| 1 |
.. _tut-io: |
| 2 |
|
| 3 |
**************** |
| 4 |
Input and Output |
| 5 |
**************** |
| 6 |
|
| 7 |
There are several ways to present the output of a program; data can be printed |
| 8 |
in a human-readable form, or written to a file for future use. This chapter will |
| 9 |
discuss some of the possibilities. |
| 10 |
|
| 11 |
|
| 12 |
.. _tut-formatting: |
| 13 |
|
| 14 |
Fancier Output Formatting |
| 15 |
========================= |
| 16 |
|
| 17 |
So far we've encountered two ways of writing values: *expression statements* and |
| 18 |
the :func:`print` function. (A third way is using the :meth:`~io.TextIOBase.write` method |
| 19 |
of file objects; the standard output file can be referenced as ``sys.stdout``. |
| 20 |
See the Library Reference for more information on this.) |
| 21 |
|
| 22 |
Often you'll want more control over the formatting of your output than simply |
| 23 |
printing space-separated values. There are several ways to format output. |
| 24 |
|
| 25 |
* To use :ref:`formatted string literals <tut-f-strings>`, begin a string |
| 26 |
with ``f`` or ``F`` before the opening quotation mark or triple quotation mark. |
| 27 |
Inside this string, you can write a Python expression between ``{`` and ``}`` |
| 28 |
characters that can refer to variables or literal values. |
| 29 |
|
| 30 |
:: |
| 31 |
|
| 32 |
>>> year = 2016 |
| 33 |
>>> event = 'Referendum' |
| 34 |
>>> f'Results of the {year} {event}' |
| 35 |
'Results of the 2016 Referendum' |
| 36 |
|
| 37 |
* The :meth:`str.format` method of strings requires more manual |
| 38 |
effort. You'll still use ``{`` and ``}`` to mark where a variable |
| 39 |
will be substituted and can provide detailed formatting directives, |
| 40 |
but you'll also need to provide the information to be formatted. In the following code |
| 41 |
block there are two examples of how to format variables: |
| 42 |
|
| 43 |
|
| 44 |
:: |
| 45 |
|
| 46 |
>>> yes_votes = 42_572_654 |
| 47 |
>>> total_votes = 85_705_149 |
| 48 |
>>> percentage = yes_votes / total_votes |
| 49 |
>>> '{:-9} YES votes {:2.2%}'.format(yes_votes, percentage) |
| 50 |
' 42572654 YES votes 49.67%' |
| 51 |
|
| 52 |
Notice how the ``yes_votes`` are padded with spaces and a negative sign only for negative numbers. |
| 53 |
The example also prints ``percentage`` multiplied by 100, with 2 decimal |
| 54 |
places and followed by a percent sign (see :ref:`formatspec` for details). |
| 55 |
|
| 56 |
|
| 57 |
* Finally, you can do all the string handling yourself by using string slicing and |
| 58 |
concatenation operations to create any layout you can imagine. The |
| 59 |
string type has some methods that perform useful operations for padding |
| 60 |
strings to a given column width. |
| 61 |
|
| 62 |
When you don't need fancy output but just want a quick display of some |
| 63 |
variables for debugging purposes, you can convert any value to a string with |
| 64 |
the :func:`repr` or :func:`str` functions. |
| 65 |
|
| 66 |
The :func:`str` function is meant to return representations of values which are |
| 67 |
fairly human-readable, while :func:`repr` is meant to generate representations |
| 68 |
which can be read by the interpreter (or will force a :exc:`SyntaxError` if |
| 69 |
there is no equivalent syntax). For objects which don't have a particular |
| 70 |
representation for human consumption, :func:`str` will return the same value as |
| 71 |
:func:`repr`. Many values, such as numbers or structures like lists and |
| 72 |
dictionaries, have the same representation using either function. Strings, in |
| 73 |
