def sortTempList(temp_list):
time = np.array([item[0] for item in temp_list])
posts = np.array([item[1] for item in temp_list])
idxs = time.argsort().tolist()
rst = [[t, p] for (t, p) in zip(time[idxs], posts[idxs])]
del time, posts
return rst
def sort_by_date(data, tpos=2):
"""
sort the data by time, assuming the second element is the time
input: data --- a list of string data
tpos --- the position of the time data
output: numpy array of 7 x m dimention
note: assume that the second element is time of the format of 2007-05-07T11:59:59
"""
mdata = [[], [], [], [], [], [], []]
time = []
for ent in data:
atemp = re.split('\s+', ent)
for k in range(0, 7):
mdata[k].append(atemp[k])
time.append(convert_time(atemp[tpos]))
mdata = numpy.array(mdata)
time = numpy.array(time)
tind = time.argsort()
for k in range(0, 7):
mdata[k] = mdata[k][tind[::]]
return mdata
help="Input HDF file.")
parser.add_option("-p", "--purge", dest="purge",
action=count, default=0,
help="remove entries that might be duplicates")
(options, args) = parser.parse_args()
if len(args) and options.inputFile == None:
options.inputFile = args[0]
if options.debug:
print options
fh = tables.openFile(options.input, mode = "a")
for tbl in fh.iterNodes("/"):
print "Sorting", tbl
time=tbl.col('time')
val=tbl.col('val')
ts=time.argsort()
time=time.take(ts)
val=val.take(ts)
tbl.modifyColumns(names=('time', 'val'), columns=(time, val))
if options.purge:
print "Purging duplicates from", tbl
print "code me"
fh.close()