setup.py

import os
import re
import random
import numpy as np
from math import sqrt, log
from pathlib import Path
import logging
import datetime
import argparse
import sys

import musicpy.Antoine as Antoine
import musicpy.Forcefield as Forcefield
now = datetime.datetime.now()

logger = logging.getLogger(__name__)
#logger.setLevel(logging.DEBUG)
#handler = logging.FileHandler('./setup.log')
#handler.setLevel(logging.DEBUG)
#handler2 = logging.StreamHandler()
#handler2.setLevel(logging.INFO)
#logger.addHandler(handler)
#logger.addHandler(handler2)
#logger.debug('-----------------------------------')
#logger.debug('Generating new files on {0}-{1}-{2} at {3}:{4}'.format(now.year, now.month, now.day, now.hour, now.minute))
#logger.debug('-----------------------------------')
#logger = logging.getLogger('foo').addHandler(logging.NullHandler())


######



#############
#Functions used here:

#Reads a pdb file to get atom types, positions, and conneciton information. Does it by finding the phrases 'HETATM' and 'CONECT', and required the lines ot have a specific number of strings in them
def dataextract(logger, species, sourcedir = '.'):
    atoms = []
    connections = []
    f = open('{0}/{1}.pdb' .format(sourcedir, species), "r")
    logger.debug('Reading from {0}{1}.pdb.'.format(sourcedir, species))
    for curline in f:
        fish = curline.split()
        #print(fish)
        if fish[0] == "HETATM": #I expect HETATM lines to look like this: HETATM index symbol residue ? x y z ? q symbol
            del(fish[8:10]) #deletes "? q " from the end fo the above
            del(fish[2:5]) #deletes "symbol residue ?" from above
            del(fish[0]) #deletes 'HETATM' from above
            atoms.append(fish)
        elif fish[0] == "CONECT": #I expect CONECT lines to be 'CONECT index1 index2 index3... where index1 is an atom and intex2 etc. are atoms bound to it
            del(fish[0]) #deletes 'CONECT from the above
            connections.append(fish)
    logger.debug('Atoms deteceted are: {0}\n Connections detected are:{1}'.format(atoms, connections))
    for thing in atoms:
        if thing[4] in element_names:
            logger.debug('Element name {0} changed to {1}'.format(thing[4], element_names[thing[4]]))
            thing[4] = element_names[thing[4]] #replaces the atom symbol from above with the element name
    return atoms, connections

#This sorts the connections detected from apdb into a list for a molfile
def connectiontypeswrite(logger, atoms, connections):
    atomnames = []
    pairset = set()
    bondlengths = {}
    for i in atoms:
        atomnames.append(i[4]) #creates a list of atom names from the stom position information from dataxtract
    logger.debug('Atom names are found to be: {0}'.format(atomnames))
    #atomstypes = dict.fromkeys(atomnames)
    indices = []
    for i in connections:
        for j in i:
            if j != i[0]:
                pairset.add(tuple(sorted([atomnames[int(i[0])-1], atomnames[int(j)-1]]))) #creates a set of tuples of bonded atom pairs, so there;s no duplicates
    logger.debug('Connections set found is: {0}'.format(pairset))
    for thing in pairset:
        bondlengths[thing] = [] #creates a dictionary of the above tuples, so you can load in your bond lengths
    for i in connections:
        for j in i:
            key = tuple(sorted([atomnames[int(i[0])-1], atomnames[int(j)-1]])) #detects the atom bonding pair from each 'i j' pair in the connections list
            if j != i[0]:
                dist = 0 #bond length. This line resets it for each iteration of the loop
                dist = sqrt((float(atoms[int(i[0])-1][1])-float(atoms[int(j)-1][1]))**2+(float(atoms[int(i[0])-1][2])-float(atoms[int(j)-1][2]))**2+(float(atoms[int(i[0])-1][3])-float(atoms[int(j)-1][3]))**2) #measures bond length in Angstrom
                dist = round(dist, 3)   #rounds to nearest 0.001 of an angstrom. Is this good enough?
                bondlengths[key].append(dist) #appends it to the list of bonds for atom pairs 'i j'
    for pair in bondlengths:
        if len(bondlengths[pair]) > 1: #if there is recorded bond length information
            for i in bondlengths[pair]:
                if float(bondlengths[pair][0])-float(i) > 0.001: #if your bond lengths vary more than this, throw a fit.
                    logger.warning('ERROR! Tolerance exceeded for bonds between atoms of types %s.' % (str(pair)))
                    sys.exit()
            bondlengths[pair] = round(sum(bondlengths[pair])/len(bondlengths[pair]), 3) 
    logger.info('Connection types and bond lengths calculated are:')
    for key in bondlengths:
        logger.info('{0}: {1} Angstrom'.format(key, bondlengths[key]))
    return bondlengths #returns the bondlength dictionary of 'i j pair': length

