Python3’s string is UTF-8 encoded by default, Python2’s is not. To change the encoding, use s.encode('ascii'). It takes optional parameters to keep unsupported characters.

String can be defined using single quotes (') or double quotes("). Multiline strings are enclosed in triple quotes (''' or """). Prefix an r to a string makes it a raw string in Python. A raw string means Python will not escape anything.

A string is like a tuple of characters. Each character can be obtained by indexing. But strings can be concatenated using + operator. String cannot concatenate with numbers without converting numbers to strings using str() first. To convert a list of characters to string, use ''.join(char_list).

String provides modification methods. They destroy old strings and replace with a new, modified string. The commonly-used APIs are:

  • multiline_string.splitlines(): returns a list of strings containing each line.
  • s.split(char, [num]): split string s from left according to the character char and return a list of substrings. Optional parameter num is the time you want to split. If char is None, it splits on any whitespaces.
  • s.rsplit(char[, num]): does the same thing except it starts from the right end.
  • char.join(list_str): concatenates a list of strings in list_str with char.
  • str.rstrip() removes tailing whitespace, str.lstrip() removes leading whitespace, and str.strip() removes both. They can take other characters as parameters so they gets stripped.
  • s.lower() & s.upper(): converts s to lower or upper case. s.swapcase changes lower-case to upper-case and upper-case to lower-case.
  • s.replace(old, new): create a copy of sand replace substring old in s with new.
  • s.count(ss): count occurrence of a substring ss in s.
  • s.find(ss): returns the index of first occurrence from left of a substring ss. If none exist, return -1. s.rfind(ss) searches from right.
  • s.index(ss) returns the index of first occurrence from left of a substring ss. If none exist, ValueError exception is raised. s.rindex(ss) searches from right.
  • ord(char): returns the integer value of a character. It is the inverse of chr().
  • str.translate(trans_table): it takes a dictionary that maps the byte representation of one character to that of another, and replaces all occurrences of the keys of the translation table with corresponding values. To delete a character, map it to None.
  • s.ljust(width), s.rjust(width), s.center(width) aligns s to left, right, and center. It takes an optional fill character. It is the same as format(s, '<20'), format(s, '>20'), format(s, '=^20') ('=' is the fill character). format works with more than strings.
  • s.isalnum() True if all characters are alphabet or numbers.
  • s.isalpha() True if all characters are alphabet.
  • s.isdigit() True if all characters are numbers.
  • s.istitle() True if all the first letters in all words are capitalized.
  • s.isspace() True if all characters are whitespaces.
  • s.islower() True if all characters are in lower case.
  • s.isupper() True if all characters are in upper case.
  • s.startswith(('.c', '.h')) True if s starts with either '.c' or '.h'. Same rule applies to s.endswith(). Note here tuple is required as input!

String Formatting

There are two ways of formatting a string.

str.format Method

a_string.format(var1, var2, ...): format string with placeholders.

a_string contains placeholder in a format of {i}, i=0,1,..., and {0} is replaced by the first argument var1, {1} is replaced by var2 etc. We can skip the number i in curly brackets and just use {} for placeholder. var1 etc. can be a list too, so placeholder becomes {0[0]}, etc. The placeholder can be any datatypes. If var1 is a dictionary, to access value from a string-type key, quotes a omitted, for example, {0[key]} instead of {0['key']}. Format Specifiers, e.g. {0:.1f}, can be used to specify the format of the variable. A colon marks the start of the format specifier.

% Operator

Use % to format a string:

str % (var1, var2, ...)  # 1
str % {key1: val1, key2: val2, ...}  # 2

In 1, a template str is given, it’s like C-type string formatter: "I'm %s. I'm %d year old". A tuple of matching replacers are given follow the % operator.

In 2, a template str is given, placeholders are keys in the following dictionary. For example, "I'm %(name)s. I'm %(age)d year old" with the dictionary {'name':'Vamei', 'age':99}.

Unicode Normalization

A single character can have multiple representations, namely composed (i.e., use a single code point if possible) and decomposed (with usage of combining characters).

unicodedata module provides a method normalize(way, str). way specifies how you want to normalize str: 'NFC' means fully composed, 'NFD' means fully decomposed.

combining(c) method checks whether c is a combining character.

Escape special characters in HTML

Use html.escape(s) in html module. To unescape it (given a bare text), use unescape(s) method in HTMLParser class in html.parser module. Normally you just need an HTML or XML parser.

textwrap module

It provides a fill(str, width) method to convert a string to a multi-line string with given column width. Its optional parameter initial_indent and subsequent_indent specifies the indent for first and rest of lines, respectively.

To make it terminal-friendly, you can use os.get_terminal_size().columns to get the width of the terminal.