particular, have two distinct representations. |
| 74 |
|
| 75 |
Some examples:: |
| 76 |
|
| 77 |
>>> s = 'Hello, world.' |
| 78 |
>>> str(s) |
| 79 |
'Hello, world.' |
| 80 |
>>> repr(s) |
| 81 |
"'Hello, world.'" |
| 82 |
>>> str(1/7) |
| 83 |
'0.14285714285714285' |
| 84 |
>>> x = 10 * 3.25 |
| 85 |
>>> y = 200 * 200 |
| 86 |
>>> s = 'The value of x is ' + repr(x) + ', and y is ' + repr(y) + '...' |
| 87 |
>>> print(s) |
| 88 |
The value of x is 32.5, and y is 40000... |
| 89 |
>>> # The repr() of a string adds string quotes and backslashes: |
| 90 |
>>> hello = 'hello, world\n' |
| 91 |
>>> hellos = repr(hello) |
| 92 |
>>> print(hellos) |
| 93 |
'hello, world\n' |
| 94 |
>>> # The argument to repr() may be any Python object: |
| 95 |
>>> repr((x, y, ('spam', 'eggs'))) |
| 96 |
"(32.5, 40000, ('spam', 'eggs'))" |
| 97 |
|
| 98 |
The :mod:`string` module contains support for a simple templating approach |
| 99 |
based upon regular expressions, via :class:`string.Template`. |
| 100 |
This offers yet another way to substitute values into strings, |
| 101 |
using placeholders like ``$x`` and replacing them with values from a dictionary. |
| 102 |
This syntax is easy to use, although it offers much less control for formatting. |
| 103 |
|
| 104 |
.. index:: |
| 105 |
single: formatted string literal |
| 106 |
single: interpolated string literal |
| 107 |
single: string; formatted literal |
| 108 |
single: string; interpolated literal |
| 109 |
single: f-string |
| 110 |
single: fstring |
| 111 |
|
| 112 |
.. _tut-f-strings: |
| 113 |
|
| 114 |
Formatted String Literals |
| 115 |
------------------------- |
| 116 |
|
| 117 |
:ref:`Formatted string literals <f-strings>` (also called f-strings for |
| 118 |
short) let you include the value of Python expressions inside a string by |
| 119 |
prefixing the string with ``f`` or ``F`` and writing expressions as |
| 120 |
``{expression}``. |
| 121 |
|
| 122 |
An optional format specifier can follow the expression. This allows greater |
| 123 |
control over how the value is formatted. The following example rounds pi to |
| 124 |
three places after the decimal:: |
| 125 |
|
| 126 |
>>> import math |
| 127 |
>>> print(f'The value of pi is approximately {math.pi:.3f}.') |
| 128 |
The value of pi is approximately 3.142. |
| 129 |
|
| 130 |
Passing an integer after the ``':'`` will cause that field to be a minimum |
| 131 |
number of characters wide. This is useful for making columns line up. :: |
| 132 |
|
| 133 |
>>> table = {'Sjoerd': 4127, 'Jack': 4098, 'Dcab': 7678} |
| 134 |
>>> for name, phone in table.items(): |
| 135 |
... print(f'{name:10} ==> {phone:10d}') |
| 136 |
... |
| 137 |
Sjoerd ==> 4127 |
| 138 |
Jack ==> 4098 |
| 139 |
Dcab ==> 7678 |
| 140 |
|
| 141 |
Other modifiers can be used to convert the value before it is formatted. |
| 142 |
``'!a'`` applies :func:`ascii`, ``'!s'`` applies :func:`str`, and ``'!r'`` |
| 143 |
applies :func:`repr`:: |
| 144 |
|
| 145 |
>>> animals = 'eels' |
| 146 |
>>> print(f'My hovercraft is full of {animals}.') |
| 147 |
My hovercraft is full of eels. |
| 148 |
>>> print(f'My hovercraft is full of {animals!r}.') |
| 149 |
My hovercraft is full of 'eels'. |
| 150 |
|
| 151 |
The ``=`` specifier can be used to expand an expression to the text of the |
| 152 |
expression, an equal sign, then the representation of the evaluated expression: |
| 153 |
|
| 154 |
>>> bugs = 'roaches' |
| 155 |
>>> count = 13 |
| 156 |
>>> area = 'living room' |
| 157 |
>>> print(f'Debugging {bugs=} {count=} {area=}') |
| 158 |
Debugging bugs='roaches' count=13 area='living room' |
| 159 |
|
| 160 |