#This writes a molecule file from the informaiton from dataextract and connectiontypes. It can overwrite your existing files though, beware!
def molwrite(logger, species, atoms, connectiontypes, connections, sourcedir = '.', forcefield = None):
    f = open("{0}/{1}.mol" .format(sourcedir, species), 'w')
    f.write("### Music molecule construction information, generated from {0}{1}.pdb by a python script. Have you proofread me? \n\n" .format(sourcedir, species)) #Explains that this is autogenerated
    f.write("#Basic molecule information \n") #Required for music?
    f.write("Molecule_Name: {0} \n" .format(species)) #definitely required for music. And it hates tabs in your lines.
    f.write("Coord_Info: listed cartesian rigid\n") #here i'm telling the moelcule file your moecule has listed atoms, cartesian coordinates, and rigid relationships between them
    f.write("{0} \n" .format(len(atoms))) #this prints the number of atoms to your file
    if len(atoms) >1:
        for i in atoms:
            logger.debug(i)
            for j in i:
                f.write('{0} ' .format(str(j))) #prints the atom index, x, y, z, name
            f.write("{0} 0 0\n" .format('0' if forcefield == None else str(Forcefield.IntParams['{0}_{1}'.format(i[-1].split('_')[0], forcefield)][2]))) #prints q, ?, ?. change the '0' in format if you want to set a q to a variable
    else:
        f.write('1 0 0 0 {0} ' .format(str(atoms[-1]))) #prints the atom index, x, y, z, name
        f.write("{0} 0 0\n" .format('0' if forcefield == None else str(Forcefield.IntParams['{0}_{1}'.format(atoms[-1].split('_')[0], forcefield)][2]))) #prints q, ?, ?. change the '0' in format if you want to set a q to a variable
    if connectiontypes:
        f.write("Connect_Info: listed\n") #declares it's listing the number of bond dypes as well as the exact bond pairs
        f.write("{0} # number of unique bond types in your molecule\n" .format(len(connectiontypes)))
        for i in connectiontypes:
            x = str(i) #makes a string of you 'i j' pair
            x = x.strip("('" "')") #cuts out the quites from the string
            x = x.split("', '") #and the comma from between them
            logger.debug(x) #now it should just be 2 atom type names
            f.write("{0} {1} {2}\n" .format(x[0], x[1], connectiontypes[i])) #now it writes atom type 1, atom type 2, length
        for i in connections: #This loops over your CONECT reading list from dataextract to write the bond list
            for j in i:
                f.write("{0} " .format(str(j)))
            f.write('\n')
    f.write("# Degrees of freedom (optional)\n") #writes out the degrees of freedom, essentialy saying you have 3 for monoatomic, and 6 otherwise
    if len(atoms) == 1:
        dof = 3
    else:
        dof = 6
    f.write("Molecule_DOF: {0} #3 for monoatomic, 5 for rigid linear and 6 for  rigid nonlinear, if flexible then generally 3*number of atoms " .format(dof))        
    f.close()
    logger.info("Molecule file {0}/{1}.mol written!" .format(sourcedir, species))

#This function checks if you already have a directory with a certain name, and makes it if not.
def directorymaker(logger, dxout = "."):
    filename = "{0}/test.txt" .format(dxout) #Test file name
    if not os.path.exists(os.path.dirname(filename)): #Checks if the test file exists
        try:
            os.makedirs(os.path.dirname(filename)) #Makes the file
        except OSError as exc: # Guard against race condition
            if exc.errno != errno.EEXIST:
                raise
    with open(filename, "w") as f:
        f.write("FOOBAR") #Writes something in the test file
    logger.debug("Directory {0} written!".format(dxout))

#This writes your atom-atom forcefield information based on the above dictionary at the start of the document
def AtmAtmMover(logger, elements, forcefield,coultype=None, hicut=18, framework='IRMOF1', dxout = "./mapgen/test", FluFra = None):
#Find the MOF element list
    try:
        Forcefield.MOF_el_list[framework]
    except KeyError:
        logger.warning('I failed to find yuor MOF element list. Go into the Forcefield file and make sure it\'s in there before continuing!')
        sys.exit(0)
    
    Frame_el = Forcefield.MOF_el_list[framework]
        
    logger.info('The elements I\'m considering in your framework are: {0}\n The elements I\'m considering in your sorbent are: {1}'.format(Frame_el, elements))

    with open("{0}/atom_atom_file" .format(dxout), "w") as f:
        f.write("""#This is an autogenereated interactions file for music, based on the python script Umbrella-v2. It's probably best to look this over before running your sims!\n""")
        f.write("Framework-Framework interactions (usually off)\nLennard-Jones\n")
        for i in range(len(Frame_el)):
            f.write(LJPrep(True, Frame_el[i], Frame_el[i], forcefield, hicut))
            for j in range(i+1, len(Frame_el)):
                f.write(LJPrep(False, Frame_el[i], Frame_el[j]))
            f.write('\n')

        f.write('Coulombic\n')
        for i in range(len(Frame_el)):
            f.write(CoulPrep(None, Frame_el[i], Frame_el[i]))
            for j in range(i+1, len(Frame_el)):
                f.write(CoulPrep(None, Frame_el[i], Frame_el[j]))
            f.write('\n')

        f.write("Fluid-Fluid interactions\nLennard-Jones\n")
        for i in range(len(elements)):
            f.write(LJPrep(True,elements[i], elements[i],forcefield, hicut))
            for j in range(i+1, len(elements)):
                f.write(LJPrep(True, elements[i], elements[j], forcefield,hicut))
            f.write('\n')

        f.write('Coulombic\n')
        for i in range(len(elements)):
            f.write(CoulPrep(coultype, elements[i], elements[i], hicut))
            for j in range(i+1, len(elements)):
                f.write(CoulPrep(coultype, elements[i], elements[j], hicut))
            f.write('\n')

        f.write("Fluid-Framework interactions\nLennard-Jones\n")
        for i in range(len(elements)):
            for j in range(len(Frame_el)):
                f.write(LJPrep(FluFra, elements[i], Frame_el[j], forcefield, hicut))
            f.write('\n')
    
        f.write('Coulombic\n')
        for i in range(len(elements)):
            for j in range(len(Frame_el)):
                f.write(CoulPrep(None, elements[i], Frame_el[j]))
            f.write('\n')
    logger.debug("Atom-Atom interaction file written in directory {0}!".format(dxout))

