Date series process

1. Pick a slice in Panda time index

   a.index  #pandas.core.indexes.datetimes.DatetimeIndex
  np.where(np.logical_and(a.index>'20190506',a.index<'20190508'))
  # but remember, the value return is a multidimension array
  
2. datetime format array and using logical_or and logical_and

time = np.array([dt.datetime.strptime('{}-{}-{}T{}:{}'.format(int(data[i,0]),int(data[i,1]),int(data[i,2]),int(data[i,4]), \
int(data[i,5])),'%Y-%m-%dT%H:%M') for i in range(len(data))])

 time_range = np.logical_or(np.logical_and(time>dt.datetime(2019,1,2),time<dt.datetime(2019,1,12,12)), \
 np.logical_and(time>dt.datetime(2019,1,30,17),time<dt.datetime(2019,2,10,8)))
 
3. <change the shape of array>
   Flatten
   reshape
    Numpy Arrays: Concatenating, Flattening and Adding Dimensions
    https://www.python-course.eu/numpy_changing_dimensions.php
4. Concatenate two array 
  https://cmdlinetips.com/2018/04/how-to-concatenate-arrays-in-numpy/
   
   
5: How to plot the pandas datetime index using matplotlib.pyplot

   See http://pandas.pydata.org/pandas-docs/stable/whatsnew.html#v0-21-0-october-27-2017 / #17710 
   and https://matplotlib.org/faq/howto_faq.html#plot-numpy-datetime64-values
   # add this before plot.
   from pandas.tseries import converter
   converter.register()
6. convert string to datetime: inclding one function convert pandas to datetime

 #https://chrisalbon.com/python/basics/strings_to_datetime/
  from dateutil.parser import parse: directly parse date string
  pd.to_datetime(df['date'])

7. calculate the delta time:
     time = np.append(time, (parse(meta['DATE-OBS'])-time_0)/datetime.timedelta(seconds=1))

8. Timezone aware and unware issu of datetime 
   ---Method 1.
   from datetime import datetime
   from datetime import timezone

   dt = datetime.now()
   dt.replace(tzinfo=timezone.utc)
   ----Method 2.
   import datetime
   import pytz

   unaware = datetime.datetime(2011, 8, 15, 8, 15, 12, 0)
   aware = datetime.datetime(2011, 8, 15, 8, 15, 12, 0, pytz.UTC)  # this is a way method 3.

   now_aware = pytz.utc.localize(unaware) # method 2
   assert aware == now_aware
   print(dt.replace(tzinfo=timezone.utc).isoformat())
   '2017-01-12T22:11:31+00:00'
   -----Case 3: pandas read and localized it
   location = pd.read_csv(file, sep = ' ', skiprows=1, index_col=0, header = None,parse_dates=True)
   import pytz
   location = location.rename_axis( 'Time')
   location.index = location.index.tz_localize(pytz.UTC)