Grammar/Grammar

# Grammar for Python

# Change log:

# 27-Feb-92:
#	Allow NEWLINE* after eval input

# 16-Jan-92:
#	Added '*' as alternative for '+' in varargs syntax
#	(Not sure which alternative is better yet.)

# 11-Jan-92:
#	Variable length argument list syntax added: def f(a, b, +rest): ...

# 8-Jan-92:
#	Allow only '==' for equality testing

# Changes since version 8:
#	Trailing commas in formal parameter lists are allowed

# Changes since version 7:
#	New syntax to specify base classes (but old syntax retained for now)
#	'global' statement (valid only in functions but not enforced here)

# Changes since version 6:
#	Add logical operators '|', '^', '&' and '~'
#	Add shift operators '<<' and '>>'

# Changes since version 5:
#	Comparison operators '<=' '>' '<>' are now 1 token
#	Also support '!=' and '==' as alternatives for '<>' and '='

# Changes compared to version 4:
#	Blank lines and lines only containing a comment are now eaten
#		by the lexer, so the NEWLINE* things in suite are gone
#		(but the 2nd NEWLINE terminating single_input stays!)
#	Semicolons can separate small statements
#	'continue' statement
#	Dictionary constructors: {key:value, key:value, ...}
#	More tests instead of exprs

# Changes compared to version 3:
#	Removed 'dir' statement.
#	Function call argument is a testlist instead of exprlist.

# Changes compared to version 2:
#	The syntax of Boolean operations is changed to use more
#	conventional priorities: or < and < not.

# Changes compared to version 1:
#	modules and scripts are unified;
#	'quit' is gone (use ^D);
#	empty_stmt is gone, replaced by explicit NEWLINE where appropriate;
#	'import' and 'def' aren't special any more;
#	added 'from' NAME option on import clause, and '*' to import all;
#	added class definition.

# Start symbols for the grammar:
#	single_input is a single interactive statement;
#	file_input is a module or sequence of commands read from an input file;
#	expr_input is the input for the input() function;
#	eval_input is the input for the eval() function.

# NB: compound_stmt in single_input is followed by extra NEWLINE!
single_input: NEWLINE | simple_stmt | compound_stmt NEWLINE
file_input: (NEWLINE | stmt)* ENDMARKER
expr_input: testlist NEWLINE
eval_input: testlist NEWLINE* ENDMARKER

funcdef: 'def' NAME parameters ':' suite
parameters: '(' [varargslist] ')'
varargslist: (fpdef ',')* ('+'|'*') NAME | fpdef (',' fpdef)* [',']
fpdef: NAME | '(' fplist ')'
fplist: fpdef (',' fpdef)* [',']

stmt: simple_stmt | compound_stmt
simple_stmt: small_stmt (';' small_stmt)* [';'] NEWLINE
small_stmt: expr_stmt | print_stmt  | del_stmt | pass_stmt | flow_stmt | import_stmt | global_stmt
expr_stmt: (exprlist '=')* exprlist
# For assignments, additional restrictions enforced by the interpreter
print_stmt: 'print' (test ',')* [test]
del_stmt: 'del' exprlist
pass_stmt: 'pass'
flow_stmt: break_stmt | continue_stmt | return_stmt | raise_stmt
break_stmt: 'break'
continue_stmt: 'continue'
return_stmt: 'return' [testlist]
raise_stmt: 'raise' test [',' test]
import_stmt: 'import' NAME (',' NAME)* | 'from' NAME 'import' ('*' | NAME (',' NAME)*)
global_stmt: 'global' NAME (',' NAME)*
compound_stmt: if_stmt | while_stmt | for_stmt | try_stmt | funcdef | classdef
if_stmt: 'if' test ':' suite ('elif' test ':' suite)* ['else' ':' suite]
while_stmt: 'while' test ':' suite ['else' ':' suite]
for_stmt: 'for' exprlist 'in' testlist ':' suite ['else' ':' suite]
try_stmt: 'try' ':' suite (except_clause ':' suite)+ | 'try' ':' suite 'finally' ':' suite
# NB compile.c makes sure that the default except clause is last
except_clause: 'except' [test [',' test]]
suite: simple_stmt | NEWLINE INDENT stmt+ DEDENT

test: and_test ('or' and_test)*
and_test: not_test ('and' not_test)*
not_test: 'not' not_test | comparison
comparison: expr (comp_op expr)*
comp_op: '<'|'>'|'=='|'>='|'<='|'<>'|'!='|'in'|'not' 'in'|'is'|'is' 'not'
expr: xor_expr ('|' xor_expr)*
xor_expr: and_expr ('^' and_expr)*
and_expr: shift_expr ('&' shift_expr)*
shift_expr: arith_expr (('<<'|'>>') arith_expr)*
arith_expr: term (('+'|'-') term)*
term: factor (('*'|'/'|'%') factor)*
factor: ('+'|'-'|'~') factor | atom trailer*
atom: '(' [testlist] ')' | '[' [testlist] ']' | '{' [dictmaker] '}' | '`' testlist '`' | NAME | NUMBER | STRING
trailer: '(' [testlist] ')' | '[' subscript ']' | '.' NAME
subscript: test | [test] ':' [test]
exprlist: expr (',' expr)* [',']
testlist: test (',' test)* [',']
dictmaker: test ':' test (',' test ':' test)* [',']

# New class syntax should be:
#	classdef: class NAME ['(' testlist ')'] ':' suite
# but merged with old syntax for compatibility it becomes:
classdef: 'class' NAME ['(' testlist ')' |'(' ')' ['=' baselist]] ':' suite
baselist: atom arguments (',' atom arguments)*
arguments: '(' ')'