def CoulPrep(type, i, j, hicut=None):
    if type == 'Wolf':
        return '{0} {1} WFCOUL  HICUT@{2}\n'.format(i, j, hicut)
    elif type == 'Ewald':
        return '{0} {1} SUM FAST EWALD HICUT@{2}\n'.format(i, j, hicut)
    elif type == None:
        return '{0} {1} COUL OFF\n'.format(i, j)
    else:
        logger.warning("Well something has gone awfully wrong in the CoulPrep function! I'm going to exit now")
        sys.exit()

def LJPrep(status, i, j, forcefield=None, hicut=None):
    tag = {}
    if len(i.split('_'))>1:
        tag[i] = '{0}_{1}'.format(i.split('_')[0], forcefield)
    else:
        tag[i] = '{0}'.format(i)
    if len(j.split('_'))>1:
        tag[j] = '{0}_{1}'.format(j.split('_')[0], forcefield)
    else:
        tag[j] = '{0}'.format(j)
    if status:
        sig = LBmixSig(Forcefield.IntParams[tag[i]][0], Forcefield.IntParams[tag[j]][0])
        eps = LBmixEps(Forcefield.IntParams[tag[i]][1], Forcefield.IntParams[tag[j]][1])
        return '{0} {1} LJ SIG@{2} EPS@{3} HICUT@{4}\n'.format(i, j, sig, eps, hicut)    
    else:
        return '{0} {1} LJ OFF\n'.format(i, j)

def LBmixSig(a, b):
    result = (float(a)+float(b))/2
    return str(round(result,3))

def LBmixEps(a, b):
    result = sqrt(float(a)*float(b))
    return str(round(result,3))
    
#This writes your sorb-sorb file telling music which intermolecular interactions are on
def SorbSorbWriter(logger, species, elements,framework, dxout = "./mapgen", pmap = True, emap = False):
    filename = "sorb_sorb_file"
    with open("{0}/{1}" .format(dxout, filename), 'w') as file:
        file.write("{0} {0} NCOUL OFF\n{0} {0} COUL OFF\n\n" .format(framework)) #MOF-MOF interactions are off
        if isinstance(species, list) == True: #This checks if your species is a list or a string, which makes it work for multiple species at once
            if pmap == True: #using a MOF-fluid LJ potential map if true
                for i in species:            
                    file.write("{0} {1} NCOUL MAP@{1} FAST ".format(i, framework)) #declares it'll use a map
                    for j in elements:
                        file.write("{0}@PMAP@{1}.{0}.p " .format(j, framework)) #writes your map names. WARNING - it's a 18 character limit on map names, and no '-' allowed!
                    file.write("\n")
            else:
                for i in species:
                    file.write("{0} {1} NCOUL BASIC LJ FAST\n".format(i, framework)) #pairwise fluid-framework interactions
            if emap == True: #using a MOF-fluid coulomb map if true
                for i in species:
                    file.write("{0} {1} COUL MAP@{1} FAST ".format(i, framework)) #declares it'll use a map
                    for j in elements:
                        file.write("{0}@EMAP@{1}.{2}.e " .format(j, framework, 'Probe')) #writes your map names. WARNING - it's a 18 character limit on map names, and no '-' allowed!
                    file.write("\n")
            else:
                for i in species:
                    file.write("\n{0} {1} COUL OFF\n" .format(i, framework)) #pairwise coulomb interactionsdon't work. Trust me, this is better.
                logger.warning("Framework-fluid coulombic interactions are off") #warns you your coulomb fluid-framework stuff isn't happeneing
            file.write("\n{0} {0} NCOUL BASIC LJ FAST\n{0} {0} COUL SUM FAST EWALD KMAX@15 KAPPA@6.7\n\n" .format(i)) #I think this line shouldn't be here. Needs testing!
        elif isinstance(species, str) == True: 
            if pmap == True: #using a MOF-fluid LJ potential map if true
                file.write("{0} {1} NCOUL MAP@{1} FAST ".format(species, framework)) #declares it'll use a map
                for j in elements:
                    file.write("{0}@PMAP@{1}.{0}.p " .format(j, framework)) #writes your map names. WARNING - it's a 18 character limit on map names, and no '-' allowed!
                file.write("\n\n")
            else:
                file.write("{0} {1} NCOUL BASIC LJ FAST".format(species, framework)) #pairwise fluid-framework interactions
                file.write("\n\n")
            if emap == True: #using a MOF-fluid coulomb map if true
                file.write("{0} {1} COUL MAP@{1} FAST ".format(species, framework)) #declares it'll use a map
                for j in elements:
                    file.write("{0}@EMAP@{1}.{2}.e " .format(j, framework, 'Probe')) #writes your map names. WARNING - it's a 18 character limit on map names, and no '-' allowed!
                file.write("\n\n")
            else:
                file.write("{0} {1} COUL OFF\n" .format(species, framework)) #pairwise coulomb interactionsdon't work. Trust me, this is better.
                logger.warning("Framework-fluid coulombic interactions are off") #warns you your coulomb fluid-framework stuff isn't happeneing
        if isinstance(species, str) == True:
            file.write("\n{0} {0} NCOUL BASIC LJ FAST\n{0} {0} COUL SUM FAST EWALD KMAX@15 KAPPA@6.7\n\n" .format(species))  #fluid fluid interactions, with Wolf coulombic interactions 
        elif isinstance(species, list) == True: 
            for i in species:
                file.write("\n{0} {0} NCOUL BASIC LJ FAST\n{0} {0} COUL SUM FAST EWALD KMAX@15 KAPPA@6.7\n\n" .format(i))  #fluid fluid interactions, with Wolf coulombic interactions 
    logger.info("Sorb-Sorb file written!")    

