New submission from Terry J. Reedy <tjre...@udel.edu>:
"ast.parse(expr, filename='<unknown>', mode='exec')
Parse an expression into an AST node. Equivalent to compile(expr, filename,
mode, ast.PyCF_ONLY_AST)."
but
"compile(source, ...)
Compile the source into a code or AST object.
...
The mode argument specifies what kind of code must be compiled; it can be
'exec' if source consists of a sequence of statements,"
and indeed, with all three versions,
>>> import ast
>>> e3=ast.parse('x=1; y=2\nif x: print(y)')
>>> o3 = compile(e3,'','exec')
>>> exec(o3)
2
>>> print(x,y)
(1, 2)
Attached patch (untested) changes doc and doc string.
Inquiry: I suspect that the filename is *not* attached to the ast. (In any
case, it must be explicitly supplied to a subsequent compile() call.) If so, it
is a useless parameter and the api and doc should be
parse(source, mode) ... Equivalent to compile(source, '<unknown>', ...
If so, and failing such a change, should the doc warn people to not bother
supplying a filename arg?
I also wonder whether the mode arg has any effect on the ast. If not, same
question.
----------
assignee: docs@python
components: Documentation
files: zast.diff
keywords: patch
messages: 126952
nosy: docs@python, georg.brandl, terry.reedy
priority: normal
severity: normal
status: open
title: Doc: ast.parse parses source, not just expressions
versions: Python 2.7, Python 3.1, Python 3.2
Added file: http://bugs.python.org/file20506/zast.diff
_______________________________________
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/issue11000>
_______________________________________
_______________________________________________
Python-bugs-list mailing list
Unsubscribe:
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
