def prepare_log_data(logdata):
ch = logdata[2]
mtp = logdata[3]
symbols = logdata[4]
xcc = logdata[5]
gcc = logdata[6]
Fcc = logdata[7]
V0 = logdata[8]
zmat = logdata[10]
# MM freq files --> z-matrix not shown in output; so get it from input
if zmat is None: zmat = gau.zmat_from_loginp(folder + log)
# z-matrix?
if zmat is None:
print("WARNING: file %s does not contains Z-matrix!" % (folder + log))
raise Exception
# data from zmat lines
(lzmat, zmatvals, zmatatoms), symbols = gau.convert_zmat(zmat)
# correct symbols (just in case)
symbols, atonums = fncs.symbols_and_atonums(symbols)
# Data without dummies
symbols_wo, xcc_wo = fncs.clean_dummies(symbols, xcc=list(xcc))
symbols_wo, gcc_wo = fncs.clean_dummies(symbols, gcc=list(gcc))
symbols_wo, Fcc_wo = fncs.clean_dummies(symbols, Fcc=list(Fcc))
# return data
return ch, mtp, V0, symbols, symbols_wo, xcc_wo, gcc_wo, Fcc_wo, lzmat, zmatvals
def read_normal_log(ofile, inpvars, case):
# keylvl from inpvars according to case
if case == "LL": keylvl = inpvars._keyll
elif case == "HL": keylvl = inpvars._keyhl
else: raise Exception
# reading data
exception_line = "ERROR! Level '%s' not found in archive section!\n" % keylvl
try:
logdata = gau.read_gaussian_log(ofile, target_level=keylvl)
except LevelNotFound:
print(" " * 11 + exception_line)
sys.exit()
commands, comment, ch, mtp, symbols, xcc, gcc, Fcc, energy, statusFRQ, ozmat, level = logdata
# it may happen that in FRQ calculation zmatrix is not found!
try:
lzmat, zmatvals, zmatatoms = gau.convert_zmat(ozmat)[0]
# put angles in (0,360) (sometimes Gaussian returns an angle > 360)
for ic in inpvars._tic:
zmatvals[ic] %= 360
# Save info
vec = TorPESpoint(tfh.zmat2vec(zmatvals, inpvars._tic),
inpvars._tlimit)
except:
zmatvals = None
zmatatoms = None
vec = None
print(" energy (%s) = %.6f hartree" % (level, energy))
print(" final vector: %s" % str(vec))
if statusFRQ >= 0: print(" num imag frq: %i" % statusFRQ)
return statusFRQ, vec, zmatvals, zmatatoms
def get_logs_vecs(inpvars, dcorr, svec2log={}):
# conformers in folder
logs = folder_logs(inpvars._dirll)
if len(logs) == 0: return svec2log
# find equivalent points
for log in logs:
svec = log.split(".")[1]
if svec in svec2log: continue
svec2log[svec] = log
if inpvars._enantio:
# (a) read log
zmatlines = gau.read_gaussian_log(inpvars._dirll + log)[10]
(lzmat, zmatvals, zmatatoms), symbols = gau.convert_zmat(zmatlines)
# (b) enantiomer
svec2 = str(enantiovec(lzmat, zmatvals, dcorr, inpvars))
svec2log[svec2] = log
# return data
return svec2log
def gen_guess_zmatrix(vec, i_zmatvals, inpvars, log, distance):
# read z-matrix from closest domain
use_default = True
try:
logdata = gau.read_gaussian_log(inpvars._dirll + log)
zmatlines = logdata[10]
(lzmat, zmatvals, zmatatoms), symbols = gau.convert_zmat(zmatlines)
sprint("using Z-matrix from stored file (%s)" % log, tvars.NIBS2 + 4)
sprint("distance to stored Z-matrix: %.0f degrees" % distance,
tvars.NIBS2 + 4)
use_default = False
except:
use_default = True
# get closest z-matrix
if use_default:
sprint("using default Z-matrix (%s)" % inpvars._zmatfile,
tvars.NIBS2 + 4)
zmatvals = {k: v for (k, v) in i_zmatvals.items()}
# Apply vector to current z-matrix
for angle, ic in zip(vec._fvec, inpvars._tic):
zmatvals[ic] = angle
# return zmatvals
return zmatvals
def status_from_log(logfile, inpvars):
'''
About statusOPT, it returns:
* -1 if the calculation failed
* 0 if the calculation succeeded
About statusFRQ, it returns:
* -1 if the frequencies were not found in output (or fails)
* num_imag if the frequencies were found
'''
assert os.path.exists(logfile)