#Writes your intramolecular file
def IntraWriter(logger, species, elements, framework, dxout = "./mapgen"):
    with open("{0}/intramolecular_file" .format(dxout), "w") as file:
        file.write("####This intramolecular interaction file is workable for making a map of {0} in a 2x2x2 IRMOF-1 framework\n" .format(species)) #outdated preamble
        if isinstance(species, list) == True: #Lets you handle multiple fluids at once
            for i in species:
                file.write("Intra: {0}\n" .format(i)) #no intramolecular interactions at all
        elif isinstance(species, str) == True: 
            file.write("Intra: {0}\n" .format(species))
        for i in list(elements): #lets you create a map of every element in your system. Having the extra lines in your production script don't matter
            file.write("Intra: {0}\n".format(i))
        file.write("Intra: {0}" .format(framework)) #No framework intramolecular interactions happening
    logger.debug("Intra file written!")

#bundles the above 3 functions into one for simplicity
def Intsetup(logger, mol_name, elements, framework, forcefield, dxout, pmap = True, emap = False):
    AtmAtmMover(logger, elements, forcefield,None, 18, framework, dxout)
    SorbSorbWriter(logger, mol_name, elements, framework, dxout, pmap, emap)
    IntraWriter(logger, mol_name, elements, framework, dxout)

#Writes a bashscript for making your maps, compatible with SLURM
def MapMakeRunWriter(logger, species, elements, parentdir, dxout = "./mapgen", filelocation='../../../../'):
    with open("{0}/run.mapmaker" .format(dxout), "w") as file:
        file.write("""#!/usr/bin/env bash 
#SBATCH --job-name={0}.map
#SBATCH --partition=batch
#SBATCH --nodes=1
#SBATCH --tasks-per-node=1
#SBATCH --time=06:00:00
#SBATCH --mail-type=END
#SBATCH --mail-user=jrhm21@bath.ac.uk #Change this to your email!
#SBATCH --account=free

# -- Set up the environment
module purge
module load group ce-molsim stack
module load music/std

# -- Find my directory
cd {2}
cd {1}\n #Change this bit to get to your directory
export ATOMSDIR={3}atoms
export MOLSDIR={3}molecules/tension/
export PMAPDIR={3}maps/{0}
export EMAPDIR={3}maps
\n\n# --Run\n""" .format(species, dxout, parentdir, filelocation))
        for count, i in enumerate(elements, 1):
            file.write("music_mapmaker makemap_{0}.ctr > logfile{1}\n" .format(i, count, parentdir))
    os.chmod("{0}/run.mapmaker" .format(dxout), 0o777) #chmod so it's fully rwx for everyone (0o denotes the following number is in octal)
    logger.debug("Runfile written!")   

#Writes a control file for making your maps
def mapctrlfilewriter(logger, species, framework, dxout = "./mapgen"):
    x = random.randint(0,99999) #makes the random seed 
    with open("{0}/makemap_{1}.ctr" .format(dxout, species), "w") as file:
        file.write("""------ General Information ------------------------------------------
{0}_pmap map probe in {2}
1               # No. of iterations
1               # No. of steps between writes to output/log file
1               # No. of steps between writes to crash file
1               # No. of steps between writes to config. file
2                    # Start numbering simulations from .
{1}              # Iseeed
4                    # specifies contents of config file,
{0}.res           # Restart File to write to
{0}.con           # Configuration File
------ Atomic Types --------------------------------------------------
5                                  # number of atomic types

{0}
{0}.atm
           
Carbon                            # atom type
Carbon.atm                        # basic atom info file

Oxygen                             # atom type
Oxygen.atm                        # basic atom info file

Hydrogen                             # atom type
Hydrogen.atm                        # basic atom info file

Zinc                         # atom type
Zinc.atm                        # basic atom info file
------ Molecule Types -------------------------------------------------
2                       # number of sorbate types

{0}                   # sorbate 
{0}.mol               # sorbate coordinates file

{2}                    # sorbate 
{2}.mol                # sorbate coordinates file
------ Simulation Cell Information --------------------------------------
{2}                   # Fundamental cell type
2, 2, 2                 # No. of unit cells in x, y, z direction
1, 1, 1                 # (1 = Periodic) in x, y, z
------ Forcefield Information -------------------------------------------
BASIC
SPC
atom_atom_file     # atom-atom interaction file 
sorb_sorb_file     # sorbate-sorbate interaction file (optional)
intramolecular_file  # intramolecular interactions
------ Mapmaker Information --------------------------------------------
1              # Number of maps to make

{2}           # Sorbate to map
{0}       # Sorbate to probe map with
NCOUL LJ       # Interaction type to map
0.2            # Approximate grid spacing (Ang)
100.0          # High end potential cutoff (kJ/mol)
{2}.{0}.p           # Map filename or AUTO
------ Configuration Initialization -------------------------------------
{0}                            # Sorbate_Type  
Molecule NULL                              # Source Filename
{2}                            # Sorbate_Type
Fixed NULL                     # Source Filename""" .format(species, x, framework))
    logger.debug("Map control file written!")

