makefile: execute a command and grep then awk

CP=cp
LIB_UNIXODBC=/usr/src/tpkgs/unixodbc/2.3.1/linux86w/lib/libodbc.so
RELEASE_DESTDIR=/bld/release/nsr/fb_mssql_linux/linux86w/source


$(CP) $(LIB_UNIXODBC) $(RELEASE_DESTDIR)/ddbda/odbc/$(shell objdump -p $(LIB_UNIXODBC) |grep
'SONAME' |awk -F ' ' '{print $$2}')

which equals to command line:

cp -f /usr/src/tpkgs/unixodbc/2.3.1/linux86w/lib/libodbc.so /bld/release/nsr/fb_mssql_linux/linux86w/source/ddbda/odbc/libodbc.so.2

Note:

1. shell to execute a command in a makefile

2. not like that in bash command line, the grep string is marked with single quotes.

3. there are double '$' in the awk statement.

pandas read_csv from https with Python 3.6.4

On Mac OSX, if you are using Python 3.6 and pandas to try to read a csv file via https:

california_housing_dataframe = pd.read_csv("https://storage.googleapis.com/mledu-datasets/california_housing_train.csv", sep=",")
california_housing_dataframe.describe()

 you may get an error like:
urllib.error.URLError:

To fix this issue:

Open a terminal and take a look at:

/Applications/Python 3.6/Install Certificates.command

Python 3.6 on MacOS uses an embedded version of OpenSSL, which does not use the system certificate store. More details here.

(To be explicit: MacOS users can probably resolve by opening Finder and double clicking Install Certificates.command)

Or read https csv with a workaround:

from io import StringIO

import pandas as pd
import requests


url = "https://storage.googleapis.com/mledu-datasets/california_housing_train.csv"
s = requests.get(url).text
c = pd.read_csv(StringIO(s))
print(c.head())

numpy.array vs numpy.asarray

Looking at the definition, you'll see the difference between them:

def asarray(a, dtype=None, order=None):
    return array(a, dtype, copy=False, order=order)

The main difference is that array (by default) will make a copy of the object, while asarray will not unless necessary.

The difference can be demonstrated by this example:

1. generate a matrix

   >>> A = numpy.matrix(np.ones((3,3)))
   >>> A
   matrix([[ 1.,  1.,  1.],
           [ 1.,  1.,  1.],
           [ 1.,  1.,  1.]])

2. use numpy.array to modify A. Doesn't work because you are modifying a copy

   >>> numpy.array(A)[2]=2
   >>> A
   matrix([[ 1.,  1.,  1.],
           [ 1.,  1.,  1.],
           [ 1.,  1.,  1.]])

3. use numpy.asarray to modify A. It worked because you are modifying A itself

   >>> numpy.asarray(A)[2]=2
   >>> A
   matrix([[ 1.,  1.,  1.],
           [ 1.,  1.,  1.],
           [ 2.,  2.,  2.]])

Install Emacs on Mac OSX with brew

$ brew cask install emacs
Reference: https://www.gnu.org/software/emacs/download.html#macos

If you run into this error while access Desktop/Documents/Downlaods directory:


Here is the fix:

Python None vs Empty list

>>> a=[]
>>> b=None
>>> type(a),type(b)
(<class 'list'>, <class 'NoneType'>)
>>> not a, not b
(True, True)
>>> a is None, b is None
(False, True)
>>> a is not None, b is not None
(True, False)
>>> a,b
([], None)

best way to check if a list is empty in Python3

Do it with:

if not a:
    print("a is an empty list.")

instead of:

if len(a):
    print("a is an empty list.")

Reference: Official Python programming recommendations
See the discussions on Stack Overflow

/ vs // in python3

Code speaks:

>>> a=b=5
>>> a,b
(5, 5)
>>> type(a), type(b)
(<class 'int'>, <class 'int'>)
>>> a/=2
>>> b//=2
>>> a,b
(2.5, 2)
>>> type(a), type(b)
(<class 'float'>, <class 'int'>)
>>> alist = list(range(10))
>>> alist
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]
>>> alist[:a]
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: slice indices must be integers or None or have an __index__ method
>>> alist[:b]
[0, 1]

Note if you are trying calculating the indices with '/' you would get trouble as showed above.