See :ref:`self-documenting expressions <bpo-36817-whatsnew>` for more information |
| 161 |
on the ``=`` specifier. For a reference on these format specifications, see |
| 162 |
the reference guide for the :ref:`formatspec`. |
| 163 |
|
| 164 |
.. _tut-string-format: |
| 165 |
|
| 166 |
The String format() Method |
| 167 |
-------------------------- |
| 168 |
|
| 169 |
Basic usage of the :meth:`str.format` method looks like this:: |
| 170 |
|
| 171 |
>>> print('We are the {} who say "{}!"'.format('knights', 'Ni')) |
| 172 |
We are the knights who say "Ni!" |
| 173 |
|
| 174 |
The brackets and characters within them (called format fields) are replaced with |
| 175 |
the objects passed into the :meth:`str.format` method. A number in the |
| 176 |
brackets can be used to refer to the position of the object passed into the |
| 177 |
:meth:`str.format` method. :: |
| 178 |
|
| 179 |
>>> print('{0} and {1}'.format('spam', 'eggs')) |
| 180 |
spam and eggs |
| 181 |
>>> print('{1} and {0}'.format('spam', 'eggs')) |
| 182 |
eggs and spam |
| 183 |
|
| 184 |
If keyword arguments are used in the :meth:`str.format` method, their values |
| 185 |
are referred to by using the name of the argument. :: |
| 186 |
|
| 187 |
>>> print('This {food} is {adjective}.'.format( |
| 188 |
... food='spam', adjective='absolutely horrible')) |
| 189 |
This spam is absolutely horrible. |
| 190 |
|
| 191 |
Positional and keyword arguments can be arbitrarily combined:: |
| 192 |
|
| 193 |
>>> print('The story of {0}, {1}, and {other}.'.format('Bill', 'Manfred', |
| 194 |
... other='Georg')) |
| 195 |
The story of Bill, Manfred, and Georg. |
| 196 |
|
| 197 |
If you have a really long format string that you don't want to split up, it |
| 198 |
would be nice if you could reference the variables to be formatted by name |
| 199 |
instead of by position. This can be done by simply passing the dict and using |
| 200 |
square brackets ``'[]'`` to access the keys. :: |
| 201 |
|
| 202 |
>>> table = {'Sjoerd': 4127, 'Jack': 4098, 'Dcab': 8637678} |
| 203 |
>>> print('Jack: {0[Jack]:d}; Sjoerd: {0[Sjoerd]:d}; ' |
| 204 |
... 'Dcab: {0[Dcab]:d}'.format(table)) |
| 205 |
Jack: 4098; Sjoerd: 4127; Dcab: 8637678 |
| 206 |
|
| 207 |
This could also be done by passing the ``table`` dictionary as keyword arguments with the ``**`` |
| 208 |
notation. :: |
| 209 |
|
| 210 |
>>> table = {'Sjoerd': 4127, 'Jack': 4098, 'Dcab': 8637678} |
| 211 |
>>> print('Jack: {Jack:d}; Sjoerd: {Sjoerd:d}; Dcab: {Dcab:d}'.format(**table)) |
| 212 |
Jack: 4098; Sjoerd: 4127; Dcab: 8637678 |
| 213 |
|
| 214 |
This is particularly useful in combination with the built-in function |
| 215 |
:func:`vars`, which returns a dictionary containing all local variables:: |
| 216 |
|
| 217 |
>>> table = {k: str(v) for k, v in vars().items()} |
| 218 |
>>> message = " ".join([f'{k}: ' + '{' + k +'};' for k in table.keys()]) |
| 219 |
>>> print(message.format(**table)) |
| 220 |
__name__: __main__; __doc__: None; __package__: None; __loader__: ... |
| 221 |
|
| 222 |
As an example, the following lines produce a tidily aligned |
| 223 |
set of columns giving integers and their squares and cubes:: |
| 224 |
|
| 225 |
>>> for x in range(1, 11): |
| 226 |
... print('{0:2d} {1:3d} {2:4d}'.format(x, x*x, x*x*x)) |
| 227 |
... |
| 228 |
1 1 1 |
| 229 |
2 4 8 |
| 230 |
3 9 27 |
| 231 |
4 16 64 |
| 232 |
5 25 125 |
| 233 |
6 36 216 |
| 234 |
7 49 343 |
| 235 |
8 64 512 |
| 236 |
9 81 729 |
| 237 |
10 100 1000 |
| 238 |
|
| 239 |
For a complete overview of string formatting with :meth:`str.format`, see |
| 240 |
:ref:`formatstrings`. |
| 241 |
|
| 242 |