#calculates your saturation pressure using the antoine equation parameters from the preamblem, so you can autogenerate pressures
def pSat(logger, species, T):#This function checks the Species is there and that you're in the right temperature range. 
    chemical = species.split("_")[0]
    logger.debug(chemical)
    press = 1 #pressure defaults to 1 kPa
    if chemical in Antoine.Antoine:                              #It then reads out your saturation pressure in kPa for the user benefit and to get the rest of the program working
        if Antoine.Antoine[chemical][3] <= T <= Antoine.Antoine[chemical][4]:
            logger.info("I have Antoine parameters for {0} at {1} K." .format(chemical, T))
        elif T < Antoine.Antoine[chemical][3]:
            logger.warning("WARNING, that temperature is too low for my Antoine parameters for {0} (my limit is {1} K). I'll persevere anyway." .format(chemical, Antoine.Antoine[chemical][3]))
        elif Antoine.Antoine[chemical][4] < T:
            logger.warning("WARNING, that temperature is too high for my Antoine parameters for {0} (my limit is {1} K). I'll persevere anyway." .format(chemical, Antoine.Antoine[chemical][4]))
        else:
            logger.warning("Something weird happened, I've probably got a bug. Oops!")
        press = 100*(10**(Antoine.Antoine[chemical][0]-(Antoine.Antoine[chemical][1]/(Antoine.Antoine[chemical][2]+T))))
        press = round(press, 3)
        logger.info('Calculated saturation pressure is {0}'.format(press))
    else:
        logger.warning("I'm sorry, I don't have that species in my database.")
    return press

#Creates a log-linear pressure isotherm using the saturation pressure from pSat and preamble defined number of points/predefined points 
def isothermcalculator(logger, pressure, n, minrelpress, maxrelpress):                                  #This function produces the isotherm as a data list, using pSat calculated by function pSat and the user defined isotherm length n
    try:
       FixedPoints[-1] 
    except NameError:
        logger.warning('I failed to find your Fixed simulation points. I\'m going to continue on with the assumption that you didn\'t want any!')
        FixedPoints = []
    m = n-len(FixedPoints)-1                                         #Pressure points re linearly distributed above 0.09 kPa to pSat, which may eb a bad idea. Who knows?
    AllPoints = [] #your list of total pressures
    for point in range(1,n+1):
        relpress = minrelpress+(point*(maxrelpress-minrelpress)/n) #Calculates your partial pressure list between minrelpress and 0. Dividing the (point/float(n+1)) statement by a number lowers your max pressure to a fraction of the sat pressure
        press = pressure*10**relpress #converts the above into absolute pressures
        press = round(press, 5) #Rounds the float to 3 decimal points, for simplicity
        AllPoints.append(press)
        AllPoints.sort() #sorts your points into ascending order
    isotherm = AllPoints
    logger.info("PSat = {0}KPa\n Isotherm = {1}".format(pressure, isotherm))
    return isotherm
    
#this function write s a .dat file based on the pressures you've calculated in isothermcalculator
def PressureFileWriter(logger, species, T, Pressure, isotherm, dirout):                 # This function writes your file to a specific directory, but currently doesn't say if your temp is out of range. Annoying!!#
#    print(isotherm)
    with open("{0}/pressure.{1}.{2}.dat" .format(dirout, species, T), "w") as file:
        file.write("{0!s} #PSat at {1!s} K is {2:1.3f} kPa. \n" .format(species, T, Pressure))
        file.write(str(len(isotherm)) + "\n")
        file.write(', '.join(str(thing) for thing in isotherm))
    logger.info('Pressure file {0}/pressure.{1}.{2}.dat successfully written!'.format(dirout, species, T))

#Writes a bashscript for taskfarming yuor simulations, compatible with SLURM
def TaskfarmRunWriter(logger, species, T, framework, parentdir, dirout, number = 16):
    with open("{0}/run.taskfarmer" .format(dirout), 'w') as file:
        file.write("""#!/usr/bin/env bash
#SBATCH --job-name={0}.{2}.{1}k
#SBATCH --partition=batch
#SBATCH --nodes=1
#SBATCH --tasks-per-node=16
#SBATCH --time=12:00:00
#SBATCH --mail-type=END
#SBATCH --mail-user=jrhm21@bath.ac.uk #Change this to your email!
#SBATCH --account=re-ce1100

# -- Set up the environment
module purge
module load group ce-molsim stack
module load taskfarmer

#locate the experiment directory
cd {4} #Change this bit to get to your directory
cd {3}

# -- Run
mpirun -np 16 taskfarmer -f taskfarm

# -- python analyse the results
python ./isothermextractor.py
#python ./ResultsParser.py""".format(species, T, framework, dirout, parentdir)) #uses my isotherm extractor script to pull all your isotherms out and put them into .csv files in the directory abocve the taskfarm run file
    os.chmod("{0}/run.taskfarmer" .format(dirout), 0o777)   #chmod so it's fully rwx for everyone (0o denotes the following number is in octal)
    with open("{0}/taskfarm" .format(dirout), 'w') as file1: #Writes the commands to taskfarm into a separate taskfram file
        for i in range(1,number+1):
            file1.write("bash ./{0:02d}/run.gcmc\n".format(i))
    with open("{0}/taskfarm.backup" .format(dirout), 'w') as file2: #writes a backup you can restore the above to if there's a bug
        for i in range(1,number+1):
            file2.write("bash ./{0:02d}/run.gcmc\n".format(i))
    logger.debug("Taskfarm things file written!")

