def main():
gb.to_log += """data in = {data_in}
data out = {data_out}\n\n""".format(data_in=gb.org_data,
data_out=gb.select_rev)
with open(log, 'w') as out:
out.write(gb.to_log)
spssaux.OpenDataFile(gb.org_data)
spss.SetOutput("off")
db = pd.DataFrame.from_csv(gb.input_csv)
db2 = db.loc[db[gb.version].notnull() | db[gb.single_items].notnull()
| db['rename'].notnull() | db['recode'].notnull()]
vars, to_log2 = mod_database(db2, gb.prefix, gb.start, gb.stop)
cmd = prepare_data(vars)
with open(syntax_save, 'w') as out:
out.write(cmd)
list_of_all_vars = vars['items'].values.tolist()
vars_in_file = spssaux.VariableDict().Variables
to_log1 = str(len(list_of_all_vars)) + ' variables extracted.\n'
print(to_log1)
with open(log, 'a') as out:
out.write(to_log1)
how_often(vars, vars_in_file, gb.prefix, gb.start, gb.stop)
reverse_items.main(vars, log, syntax_save)
cmd = extract_vars(list_of_all_vars, gb.urvalsinfo, gb.select_rev,
gb.keep_vars)
with open(syntax_save, 'a') as out:
out.write(cmd + '\n')
#spss.Submit(cmd)
with open(log, 'a') as out:
out.write(to_log2)
print("--- %s seconds ---" % (time.time() - start_time))
def main():
spssaux.OpenDataFile(data)
spss.SetOutput("off")
spss.Submit('SET TNumbers=Values ONumbers=Labels OVars=Labels.')
db_prefixed = mod_database(db_glob, prefix, start, stop)
#print(recode_cut(db_prefixed))
spss.Submit(recode_cut(db_prefixed))
#print(make_long_exp_vars())
spss.Submit(make_long_exp_vars())
spss.Submit("SAVE OUTFILE='%s'\n/COMPRESSED." % save_data)
def recode():
cmd = ''
spssaux.OpenDataFile(imputations + '/COMPLETE.SAV')
#cmd += ("""SORT CASES BY Imputation_.
#SPLIT FILE LAYERED BY Imputation_.\n""")
recode_db = db_glob.loc[db_glob['recode2'].notnull()]
cmd += '\n'.join([
'RECODE {prefix}{var} {recode}.'.format(var=pos[0],
recode=pos[1],
prefix=gb.prefix + str(i))
for pos in recode_db[['items', 'recode2']].values.tolist()
for i in range(gb.start, gb.stop)
])
cmd += '\nEXECUTE.\n'
print(cmd)
return (cmd)
def main():
spssaux.OpenDataFile(data)
spss.SetOutput("off")
spss.Submit('SET TNumbers=Values ONumbers=Labels OVars=Labels.')
spss.Submit('SORT CASES BY Imputation_.\nSPLIT FILE LAYERED BY Imputation_.')
db_prefixed = mod_database(db_glob,prefix,start,stop)
get_median_cut()
input_indep = db_prefixed.query('dependent != 1')
spss.Submit(dicotomize(input_indep))
#spss.SetOutput('on')
spss.Submit(make_analysis_exposure(input_indep))
spss.Submit("SAVE OUTFILE='COMPLETE_RECODED_BINARY.sav'\n/COMPRESSED.")
outdb,outdb_err = calculate_odds(db_prefixed,start,stop,prefix)
outdb.to_csv('{version}_{cutmod}.csv'.format(version=version,cutmod=cut_mod))
outdb_err.to_csv('error_{version}_{cutmod}.csv'.format(version=version, cutmod=cut_mod))
#save_output(outdb)
spss.Submit("SAVE OUTFILE='%s'\n/COMPRESSED." % save_data)
def main():
os.chdir(imputations)
spssaux.OpenDataFile(gb.select_rev)
cmd = 'DATASET NAME orginal.\n'
cmd += 'cd "{cwd}".\n'.format(cwd=imputations)
spss.SetOutput("off")
cmd += fix_value_lables()
with open(imputation_syntax, 'w') as out:
out.write(cmd)
spss.Submit(cmd)
run_cmd = impute_item_for_item()
with open(imputation_syntax, 'a') as out:
out.write('\n' + run_cmd)
spss.Submit(run_cmd)
match_files()
cmd = recode()
with open(imputation_syntax, 'a') as out:
out.write('\n' + cmd)
spss.Submit("""SAVE OUTFILE = '{imputed_data}'
/COMPRESSED.\n""".format(imputed_data=gb.imputed_data))
def add_suffix(folder):
for item in ["/no_strings", "/strings"]:
datafiles = get_filelist(folder + item, 'sav')
for file in datafiles:
print(file)
exclude = ['kod_id'] #Ange namnet på id_variablen
spssaux.OpenDataFile(file)
basename = os.path.basename(file).strip('.sav')
suffix = basename #önskat suffix
print(basename)
vars = spssaux.VariableDict().variables
for i in exclude:
if i in vars:
vars.remove(i)
oldnames = spssaux.VariableDict().expand(vars)
newnames = [varnam + "_" + suffix for varnam in oldnames]
spss.Submit('rename variables (%s=%s).' %
('\n'.join(oldnames), '\n'.join(newnames)))
spss.Submit("""
SAVE OUTFILE = "%s%s".
