[IUCr Home Page] [CIF Home Page] [ciftbx home] [cyclops.src] [ciftbx.src]


Information for cif2cif 1.0.1, 7 April 2005

Before using this software, please read the
and please read the IUCr
on the Use of the Crystallographic Information File (CIF)

    \ | /      \ | /
     \|/        \|/ 
    -- -->>>>>>-- --               c i f 2 c i f  ...... CIF COPY PROGRAM
     /|\        /|\
    / | \      / | \                          Version 1.0.1
                                               7 April 2005

 cif2cif  is a fortran program using CIFtbx2 to copy a CIF on standard
 -------  input to an equivalent CIF on standard output, while checking
          data names against dictionaries and reformating numbers with
          esd's to conform to the rule of 19.  A request list may be 


                  Copyright © 1997, 1998, 2000, 2005
                  Herbert J. Bernstein (yaya@bernstein-plus-sons.com)
                  Bernstein + Sons
                  5 Brewster Lane
                  Bellport, NY 11713, U.S.A.

           based on suggestions by

                  Sydney R. Hall (syd@crystal.uwa.edu.au)
                  Crystallography Centre
                  University of Western Australia
                  Nedlands 6009, AUSTRALIA

           with the request list handling suggested by the program
           QUASAR by Sydney R. Hall and

                  Rolf Sievers (unc411@dbnrhrz1.bitnet)
                  Institut fuer Anorganische-Chemisches der Universitaet
                  Bonn, GERMANY

In order to ensure continuing availability of source code and documentation cif2cif and its documentation are subject to copyright. This does not prevent you from using the program, from making copies and changes, but prevents the creation of "closed source" versions out of the open source versions. See NOTICE.

Science is best served when the tools we use are fully understood by those who wield those tools and by those who make used of results obtained with those tools. When a scientific tool exists as software, access to source code is an important element in achieving full understanding of that tool. As our field evolves and new versions of software are required, access to source allows us to adapt our tools quickly and effectively.

In the early days of software development, most scientific software source code was freely and openly shared with a minimum of formalities. These days, it appears that carefully drawn legal documents are necessary to protect free access to the source code of scientific software. We are all deeply indebted to Richard Stallman for showing us how a creative combination of copyrights and seemingly restrictive licenses could give us truly unfettered freedom to use programs, to read their source code and to develop new versions. The GNU project, and the Linux project have shown that an open source approach works. We use the GNU General Public License (the "GPL") for our program starting with the release of CIFtbx3. Older versions use the license from OpenRasmol. The OpenRasMol conditions for use have correctly been called "GPL-like".

If you are a user of this program, you will find that the copyrights and notices ask little more of you than that you avoid mistakes by others by keeping the notices with copies, display scientific integrity by citing your sources properly and treating this like other shared scientific developments by not inferring a warranty. If you are a software developer and wish to incorporate what you find here into new code, or to pick up bits and pieces and used them in another context, the situation becomes more complex. Read the copyrights and notices carefully. You will find that they are "infectious". Whatever you make from our Open Source code must itself be offered as Open Source code. In addition, in order to allow users to understand what has changed and to ensure orderly development you have to describe your changes.

cif2cif reads the input CIF from the standard input device (normally device 5). An optional STAR data name dictionary (in DDL ) is opened. A reformatted copy of the input CIF is written to standard output (device 6). Messages are output to the standard error device (normally device 0). Note that the PARAMETER 'MAXBUF' should contain the maximum number of char- acters contained on a single text line. The default value is 2048.

In a unix-like environment, the program is run as:

      cif2cif [-i input_cif] [-o output_cif] [-d dictionary] [-c catck] \
         [-a aliaso_] [-e esdlim_] [-f command_file] [-g guess_type] \
         [-m maxline] [-p prefix] [-q request_list] [-t tabl_] \
         [-u unfold] [-w wrap]
         [input_cif [output_cif [dictionary [request_list]]]]
               input_cif defaults to $CIF2CIF_INPUT_CIF or stdin
               output_cif defaults to $CIF2CIF_OUTPUT_CIF or stdout
               dictionary defaults to $CIF2CIF_CHECK_DICTIONARY
               multiple dictionaries may be specified 
               request_list defaults to $CIF2CIF_REQUEST_LIST
               input_cif of "-" is stdin, output_cif of "-" is stdout
               request_list of "-" is stdin
               -a has values of t or 1 or y vs. f or 0 or n,
               -c has values of t or 1 or y vs. f or 0 or n,
               -e has integer values (e.g. 9, 19(default) or 29),
               -g has values of t, 1 or y vs. f, 0 or n (default t),
               -m has values from 80 to 2048 for a maximum line width
               -p has a string value in which "_" is replaced by blank,
               -t has values of t, 1 or y vs. f, 0 or n (default f),
               -u has values of t, 1 or y vs. f, 0 or n (default f),
               -w has values of 0 for no wrap or a column to wrap on

The program can reformat CIFs with long lines to CIFs with shorter line by use of the '-w wrap' option, and can reformat CIFs that have been wrapped to a short line length to a longer line length with the '-u y' option. Numbers with standard uncertainties can be reformatted to different rules (e.g. from rule of 9 to rule of 29), provided the type of the data name is known to be numeric or can be guessed from context. The guessing of types for data names not specified in a dictionary can be suppressed by use of the '-g n' option.

The extraction of selected fields using a request list is supported by the '-q request_list' option. If a data name is requested that does not appear in the input CIF, a dummy entry for the requested name will be placed in the output CIF.


Here is the recommended procedure for implementing and testing this version of cif2cif.

1.0. Before you try to install this version of cif2cif

     *** ========================================================== ***
     *** ========================================================== ***
     *** ==>>> You must have ciftbx version 3.0.0 or greater  <<<== ***
     *** ==>>> installed in a directory named ciftbx.src.     <<<== ***
     *** ==>>> The scripts mkdecompln and rmdecompln, which   <<<== ***
     *** ==>>> come with ciftbx, must be installed in the     <<<== ***
     *** ==>>> top level directory and executable.            <<<== ***
     *** ==>>> To test cif2cif, you must have a compressed    <<<== ***
     *** ==>>> copy of the dictionary cif_mm.dic in a         <<<== ***
     *** ==>>> directory named dictionaries.                  <<<== ***
     *** ========================================================== ***
     *** ========================================================== ***
The directory structure within which you will work is
                  top level directory
            |              |             | 
       dictionaries   ciftbx.src     cif2cif.src
       ------------   ----------     -----------

You may have acquired this package in one of several forms. The most likely are as a "C-shell Archive," a "Shell Archive", or as separate files. The idea is to get to separate files, all in the same directory, named cif2cif.src, parallel to the directory ciftbx.src, but let's start with the possibility that you got the package as one big file, i.e. in one of the archive file formats. Place the archive in the top level directory.

   *** ========================================================== ***
   *** ========================================================== ***
   *** ==>>> The files in this kit will unpack into a       <<<== ***
   *** ==>>> directory named cif2cif.src.  It is a good idea<<<== ***
   *** ==>>> to save the current contents of cif2cif.src    <<<== ***
   *** ==>>> and then to make the directory empty           <<<== ***
   *** ========================================================== ***
   *** ========================================================== ***
If you are on a machine which does not provide a unix-like shell, you will need to take apart the archive by hand using a text editor. We'll get to that in a moment.


If you have the shell archive on a unix machine, follow the instructions at the front of the archive, i.e. save the uncompressed archive file as "file", then, if the archive is a "Shell Archive" execute "sh file". If the archive is a "C-Shell Archive" execute "csh file".


If sh or csh are not available, then it is best to start with the "C-Shell Archive" and do the steps that follow. If you must use the "Shell Archive" you should be aware that the lines you want to extract have been prefixed with "X", while most of the lines you want to discard have not. For a "C-Shell Archive" such prefixes are rare and the file is easier to read. Assume you have a "C-Shell Archive".

Use your editor to separate the different parts of the file into individual files in your workspace. Each part starts with a lot of unixisms, then several blank lines and then two lines which identify the file, and most importantly, contain the text "CUT_HERE_CUT_HERE_CUT_HERE" You can look at the line before and the line after to see if you are at the head or tail of a file. Use your editor to search for the "CUT_HERE" lines. Each part is carefully labeled and indicates the recommended filename for the separated file. On some machines these filenames may need to be altered to suit the OS or compiler.


The partitions are as follows:

   part  filename                   description

     1  gpl.txt                          GPL (GNU General Public License)
     2  NOTICE                           Notices
     3  cif2cif.src/README.cif2cif       additional information on cif2cif
     4  cif2cif.src/MANIFEST             a list of files in the kit
     5  cif2cif.src/Makefile             a preliminary control file for make
     6  cif2cif.src/4ins.cif             example mmcif file used to test cif2cif
     7  cif2cif.src/4ins.out             example mmcif output from test of cif2cif
     8  cif2cif.src/4ins.prt             example mmcif list file from test of cif2cif
     9  cif2cif.src/4insuw.out           example mmcif unwrapped long output
    10  cif2cif.src/4insuw.prt           example mmcif list unwrapped long output
    11  cif2cif.src/4insw.out            example mmcif wrapped output
    12  cif2cif.src/4insw.prt            example mmcif list wrapped output
    13  cif2cif.src/cif2cif.cmn          cif2cif common block
    14  cif2cif.src/cif2cif.f            cif2cif fortran source
    15  cif2cif.src/qtest.cif            quasar mode test input cif
    16  cif2cif.src/qtest.out            quasar mode test output cif
    17  cif2cif.src/qtest.req            quasar mode test request file
    18  cif2cif.src/qtest.prt            example cif file used to test cif2cif
    19  cif2cif.src/xtalt2.cif           example cif file used to test cif2cif
    20  cif2cif.src/xtalt2.out           example cif output from test of cif2cif
    21  cif2cif.src/xte29.out            example cif output from test of cif2cif
    22  cif2cif.src/xttne9.out           example cif output from test of cif2cif


Here are the recommended steps for a UNIX system. Vary this according to the requirements of your OS and compiler. You will simplest to work if you place the cif2cif files together in a common subdirectory named 'cif2cif.src'. Be very careful if you place them in a directory with other files, since some of the build and test instructions remove or overwrite existing files, especially with extensions such as '.o', '.lst', or '.diff'. On a UNIX system, most of what you need to do to build and test cif2cif is laid out in Makefile. *** Be sure to examine and edit this file appropriately before using it.*** But, once properly edited, all you should need to do is 'make clean' to remove old object files, 'make all' to build new version of 'cif2cif' and 'make tests' to test what you have built.

For non-UNIX-like environments, you will have to provide replacements for iargc, getarg and getenv. The following are reasonable possibilities:

         integer function iargc(dummy)

         subroutine getarg(narg,string)
         integer narg
         character*(*) string

         subroutine getenv(evar,string)
         character*(*) evar,string
        *  string='INPCIF.CIF'//char(0)
        *  string='OUTCIF.CIF'//char(0)
        *  string='CIF_CORE.DIC'//char(0)
This combination of substitute routines would "wire-in" cif2cif to read its input cif from a file named INPCIF.CIF, write its output cif to a file named OUTCIF.CIF, and check names against CIF_CORE.DIC


       dictionary input         input   on device 2
       Reformatted CIF          output  on device 6 ('stdout')
       Input CIF                input   on device 2, if a file, 5  if 'stdin'
       Message device           output  on device 0 ('stderr')
       Direct access            in/out  on device 3

TEST files

Three test CIFs are provided. xtal2.cif is a test file borrowed from xtal_gx (file xtest2.cif at www.iucr.org/iucr-top/cif/software/xtal/gx/, provided by S. R. Hall. 4ins.cif is an mmCIF file created from the PDB entry 4INS by G.G. Dodson, E. J. Dodson, D. C. Hodgkin, N.W. Isaacs and M. Vijayan (1989) by the program pdb2cif (P.E. Bourne, F.C. Bernstein and H.J. Bernstein, 1996, see www.bernstein-plus-sons.com/software/pdb2cif). qtest.cif is the test cif from the quasar release by Hall and Sievers (1994). A modified version of the request list, qtest.req, is also included.

xtalt2.cif provides good test cases for the conversion of esd's. The command

    cif2cif -t y < xtalt2.cif > xtalt2.new
ensures that all esd's follow the rule of 19, while
    cif2cif -t y -e 29 < xtalt2.cif > xte29.new
converts esd's to the rule of 29. The difference between the two rules is that for the rule of 19, all esd's lie between 2 and 19, so that an esd of (1) has to be converted to (10), while for the rule of 29, all esd's lie between 3 and 29, so that an esd of (2) also has to be converted, in this case to (20). The option "-t y" tidies the output to tab stops.

One last test with this file is to use the command

   cif2cif -e 9 < xtalt2.cif > xttne9.new
to copy the original cif spacing and to use the rule of 9 on esd's

4ins.cif has many comments, text fields and dense loops. The test in the Makefile tests handling of these items and adds the additional complication of processing a prefix ".._" with the command

   cif2cif -t y -p .._ < 4ins.cif > 4ins.new
The output spacing is controlled by the program.

Two additional tests are provided. The command

   cif2cif -gn -u -m2048 -i 4ins.cif > 4insuw.new
unwraps the 80 character version of 4ins.cif to the maximum line length, not using a dictionary and suppressing the guessing of numeric data types. This is followed by a pass wrapping the resulting cif to 120 columns with the command
   cif2cif -d cif_mm.dic -w 120 -i 4insuw.new > 4insw.new

The quasar mode may be tested by the command

  cif2cif -i qtest.cif -o qtest.new -q qtest.req


Version 1.0.1 (7 April 2005) added the feature of including a dummy entry in the output cif for each name that appears on a request list,l but not on the input cif. The new command line option -gn adds the ability to suppress guessing of data types for names not defined in a dictionary. Formatting of unwrapped output cifs was improved. Work supported in part by the International Union of Crystallography.

Version 1.0.0 (2 April 2005) added support for line folding and unfolding using -w (for wrap), -u (for unfold) and -m (for maximum line length). Work supported in part by the International Union of Crystallography.

Version 0.1.0 (8 July 2000) fixed a parse error in the command line and an error in the comments.

Version 0.0.9 (30 May 2000) changed the variable named tab to xxtab to avoid a conflict.

Version 0.0.8 (2 April 1998) added code to preserve the distinction between esds which are numerically identical but presented with varying numbers of trailing zeros. To enable this feature, the command line argument -e must be given a negative value large enough to span the desired range, e.g. "-e -19" to allow both .6870(10) and .687(1) to be handled. The variables esddig_ and pesddig_ introduced with ciftbx 2.6 are used. A bug in the preserving leading zeros of unlooped numeric values was also fixed.

Version 0.0.7 (7 August 1997) added support for quasar-style request lists. In addition, a special form of request for "data_which_contains:" will find the data block which contains at least one of the following tags.

Version 0.0.6 (12 May 1997) added support of negative values of esdlim_, indicating a range of esd's from 1 to -esdlim_. The practical use of this change is to use the command line parameter -e-9999 to copy most esds unchanged. The new ciftbx variables decp_, pdecp_, lzero_ and plzero_ are used internally to copy decimal points and leading zeros more faithfully. The 4ins example has been updated to use cif_mm.dic (0.9.01). The command-line option "-c catck" was added. The default is not to check categories.

Version 0.0.5 (2 December 96) adds support for copying of global_ sections and corrects a typo in the error message issued when the output CIF cannot be opened.

Version 0.0.4 (24 September 96) adapts to the 'ciftbx.cmn' and 'ciftbx.sys' reorganization in ciftbx 2.5.1.

Version 0.0.3 (24 July 96) used ciftbx 2.5.0 to preserve tabs, fixed a case in which string quoted in the input cif may not have been quoted in the output cif, preserved more white space, and copied comments within loop headers. The lateral position of "loop_" is preserved. A bug in copying final comments which caused the a comment to be duplicated or some final comments to be lost was fixed.

Version 0.0.2 (24 June 96) changed the default for -t to n instead of y, and used ciftbx 2.4.6 to increase the faithfulness of the copy.


cif2cif does not copy white space exactly, and will reformat some data values. Always compare the original to the output.

Updated 24 July 2005