#Writes a python script to extract data from your experiments after taskfarming
def IsothermExtractMover(logger, species, T, framework, dirout, n = 20):
    with open("{0}/isothermextractor.py" .format(dirout), 'w') as file:
        file.write("""from os import listdir
import numpy as np

isotherms = np.zeros((17, {3})) # this is a matrix of a 16x {3}-point isotherms with identical pressure values

def isothermextract(page): #this function scrapes subdirectories of name for isotherm data for a given isotherm file name
    z = 0
    z = int(page) #sets z to be your taskfarm run
#    print(z)
    drx = str("./{{0:02d}}/isotherm.{0}" .format(page)) # defines the isotherm
#    print(drx)
    f = open(drx, "r")
    x = []
    page = [] #redefines your input variable, but somehow still works
    for curline in f:
        if "#" not in curline:
            fish = curline.split()
            if len(fish) == 2:
                x.append(fish[0])
                page.append(fish[1])
    f.close()
    isotherms[0] = x #overwrites the first column of your np.zeros matrix to be your pressure oiubts
    isotherms[z] = page #overwrites column z in your np.zeros matrix with resutls from isotherm z
    return isotherms
#####################################################################################################################################################    
for i in range (1,17): #loops over your 16 isotherms
    SuccessCount = 0 #this variable preents issues downstream if some of your simulations fail
    filenames = listdir("./{{0:02d}}/" .format(i))
    for file in filenames:
        if file == "isotherm.{0}": #change the np.zeroes to be your results
            isothermextract(i)
            SuccessCount += 1
    if SuccessCount != 1: #Checks the abovle was successful
        print("WARNING! Something went wrong in simulation {{0:02d}}" .format(i)) #complains if not

x = isotherms[0] #This is now getting a mean and stdev of your results. Still works even if some of your simulations don't!
avg=[]
stdev=[]

avgisotherms = np.zeros((3, {3})) #now makes a 3x{3} matrix to hold your averaged data and stdevs
for i in range (0, {3}):
    spread = []
    for thing in range (2,17):
        if isotherms.item((thing, i)) != 0: #prevents your failed simulations from messing things up
            spread.append(isotherms.item((thing, i)))
    avg.append(np.mean(spread)) #gets your mean
    stdev.append(np.std(spread)) #gets your stdev
avgisotherms[0] = x
avgisotherms[1] = avg
avgisotherms[2] = stdev
np.savetxt("../{2}.{1}.{0}.results.csv", avgisotherms, delimiter=",") #output it to this file""" .format(species, T, framework, n))
    logger.debug("Python isotherm extractor written!")

#Writes a .ctr file for your gcmc simulations (production and for getting your final configuarions at the end). This currently cannot cope with multiple sorbents at once, creates a 2x2x2 cell, and has bias insert atom softcoded in
def GcmcControlChanger(logger, species, elements, T, n, framework, dirout, iterations = '750000', Restart = None, name = 'gcmc.ctr', pressure = 'file'):
    logger.debug(elements)
    x = random.randint(0,99999) #sets your random seed
    if isinstance(elements, list): #lets you use multi-element sorbent molecules
        n_species = len(Forcefield.MOF_el_list[framework])+len(elements) #sets the total number of element types in your gcmc
    elif isinstance(elements, int): #lets you use single-element sorbents
        n_species = len(Forcefield.MOF_el_list[framework])+elements
    with open("{0}/{1}" .format(dirout, name), 'w') as file:
        file.write("""#This control file was written by the python scrpt umbrellav2-6 You probably ought to check me before use!
------ General Information ------------------------------------------
{0} molecule in {1} 
{4}              # No. of iterations, defaults to 750000
50000                # No. of steps between writes to output/log file
100000                # No. of steps between writes to crash file
2500                  # No. of steps between writes to config. file
{5}                   # Start numbering simulations fromhere. defaults to 1 for production run, 30 for config runs
{2}                #random seed
3                    # specifies contents of config file, outdated?
{1}.{0}.res         # Restart File to write to
{1}.{0}.con          # Configuration File
------ Atomic Types --------------------------------------------------
{3}                    # number of atomic types            \n\n""".format(species, framework, x, n_species, iterations, 1 if Restart == None else 30))
        for i in elements:
            file.write("""{0}   #atom type\n{0}.atm #atom file name\n\n""".format(i))
        for i in Forcefield.MOF_el_list[framework]:
            file.write("""{0}   #atom type\n{0}.atm #atom file name\n\n""".format(i))
        file.write("""------ Molecule Types -------------------------------------------------
2                    # number of sorbate types

{0}               # sorbate
{0}.mol           # sorbate coordinates file

{1}                # sorbate
{1}.mol             # sorbate coordinates file
------ Simulation Cell Information ------------------------------------
{1}                # Fundamental cell file
2, 2, 2              # No. of unit cells in x, y, z direction
1, 1, 1              # (1 = Periodic) in x, y, z
------ Forcefield Information -------------------------------------------
BASIC
SPC
atom_atom_file       # atom-atom interaction file
sorb_sorb_file       # sorbate-sorbate interaction file
intramolecular_file  # intramolecular interaction file/specification
------ Ideal Parameters -----------------------------------------------
Ideal                # Equation of State
1                    # no. of sorbates
{0}              # Sorbate Name
------ GCMC Information -----------------------------------------------
1                 # No. of iterations
{2}              # temperature
Ideal Parameters   # Tag for the equation of state (NULL = Ideal Gas)
{3}                  # No. of simulation points
5000                # Block size for statistics
1                  # no. of sorbates
          -------------------------
{0}            # Sorbate Name
{5}           #  pressure
Null               # sitemap filename (Null = no sitemap)
4                  # no of gcmc movetypes
1.0, 1.0, 1.0, 1.0      # move type weights
RINSERT                   # type of move.1
RDELETE                   # type of move.2
RTRANSLATE                # type of move.4
0.2, 1                    # Delta Translate, adjust delta option (0=NO, 1=YES)
RROTATE
0.2, 1
------ Configuration Initialization -------------------------------------
{0}             # Sorbate_Type\n""".format(species, framework, T, n, elements[-1], 'pressure.{0}.{1}.dat'.format(species, T) if pressure == 'file' else pressure)) #lets you define your pressure points manually, defaults to the values generated from the isothermgenerator
        if Restart == None: #for if you're starting a new isotherm
            file.write("""GCMC NULL
{0}              # Sorbate_Type
FIXED NULL
--------  Main Datafile Information --------
Energy, position, pair_energy  # contents of datafile""".format(framework))
        else: #lets you get configs from a main isotherm res file
            file.write("""{0}
{1}              # Sorbate_Type
FIXED NULL
--------  Main Datafile Information --------
Energy, position, pair_energy  # contents of datafile""".format(Restart, framework))
    logger.debug("GCMC control file written!")

