def gda2csv(gda, dest):
if isinstance(gda, basestring):
with open(gda, 'rb') as f:
gda, header = gff4.read_gff4(f)
elif not isinstance(gda, gff4.Structure):
gda, header = gff4.read_gff4(gda)
close = False
if isinstance(dest, basestring):
f = open(dest, 'wb')
close = True
try:
csvw = csv.writer(f)
csvw.writerow([colname(col) for col in gda[10002]])
csvw.writerow([coltypes[col[10999]] for col in gda[10002]])
for row in gda[10003]:
csvw.writerow([colvalue(v) for k, v in sorted(row.iteritems())])
finally:
if close:
f.close()
n = data[index]
shift = 0
if e == -1:
break
sb.append(unichr(-e - 1))
return sb.tounicode()
def _dump(htlk, out):
table = htlk[19007]
e = len(table) / 2 - 1
for n in htlk[19008]:
for bit in xrange(32):
e = table[e * 2 + (n & 1)]
n >>= 1
if e < 0:
out.write(unichr(-e - 1).encode('utf_16le'))
e = len(table) / 2 - 1
if __name__ == '__main__':
from gff4 import read_gff4, write_gff4
import sys
with open(sys.argv[1], 'rb') as f:
g, h = read_gff4(f)
tlk = htlk2tlk(g)
with open(sys.argv[2], 'wb') as f:
write_gff4(f, tlk, tlkheader)
return '$%08x' % col[10001]
coltypes = {
0: 'string',
1: 'int',
2: 'float',
3: 'bool',
4: 'resource',
}
def colvalue(v):
if v is None:
return '****'
elif isinstance(v, unicode):
return v.encode('utf_8')
else:
return v
with open(sys.argv[1], 'rb') as f:
gff, header = gff4.read_gff4(f)
with open(sys.argv[2], 'wb') as f:
csv = csv.writer(f)
csv.writerow([colname(col) for col in gff[10002]])
csv.writerow([coltypes[col[10999]] for col in gff[10002]])
for row in gff[10003]:
csv.writerow([colvalue(v) for k, v in sorted(row.iteritems())])
Structrows = gff4._structtype(rowtype.fourcc, rowtype.fields, rowtype.size)
field10003 = gff4.Field(field.label, gff4._listtype(Structrows, False), False, field.offset)
elif field.label == 10002:
field10002 = gff4.Field(field.label, ListStructcolm, False, field.offset)
Structgtop = gff4._structtype(header.structs[0].fourcc, (field10002, field10003), header.structs[0].size)
header = header._replace(structs=(Structgtop, Structcolm, Structrows))
gtop = Structgtop()
colms = gtop[10002]
existing = []
for colm in data[10002]:
colm = Structcolm(colm)
if not colm[10000]:
colname = lookupcolname(colm[10001], cache)
if colname is not None:
colm[10000] = colname
else:
existing.append(colm[10000])
colms.append(colm)
gtop[10003].extend(Structrows(row) for row in data[10003])
addnames(existing)
return gtop, header
if __name__ == '__main__':
import sys
with open(sys.argv[1], 'rb') as f:
g, h = gff4.read_gff4(f)
annotate(g, h)
gff4._DEBUG_COMPARISONS = True
t = clock()
with open(filename, 'rb') as f:
mem = StringIO(f.read())
print 'file read in %.2f seconds' % (clock() - t)
t = clock()
gff = LazyGFF4(mem)
data, header = gff.root, gff.header
print 'data loaded in %.2f seconds' % (clock() - t)
#from gff4.toyaml import gff2yamlevents
#import yaml
#yaml.emit(gff2yamlevents(data, header, True), sys.stdout)
t = clock()
hold = StringIO()
write_gff4(hold, data, header)
print 'data dumped in %.2f seconds' % (clock() - t)
t = clock()
hold.seek(0)
data2, header2 = read_gff4(hold)
print 'data reloaded in %.2f seconds' % (clock() - t)
t = clock()
if data != data2:
print 'data dumped is not the same as that that was loaded'
print 'data compared in %.2f seconds' % (clock() - t)