9
single: lexical analysis
13
A Python program is read by a *parser*. Input to the parser is a stream of
14
*tokens*, generated by the *lexical analyzer*. This chapter describes how the
15
lexical analyzer breaks a file into tokens.
17
Python uses the 7-bit ASCII character set for program text.
20
An encoding declaration can be used to indicate that string literals and
21
comments use an encoding different from ASCII.
23
For compatibility with older versions, Python only warns if it finds 8-bit
24
characters; those warnings should be corrected by either declaring an explicit
25
encoding, or using escape sequences if those bytes are binary data, instead of
28
The run-time character set depends on the I/O devices connected to the program
29
but is generally a superset of ASCII.
31
**Future compatibility note:** It may be tempting to assume that the character
32
set for 8-bit characters is ISO Latin-1 (an ASCII superset that covers most
33
western languages that use the Latin alphabet), but it is possible that in the
34
future Unicode text editors will become common. These generally use the UTF-8
35
encoding, which is also an ASCII superset, but with very different use for the
36
characters with ordinals 128-255. While there is no consensus on this subject
37
yet, it is unwise to assume either Latin-1 or UTF-8, even though the current
38
implementation appears to favor Latin-1. This applies both to the source
39
character set and the run-time character set.
47
.. index:: single: line structure
49
A Python program is divided into a number of *logical lines*.
63
The end of a logical line is represented by the token NEWLINE. Statements
64
cannot cross logical line boundaries except where NEWLINE is allowed by the
65
syntax (e.g., between statements in compound statements). A logical line is
66
constructed from one or more *physical lines* by following the explicit or
67
implicit *line joining* rules.
75
A physical line is a sequence of characters terminated by an end-of-line
76
sequence. In source files, any of the standard platform line termination
77
sequences can be used - the Unix form using ASCII LF (linefeed), the Windows
78
form using the ASCII sequence CR LF (return followed by linefeed), or the old
79
Macintosh form using the ASCII CR (return) character. All of these forms can be
80
used equally, regardless of platform.
82
When embedding Python, source code strings should be passed to Python APIs using
83
the standard C conventions for newline characters (the ``\n`` character,
84
representing ASCII LF, is the line terminator).
94
single: hash character
96
A comment starts with a hash character (``#``) that is not part of a string
97
literal, and ends at the end of the physical line. A comment signifies the end
98
of the logical line unless the implicit line joining rules are invoked. Comments
99
are ignored by the syntax; they are not tokens.
104
Encoding declarations
105
---------------------
107
.. index:: source character set, encoding declarations (source file)
109
If a comment in the first or second line of the Python script matches the
110
regular expression ``coding[=:]\s*([-\w.]+)``, this comment is processed as an
111
encoding declaration; the first group of this expression names the encoding of
112
the source code file. The encoding declaration must appear on a line of its
113
own. If it is the second line, the first line must also be a comment-only line.
114
The recommended forms of an encoding expression are ::
116
# -*- coding: <encoding-name> -*-
118
which is recognized also by GNU Emacs, and ::
120
# vim:fileencoding=<encoding-name>
122
which is recognized by Bram Moolenaar's VIM. In addition, if the first bytes of
123
the file are the UTF-8 byte-order mark (``'\xef\xbb\xbf'``), the declared file
124
encoding is UTF-8 (this is supported, among others, by Microsoft's
127
If an encoding is declared, the encoding name must be recognized by Python. The
128
encoding is used for all lexical analysis, in particular to find the end of a
129
string, and to interpret the contents of Unicode literals. String literals are
130
converted to Unicode for syntactical analysis, then converted back to their
131
original encoding before interpretation starts.
133
.. XXX there should be a list of supported encodings.