|
| 243 |
Manual String Formatting |
| 244 |
------------------------ |
| 245 |
|
| 246 |
Here's the same table of squares and cubes, formatted manually:: |
| 247 |
|
| 248 |
>>> for x in range(1, 11): |
| 249 |
... print(repr(x).rjust(2), repr(x*x).rjust(3), end=' ') |
| 250 |
... # Note use of 'end' on previous line |
| 251 |
... print(repr(x*x*x).rjust(4)) |
| 252 |
... |
| 253 |
1 1 1 |
| 254 |
2 4 8 |
| 255 |
3 9 27 |
| 256 |
4 16 64 |
| 257 |
5 25 125 |
| 258 |
6 36 216 |
| 259 |
7 49 343 |
| 260 |
8 64 512 |
| 261 |
9 81 729 |
| 262 |
10 100 1000 |
| 263 |
|
| 264 |
(Note that the one space between each column was added by the |
| 265 |
way :func:`print` works: it always adds spaces between its arguments.) |
| 266 |
|
| 267 |
The :meth:`str.rjust` method of string objects right-justifies a string in a |
| 268 |
field of a given width by padding it with spaces on the left. There are |
| 269 |
similar methods :meth:`str.ljust` and :meth:`str.center`. These methods do |
| 270 |
not write anything, they just return a new string. If the input string is too |
| 271 |
long, they don't truncate it, but return it unchanged; this will mess up your |
| 272 |
column lay-out but that's usually better than the alternative, which would be |
| 273 |
lying about a value. (If you really want truncation you can always add a |
| 274 |
slice operation, as in ``x.ljust(n)[:n]``.) |
| 275 |
|
| 276 |
There is another method, :meth:`str.zfill`, which pads a numeric string on the |
| 277 |
left with zeros. It understands about plus and minus signs:: |
| 278 |
|
| 279 |
>>> '12'.zfill(5) |
| 280 |
'00012' |
| 281 |
>>> '-3.14'.zfill(7) |
| 282 |
'-003.14' |
| 283 |
>>> '3.14159265359'.zfill(5) |
| 284 |
'3.14159265359' |
| 285 |
|
| 286 |
|
| 287 |
Old string formatting |
| 288 |
--------------------- |
| 289 |
|
| 290 |
The % operator (modulo) can also be used for string formatting. |
| 291 |
Given ``format % values`` (where *format* is a string), |
| 292 |
``%`` conversion specifications in *format* are replaced with |
| 293 |
zero or more elements of *values*. |
| 294 |
This operation is commonly known as string |
| 295 |
interpolation. For example:: |
| 296 |
|
| 297 |
>>> import math |
| 298 |
>>> print('The value of pi is approximately %5.3f.' % math.pi) |
| 299 |
The value of pi is approximately 3.142. |
| 300 |
|
| 301 |
More information can be found in the :ref:`old-string-formatting` section. |
| 302 |
|
| 303 |
|
| 304 |
.. _tut-files: |
| 305 |
|
| 306 |
Reading and Writing Files |
| 307 |
========================= |
| 308 |
|
| 309 |
.. index:: |
| 310 |
pair: built-in function; open |
| 311 |
pair: object; file |
| 312 |
|
| 313 |
:func:`open` returns a :term:`file object`, and is most commonly used with |
| 314 |
two positional arguments and one keyword argument: |
| 315 |
``open(filename, mode, encoding=None)`` |
| 316 |
|
| 317 |
:: |
| 318 |
|
| 319 |
>>> f = open('workfile', 'w', encoding="utf-8") |
| 320 |
|
| 321 |
.. XXX str(f) is <io.TextIOWrapper object at 0x82e8dc4> |
| 322 |
|
| 323 |
>>> print(f) |
| 324 |
<open file 'workfile', mode 'w' at 80a0960> |
| 325 |
|
| 326 |
The first argument is a string containing the filename. The second argument is |
| 327 |
another string containing a few characters describing the way in which the file |
| 328 |
will be used. *mode* can be ``'r'`` when the file will only be read, ``'w'`` |
| 329 |
for only writing (an existing file with the same name will be erased), and |
| 330 |
``'a'`` opens the file for appending; any data written to the file is |
| 331 |
automatically added to the end. ``'r+'`` opens the file for both reading and |
