Shantanu's Blog

Database Consultant

August 28, 2012

monitoring mysql error log

The most important aspect of mysql that needs to be monitored is error log file. MySQL will write almost all important errors to this file. Here is a cron that will monitor it every hour and let the DBA know when the file is updated.

02 * * * * tail -1000 `mysqladmin variables | grep log_error | awk '{print $4}'` | grep -e "`date +'\%y\%m\%d \%k' --date='1 hour ago'`" | mail -s "crashed lo from `hostname`" dba@company.com > /home/alert_mail_success.txt 2> /home/alert_mail_err.txt

1) Replace "mail -s" with "mail -E -s" so that blank (empty) mails will not be sent.
2) The "-E" option may not work with older mail version. But the following cron should work...

02 * * * * tail -1000 `mysqladmin variables | grep log_error | awk '{print $4}'` | grep -e "`date +'\%y\%m\%d \%k' --date='1 hour ago'`" > myerror.txt; if [[ -s myerror.txt ]]; then cat myerror.txt | mail -s "last hour error log from `hostname`" dba@company.com ; fi > /home/alert_mail_success.txt 2> /home/alert_mail_err.txt

It will not send a mail if the error log extract file (myerror.txt) is empty.

posted by shantanu # Tuesday, August 28, 2012 0 comments

August 24, 2012

Online Utilities

Regular Expressions demystified

There are times when using regular expression is the only way to find the text from a given string.
Here is a really useful utility that will explain and at the same time help us to write complex regular expressions.

http://gskinner.com/RegExr/?31u07
_____

SQL formatter

The following utility will help to format a complex SQL query.

http://www.dpriver.com/pp/sqlformat.htm

Labels: mysql tips, usability

posted by shantanu # Friday, August 24, 2012 0 comments

August 22, 2012

Python tips 9 - SQLite DB

import sqlite3 as sqlite

conn = sqlite.connect(":memory:")
conn.execute("CREATE TABLE A (name integer PRIMARY KEY AUTOINCREMENT, address varchar(100))")
conn.execute("insert into A values (22, 'india')")
conn.execute("insert into A values (23, 'usa')")
a = conn.execute("SELECT * FROM A").fetchall()

for i in a:
print i[1]

# will print this...
#india
#usa

# The cursor retains the id of the last row inserted in the lastrowid attribute

import sqlite3
connection = sqlite3.connect(':memory:')
cursor = connection.cursor()
cursor.execute('CREATE TABLE foo (id integer primary key autoincrement ,
username varchar(50),
password varchar(50))')
cursor.execute('INSERT INTO foo (username,password) VALUES (?,?)',
('test', 'test'))
print(cursor.lastrowid)

# fetchone() returns the first row from the select statement as a tuple, so fetchone()[0] will return the first (and only) column in the first (and only) row, i.e. the max(id).

cursor = sqlite3.execute('SELECT max(id) FROM table_name')
max_id = cursor.fetchone()[0]

_____

# it is also possible to update a value based on some other column

cursor.execute('SELECT column2 FROM Table ORDER BY column2')
column2 = [row[0] for row in cursor]
for j, c in zip(lst, column2):
cursor.execute('UPDATE Table SET column = ? WHERE column2 = ?', [j, c])

Labels: python

posted by shantanu # Wednesday, August 22, 2012 0 comments

Python tips 8 - import module

It is possible to directly import a module and initialize the class.

>>> exec(urllib2.urlopen('http://www.ics.uci.edu/~eppstein/PADS/UnionFind.py').read())
>>> uf = UnionFind()

We can use wget or curl
$ wget http://www.ics.uci.edu/~eppstein/PADS/UnionFind.py

>>> from UnionFind import UnoinFind
>>> uf = UnionFind()

Labels: python

posted by shantanu # Wednesday, August 22, 2012 0 comments

August 20, 2012

python tips 7 - List comprehension

If you are looking for concise, pythonic and elegant way of writing a loop then list comprehension is the right choice.

