useful for specifying your own defaults for Dataplot. The default version for Linux is essentially a place holder (i.e., it doesn't really do anything).

    If you would like to create your own custom file , then do the following

      cp /usr/local/lib/dplogf.tex ~/dplogf.tex

    Then edit ~/dplogf.tex using any ASCII editor. If you have a dplogf.tex file in your home directory, this will override the system version.

    How much you would like to utilize dplogf.tex is a personal preference.

  • The "/usr/local/lib/frscript/xdpConfig" file is a configuration file for the GUI. In general, you will not need to edit this file. However, if you would like to create your own custom version, then

      cp /usr/local/lib/frscript/xdpConfig ~/xdpConfig

    This is an ASCII file so edit ~/xdpConfig using your preferred ASCII editor. As with dplogf.tex, a copy of xdpConfig in your home directory will override the system version. See the comments in xdpConfig for guidance in editing this file.

Privacy Policy/Security Notice
Disclaimer | FOIA

NIST is an agency of the U.S. Commerce Department.

Date created: 06/05/2001
Last updated: 05/27/2020

Please email comments on this WWW page to alan.heckert@nist.gov.