# Log file does not end with Normal termination?
if not gau.normal_termination(logfile):
statusOPT = -1
statusFRQ = -1
vec = None
# Ends with normal termination
else:
logdata = gau.read_gaussian_log(logfile)
statusOPT = 0
statusFRQ = logdata[9]
zmatlines = logdata[10]
if zmatlines is None:
statusOPT = -1
statusFRQ = -1
vec = None
else:
(lzmat, zmatvals, zmatatoms), symbols = gau.convert_zmat(zmatlines)
# put angles in (0,360) (sometimes Gaussian returns an angle > 360)
for ic in inpvars._tic:
zmatvals[ic] %= 360
# final vector
vec = TorPESpoint(tfh.zmat2vec(zmatvals, inpvars._tic),
inpvars._tlimit)
# Return data
return statusOPT, statusFRQ, vec
for log, point in zip(logs, points):
logdata = gau.read_gaussian_log(folder + log, target_level=None)
ch = logdata[2]
mtp = logdata[3]
symbols = logdata[4]
xcc = logdata[5]
gcc = logdata[6]
Fcc = logdata[7]
V0 = logdata[8]
zmat = logdata[10]
if zmat is None:
zmat = gau.zmat_from_loginp(folder + log)
(lzmat, zmatvals, zmatatoms), symbols = gau.convert_zmat(zmat)
# same current point
strconfs.append(str(point))
allconfs.append(point)
# correct symbols (just in case)
symbols, atonums = fncs.symbols_and_atonums(symbols)
# Data without dummies
symbols_wo, xcc_wo = fncs.clean_dummies(symbols, xcc=list(xcc))
symbols_wo, gcc_wo = fncs.clean_dummies(symbols, gcc=list(gcc))
symbols_wo, Fcc_wo = fncs.clean_dummies(symbols, Fcc=list(Fcc))
# calculate weight
atonums_wo = fncs.symbols2atonums(symbols_wo)
masses_wo = fncs.symbols2masses(symbols_wo)
pgroup, rotsigma = get_pgs(atonums_wo, masses_wo, xcc_wo)
def highlevel_reopt(inpvars,cmatrix,\
lowerconformer=0,\
upperconformer=float("inf"),\
calculate=False,\
lNH2=[]):
dirll, dirhl = inpvars._dirll , inpvars._dirhl
if not os.path.exists(dirll):
print(" Folder NOT FOUND: %s\n"%dirll)
return
# Create folders
if not os.path.exists(dirhl):
try : mkdir_recursive(dirhl)
except: pp.print_createdir(dirhl,tvars.NIBS2); exit()
# Check Gaussian software
if calculate:
end_program = itf.check_executable()
if end_program: return
# print status of tests
pp.print_tests(inpvars._tests,1)
pp.print_ifreqconstr(inpvars._ifqrangeHL)