| 332 |
writing. The *mode* argument is optional; ``'r'`` will be assumed if it's |
| 333 |
omitted. |
| 334 |
|
| 335 |
Normally, files are opened in :dfn:`text mode`, that means, you read and write |
| 336 |
strings from and to the file, which are encoded in a specific *encoding*. |
| 337 |
If *encoding* is not specified, the default is platform dependent |
| 338 |
(see :func:`open`). |
| 339 |
Because UTF-8 is the modern de-facto standard, ``encoding="utf-8"`` is |
| 340 |
recommended unless you know that you need to use a different encoding. |
| 341 |
Appending a ``'b'`` to the mode opens the file in :dfn:`binary mode`. |
| 342 |
Binary mode data is read and written as :class:`bytes` objects. |
| 343 |
You can not specify *encoding* when opening file in binary mode. |
| 344 |
|
| 345 |
In text mode, the default when reading is to convert platform-specific line |
| 346 |
endings (``\n`` on Unix, ``\r\n`` on Windows) to just ``\n``. When writing in |
| 347 |
text mode, the default is to convert occurrences of ``\n`` back to |
| 348 |
platform-specific line endings. This behind-the-scenes modification |
| 349 |
to file data is fine for text files, but will corrupt binary data like that in |
| 350 |
:file:`JPEG` or :file:`EXE` files. Be very careful to use binary mode when |
| 351 |
reading and writing such files. |
| 352 |
|
| 353 |
It is good practice to use the :keyword:`with` keyword when dealing |
| 354 |
with file objects. The advantage is that the file is properly closed |
| 355 |
after its suite finishes, even if an exception is raised at some |
| 356 |
point. Using :keyword:`!with` is also much shorter than writing |
| 357 |
equivalent :keyword:`try`\ -\ :keyword:`finally` blocks:: |
| 358 |
|
| 359 |
>>> with open('workfile', encoding="utf-8") as f: |
| 360 |
... read_data = f.read() |
| 361 |
|
| 362 |
>>> # We can check that the file has been automatically closed. |
| 363 |
>>> f.closed |
| 364 |
True |
| 365 |
|
| 366 |
If you're not using the :keyword:`with` keyword, then you should call |
| 367 |
``f.close()`` to close the file and immediately free up any system |
| 368 |
resources used by it. |
| 369 |
|
| 370 |
.. warning:: |
| 371 |
Calling ``f.write()`` without using the :keyword:`!with` keyword or calling |
| 372 |
``f.close()`` **might** result in the arguments |
| 373 |
of ``f.write()`` not being completely written to the disk, even if the |
| 374 |
program exits successfully. |
| 375 |
|
| 376 |
.. |
| 377 |
See also https://bugs.python.org/issue17852 |
| 378 |
|
| 379 |
After a file object is closed, either by a :keyword:`with` statement |
| 380 |
or by calling ``f.close()``, attempts to use the file object will |
| 381 |
automatically fail. :: |
| 382 |
|
| 383 |
>>> f.close() |
| 384 |
>>> f.read() |
| 385 |
Traceback (most recent call last): |
| 386 |
File "<stdin>", line 1, in <module> |
| 387 |
ValueError: I/O operation on closed file. |
| 388 |
|
| 389 |
|
| 390 |
.. _tut-filemethods: |
| 391 |
|
| 392 |
Methods of File Objects |
| 393 |
----------------------- |
| 394 |
|
| 395 |
The rest of the examples in this section will assume that a file object called |
| 396 |
``f`` has already been created. |
| 397 |
|
| 398 |
To read a file's contents, call ``f.read(size)``, which reads some quantity of |
| 399 |
data and returns it as a string (in text mode) or bytes object (in binary mode). |
| 400 |
*size* is an optional numeric argument. When *size* is omitted or negative, the |
| 401 |
entire contents of the file will be read and returned; it's your problem if the |
| 402 |