DATASET CLOSE ALL.
NEW FILE.
""" % (folder + item + '/suffix/', basename + '.sav'))
def fix_files(folder, datevars, id_variable, date):
datafiles = get_filelist(folder, "sav")
for fil in datafiles:
exclude = []
spssaux.OpenDataFile(fil)
vars_in_file = spssaux.VariableDict().variables
rename_vars(vars_in_file)
spss.Submit("ALTER TYPE %s (f8)." % id_variable)
stringvars = get_stringvars()
print(stringvars)
vars_in_file = spssaux.VariableDict().variables
base, ext = os.path.basename(fil).split('.')
for var in stringvars:
if var in vars_in_file:
exclude.append(var)
if "AnswerDate" in vars_in_file:
spss.Submit("""
COMPUTE time_days = DATEDIFF(AnswerDate, DATE.MDY(%s), "day").
EXECUTE.
""" % date)
for var in vars_in_file:
if 'mean' not in var and var not in exclude + datevars:
spss.Submit("ALTER TYPE %s (f8)." % var)
spss.Submit("VARIABLE ALIGNMENT %s(right)." % var)
spss.Submit("VARIABLE LEVEL %s(scale)." % var)
if var != id_variable:
spss.Submit("RECODE %s (SYSMIS=999)." % var)
spss.Submit("MISSING VALUES %s (999)." % var)
spss.Submit("SORT CASES BY %s (A)." % id_variable)
spss.Submit("""COMPUTE data_from_week=%s.
EXECUTE.
ALTER TYPE data_from_week (f8).
ALTER TYPE time_days (f8).
""" % base)
spss.Submit(save_and_close(base, exclude, folder))
spss.Submit(("NEW FILE."))
def main():
spssaux.OpenDataFile(data)
#spss.SetOutput("off")
spss.Submit('SET TNumbers=Values ONumbers=Labels OVars=Labels.')
outdb = calculate_odds(start, stop, prefix)
outdb.to_csv('long_vars_ut.csv')
import time
start_time = time.time()
"""
Checks availability of variables in dataset and makes new dataset from selection variable and list of variables.
"""
from vars_and_functions import global_vars as gb
from vars_and_functions import reverse_items
import spss
import spssaux
import pandas as pd
import time
log = gb.folder + 'out/check_variables_log.txt'
syntax_save = gb.folder + 'out/check_variables.sps'
spssaux.OpenDataFile(gb.org_data)
test = spssaux.VariableDict()
print(test)
vars = ['x4_6aevyt01', 'x4_6aevyt02']
for var in reversed(vars):
i = test[var].index
dataCursor = spss.Cursor([i])
oneVar = dataCursor.fetchall()
#extending the example to get the actual list of values.
uniqueList = list((set(oneVar)))
uniq_vals = [
int(x[0]) for x in uniqueList
if x[0] != None and x[0] not in gb.missing_int
]
uniq_vals_rev = reversed(uniq_vals)
import SpssClient
import spssaux
import spss
import csv
import re
SpssClient.StartClient()
# User defined variables
input_file = r'user_input_1'
output_file = r'user_input_2'
output_var_vals_file = r'user_input_3'
output_var_info_file = r'user_input_4'
# Load sav-file.
spssaux.OpenDataFile(input_file)
# Count number of columns
varCount = spss.GetVariableCount()
caseCount = spss.GetCaseCount()
print('There are %d variables in this file' % varCount)
print('There are %d cases. Please check this matches number of cases in the output file' % spss.GetCaseCount())
# Clean file: only string columns
for ind in range(varCount):
varName = spss.GetVariableName(ind)
if spss.GetVariableType(ind) > 0:
print("String Variable: %s" % varName)
for ascii_ind in range(33):
spss.Submit('compute %s = REPLACE(%s, STRING(%d,PIB), " ")' % (varName, varName, ascii_ind))
spss.Submit('EXECUTE.')