#------------------------#
# Templates for Gaussian #
#------------------------#
# HL opt & HL freq templates for MIN
if not inpvars._ts: file1,file2 = tvars.TEMPLMINOPTHL,tvars.TEMPLMINFRQHL
# HL opt & HL freq templates for TS
else : file1,file2 = tvars.TEMPLTSOPTHL,tvars.TEMPLTSFRQHL
# (a) do templates exist??
if not os.path.exists(file1): pp.print_filenotfound(file1,tvars.NIBS2,1); raise exc.END
if not os.path.exists(file2): pp.print_filenotfound(file2,tvars.NIBS2,1); raise exc.END
# (b) read templates
with open(file1,'r') as asdf: loptHL = asdf.readlines()
pp.print_template(loptHL,file1,"HL optimization",tvars.NIBS2)
with open(file2,'r') as asdf: lfrqHL = asdf.readlines()
pp.print_template(lfrqHL,file2,"HL frequency calculation",tvars.NIBS2)
# (c) temporal folder
if not os.path.exists(inpvars._tmphl):
try : mkdir_recursive(inpvars._tmphl)
except: pp.print_createdir(inpvars._tmphl,tvars.NIBS2); exit()
# read LL energies
if not os.path.exists(tvars.ENERGYSUMLL):
sprint("- file NOT found: %s"%tvars.ENERGYSUMLL,tvars.NIBS2)
sprint(" execute %s using --msho ll"%tvars.PROGNAMEnopy,tvars.NIBS2)
sprint("")
return
sprint("- reading file: %s"%tvars.ENERGYSUMLL,tvars.NIBS2)
sprint("")
llenergies, lltemp = rw.read_llenergies()
# check temperature
if abs(lltemp-inpvars._temp) > 1e-3:
sprint(" something went wrong!",tvars.NIBS2)
sprint("")
sprint(" temperature in file : %.3f K"%(lltemp),tvars.NIBS2)
sprint(" temperature in %s: %.3f K"%(tvars.IFILE,inpvars._temp),tvars.NIBS2)
sprint("")
return
sprint(" temperature for Gibbs free energy: %.3f K"%lltemp,tvars.NIBS2)
# check number of conformers
logs = [fname for fname in os.listdir(inpvars._dirll) if fname.endswith(".log")]
if len(logs) != len(llenergies):
sprint(" something went wrong!",tvars.NIBS2)
sprint(" number of conformer differs!",tvars.NIBS2)
sprint("")
sprint(" # of conformers in file : %i"%len(llenergies),tvars.NIBS2)
sprint(" # of conformers in LL folder: %i"%len(logs),tvars.NIBS2)
sprint("")
return
sprint(" number of low-level conformers : %i"%len(llenergies),tvars.NIBS2)
sprint("")
# check conformers
svecs_llenergies = [svec for count,V0,G,svec in llenergies]
for log in logs:
svec_i = log.split(".")[1]
if svec_i not in svecs_llenergies:
sprint(" something went wrong!",tvars.NIBS2)
sprint("")
sprint(" conformer NOT included: %s"%log,tvars.NIBS2)
sprint("")
return
#---------------------#
# Create HL gjf files #
#---------------------#
nopt = 0
nfrq = 0
nsp = 0
minV0 = llenergies[0][0]
for count,relV0_kcalmol,relG_kcalmol,svec in llenergies:
if not (lowerconformer <= count <= upperconformer): continue
# read LL/HL-correlation file
correlations = rw.read_llhlcorr()
# LL log file
log_prec = "prec.%s.log"%svec
log_stoc = "stoc.%s.log"%svec
if os.path.exists(inpvars._dirll+log_prec):
log = log_prec
prefix1 = "optprec."
prefix2 = "frqprec."
prefix3 = "prec."
elif os.path.exists(inpvars._dirll+log_stoc):
log = log_stoc
prefix1 = "optstoc."
prefix2 = "frqstoc."
prefix3 = "stoc."