file is twice as large as your machine's memory. Otherwise, at most *size* |
| 403 |
characters (in text mode) or *size* bytes (in binary mode) are read and returned. |
| 404 |
If the end of the file has been reached, ``f.read()`` will return an empty |
| 405 |
string (``''``). :: |
| 406 |
|
| 407 |
>>> f.read() |
| 408 |
'This is the entire file.\n' |
| 409 |
>>> f.read() |
| 410 |
'' |
| 411 |
|
| 412 |
``f.readline()`` reads a single line from the file; a newline character (``\n``) |
| 413 |
is left at the end of the string, and is only omitted on the last line of the |
| 414 |
file if the file doesn't end in a newline. This makes the return value |
| 415 |
unambiguous; if ``f.readline()`` returns an empty string, the end of the file |
| 416 |
has been reached, while a blank line is represented by ``'\n'``, a string |
| 417 |
containing only a single newline. :: |
| 418 |
|
| 419 |
>>> f.readline() |
| 420 |
'This is the first line of the file.\n' |
| 421 |
>>> f.readline() |
| 422 |
'Second line of the file\n' |
| 423 |
>>> f.readline() |
| 424 |
'' |
| 425 |
|
| 426 |
For reading lines from a file, you can loop over the file object. This is memory |
| 427 |
efficient, fast, and leads to simple code:: |
| 428 |
|
| 429 |
>>> for line in f: |
| 430 |
... print(line, end='') |
| 431 |
... |
| 432 |
This is the first line of the file. |
| 433 |
Second line of the file |
| 434 |
|
| 435 |
If you want to read all the lines of a file in a list you can also use |
| 436 |
``list(f)`` or ``f.readlines()``. |
| 437 |
|
| 438 |
``f.write(string)`` writes the contents of *string* to the file, returning |
| 439 |
the number of characters written. :: |
| 440 |
|
| 441 |
>>> f.write('This is a test\n') |
| 442 |
15 |
| 443 |
|
| 444 |
Other types of objects need to be converted -- either to a string (in text mode) |
| 445 |
or a bytes object (in binary mode) -- before writing them:: |
| 446 |
|
| 447 |
>>> value = ('the answer', 42) |
| 448 |
>>> s = str(value) # convert the tuple to string |
| 449 |
>>> f.write(s) |
| 450 |
18 |
| 451 |
|
| 452 |
``f.tell()`` returns an integer giving the file object's current position in the file |
| 453 |
represented as number of bytes from the beginning of the file when in binary mode and |
| 454 |
an opaque number when in text mode. |
| 455 |
|
| 456 |
To change the file object's position, use ``f.seek(offset, whence)``. The position is computed |
| 457 |
from adding *offset* to a reference point; the reference point is selected by |
| 458 |
the *whence* argument. A *whence* value of 0 measures from the beginning |
| 459 |
of the file, 1 uses the current file position, and 2 uses the end of the file as |
| 460 |
the reference point. *whence* can be omitted and defaults to 0, using the |
| 461 |
beginning of the file as the reference point. :: |
| 462 |
|
| 463 |
>>> f = open('workfile', 'rb+') |
| 464 |
>>> f.write(b'0123456789abcdef') |
| 465 |
16 |
| 466 |
>>> f.seek(5) # Go to the 6th byte in the file |
| 467 |
5 |
| 468 |
>>> f.read(1) |
| 469 |
b'5' |
| 470 |
>>> f.seek(-3, 2) # Go to the 3rd byte before the end |
| 471 |
13 |
| 472 |
>>> f.read(1) |
| 473 |
b'd' |
| 474 |
|
| 475 |
In text files (those opened without a ``b`` in the mode string), only seeks |
| 476 |
relative to the beginning of the file are allowed (the exception being seeking |
| 477 |
to the very file end with ``seek(0, 2)``) and the only valid *offset* values are |
| 478 |
those returned from the ``f.tell()``, or zero. Any other *offset* value produces |
| 479 |
undefined behaviour. |
| 480 |
|
| 481 |
File objects have some additional methods, such as :meth:`~io.IOBase.isatty` and |
