def nycep_acs_extract_tracts(dpath=None):
# -- defaults
if dpath==None:
print("Must set ACS data path!!!")
return
d5path = os.path.join(dpath,'2012','5')
# -- utilities
opath = os.path.join('../output','acs','2012','5')
# -- get the file list
flist = sorted([i for i in os.listdir(d5path) if '000.txt' in i])
# -- get the Logical Record Numbers for NYC census tracts
logrecno = nycep_acs_tract_logrecno(dpath=dpath)
# -- loop through the files and extract the appropriate lines
for fl in flist:
in_file = open(os.path.join(d5path,fl),'r')
out_file = open(os.path.join(opath,fl),'w')
print("Writing file {0}...".format(os.path.join(opath,fl)))
dum = [out_file.write(i) for i in in_file if
i.split(',')[5] in logrecno]
in_file.close()
out_file.close()
return
def nycep_acs_read_table(tlabel, year=2012, summary=5, dpath=None,
margins=False):
# -- check path
if dpath==None:
print("Must set data path to ACS data!")
return
# -- utilities
eom = 'm' if margins else 'e'
sfile = os.path.join(dpath,str(year),str(summary),
'Sequence_Number_and_Table_Number_Lookup.txt')
# -- read in the sequence number file
fopen = open(sfile,'r')
slines = [line for line in fopen if 'CELL' in line]
fopen.close()
# -- get the sequence number, start, and # of cells for the requested table
recs = slines[[line.split(',')[1] for line in slines].index(tlabel)
].split(',')
seqnum = recs[2]
cstart = int(recs[4])-1
ncell = int(''.join([i for i in recs[5] if i.isdigit()]))
# -- read the estimates (or margins) file
dfile = os.path.join(dpath,str(year),str(summary),
eom+str(year)+str(summary)+'ny'+seqnum+'000.txt')
fopen = open(dfile,'r')
data = {'fileid' : '',
'filetype' : '',
'stusab' : '',
'chariter' : '',
'sequence' : '',
'logrecno' : [],
'vals' : []}
if summary==5:
logrecno = nycep_acs_tract_logrecno(dpath=dpath)
dlines = []
for line in fopen:
recs = line.split(',')
if recs[5] in logrecno:
data['fileid'] = recs[0]
data['filetype'] = recs[1]
data['stusab'] = recs[2]
data['chariter'] = recs[3]
data['sequence'] = recs[4]
data['logrecno'].append(recs[5])
data['vals'].append(recs[cstart:cstart+ncell])
elif summary==1:
data = [line.split(',')[cstart:cstart+ncell] for line in fopen]
else:
print("Only 1 and 5 year summaries supported!!!")
return
fopen.close()
# -- return data
return data