else: raise Exception
# read log file
conftuple,symbols,string_fccards = itf.log_data(log,inpvars,inpvars._freqscalLL,dirll)
# unpack conftuple
vecA,V0,V1,G,weight,Qrv,ifreq,lzmat,zmatvals,log = conftuple
sprint("(%i) %s [%s]"%(count,str(vecA),prefix3[:-1]),tvars.NIBS2)
sprint("")
# ONLY CONFORMERS IN ENERGY WINDOWS
sprint("LL electronic energy: %.4f kcal/mol"%relV0_kcalmol,tvars.NIBS2+4)
# ONLY CONFORMERS BELOW HLCUTOFF IN GIBBS
if inpvars._hlcutoff is not None:
relG = (G-minG)*KCALMOL
sprint("Gibbs energy: %.3f kcal/mol"%relG,tvars.NIBS2+4)
if relG > inpvars._hlcutoff:
sprint("hlcutoff : %.3f kcal/mol"%inpvars._hlcutoff,tvars.NIBS2+4)
sprint("HL optimization will not be carried out!",tvars.NIBS2+4)
sprint()
continue
sprint()
# SKIP IF ALREADY CALCULATED
if str(vecA) in correlations.keys():
vecB = correlations[str(vecA)]
# possible names for log files
saved_log1 = inpvars._dirhl+"prec.%s.log"%str(vecB)
saved_log2 = inpvars._dirhl+"stoc.%s.log"%str(vecB)
# check existence
bool1 = os.path.exists(saved_log1)
bool2 = os.path.exists(saved_log2)
# exists? then skip
if bool1 or bool2:
sprint("already in %s"%inpvars._dirhl,tvars.NIBS2+4)
sprint("optimized conformer: %s"%str(vecB),tvars.NIBS2+4)
sprint()
continue
sprint()
#-----------------#
# OPT CALCULATION #
#-----------------#
# Perform HL optimization
data = (loptHL,prefix1+str(vecA),inpvars,lzmat,zmatvals,"HL",string_fccards)
if calculate:
sprint("optimization...",tvars.NIBS2+4)
# execute Gaussian
ofileOPT,statusOPT,dummy,vecB,zmatvals,zmatatoms = itf.execute(*data)
sprint()
# opt ends ok?
if statusOPT == -1: continue
else:
sprint("optimization file:",tvars.NIBS2+4)
generated, ifile, ofile, err = itf.generate_gjf(*data)
sprint("- %s"%ifile,tvars.NIBS2+8)
if generated:
sprint("- generated!",tvars.NIBS2+8,1)
nopt += 1
continue
sprint("- not generated!",tvars.NIBS2+8,1)
logdata = gau.read_gaussian_log(ofile)
commands,comment,ch,mtp,symbols,xcc,gcc,Fcc,energy,statusFRQ,ozmat,level = logdata
lzmat,zmatvals,zmatatoms = gau.convert_zmat(ozmat)[0]
for ic in inpvars._tic: zmatvals[ic] %= 360
vecB = TorPESpoint(tfh.zmat2vec(zmatvals,inpvars._tic),inpvars._tlimit)
# The optimization succedeed!
sprint("analysing geometry...",tvars.NIBS2+4)
sprint("- optimized point: %s"%str(vecB),tvars.NIBS+8,1)
# NH2 inversion
zmatvals,inversion = tfh.correct_NH2_inversion(lzmat,zmatvals,zmatatoms,lNH2)
if inversion:
svecB1 = str(vecB)
vecB = TorPESpoint(tfh.zmat2vec(zmatvals,inpvars._tic),inpvars._tlimit)
svecB2 = str(vecB)
if svecB1 != svecB2:
sprint("- NH2 inversion detected! Exchanging H atoms...",tvars.NIBS+8)
sprint("- final vector: %s"%svecB2,tvars.NIBS+8,1)
bools = tfh.precheck_geom(lzmat,zmatvals,cmatrix,inpvars)
if (inpvars._tests[1][0] == 1) and not bools[0]:
sprint("- connectivity test is negative!",tvars.NIBS+8,1)
continue
if (inpvars._tests[1][1] == 1) and not bools[1]:
sprint("- torsion angle out of domain!",tvars.NIBS+8,1)
continue
if (inpvars._tests[1][2] == 1) and not bools[2]:
sprint("- hard constraint test is negative!",tvars.NIBS+8,1)
continue
if (inpvars._tests[1][3] == 1) and not bools[3]:
sprint("- soft constraint test is negative!",tvars.NIBS+8,1)