## normal loop to print numbers till 10 except 2 and 3
# expected output
# [1, 4, 5, 6, 7, 8, 9, 10]

t = 0
l = []
while t < 10:
    t += 1
    if t == 2 or t == 3: continue
    l.append(t)

## list comprehension # 6 lines in 1
a = [x for x in range(1,10) if x != 2 and x != 3]
## a is now a list and can be used in for x in a:

# use unpacking in list comprehension
dict((s, v) for (o, s), v in mydict.iteritems() if o is co1)

Labels: python

posted by shantanu # Monday, August 20, 2012 0 comments

August 19, 2012

Python tips 6 - loops

# following has no loop for it is printed only once
def thisLines(n):
    mv = 0
    if n > mv:
        print "testme22"
        n += 1

thisLines(10)

# Using while loop
def printLines(nr):
    myva = 0
    while nr > myva:
        print "testme"
        myva += 1

printLines(10)

# Using recursive function:

def nLines(n):
    if n > 0:
        print "test"
        nLines(n-1)

nLines(10)

Labels: python

posted by shantanu # Sunday, August 19, 2012 0 comments

August 18, 2012

python tips 5

'''
save the URL list in a file and then copy the source code to a text file
'''

import urllib

class mydict():
   def myurllist():
           fp = open('URLS-HERE.txt', "w")
           fp.write('http://google.com/index.html')
           fp.close()

   def read_myurllist():
           for line in open('URLS-HERE.txt','r'):
              if line.startswith('http'):
                     fichier = open("other.txt", "w")
                     allhtml = urllib.urlopen(line)
                     fichier.write(allhtml.read())
                     fichier.close()

m = mydict()
m.myurllist
m.read_myurllist

fp1 = open('other.txt', 'r')
for myline in fp1.readlines():
   print myline
fp1.close()

_____

parse_qs is used to Parse a query string. Data are returned as a dictionary.

import urllib2
import urlparse

url = 'http://www.tip.it/runescape/gec/price_graph.php?avg=1&start=1327715574&mainitem=10350&item=10350'
response = urllib2.urlopen(url)
content = response.read()
params = urlparse.parse_qs(content)
print(params['values'])

posted by shantanu # Saturday, August 18, 2012 0 comments

August 16, 2012

python with NumPy

NumPy for windows can be downloaded from...
http://sourceforge.net/projects/numpy/files/NumPy/

## np allows to construct array differently:
import numpy as np
z = np.array(zip([1,2,3,4,5], ['a','b','c','d','e']), dtype=[('int', int), ('str', '|S1')])
np.savetxt('test.txt', z, fmt='%i %s')

posted by shantanu # Thursday, August 16, 2012 0 comments

August 10, 2012

Python tips 4

# Find the ten most common words in abc.txt

def cat(filename):
f = open(filename, 'rU')
text = f.read()
wordlist = text.split()
for myword in wordlist:
oldval = d.get(myword, 0)
d[myword] = oldval + 1

t = sorted(d.iteritems(), key=lambda x:-x[1])[:3]
for x in t: print "{0}: {1}".format(*x)

# Better

def cat(filename):
f = open(filename, 'rU')
text = f.read()
wordlist = text.split()
return collections.Counter(wordlist).most_common(3)

# Best
import re
wordlist = re.findall('\w+', open('abc.txt').read().lower())
collections.Counter(wordlist).most_common(3)

## Counter ##

>>> from collections import Counter

>>> c=Counter({2:2, 3:1, 5:1,9:4})
>>> type(c)

>>> list(c.elements())
[9, 9, 9, 9, 2, 2, 3, 5]

_____

If I need to total the values for each key, I can use the counter method available in collections.

d = {'apple': {'a': 1, 'b': 4, 'c': 2}, 'orange': {'a': 4, 'c': 5}, 'pear': {'a': 1, 'b': 2}}
# expected results
# Counter({'c': 7, 'a': 6, 'b': 6})

from collections import Counter
one = sum(map(Counter, d.values()), Counter())
two = sum((Counter(v) for v in d.itervalues()), Counter())
# using loop to count
three = Counter()
for v in d.values():
    three.update(v)

Or it is also possible to use default dictionary method as shown below.

from collections import defaultdict
de = defaultdict(int)
for dct in d.values():
    for k,v in dct.items():
        de[k] += v
print(de)

Labels: python

posted by shantanu # Friday, August 10, 2012 0 comments

python tips 3

## list ##
## A list can contain anything like integer, string, tuple, another list etc.

>>> a = [1,2,'abc']
>>> a
[1, 2, 'abc']

## Variables are only containers or pointers to the real stuff.
## So when a is appended, b will also change. But when a is removed b is not removed because it will still point to the old data.

