csh to bash script conversion gives unexpected eof error

[labmonkey]

Hello forum members, I am trying to adapt a script originally written by programmers for csh to run in bash. The script calls a number of functions to calculate protein conformational energy.
I have changed a number of non-bash commands, such as setenv to export, and after
many hours apt-getting new code, got to the stage where I could do a makefile and then source the .bashrc file (even though this has to be manually at the start of each session).

However, there is a dreaded 'unexpected end of file' error flagged up when I do the
following.........(my comments are marked with ##).

##source the .bashrc file##

research1(at)heksaeder:~$ source /home/research1/.bashrc.eceppak

##OK so far, but then run the script recepp.s with appropriate arguments##

research1(at)heksaeder:~$ recepp.s ENERGY H_alpha OUT x x 1

##The recepp-LNX.s file is pointed to by the generic recepp.s script, however I get the
following error##

/home/research1/eceppak/bin/recepp-LNX.s: line 238: syntax error: unexpected end of file

##I know that this refers probably to an incomplete if else endif sequence, but I cannot see
where the problem might be...is if else endif the correct syntax for bash?##

##I also know that this is a common error that has been mentioned in the forum before, but
I haven't found anything that would point me in the right direction I'm afraid##

##and the normal shell commands that are available before this script is run, such as ls and cd
and the editor (nano) are no longer available, I have to re-open a new terminal window##

research1(at)heksaeder:~$ nano /home/research1/eceppak/bin/recepp-LNX.s
bash: nano: command not found

##The script recepp-LNX.s is listed below, sorry I could of course try reducing it whilst retaining
the error, but I'm not a bash scripter, rather someone wanting to use a script that needs to work in a bash
environment.##

##The recepp-LNX.s script lists as follows, and apologies for its length##


#!/bin/bash -f
# procedure to run new CSearch
if ( $#argv == 0 ) then
echo "You must specified the type of Run: recepp.s RUN_TYPE (+ ARGUMENTS)"
echo "Where RUN_TYPE is one of the followings: "
echo " ENERGY"
echo " CHECKGRAD"
echo " MINIMIZE"
echo " EDMC"
echo " RMS_FIT"
echo " BOUNDS"
echo " VTF"
exit 1
endif
export RUNTYP=$1
export RUN="TYPE0"
if ( $1 == "ENERGY" ) then
export RUN="TYPE1"
endif
if ( $1 == "CHECKGRAD" ) then
export RUN="TYPE1"
endif
if ( $1 == "MINIMIZE" ) then
export RUN="TYPE1"
endif
if ( $1 == "EDMC" ) then
export RUN="TYPE1"
endif
if ( $1 == "RMS_FIT" ) then
export RUN="TYPE2"
endif
if ( $1 == "BOUNDS" ) then
export RUN="TYPE2"
endif
if ( $1 == "VTF" ) then
export RUN="TYPE1"
endif
if ( $RUN == "TYPE0" ) then
echo "Unrecognized option " $1
echo "You must specified: recepp.s RUN_TYPE (+ ARGUMENTS)"
echo "Where RUN_TYPE is one of the followings: "
echo " ENERGY"
echo " CHECKGRAD"
echo " MINIMIZE"
echo " EDMC"
echo " RMS_FIT"
echo " BOUNDS"
echo " VTF"
exit 1
endif
if ( $RUN == "TYPE1" ) then
if ( $#argv < 5 ) then
echo "Missing argument(s)! "
echo "You must enter: recepp.s " $1 " ARG2 ARG3 ARG4 ARG5 "
echo "ARG2 is the FILENAME of INPUT (*.inp) file (Input for the run)."
echo "ARG3 is the FILETYPE of OUTPUT files."
echo "ARG4 is the FILENAME of the ECEPP (outo.*) input file (if any) "
echo " or a Dummy Argument "
echo "ARG5 is the FILETYPE of Constraints (bounds.*) files or a "
echo " Dummy argument if no constraints are used."
exit 1
endif
export BNDF=$5
export ANAF=$4
endif
if ( $RUN == "TYPE2" ) then
if ( $1 == "RMS_FIT" ) then
if ( $#argv < 5 ) then
echo "Missing argument(s)! "
echo "You must enter: recepp.s " $1 " ARG2 ARG3 ARG4 ARG5 "
echo "ARG2 is the FILENAME of INPUT (*.inp) file (Input for the run)."
echo "ARG3 is the FILETYPE of OUTPUT files."
echo "ARG4 is the FILENAME of the PDB or ECEPP (outo.*) input file."
echo "ARG5 is the FILENAME of the Reference (xray.*) file."
exit 1
endif
export ANAF=$4
export REFF=$5
endif
if ( $1 == "BOUNDS" ) then
if ( $#argv < 4 ) then
echo "Missing argument(s)! "
echo "You must enter: recepp.s " $1 " ARG2 ARG3 ARG4 "
echo "ARG2 is the FILENAME of INPUT (*.inp) file (Input for the run)."
echo "ARG3 is the FILETYPE of OUTPUT files."
echo "ARG4 is the FILENAME of the PDB input file."
exit 1
endif
export ANAF=$4
endif
endif