continue
# Already saved?
pointsHL = tfh.folder_points(inpvars._dirhl)
inlist, equalto = vecB.is_in_list(pointsHL,inpvars._epsdeg)
if inlist:
# exists?
sprint("already in %s (%s)"%(inpvars._dirhl,str(equalto)),tvars.NIBS2+4)
sprint()
# rename stoc --> prec if required!
for ext in "log,zmat".split(","):
# possible names for log files
saved_log1 = inpvars._dirhl+"prec.%s.%s"%(str(equalto),ext)
saved_log2 = inpvars._dirhl+"stoc.%s.%s"%(str(equalto),ext)
# check existence
bool1 = os.path.exists(saved_log1)
bool2 = os.path.exists(saved_log2)
# if this is prec and the one obtained is stoc, replace name!!
if bool2 and "prec" in prefix3: move(saved_log2,saved_log1)
# save LL-->HL correlation
correlations[str(vecA)] = str(vecB)
rw.add_llhlcorr(str(vecA),str(vecB))
continue
sprint()
#-----------------#
# FRQ CALCULATION #
#-----------------#
data = (lfrqHL,prefix2+str(vecA),inpvars,lzmat,zmatvals,"HL")
if not calculate:
sprint("frequency file:",tvars.NIBS2+4)
generated, ifile, ofile, err = itf.generate_gjf(*data)
sprint("- %s"%ifile,tvars.NIBS2+8)
if generated:
sprint("- generated!\n",tvars.NIBS2+8,1)
nfrq += 1
continue
# log with normal termination!
else:
sprint("- not generated!\n",tvars.NIBS2+8,1)
statusFRQ,vecB_frq,dummy1,dummy2 = itf.read_normal_log(ofile,inpvars,"HL")
ofileFRQ = ofile
# Perform HL frequency
else:
sprint("frequency calculation...",tvars.NIBS2+4)
ofileFRQ,dummy,statusFRQ,vecB_frq,dummy,dummy = itf.execute(*data)
#============#
# Save files #
#============#
bool1 = (statusFRQ == 0 and not inpvars._ts)
bool2 = (statusFRQ == 1 and inpvars._ts)
# check imag freq
if bool2 and inpvars._ifqrangeHL != []:
ifreq = abs( fncs.afreq2cm(itf.get_imag_freq(ofileFRQ,inpvars._freqscalHL)) )
isok = fncs.float_in_domain(ifreq,inpvars._ifqrangeHL)
if not isok:
sprint("imaginary frequency is: %.2fi cm^-1"%ifreq,tvars.NIBS2+4)
bool2 = False
if bool1 or bool2:
nsp += 1
# create z-matrix file
dst = inpvars._dirhl+prefix3+"%s.zmat"%str(vecB)
write_zmat(dst,lzmat, zmatvals)
# copy frq file
src = ofileFRQ
dst = inpvars._dirhl+prefix3+"%s.log"%str(vecB)
if not os.path.exists(dst): copyfile(src, dst)
# save LL-->HL correlation
correlations[str(vecA)] = str(vecB)
rw.add_llhlcorr(str(vecA),str(vecB))
sprint()
# Print number of calculations
pp.print_numcalcs(count,nopt,nfrq,nsp)
if not calculate:
sprint("Number of opt gjf files generated: %i"%nopt,tvars.NIBS2)
sprint("Number of frq gjf files generated: %i"%nfrq,tvars.NIBS2)
sprint()
else:
# rewrite correlations to remove repetitions
correlations = rw.read_llhlcorr()
rw.write_llhlcorr(correlations)