136
.. _explicit-joining:
138
Explicit line joining
139
---------------------
142
single: physical line
144
single: line continuation
145
single: backslash character
147
Two or more physical lines may be joined into logical lines using backslash
148
characters (``\``), as follows: when a physical line ends in a backslash that is
149
not part of a string literal or comment, it is joined with the following forming
150
a single logical line, deleting the backslash and the following end-of-line
151
character. For example::
153
if 1900 < year < 2100 and 1 <= month <= 12 \
154
and 1 <= day <= 31 and 0 <= hour < 24 \
155
and 0 <= minute < 60 and 0 <= second < 60: # Looks like a valid date
158
A line ending in a backslash cannot carry a comment. A backslash does not
159
continue a comment. A backslash does not continue a token except for string
160
literals (i.e., tokens other than string literals cannot be split across
161
physical lines using a backslash). A backslash is illegal elsewhere on a line
162
outside a string literal.
165
.. _implicit-joining:
167
Implicit line joining
168
---------------------
170
Expressions in parentheses, square brackets or curly braces can be split over
171
more than one physical line without using backslashes. For example::
173
month_names = ['Januari', 'Februari', 'Maart', # These are the
174
'April', 'Mei', 'Juni', # Dutch names
175
'Juli', 'Augustus', 'September', # for the months
176
'Oktober', 'November', 'December'] # of the year
178
Implicitly continued lines can carry comments. The indentation of the
179
continuation lines is not important. Blank continuation lines are allowed.
180
There is no NEWLINE token between implicit continuation lines. Implicitly
181
continued lines can also occur within triple-quoted strings (see below); in that
182
case they cannot carry comments.
190
.. index:: single: blank line
192
A logical line that contains only spaces, tabs, formfeeds and possibly a
193
comment, is ignored (i.e., no NEWLINE token is generated). During interactive
194
input of statements, handling of a blank line may differ depending on the
195
implementation of the read-eval-print loop. In the standard implementation, an
196
entirely blank logical line (i.e. one containing not even whitespace or a
197
comment) terminates a multi-line statement.
208
single: leading whitespace
212
single: statement grouping
214
Leading whitespace (spaces and tabs) at the beginning of a logical line is used
215
to compute the indentation level of the line, which in turn is used to determine
216
the grouping of statements.
218
First, tabs are replaced (from left to right) by one to eight spaces such that
219
the total number of characters up to and including the replacement is a multiple
220
of eight (this is intended to be the same rule as used by Unix). The total
221
number of spaces preceding the first non-blank character then determines the
222
line's indentation. Indentation cannot be split over multiple physical lines
223
using backslashes; the whitespace up to the first backslash determines the
226
**Cross-platform compatibility note:** because of the nature of text editors on
227
non-UNIX platforms, it is unwise to use a mixture of spaces and tabs for the
228
indentation in a single source file. It should also be noted that different
229
platforms may explicitly limit the maximum indentation level.
231
A formfeed character may be present at the start of the line; it will be ignored
232
for the indentation calculations above. Formfeed characters occurring elsewhere
233
in the leading whitespace have an undefined effect (for instance, they may reset
234
the space count to zero).
240
The indentation levels of consecutive lines are used to generate INDENT and
241
DEDENT tokens, using a stack, as follows.
243
Before the first line of the file is read, a single zero is pushed on the stack;
244
this will never be popped off again. The numbers pushed on the stack will
245
always be strictly increasing from bottom to top. At the beginning of each
246
logical line, the line's indentation level is compared to the top of the stack.
247
If it is equal, nothing happens. If it is larger, it is pushed on the stack, and
248
one INDENT token is generated. If it is smaller, it *must* be one of the
249
numbers occurring on the stack; all numbers on the stack that are larger are
250
popped off, and for each number popped off a DEDENT token is generated. At the
251
end of the file, a DEDENT token is generated for each number remaining on the
252
stack that is larger than zero.
254
Here is an example of a correctly (though confusingly) indented piece of Python
258
# Compute the list of all permutations of l
262
for i in range(len(l)):
266
r.append(l[i:i+1] + x)
269
The following example shows various indentation errors::
271
def perm(l): # error: first line indented
272
for i in range(len(l)): # error: not indented
274
p = perm(l[:i] + l[i+1:]) # error: unexpected indent
276
r.append(l[i:i+1] + x)
277
return r # error: inconsistent dedent
279
(Actually, the first three errors are detected by the parser; only the last
280
error is found by the lexical analyzer --- the indentation of ``return r`` does
281
not match a level popped off the stack.)