>>> b = a
>>> a.append(4)
>>> b
[1, 2, 'abc', 4]
>>> del a
>>> a

Traceback (most recent call last):
File "", line 1, in a
NameError: name 'a' is not defined

>>> b
[1, 2, 'abc', 4]

## sort ##

>>> sort
Traceback (most recent call last):
File "", line 1, in sort
NameError: name 'sort' is not defined

>>> sorted
built-in function sorted

>>>help sorted
SyntaxError: invalid syntax
>>> help (sorted)
Help on built-in function sorted in module __builtin__:
sorted(...)
sorted(iterable, cmp=None, key=None, reverse=False) --> new sorted list

## sorted is a function that returns the sorted list while a.sort() is a method that will sort data in-place.

>>> a = ['amar', 'akbar', 'anthony']
>>> sorted(a)
['akbar', 'amar', 'anthony']
>>> sorted(a, reverse=True)
['anthony', 'amar', 'akbar']
>>> def Last(s): return s[-1]
>>> sorted(a, key=Last, reverse=True)
['anthony', 'amar', 'akbar']

## reverse sort is possible as well as custom sort can be achived using a function like "Last" in this case

## slicing ##

>>> b = ':'.join(a)
>>> b
'amar:akbar:anthony'
>>> b.split(':')
['amar', 'akbar', 'anthony']

## tuples ##
>>> x = (1,2,3)
>>> type(x)
type 'tuple'
>>> a = (1,2,3)
>>> len(a)
3
>>> a[0]
1

## looping tuple values in a list called result
>>> result = []
>>> type (result)
type 'list'
>>> for s in x: result.append(s)
>>> result
[1, 2, 3]

## or simply use the built-in function list to convert a tuple to list

>>> x = (1,2,3)
>>> y = list(x)

>>> type (x)
type 'tuple'
>>> type (y)
type 'str'

## Lists are mutable while tuples and strings are not.
## it means tuple values can not be changed just like list values

>>> x[0] = 9

Traceback (most recent call last):
File "", line 1, in
a[0] = 9
TypeError: 'tuple' object does not support item assignment

## dictionary ##

>>> d = {}
>>> d['a'] = 'alpha'
>>> d['o'] = 'omega'
>>> d['g'] = 'gamma'
>>> d
{'a': 'alpha', 'g': 'gamma', 'o': 'omega'}
>>> d['a']
'alpha'
>>> d.get('a')
'alpha'

>>> d['x']
Traceback (most recent call last):
File "", line 1, in d['x']
KeyError: 'x'

>>> d.get('x')
>>> d.get('x', 'mydefault')
'mydefault'

>>> 'a' in d
True
>>> 'x' in d
False

>>> d.keys()
['a', 'g', 'o']

>>> d.values()
['alpha', 'gamma', 'omega']

>>> d.items()
[('a', 'alpha'), ('g', 'gamma'), ('o', 'omega')]

>>> for k in d: print 'key:', k, '->', d[k]
key: a -> alpha
key: g -> gamma
key: o -> omega

>>> for tuple in d.items(): print tuple
('a', 'alpha')
('g', 'gamma')
('o', 'omega')

>>> for k in sorted(d.keys()): print 'key:', k, '->', d[k]
key: a -> alpha
key: g -> gamma
key: o -> omega

posted by shantanu # Friday, August 10, 2012 0 comments

August 09, 2012

python tips 2

# first, second and third ranks
rank = range(1,4)
# player names
player = ['amar','akbar','anthony']
# link the ranks with players
result = dict(zip(rank,player))
# find who is third
result[3]
# find what is the rank of amar
[k for k, v in result.iteritems() if v == 'amar'][0]

# the values can also be assigned to a tuple
(amar, akbar, anthony) = range(1,4)

zip can also be used to transpose data
>>> rows = [[1, 1, 2, 3], [4, 5, 6, 7], [1, 2, 5, 7], [6, 9, 8, 5]]

>>> zip(*rows)
[(1, 4, 1, 6), (1, 5, 2, 9), (2, 6, 5, 8), (3, 7, 7, 5)]

Labels: python

posted by shantanu # Thursday, August 09, 2012 0 comments

Shantanu's Blog

August 28, 2012

monitoring mysql error log

August 24, 2012

Online Utilities

August 22, 2012

Python tips 9 - SQLite DB

Python tips 8 - import module

August 20, 2012

python tips 7 - List comprehension

August 19, 2012

Python tips 6 - loops

August 18, 2012

python tips 5

August 16, 2012

python with NumPy

August 10, 2012

Python tips 4

python tips 3

August 09, 2012

python tips 2

Archives