Home

My BADC

Data

Search

Community

Help

Guided Tour

Getting data

Formats

Software

10 FAQ

Xconv/Convsh - Software for file manipulation/conversion

Contents

• Introduction
• Xconv
• Convsh
• Subsetting utilities
• Downloads
• Swapbytes (re-arranges byte order in some PP files, allowing XCONV to read them)
• Unpacking WGDOS packed pp files
• Credits

Introduction

This page contains binary utilities which allow the user to access, subset, interpolate, manipulate, convert and visualise data files in the following formats:

• NetCDF format
• GRIB format
• GrADS format
• UK Met Office Unified Model Data Output format
• UK Met Office PP format
• DRS format

Please note that these binaries are supplied "as is", and the BADC cannot offer technical support, such as fixing bugs, adding extra features etc.

Xconv is also available on the portable UM CD-ROM.

Documentation for Xconv (version 1.00) and Convsh is available via the world wide web.

Xconv

Xconv is an X windows utility which allows the user to interactively manipulate the data and produce an on-screen plot of the data field. It has an intuitive, user-friendly interface and is able to read and write a wide variety of different data formats.

Convsh

Convsh is the command line equivalent of Xconv, and allows GRIB files to be processed in 'batch' mode. The BADC has developed some Convsh scripts to batch process files related to individual datasets.

Subsetting utilities

Downloads

This section contains the binary versions of the utilities for use on a number of platforms. The current version is

1.91. Versions from 1.90 onwards can deal with all data files natively (without the need for byte-swapping).

Version 1.91 (latest version as of 19/8/09)

An updated release of xconv is now available, containing a number of bug fixes, for both the 64 and 32 bit linux versions ("Linux x86_32" and "Linux x86_x64" in the list below).

NOTE: Users should update their versions to the latest release.

Changes since original 1.91 release

read.c - fix to allow non-atmos multi-level UM ancil files to work.

read.c - fix to allow Lambert type grid in GRIB files to work.

read.c - fixed bug in calculation of date in GRIB files.

xtra.c,tcl_ext_xconv.c,tcl_ext_convsh.c - changed code for creating temporary files to fix problems running under windows.

gradsio.c - fixed bug in getting time from grads files (strcat -> strcpy).

header.h - Increased size of GRIB buffer from 600000 to 1100000, for high res ecmwf data.

interp.f - Make code work better with rotated limited area grids which cross over 180 deg. Added in code from UM 6.1 for rotated grids with a north pole with a negative latitude.

libutil - header files didn't define LITTLE__ENDIAN for x86_64 linux, needed to
          add #if defined __x86_64

read.c - fix in fillhead7 to allow large (>2GB) grads files to be read.

header.h, read.c - Need to use 8 byte INTEGER for very large UM files (> 8*2**31 bytes), various ints need to be changed to INTEGER and make c version of Fortran getlat.

The problem with libutil seems to be why your version of xconv doesn't work correctly with WG-DOS packed UM files.

Also a problem with the 32 bit version of xconv not working on recent versions of suse linux (and maybe others). This has been rebuild with a different version of tclkit.

The installation process remains as detailled below.

Version 1.91 (old version from 17/5/06)

This version has a number of bug fixes, and for most platforms, convsh functionaltiy is made possible by creating a symbolic link. The symbolic link name must contain the word convsh. e.g. for xconvR81.91:

ln -s xconvR81.91 convshR81.91

Notes:

• Please ensure that space is available on /tmp for xconv/convsh to function properly on linux/unix systems.

• The Windows versions shown above contain a fix for problems previously reported by some users (using Windows Server 2003), although they have been released under the same version number as before (1.91). These versions were updated on 14 February 2007, (previous versions: old xconv and old convsh).

• The following explanation has been provided for the Starkit file:

  That is a Starkit file, the executables are starpacks. It is a way of packaging up all the tcl files, shared libraries etc. into one standalone executable. If you have a program called tclkit installed on your system then you should be able to run xconv using the kit file. Tclkit is basically the whole tcl/tk code in one executable. The starpacks are just the kit file with tclkit added in so it works on all machines whether or not you've installed tclkit or tcl/tk.

• More information about how to get Xconv and Convsh to work on Windows is available here.

Version 1.90

Binaries for the other versions can be found below:

An xconv icon and png image are also available for download.

Byte swapping

If you are using an older versions of xconv than 1.90, then it is possible that byte swapping of the input data files may be required before they can be read on your systems using xconv/convsh. This may be done using a (unix) based byte-swapping utility called swapbytes. This allows a 4 byte word file to be converted from big to little endian, and vice versa. The tar file contains the source code and a basic makefile. (Should you require byte-swapping of 8-byte word files, use this utility instead; usage: swap8 < infile > outfile.)

Unpacking WGDOS packed pp Files

Before use in MDIAG, the BADC NWP data will need to be unpacked MDIAG. We are currently investigating new code that we hope to make available to unpack the NWP data files.

Please note that additional UM software is listed on the UWERN UM web pages: http://www.met.rdg.ac.uk/~sws00cw2/othertools/ .

Home

Contact

Disclaimer

Last Modified: