srs4.05 is ready

Thure Etzold etzold
Tue May 30 09:43:32 EST 1995

dear all,

release 4.05 of SRS is available now. Get the file
pub/software/unix/srs/srs4_05.tar.gz via anonymous ftp from

There were some problems installing SRS with the new httpd from NCSA as
reported by Jack Leunissen and Steve Taylor which should be fixed now. The
cgi's and the document files are declared to reside now in two different 
directories (see the installation procedure in the README of the distribution or
http://www/srs/srsman_install.html. Please tell me if it works or if there are
still problems with it.

I added Tom Boutell's 'gd' library for creating GIF images and a new program
'srspict' that creates a picture with the databank network and a link diagram.
for the network it needs a file with the coordinates of all databanks which you
can create with a tcl/tk script (primitive graph editor) ...instructions in README.

The last important change is the complete replacement of the algorithm for finding
the shortest path in the library network ...which was about to crack on large
systems. The new algorithm is much faster and should work even with several hundred

Here is the complete list of changes since srs4.04:

   if a flat file is not accessible, eg, the file est.dat from
   embl then this will be mentioned only once ...eg during
   listing a list of embl entries together with selected
   fields ...still the names for these entries can be

   Added function LibIsDataType which tells whether a
   databank contains "protein" or "dna" sequences
   ...function can be easily extended for other data types 

   GCG formatted list file for features does not list
   "Strand" for protein sequences 

   enhanced srscheck to look at all single indices if they
   need rebuilding or compression 

   fixed problem with retrieving integer ranges ...integers
   are no longer treated as floats 

   prep_srs puts the srs directories at the begin of the
   PATH variable 

   the '?' for begin and end positions in swissprot are now

   a number of blast files can now be indexed - each entry
   witll receive a name composed of filename, number
   within that file, and the sequence found (eg, swissprot
   name). The file names are stored in the IDX file 

   added new attribute isSystem for #field - all fields
   marked with that should be ignored by user interfaces -
   use LibIsField (field, "system") to find out. 

   Replaced the algorithm for finding the shortest path in
   the library network ...now there is NO limit left to
   hinder its growth 

   Added buffer function for matching regular

   Changed script module so that it understands variables
   in labels and in the text 

   the script module provides now a utility to check
   number and type of arguments when printing from a
   label with print arguments 

   executables and document files can now be in different

   stdout for SRSWWW is no longer buffered. This fixes
   a problem with the httpd on OSF1 

..have fun

Thure Etzold                                   | EMBL
E-mail: etzold at embl-heidelberg.de              | Postfach 10.2209
Tel: (49) 6221 387529                          | 69012 Heidelberg
Fax: (49) 6221 387517                          | Germany

More information about the Bio-srs mailing list

Send comments to us at biosci-help [At] net.bio.net