def GcmcControlChanger2(dirout, filename, species, sorbent, n_iter, start_num, sorbent_elements, framework_elements, sorbent_molecules, 
framework_molecules,n_a, n_b, n_c,aa_filename, mm_filename, intra_filename, T,isotherm_length, pressure_points, restartfile, movie_name):
    with open("{0}/{1}" .format(dirout, filename), 'w') as f:
        f.write(GenInfo(species, sorbent, n_iter, start_num))
        f.write(AtomTypes(sorbent_elements, framework_elements))
        f.write(MoleculeTypes(sorbent_molecules, framework_molecules))
        f.write(SimCell(framework_molecules, n_a, n_b, n_c))
        f.write(ForcefieldInfo(aa_filename, mm_filename, intra_filename))
        f.write(IdealParams(sorbent_molecules))
        f.write(GCMCInfo(T, isotherm_length, sorbent_molecules, pressure_points))
        f.write(ConfigInit(sorbent_molecules, framework_molecules, restartfile = None))
        f.write(DatafileInfo())
        f.write(Movie(movie_name, n_iter))
            
def GenInfo(species, sorbent, n_iter, start_num):
    seed = random.randint(0,99999) #sets your random seed
    output = '''#This control file was written by the python scrpt umbrellav2-6 You probably ought to check me before use!
------ General Information ------------------------------------------
{0} molecule in {1} 
{2}              # No. of iterations, defaults to 750000
{3}                # No. of steps between writes to output/log file
{4}                # No. of steps between writes to crash file
{5}                  # No. of steps between writes to config. file
{6}                   # Start numbering simulations fromhere. defaults to 1 for production run, 30 for config runs
{7}                #random seed
3                    # specifies contents of config file, outdated?
{1}.{0}.res         # Restart File to write to
{1}.{0}.con          # Configuration File\n'''.format(species, sorbent, n_iter, max(10000, n_iter/20), max(100000, n_iter/10), max(1000, n_iter/1000), start_num, seed)
    return output

def AtomTypes(sorbent_elements, framework_elements):
    output = []
    output.append('''------ Atomic Types --------------------------------------------------
{0} #total number of atom types\n\n'''.format(len(sorbent_elements)+len(framework_elements)))
    for i in sorbent_elements:
        output.append("""{0}   #atom type\n{0}.atm #atom file name\n\n""".format(i))
    for i in framework_elements:
        output.append("""{0}   #atom type\n{0}.atm #atom file name\n\n""".format(i))
    return ''.join(output)

def MoleculeTypes(sorbent_molecules, framework_molecules):
    if isinstance(sorbent_molecules, str):
        sorbent_molecules = [sorbent_molecules]
    if isinstance(framework_molecules, str):
        sorbent_molecules = [framework_molecules]
    output = []
    output.append('''------ Molecule Types -------------------------------------------------
{0}                    # number of sorbate types\n\n'''.format(len(sorbent_molecules)+len(framework_molecules)))
    for i in sorbent_molecules:
        output.append("""{0}   #mol type\n{0}.mol #mol file name\n\n""".format(i))
    for i in framework_molecules:
        output.append("""{0}   #mol type\n{0}.mol #mol file name\n\n""".format(i))
    return ''.join(output)

def SimCell(framework_molecules, n_a, n_b, n_c):
    output = '''------ Simulation Cell Information ------------------------------------
{0}.mol                # Fundamental cell file
{1}, {2}, {3}              # No. of unit cells in x, y, z direction
1, 1, 1              # (1 = Periodic) in x, y, z\n'''.format(framework_molecules, n_a, n_b, n_c)
    return output

def ForcefieldInfo(aa_filename, mm_filename, intra_filename):
    output = '''------ Forcefield Information -------------------------------------------
BASIC
SPC
{0}       # atom-atom interaction file
{1}       # sorbate-sorbate interaction file
{2}  # intramolecular interaction file/specification\n'''.format(aa_filename, mm_filename, intra_filename)
    return output

def IdealParams(sorbents_used):
    output = '''------ Ideal Parameters -----------------------------------------------
Ideal                # Equation of State
{0}                    # no. of sorbates
{1}              # Sorbate Name\n'''.format(len(sorbents_used, ', '.join[sorbents_used]))
    return output

def GCMCInfo(T, isotherm_length, sorbents_used, pressure_points):
    output = []
    output.append('''------ GCMC Information -----------------------------------------------
1                 # No. of iterations
{0}              # temperature
Ideal Parameters   # Tag for the equation of state (NULL = Ideal Gas)
{1}                  # No. of simulation points
5000                # Block size for statistics
{2}                  # no. of sorbates\n\n'''.format(T, isotherm_length, len(sorbents_used)))
    for i in sorbents_used:
        output.append('''{0}            # Sorbate Name
{1}           #  pressure
Null               # sitemap filename (Null = no sitemap)
4                  # no of gcmc movetypes
1.0, 1.0, 1.0, 1.0      # move type weights
RINSERT                   # type of move.1
RDELETE                   # type of move.2
RTRANSLATE                # type of move.4
0.2, 1                    # Delta Translate, adjust delta option (0=NO, 1=YES)
RROTATE
0.2, 1\n\n'''.format(i, pressure_points))
    return ''.join(output)