286
Whitespace between tokens
287
-------------------------
289
Except at the beginning of a logical line or in string literals, the whitespace
290
characters space, tab and formfeed can be used interchangeably to separate
291
tokens. Whitespace is needed between two tokens only if their concatenation
292
could otherwise be interpreted as a different token (e.g., ab is one token, but
301
Besides NEWLINE, INDENT and DEDENT, the following categories of tokens exist:
302
*identifiers*, *keywords*, *literals*, *operators*, and *delimiters*. Whitespace
303
characters (other than line terminators, discussed earlier) are not tokens, but
304
serve to delimit tokens. Where ambiguity exists, a token comprises the longest
305
possible string that forms a legal token, when read from left to right.
310
Identifiers and keywords
311
========================
317
Identifiers (also referred to as *names*) are described by the following lexical
321
identifier: (`letter`|"_") (`letter` | `digit` | "_")*
322
letter: `lowercase` | `uppercase`
327
Identifiers are unlimited in length. Case is significant.
337
single: reserved word
339
The following identifiers are used as reserved words, or *keywords* of the
340
language, and cannot be used as ordinary identifiers. They must be spelled
341
exactly as written here:
345
and del from not while
346
as elif global or with
347
assert else if pass yield
348
break except import print
350
continue finally is return
353
.. versionchanged:: 2.4
354
:const:`None` became a constant and is now recognized by the compiler as a name
355
for the built-in object :const:`None`. Although it is not a keyword, you cannot
356
assign a different object to it.
358
.. versionchanged:: 2.5
359
Using :keyword:`as` and :keyword:`with` as identifiers triggers a warning. To
360
use them as keywords, enable the ``with_statement`` future feature .
362
.. versionchanged:: 2.6
363
:keyword:`as` and :keyword:`with` are full keywords.
368
Reserved classes of identifiers
369
-------------------------------
371
Certain classes of identifiers (besides keywords) have special meanings. These
372
classes are identified by the patterns of leading and trailing underscore
376
Not imported by ``from module import *``. The special identifier ``_`` is used
377
in the interactive interpreter to store the result of the last evaluation; it is
378
stored in the :mod:`__builtin__` module. When not in interactive mode, ``_``
379
has no special meaning and is not defined. See section :ref:`import`.
383
The name ``_`` is often used in conjunction with internationalization;
384
refer to the documentation for the :mod:`gettext` module for more
385
information on this convention.
388
System-defined names. These names are defined by the interpreter and its
389
implementation (including the standard library). Current system names are
390
discussed in the :ref:`specialnames` section and elsewhere. More will likely
391
be defined in future versions of Python. *Any* use of ``__*__`` names, in
392
any context, that does not follow explicitly documented use, is subject to
393
breakage without warning.
396
Class-private names. Names in this category, when used within the context of a
397
class definition, are re-written to use a mangled form to help avoid name
398
clashes between "private" attributes of base and derived classes. See section
399
:ref:`atom-identifiers`.
411
Literals are notations for constant values of some built-in types.
