Python:Strings

Strings are immutable, all methods return a new string

Basics

str1 + str2

Return concatenation of str1 and str2

str1 += str2

Append str2 to str1

str1 * 3

Return str1 3 times

str(<anything>)

Convert <anything> to a string

NOTE: str(Decimal(9.0)) has a different result ('9') than str(Decimal('9.0')) ('9.0') as the specified precision will be in the string too. Floats (9.0) are not exact enough to do this.

Formatting

Basic

str1.replace(old,new[,cnt])

Return str1 with old replaced by new (cnt times).

For more complex replacements, like regular expressions, see re.sub below.

str1.strip(<chars>)

Return str1 with all trailing and leading <chars> removed. If <chars> is omitted all trailing and leading whitespaces are removed.

str1.rstrip('\r\n')

Return str1 with all newline characters (windows, mac or unix) stripped from the end of str1, like perl 'chomp' does.

str1.lstrip('<char>')

lstrip removes characters from the beginning of str1

Both rstrip and lstrip remove all whitespaces if nothing is specified

str1.upper() str1.lower() str1.title()

Return str1 in uppercase, lowercase or with only all first characters in uppercase

str1.join(list)

str1.join(map(str,list))

str1.join(str(e) for e in list)

Join list (or set or other sequence) into a string with str1 as separator. The second and third form makes sure all elements are converted to string before they are joined.

str1.split(sep[,max])

Split string into a list on sep into max + 1 elements (remainder is put in last element)

str1.splitlines([keepends])

Split on newline, with 'keepends' the newline is preserved.

str1.center(w)

str1.ljust(w)

str1.rjust(w)

Put spaces around str1 until length 'w' is reached.

str1.expandtabs(size)

Replace tabs by 'size' number of spaces.

Using format(string)

str1.format(values)

Fill in 'values' in str1-fields ({}). By numbering the fields they can be in a different order than the values.

If values are in a dict, they can be addressed by their key.

Code Example

 
"Value 1: {}, Value2: {}".format(1,2)
"Value 2: {1}, Value1: {0}".format(1,2)

dict1 = {'value1':1, 'value2':2}
"Value 2: {value2}, Value1: {value1}".format(**dict1)

list1 = {'value1', 'value2'}
"Value 1: {}, Value2: {}".format(*list1)

version=3.6
f"Format strings exist since version {version}"

NOTE: Using format-stings (f"string with {variable}") you do not need .format anymore.

fstring = 'f"Convert a string to a format-string'
compiled_fstring = compile(fstring,fstring, 'eval')
print(eval(compiled_fstring))

{value:formatspec}

The format can be specified after the (optional) fieldnumber.

[[fill]align][sign][#][0][width][grouping_option][.precision][type]

Generic format specification. Anything not needed can be left out [1].

e.g. "{:07d}".format(5) fill out with 0 in front to 7 digits -> '0000005'

"{:010.6f}".format(5.7647) floating point with precision 6 and total width 10 left fill with 0 -> '005.764700'

"{:10,.2f}".format(12345.7647) floating point with precision 2,total width 10 and comma as thousand separator -> ' 12,345.76'

"{:#=+15_.3f}".format(10000) # as padding, padding after sign, always show sign, 15 characters in total, _ as thousand separator, floating point with precision 3 -> +####10_000.000

Alignment

Always show sign

<	Left
>	Right
^	Center
=	Padding (after sign)
#	Prepend for x, o and b types

Types

s	String
c	Character
d	decimal
f	Float
%	Percent
o	Octal
x	Hexadecimal
b	Binary
e	Exponent
g	Python chooses between decimal, float or exponent with lowercase e
G	Python chooses between decimal, float or exponent with capital E

Datastructures

Below can be used on datatypes like lists and tuples too, not on sets.

json.dumps(dict, indent=4)

Convert a dict into a json string nicely formatted. Indent each level with 4 spaces.

pprint provides similar functionality

Searching

Basic

if <search> in str1

True if <search> is in str1

str1.count(<search>)

Return how many times <search> is in str1

str1.find(<search>)

str1.index(<search>)

Return where <search> is found in str1. If not found -1 with find, throw exception with index.

str1.endswith(<search>)

str.startswith(<search>)

Return True if str1 ends/starts with <search> (else returns False).

Regular Expressions (regexp)

Cheatography has a nice overview[2].

import re

The re module provides Perl-like Regular Expressions matching for string and byte objects.

re1 = re.compile(regexp)

Create regular expression object to use for matching. This is more efficient if the regular expression is used several times in a program.

NOTE: In all examples below regexp can be replaced by the object from re.compile

re.sub(regexp,new,str1)

Return str1 with all parts matching regexp replaced with new.

NOTE1: str1 remains unchanged.

NOTE2: re.sub is much more expensive than string.replace

NOTE3: if you want to use search modifiers like DOTALL you must use 'flags=<modifier>' or the object from re.compile.

re.split(regexp,str1,max)

Split str1 into a list on regexp, like with split above superfluous elements will be in the max + 1 element.

match = re1.match(str1)

match = re.match(regexp,str1)

Find 'regexp' at the beginning of 'str1'. Return match object if found, else return None-object

match = re.search(regexp,str1)

Find first occurrence of 'regexp' in 'str1'. Return match object if found, else return None-object

matchlist = re.findall(regexp,str1)

Find all occurrences of 'regexp' in 'str1'. Return a list of strings.

NOTE1: This may be very slow on regular expressions with multiple '.*?' clauses. Use finditer to prevent this.

NOTE2: When using a subexpression findall returns a list of tuples for the subexpression matches.

Use finditer or, below solution to get the first subexpression matched:

ListOfTuples = re.findall('(regexppart1 (subexpression1) regexppart2)',str1)
lst1 = [ a for a,b in ListOfTuples ]

matches = re.finditer(regexp,str1)

Find all occurrences of 'regexp' in 'str1'. Return a list of match objects.

Match Objects

match.group()

match.group(0)

The matched string in match object 'match'

match.group(1)

First submatch in the matched string in 'match'. The first match is the first ( in the expression.

match.start()

The start position of the matched string in 'match'

match.end()

The end position of the matched string in 'match'

match.span()

Tuple with start and end position of the matched string in 'match'

Search Modifieres

re1 = re.search(regexp,str1,modifier)

re1 = re.compile(regexp,modifier)

Modify how matching is done

re.DOTALL

The . matches all characters (default is all characters except newline). Use for searching in web or book pages.

re.I

Ignore case

re.M

Multiline mode, ^ matches all line beginnings and $ all line endings.

Code Example:

import re
str1 = "The thing to cut in pieces"
rel1 = re.compile('h.*n')
print "Matching"
match  = rel1.match(str1)

if match:
 print match.group()
 print match.start()
 print match.end()
 print match.span()
else:
 print "no match at beginning of string"
print

print "Searching"
match = re.search('t.*n',str1)
if match:
 print match.group()
 print match.start()
 print match.end()
 print match.span()

print "Searching case insensitive"
match = re.search('h.*n',str1,re.I)
if match:
 print match.group()
 print match.start()
 print match.end()
 print match.span()


print "findall"
re1 = re.compile('t')
matchlist = re1.findall(str1)
for str2 in matchlist:
  print str2
print

print "finditer"
re1 = re.compile('i.')
matches = re1.finditer(str1)
if matches:
 for match in matches:
 	print match.group()
 	print match.start()
 	print match.end()
 	print match.span()
print