def ConfigInit(sorbents_used, framework, restartfile = None):
    output = []
    output.append('''------ Configuration Initialization -------------------------------------\n''')
    for i in sorbents_used:
        output.append('''{0}             # Sorbate_Type
        {1}\n'''.format(i, 'GCMC NULL' if restartfile == None else restartfile))
    output.append('''{0}             # Sorbent
        FIXED NULL\n'''.format(framework))
    return ''.join(output)
    
def DatafileInfo():
    output = '''--------  Main Datafile Information --------
Energy, position, pair_energy  # contents of datafile\n\n'''
    return output

def Movie(movie_name, n_iter):
    output = '''------ Movie Information ----------------------------------------------
{0}           # Movie filename (output is in xyz format)
0, {1}            # Starting step, ending step
{2}                    # Steps between frames
No                   # Include zeolite in movie: Yes or No
1, 1, 1             # Number of unit cell repeates to dump in x, y, z directions
-----------------------------------------------------------------------\n\n
'''.format(movie_name, n_iter, max(100,n_iter, n_iter/1000))
    return(output)

#Writes a .ctr file for your postprocessing. There;s orobably a better way than using music_post, but I'm not there yet
def PostControlChanger(logger, species, n, framework, dirout, prefix, cutoff = '0'):
    with open("{0}/{1}.post.ctr" .format(dirout, 'full' if cutoff == '0' else 'trunc'), 'w') as file:
        file.write("""####This section is apparently required for working with any post code. Who knows why? not me!
#
#
------------------------------------------------------------
   ### Required section ######
-- Post Processor Information ------------
GCMC                            # Type of simulation GCMC, NVTMC , MD ....
./{1}.{0}.con                    # basename for config files
1, {2}                          # first and last file numbers
{5}.post.ctr.out                       # name for new ctrlfile that will regenerated
{3:02d}.{5}.postfile          # Base name for output files
{4}, 0                         # Percentages of data to skipped at start and end 


# The sections below are necessary only if you want the corresponding 
# analysis performed
# ---------------- ALL OF THEM ARE OPTIONAL ------------------------


####    This section is reqd for energy averages in your post code output files
####    as of now only total enrgies vs sim. step
------ Post : Energy Average Info -----------------------------------
100       # Number of blocks into which data should be divided for stats

####    This section is reqd for Loading averages in your post code outputfiles
####    as of now only species loading vs sim. step (for all species)
------ Post : Loading Average Info -----------------------------------
100       # Number of blocks into which data should be divided for stats""".format(species, framework, n, prefix, cutoff, 'full' if cutoff == '0' else 'trunc'))
    logger.debug("Post control file written!")

#Writes a bashscript for actually running your simulations, compatible with SLURM
def GcmcRunWriter(logger, species, T, framework, parentdir, dirout, isotherm, iteration, intlocation='../', atomfilelocation='../../../../', molfilelocation='../../../../',mapfilelocation='../../../../'):
    with open("{0}/run.gcmc".format(dirout), 'w') as file:
        file.write("""#!/usr/bin/env bash
#SBATCH --job-name={0}.{1}
#SBATCH --partition=batch
#SBATCH --nodes=1
#SBATCH --tasks-per-node=1
#SBATCH --time=06:00:00
#SBATCH --mail-type=END
#SBATCH --mail-user=jrhm21@bath.ac.uk #Change this to your email!
#SBATCH --account=free

# -- Set up the environment
module purge
module load group ce-molsim stack
module load music/std

#locate the experiment directory
cd {3} #Change this bit to get to your directory
cd {2}

#create symbolic links to your interactions files and pressure file
ln -s {6} atoms
ln -s {7} molecules
ln -s {8} maps

ln -s {5}atom_atom_file atom_atom_file
ln -s {5}intramolecular_file intramolecular_file
ln -s {5}sorb_sorb_file sorb_sorb_file
ln -s {5}pressure.{0}.{1}.dat pressure.{0}.{1}.dat

#set the paths for atoms, molecules, pmaps, and emaps WARNING softcoded in umbrella, check it's right!
export ATOMSDIR=atoms
export MOLSDIR=molecules
export PMAPDIR=maps
export EMAPDIR=maps


# -- Run
music_gcmc gcmc.ctr > logfile.gcmc #runs your main simulation
music_post full.post.ctr > logfile.post.full #runs your postprocessing
cp {4:02d}.full.postfile {5}fullpostfiles/
music_post trunc.post.ctr > logfile.post.trunc #runs your postprocessing
cp {4:02d}.trunc.postfile {5}truncpostfiles/


""".format(species, T, dirout, parentdir, int(iteration), intlocation, atomfilelocation, molfilelocation, mapfilelocation))
        for i, value in enumerate(isotherm): #for each isotherm point you're simulating
            file.write('music_gcmc {0}kpa_restart.ctr > {1}_restart.logfile\nmv finalconfig.xyz {3:02d}.{2}.{0}kpa.xyz\n'.format(value, int(i)+1, framework, int(iteration))) #set up a simulation to generate a new xyz file and move it to be names after the pressure point 
    os.chmod("{0}run.gcmc" .format(dirout), 0o777) #chmod so it's fully rwx for everyone (0o denotes the following number is in octal)
    logger.debug("Run file written!")