419
.. index:: single: string literal
421
String literals are described by the following lexical definitions:
423
.. index:: single: ASCII@ASCII
426
stringliteral: [`stringprefix`](`shortstring` | `longstring`)
427
stringprefix: "r" | "u" | "ur" | "R" | "U" | "UR" | "Ur" | "uR"
428
: | "b" | "B" | "br" | "Br" | "bR" | "BR"
429
shortstring: "'" `shortstringitem`* "'" | '"' `shortstringitem`* '"'
430
longstring: "'''" `longstringitem`* "'''"
431
: | '"""' `longstringitem`* '"""'
432
shortstringitem: `shortstringchar` | `escapeseq`
433
longstringitem: `longstringchar` | `escapeseq`
434
shortstringchar: <any source character except "\" or newline or the quote>
435
longstringchar: <any source character except "\">
436
escapeseq: "\" <any ASCII character>
438
One syntactic restriction not indicated by these productions is that whitespace
439
is not allowed between the :token:`stringprefix` and the rest of the string
440
literal. The source character set is defined by the encoding declaration; it is
441
ASCII if no encoding declaration is given in the source file; see section
445
single: triple-quoted string
446
single: Unicode Consortium
447
single: string; Unicode
450
In plain English: String literals can be enclosed in matching single quotes
451
(``'``) or double quotes (``"``). They can also be enclosed in matching groups
452
of three single or double quotes (these are generally referred to as
453
*triple-quoted strings*). The backslash (``\``) character is used to escape
454
characters that otherwise have a special meaning, such as newline, backslash
455
itself, or the quote character. String literals may optionally be prefixed with
456
a letter ``'r'`` or ``'R'``; such strings are called :dfn:`raw strings` and use
457
different rules for interpreting backslash escape sequences. A prefix of
458
``'u'`` or ``'U'`` makes the string a Unicode string. Unicode strings use the
459
Unicode character set as defined by the Unicode Consortium and ISO 10646. Some
460
additional escape sequences, described below, are available in Unicode strings.
461
A prefix of ``'b'`` or ``'B'`` is ignored in Python 2; it indicates that the
462
literal should become a bytes literal in Python 3 (e.g. when code is
463
automatically converted with 2to3). A ``'u'`` or ``'b'`` prefix may be followed
464
by an ``'r'`` prefix.
466
In triple-quoted strings, unescaped newlines and quotes are allowed (and are
467
retained), except that three unescaped quotes in a row terminate the string. (A
468
"quote" is the character used to open the string, i.e. either ``'`` or ``"``.)
471
single: physical line
472
single: escape sequence
476
Unless an ``'r'`` or ``'R'`` prefix is present, escape sequences in strings are
477
interpreted according to rules similar to those used by Standard C. The
478
recognized escape sequences are:
480
+-----------------+---------------------------------+-------+
481
| Escape Sequence | Meaning | Notes |
482
+=================+=================================+=======+
483
| ``\newline`` | Ignored | |
484
+-----------------+---------------------------------+-------+
485
| ``\\`` | Backslash (``\``) | |
486
+-----------------+---------------------------------+-------+
487
| ``\'`` | Single quote (``'``) | |
488
+-----------------+---------------------------------+-------+
489
| ``\"`` | Double quote (``"``) | |
490
+-----------------+---------------------------------+-------+
491
| ``\a`` | ASCII Bell (BEL) | |
492
+-----------------+---------------------------------+-------+
493
| ``\b`` | ASCII Backspace (BS) | |
494
+-----------------+---------------------------------+-------+
495
| ``\f`` | ASCII Formfeed (FF) | |
496
+-----------------+---------------------------------+-------+
497
| ``\n`` | ASCII Linefeed (LF) | |
498
+-----------------+---------------------------------+-------+
499
| ``\N{name}`` | Character named *name* in the | |
500
| | Unicode database (Unicode only) | |
501
+-----------------+---------------------------------+-------+
502
| ``\r`` | ASCII Carriage Return (CR) | |
503
+-----------------+---------------------------------+-------+
504
| ``\t`` | ASCII Horizontal Tab (TAB) | |
505
+-----------------+---------------------------------+-------+
506
| ``\uxxxx`` | Character with 16-bit hex value | \(1) |
507
| | *xxxx* (Unicode only) | |
508
+-----------------+---------------------------------+-------+
509
| ``\Uxxxxxxxx`` | Character with 32-bit hex value | \(2) |
510
| | *xxxxxxxx* (Unicode only) | |
511
+-----------------+---------------------------------+-------+
512
| ``\v`` | ASCII Vertical Tab (VT) | |
513
+-----------------+---------------------------------+-------+
514
| ``\ooo`` | Character with octal value | (3,5) |
516
+-----------------+---------------------------------+-------+
517
| ``\xhh`` | Character with hex value *hh* | (4,5) |
518
+-----------------+---------------------------------+-------+
520
.. index:: single: ASCII@ASCII
525
Individual code units which form parts of a surrogate pair can be encoded using
526
this escape sequence.
