Shantanu's Blog

There are a lot of images you must be having in gmail. One way to find those images is to search for attachments and then browse through the emails to find an image that you are looking for.
Welcome to a chrome extension called "gmail to drive". There will be a newly added Save To Drive link next to each image.

https://chrome.google.com/webstore/detail/gmail-attachments-to-driv/epoohehjbaenldfbahgcegdmlogakgin

1) Enable big query API from console as shown in the image. You may need to enable "google cloud storage" as well for file uploads.

Enable big query API

2) Goto big query home page, web interface to big query...

https://bigquery.cloud.google.com/

3) If you need to share a dataset with someone, choose "share dataset" link on the drop down arrow next to dataset name.

4) If someone has shared a dataset with you, use the colon in the table path as shown below.

select date_time, requests from 567402616005:company.JUL23B limit 10;

You can also use an alias like shantanuoak:company.JUL23B

Python copies references not the data items.

>>> a = [2,3,4]
>>> c = a
>>> b = a[:]

When I append the list a, c will will be appended as well but not b. Because a and c share the same object.
>>> a.append(11)
>>> a
[2, 3, 4, 11]
>>> b
[2, 3, 4]
>>> c
[2, 3, 4, 11]

And therefore when I delete object a, b and c does not get deleted.
"b" does not get deleted because it has its own identity as seen the example mentioned above. But "c" is the copy of a and if c is appended along with a, it should also get deleted once a is removed. right?

Variables are only pointers to the data object. a and c points to the same list and when a is removed, c is still pointing to it and therefore the object is not removed. This is not a bug, python is designed to save memory.

The concept of cloud service is not new. But buiding a service based on another cloud service is something interesting. sky sql will initiate an instance in AWS ec2 cloud after taking permission from you.

https://config.skysql.com

In order to protect your security, it is better to create a new Access Key that can be shared with the company. You can select the Security Credentials option under "My Account" menu to get the access and security keys.

This will save time of starting and configuring instances. You will like the service if you trust the company :)

#skysql and #xeround are 2 cloud database solutions worth checking out.

I have a text file that I need to change the column sequence of.

# cat customer.csv
customerno, firstname, lastname, sales
23242, john, doe, 2345.00
23253, jane, doe, 1234.00
23221, greg, johnson, 2345.00
23210, howard, gardner, 2345.00

I need the last column first and so on. Data should look something like this...
sales, lastname, firstname, customerno
2345.00, doe, john, 23242
1234.00, doe, jane, 23253
2345.00, johnson, greg, 23221
2345.00, gardner, howard, 23210
_____

The 18 lines PHP script to exchange CSV columns is shown below.

$fp_out = fopen("customer2.csv", "w+");

if (($fp_in = fopen("customer.csv", "r")) !== FALSE)
{
    $exchange = array("2-3", "1-4");

    while (($data = fgetcsv($fp_in, 1000, ",")) !== FALSE)
    {
        foreach($exchange as $e)
        {
             $fields = explode("-", $e);
             $temp = $data[$fields[0] - 1];
             $data[$fields[0] - 1] = $data[$fields[1] - 1];
             $data[$fields[1] - 1] = $temp;
        }   

        fputcsv($fp_out,$data,',', '"');
    }

    fclose($fp_in);
    fclose($fp_out);
}

_____

As you can see there are a lots of functions used and arrays are not easy to understand.
Here is 10 feet long python if you are not afraid of forests !

f = open('customer.csv')
out = ''
for line in f:
    line = line.strip().split(', ')
    line.reverse()
    out += ', '.join(line) + '\n'
f.close()

f = open('customer2.csv', 'w')
f.write(out)
f.close()

A more sophisticated example will be something like this...

import csv
with open('customer2.csv', 'wb') as output:
    input = csv.reader(open('customer.csv', 'rb'))
    output = csv.writer(output, dialect=input.dialect)
    for line in input:
        try:
            (customerno, firstname, lastname, sales) = line
        except ValueError:
            print 'valueerror'
        else:
            outLine = (sales, firstname, lastname, customerno)
            output.writerow(outLine)

_____

And one liner awk is not so bad :)

awk -F", " '{print $4",", $3",", $2",", $1}' customer.csv > customer2.csv

>>> l1 = [1,2,3]
>>> l2 = ["a", "b", "c", "d"]

# matching tuples
>>> zip(l1, l2)
[(1, 'a'), (2, 'b'), (3, 'c')]

# match the list count
>>> [i for i in itertools.izip_longest(l1,l2)]
[(1, 'a'), (2, 'b'), (3, 'c'), (None, 'd')]

# make it a list
>>> [i for i in itertools.chain(*itertools.izip_longest(l1,l2))]
[1, 'a', 2, 'b', 3, 'c', None, 'd']

# remove None
>>> [i for i in itertools.chain(*itertools.izip_longest(l1,l2)) if i is not None]
[1, 'a', 2, 'b', 3, 'c', 'd']