export INPF=$2
export OUTF=$3

echo $1 " CALCULATION"

# IF names of files are too large (more than 64 characters) program won't run. In this case,
# it is better to create a symbolic link to the appropriate directory.
ln -s ${ECEPP_ROOT} ECEPP_ROOTA

# ******** GENERAL INPUT FILES *************

# Input for distance constraints
# ln -s bounds.$BNDF fort.7

# The following is the standard input file.
# ln -s $INPF.inp fort.8

# Input file containing the series of conformations whose energies
# must be evaluated or minimized ("MULT_CONF" control card).
# ln -s outo.$ANAF fort.11

# INPUTS FOR SURFACE AREA AND HYDRATION MODELS
# Parameters giving the free energy density for various solute atoms radii.
export SOLVDBS=ECEPP_ROOTA/Data/Hydration/
export SOLVFILE=$SOLVDBS/srfopt.set
# Joe's solvation model.
# export SOLVFILE=ECEPP_ROOTA/Data/Hydration/volume.set
export FXVOLFILE=ECEPP_ROOTA/Data/Hydration/fix_vol_tst.parm

# rsdata is a version of the STANDARD AMINO ACID DICTIONARY
# containing the ECEPP/3 residues and other non-standard ones.
# NP in PRO's was replaced by N to increase compatibility
# with PDB files.
#
# Atom type of protons in COOH groups ( ASP, GLU, meASP, meGLU
# and Carboxyl end terminal) changed to type 1, as in ECEPP/3
# (no H-bonding allowed).
#
# passing database directory to program
export RSDBS=ECEPP_ROOTA/Data/Residue/

export RSDATA=ECEPP_ROOTA/Data/Residue/rsdata

# File defining proton labels for each residue (for NMR purposes).
# ln -s $ECEPP_ROOT/Data/Residue/nmr_protonsT.data fort.33
export HLABELFILE=ECEPP_ROOTA/Data/Residue/nmr_protons.data

export DIHDEFILE=ECEPP_ROOTA/Data/Residue/dihedral_def

# Parameters for the unified residue approximation, the radius of
# each residue in the sequence. NOT IMPLEMENTED ON THIS VERSION DR 2-94
# RESRAD = UNIT 37
# ln -s resrad.data fort.37

# Atomic charges for IBEM
export IBEMCHG=ECEPP_ROOTA/Data/IBEM/resibem

# Atomic radii for IBEM

# ln -s ECEPP_ROOTA/data/IBEM/atrad.siz fort.29
# export ATMRAD=ECEPP_ROOTA/Data/IBEM/atrad.siz

# Run parameters for IBEM

# ln -s ECEPP_ROOTA/Data/IBEM/input.ibe fort.1
# export IBEINFILE=ECEPP_ROOTA/Data/IBEM/input.ibe

# **********************************************

# / ********* GENERAL OUTPUT FILES *********/

# ln -s outo.$OUTF fort.4
# ln -s standout.$OUTF fort.6
# ln -s sumsl.$OUTF fort.9

# Output file for BOUNDS runs.
# ln -s bounds.$OUTF fort.12
# ln -s main_out.$OUTF fort.20

# IBEM List file

# ln -s ibe.$OUTF fort.2
# export IBEOUTFILE=$OUTF

# Be aware that UNIT 30 is used to print cartesian coordinates in OUTNW
# routine. It is not assigned by this script.

# The following definition can be used to print the APS matrix (Ken Gibson's
# compact form of the VDW and RO parameters), from BD11 subroutine.
# Write statement is currently commented out.
# ln -s APS.$2 FILE.FT91F001