529
Any Unicode character can be encoded this way, but characters outside the Basic
530
Multilingual Plane (BMP) will be encoded using a surrogate pair if Python is
531
compiled to use 16-bit code units (the default).
534
As in Standard C, up to three octal digits are accepted.
537
Unlike in Standard C, exactly two hex digits are required.
540
In a string literal, hexadecimal and octal escapes denote the byte with the
541
given value; it is not necessary that the byte encodes a character in the source
542
character set. In a Unicode literal, these escapes denote a Unicode character
543
with the given value.
545
.. index:: single: unrecognized escape sequence
547
Unlike Standard C, all unrecognized escape sequences are left in the string
548
unchanged, i.e., *the backslash is left in the string*. (This behavior is
549
useful when debugging: if an escape sequence is mistyped, the resulting output
550
is more easily recognized as broken.) It is also important to note that the
551
escape sequences marked as "(Unicode only)" in the table above fall into the
552
category of unrecognized escapes for non-Unicode string literals.
554
When an ``'r'`` or ``'R'`` prefix is present, a character following a backslash
555
is included in the string without change, and *all backslashes are left in the
556
string*. For example, the string literal ``r"\n"`` consists of two characters:
557
a backslash and a lowercase ``'n'``. String quotes can be escaped with a
558
backslash, but the backslash remains in the string; for example, ``r"\""`` is a
559
valid string literal consisting of two characters: a backslash and a double
560
quote; ``r"\"`` is not a valid string literal (even a raw string cannot end in
561
an odd number of backslashes). Specifically, *a raw string cannot end in a
562
single backslash* (since the backslash would escape the following quote
563
character). Note also that a single backslash followed by a newline is
564
interpreted as those two characters as part of the string, *not* as a line
567
When an ``'r'`` or ``'R'`` prefix is used in conjunction with a ``'u'`` or
568
``'U'`` prefix, then the ``\uXXXX`` and ``\UXXXXXXXX`` escape sequences are
569
processed while *all other backslashes are left in the string*. For example,
570
the string literal ``ur"\u0062\n"`` consists of three Unicode characters: 'LATIN
571
SMALL LETTER B', 'REVERSE SOLIDUS', and 'LATIN SMALL LETTER N'. Backslashes can
572
be escaped with a preceding backslash; however, both remain in the string. As a
573
result, ``\uXXXX`` escape sequences are only recognized when there are an odd
574
number of backslashes.
577
.. _string-catenation:
579
String literal concatenation
580
----------------------------
582
Multiple adjacent string literals (delimited by whitespace), possibly using
583
different quoting conventions, are allowed, and their meaning is the same as
584
their concatenation. Thus, ``"hello" 'world'`` is equivalent to
585
``"helloworld"``. This feature can be used to reduce the number of backslashes
586
needed, to split long strings conveniently across long lines, or even to add
587
comments to parts of strings, for example::
589
re.compile("[A-Za-z_]" # letter or underscore
590
"[A-Za-z0-9_]*" # letter, digit or underscore
593
Note that this feature is defined at the syntactical level, but implemented at
594
compile time. The '+' operator must be used to concatenate string expressions
595
at run time. Also note that literal concatenation can use different quoting
596
styles for each component (even mixing raw strings and triple quoted strings).
606
single: numeric literal
607
single: integer literal
608
single: plain integer literal
609
single: long integer literal
610
single: floating point literal
611
single: hexadecimal literal
612
single: binary literal
613
single: octal literal
614
single: decimal literal
615
single: imaginary literal
616
single: complex; literal
618
There are four types of numeric literals: plain integers, long integers,
619
floating point numbers, and imaginary numbers. There are no complex literals
620
(complex numbers can be formed by adding a real number and an imaginary number).