| 482 |
:meth:`~io.IOBase.truncate` which are less frequently used; consult the Library |
| 483 |
Reference for a complete guide to file objects. |
| 484 |
|
| 485 |
|
| 486 |
.. _tut-json: |
| 487 |
|
| 488 |
Saving structured data with :mod:`json` |
| 489 |
--------------------------------------- |
| 490 |
|
| 491 |
.. index:: pair: module; json |
| 492 |
|
| 493 |
Strings can easily be written to and read from a file. Numbers take a bit more |
| 494 |
effort, since the :meth:`~io.TextIOBase.read` method only returns strings, which will have to |
| 495 |
be passed to a function like :func:`int`, which takes a string like ``'123'`` |
| 496 |
and returns its numeric value 123. When you want to save more complex data |
| 497 |
types like nested lists and dictionaries, parsing and serializing by hand |
| 498 |
becomes complicated. |
| 499 |
|
| 500 |
Rather than having users constantly writing and debugging code to save |
| 501 |
complicated data types to files, Python allows you to use the popular data |
| 502 |
interchange format called `JSON (JavaScript Object Notation) |
| 503 |
<https://json.org>`_. The standard module called :mod:`json` can take Python |
| 504 |
data hierarchies, and convert them to string representations; this process is |
| 505 |
called :dfn:`serializing`. Reconstructing the data from the string representation |
| 506 |
is called :dfn:`deserializing`. Between serializing and deserializing, the |
| 507 |
string representing the object may have been stored in a file or data, or |
| 508 |
sent over a network connection to some distant machine. |
| 509 |
|
| 510 |
.. note:: |
| 511 |
The JSON format is commonly used by modern applications to allow for data |
| 512 |
exchange. Many programmers are already familiar with it, which makes |
| 513 |
it a good choice for interoperability. |
| 514 |
|
| 515 |
If you have an object ``x``, you can view its JSON string representation with a |
| 516 |
simple line of code:: |
| 517 |
|
| 518 |
>>> import json |
| 519 |
>>> x = [1, 'simple', 'list'] |
| 520 |
>>> json.dumps(x) |
| 521 |
'[1, "simple", "list"]' |
| 522 |
|
| 523 |
Another variant of the :func:`~json.dumps` function, called :func:`~json.dump`, |
| 524 |
simply serializes the object to a :term:`text file`. So if ``f`` is a |
| 525 |
:term:`text file` object opened for writing, we can do this:: |
| 526 |
|
| 527 |
json.dump(x, f) |
| 528 |
|
| 529 |
To decode the object again, if ``f`` is a :term:`binary file` or |
| 530 |
:term:`text file` object which has been opened for reading:: |
| 531 |
|
| 532 |
x = json.load(f) |
| 533 |
|
| 534 |
.. note:: |
| 535 |
JSON files must be encoded in UTF-8. Use ``encoding="utf-8"`` when opening |
| 536 |
JSON file as a :term:`text file` for both of reading and writing. |
| 537 |
|
| 538 |
This simple serialization technique can handle lists and dictionaries, but |
| 539 |
serializing arbitrary class instances in JSON requires a bit of extra effort. |
| 540 |
The reference for the :mod:`json` module contains an explanation of this. |
| 541 |
|
| 542 |
.. seealso:: |
| 543 |
|
| 544 |
:mod:`pickle` - the pickle module |
| 545 |
|
| 546 |
Contrary to :ref:`JSON <tut-json>`, *pickle* is a protocol which allows |
| 547 |
the serialization of arbitrarily complex Python objects. As such, it is |
| 548 |
specific to Python and cannot be used to communicate with applications |
| 549 |
written in other languages. It is also insecure by default: |
| 550 |
deserializing pickle data coming from an untrusted source can execute |
| 551 |
arbitrary code, if the data was crafted by a skilled attacker. |