# print identification of files
echo "input file $INPF.inp "
echo "Identification Argument of output files " $OUTF


# Area optimization, the target areas for each residue in the sequence.
# TGTARS = UNIT 38
export TGTARS=TARGET.tgtars

#
#/* ******* INPUT-OUTPUT FILES ******** */
# ln -s edmc.rstart fort.17
export RESTFILE=edmc.rstart

# ln -s edmc.rndom fort.18
setenv RANDFILE edmc.rndom


# if ( $RUN == "TYPE2" ) then

# **********************************************
# /* INPUT FILES RMS */
#
# Input conformation used to generate a REFERENCE file.
# ln -s $REFF.pdb fort.27

# This file contains the REFERENCE conformation.
# ln -s xray.$REFF fort.57

# When comparing two pdb files the input file is assigned to unit 44.
# The program will compare the conformation on this file with (a) the one
# in UNIT 57 (xray.*); or (b) that in UNIT 27.
# ln -s $ANAF.pdb fort.44

# /* OUTPUT FILES RMS */

# ln -s sidev.$OUTF fort.22
# ln -s hedev.$OUTF fort.26
# ln -s badev.$OUTF fort.31
# ln -s rgyr.$OUTF fort.34
# ln -s xray.$OUTF fort.39
# ln -s chefol.$OUTF fort.46
# ln -s comso.$OUTF fort.47
# ln -s cmsdk.$OUTF fort.48
# ln -s kmsba.$OUTF fort.49
# **********************************************
# endif

# The name of the ECEPPAK executable goes here:
nice ECEPP_ROOTA/bin/eceppak-${ECEPP_ARCH}

endif
# it is possible to create a symbolic link to the appropriate directory.
/bin/rm ECEPP_ROOTA

##thanks in advance if anyone has the patience to look through this script and to suggest
what my be a problem and worth testing##


Steve

[robertdaleweir]

I ran a 'grep -w' on your code to count and identify the number of 'if' s and 'endif' s. If I ran on the precise code you setout properly I count 17 'endif' s and only 16 'if' s. I counted only non-commented pairs. Not sure this is the only problem with the code but may be one of them. Cheers..
Robert

[labmonkey]

Robert,

thanks for the quick reply...I looked through the script and removed the last endif statement but to no avail.
I know that sometimes there are hidden characters or spaces where there shouldn't be in scripts like this, but essentially this script has worked for other researchers using presumably the csh...but your reply was much appreciated thanks and welcome to the forum!!

Steve

[robertdaleweir]

Hi labmonkey
I see that you do have an instance of setenv, line 200 in your script, not changed to export. This will usually kil a script at runtime. Cheers...
Robert

[lomcevak]

First I suggest you change the shebang line to:

#!/bin/bash -fvx

This will list the commands as they are being executed. This will help in debugging the script.

Next the syntax for the 'if' command is as follows as according to the bash manpage:

if list; then list; [ elif list; then list; ] ... [ else list; ] fi

Therefore you have to change all the 'endif' to 'fi'. This will at least get you a little farther.

I don't know csh so I can't help more than this. You'll have to look at each error one at a time and figure out what the csh command is equivalent in bash.

[labmonkey]

Hi,

argh always one that slips through....but changing the setenv to export didn't cure the problem...something else lurks within....thanks for your continued help though

Steve

[labmonkey]

OK. I wasn't sure about the if else endif syntax so I will change these and see what that brings...also thanks for the suggestion to make the command execution verbose. Thanks for your help.

Steve