622
Note that numeric literals do not include a sign; a phrase like ``-1`` is
623
actually an expression composed of the unary operator '``-``' and the literal
629
Integer and long integer literals
630
---------------------------------
632
Integer and long integer literals are described by the following lexical
636
longinteger: `integer` ("l" | "L")
637
integer: `decimalinteger` | `octinteger` | `hexinteger` | `bininteger`
638
decimalinteger: `nonzerodigit` `digit`* | "0"
639
octinteger: "0" ("o" | "O") `octdigit`+ | "0" `octdigit`+
640
hexinteger: "0" ("x" | "X") `hexdigit`+
641
bininteger: "0" ("b" | "B") `bindigit`+
642
nonzerodigit: "1"..."9"
645
hexdigit: `digit` | "a"..."f" | "A"..."F"
647
Although both lower case ``'l'`` and upper case ``'L'`` are allowed as suffix
648
for long integers, it is strongly recommended to always use ``'L'``, since the
649
letter ``'l'`` looks too much like the digit ``'1'``.
651
Plain integer literals that are above the largest representable plain integer
652
(e.g., 2147483647 when using 32-bit arithmetic) are accepted as if they were
653
long integers instead. [#]_ There is no limit for long integer literals apart
654
from what can be stored in available memory.
656
Some examples of plain integer literals (first row) and long integer literals
657
(second and third rows)::
660
3L 79228162514264337593543950336L 0377L 0x100000000L
661
79228162514264337593543950336 0xdeadbeef
666
Floating point literals
667
-----------------------
669
Floating point literals are described by the following lexical definitions:
672
floatnumber: `pointfloat` | `exponentfloat`
673
pointfloat: [`intpart`] `fraction` | `intpart` "."
674
exponentfloat: (`intpart` | `pointfloat`) `exponent`
676
fraction: "." `digit`+
677
exponent: ("e" | "E") ["+" | "-"] `digit`+
679
Note that the integer and exponent parts of floating point numbers can look like
680
octal integers, but are interpreted using radix 10. For example, ``077e010`` is
681
legal, and denotes the same number as ``77e10``. The allowed range of floating
682
point literals is implementation-dependent. Some examples of floating point
685
3.14 10. .001 1e100 3.14e-10 0e0
687
Note that numeric literals do not include a sign; a phrase like ``-1`` is
688
actually an expression composed of the unary operator ``-`` and the literal
697
Imaginary literals are described by the following lexical definitions:
700
imagnumber: (`floatnumber` | `intpart`) ("j" | "J")
702
An imaginary literal yields a complex number with a real part of 0.0. Complex
703
numbers are represented as a pair of floating point numbers and have the same
704
restrictions on their range. To create a complex number with a nonzero real
705
part, add a floating point number to it, e.g., ``(3+4j)``. Some examples of
708
3.14j 10.j 10j .001j 1e100j 3.14e-10j
716
.. index:: single: operators
718
The following tokens are operators:
727
The comparison operators ``<>`` and ``!=`` are alternate spellings of the same
728
operator. ``!=`` is the preferred spelling; ``<>`` is obsolescent.
736
.. index:: single: delimiters
738
The following tokens serve as delimiters in the grammar:
747
The period can also occur in floating-point and imaginary literals. A sequence
748
of three periods has a special meaning as an ellipsis in slices. The second half
749
of the list, the augmented assignment operators, serve lexically as delimiters,
750
but also perform an operation.
752
The following printing ASCII characters have special meaning as part of other
753
tokens or are otherwise significant to the lexical analyzer:
759
.. index:: single: ASCII@ASCII
761
The following printing ASCII characters are not used in Python. Their
762
occurrence outside string literals and comments is an unconditional error:
768
.. rubric:: Footnotes
770
.. [#] In versions of Python prior to 2.4, octal and hexadecimal literals in the range
771
just above the largest representable plain integer but below the largest
772
unsigned 32-bit number (on a machine using 32-bit arithmetic), 4294967296, were
773
taken as the negative plain integer obtained by subtracting 4294967296 from
774
their unsigned value.