iSharkFly-Open
/
python-peps
mirror of https://github.com/python/peps.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Lotsa changes: more details and examples covering recent decisions wrt 

return stmts, try/except, try/finally; more Q&A.  I expect this is the
last major revision of this PEP, so I'm going to post it again too.
This commit is contained in:
Tim Peters 2001-06-23 08:53:21 +00:00
parent 0c7da3beb7
commit 1eee9e2d60
1 changed files with 115 additions and 39 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

154
pep-0255.txt
View File

@ -10,7 +10,7 @@ Type: Standards Track
Requires: 234
Created: 18-May-2001
Python-Version: 2.2
Post-History: 14-Jun-2001
Post-History: 14-Jun-2001, 23-Jun-2001
Abstract
@ -118,17 +118,22 @@ Motivation
of Python iterator[1], but of an especially powerful kind.
Specification
Specification: Yield
A new statement is introduced:
yield_stmt: "yield" expression_list
"yield" is a new keyword, so a future statement[8] is needed to phase
this in. [XXX spell this out]
this in. [XXX spell this out -- but new keywords have ripple effects
across tools too, and it's not clear this can be forced into the future
framework at all -- it's not even clear that Python's parser alone can
be taught to swing both ways based on a future stmt]
The yield statement may only be used inside functions. A function that
contains a yield statement is called a generator function.
contains a yield statement is called a generator function. A generator
function is an ordinary function object in all respects, but has the
new CO_GENERATOR flag set in the code object's co_flags member.
When a generator function is called, the actual arguments are bound to
function-local formal argument names in the usual way, but no code in
@ -153,6 +158,15 @@ Specification
proceed exactly as if the yield statement were just another external
call.
Restriction: A yield statement is not allowed in the try clause of a
try/finally construct. The difficulty is that there's no guarantee
the generator will ever be resumed, hence no guarantee that the finally
block will ever get executed; that's too much a violation of finally's
purpose to bear.
Specification: Return
A generator function can also contain return statements of the form:
"return"
@ -161,15 +175,42 @@ Specification
in the body of a generator (although, of course, they may appear in
the bodies of non-generator functions nested within the generator).
When a return statement is encountered, nothing is returned, but a
StopIteration exception is raised, signalling that the iterator is
exhausted. The same is true if control flows off the end of the
function. Note that return means "I'm done, and have nothing
interesting to return", for both generator functions and non-generator
functions.
When a return statement is encountered, control proceeds as in any
function return, executing the appropriate finally clauses (if any
exist). Then a StopIteration exception is raised, signalling that the
iterator is exhausted. A StopIteration exception is also raised if
control flows off the end of the generator without an explict return.
Note that return means "I'm done, and have nothing interesting to
return", for both generator functions and non-generator functions.
Note that return isn't always equivalent to raising StopIteration: the
difference lies in how enclosing try/except constructs are treated.
For example,
>>> def f1():
... try:
... return
... except:
... yield 1
>>> print list(f1())
[]
because, as in any function, return simply exits, but
>>> def f2():
... try:
... raise StopIteration
... except:
... yield 42
>>> print list(f2())
[42]
because StopIteration is captured by a bare "except", as is any
exception.
Generators and Exception Propagation
Specification: Generators and Exception Propagation
If an unhandled exception-- including, but not limited to,
StopIteration --is raised by, or passes through, a generator function,
@ -192,44 +233,45 @@ Generators and Exception Propagation
File "<stdin>", line 2, in g
File "<stdin>", line 2, in f
ZeroDivisionError: integer division or modulo by zero
>>> k.next() # and the generator function cannot be resumed
>>> k.next() # and the generator cannot be resumed
Traceback (most recent call last):
File "<stdin>", line 1, in ?
StopIteration
>>>
Yield and Try/Except/Finally
Specification: Try/Except/Finally
While "yield" is a control-flow statement, and in most respects acts
like a "return" statement from the caller's point of view, within a
generator it acts more like a callback function. In particular, it has
no special semantics with respect to try/except/finally. This is best
illustrated by a contrived example; the primary lesson to take from
this is that using yield in a finally block is a dubious idea!
As noted earlier, yield is not allowed in the try clause of a try/
finally construct. A consequence is that generators should allocate
critical resources with great care. There is no restriction on yield
otherwise appearing in finally clauses, except clauses, or in the try
clause of a try/except construct:
>>> def g():
>>> def f():
... try:
... yield 1
... 1/0 # raises exception
... yield 2 # we never get here
... try:
... yield 2
... 1/0
... yield 3 # never get here
... except ZeroDivisionError:
... yield 4
... yield 5
... raise
... except:
... yield 6
... yield 7 # the "raise" above stops this
... except:
... yield 8
... yield 9
... try:
... x = 12
... finally:
... yield 3 # yields, and we raise the exception *next* time
... yield 4 # we never get here
>>> k = g()
>>> k.next()
1
>>> k.next()
3
>>> k.next() # as if "yield 3" were a callback, exception raised now
Traceback (most recent call last):
File "<stdin>", line 1, in ?
File "<stdin>", line 4, in g
ZeroDivisionError: integer division or modulo by zero
>>> k.next() # unhandled exception terminated the generator
Traceback (most recent call last):
File "<stdin>", line 1, in ?
StopIteration
... yield 10
... yield 11
>>> print list(f())
[1, 2, 4, 5, 8, 9, 10, 11]
>>>
@ -299,6 +341,10 @@ Example
print x,
print
Both output blocks display:
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
Q & A
@ -312,7 +358,34 @@ Q & A
yield is a control construct. It's also believed that efficient
implementation in Jython requires that the compiler be able to
determine potential suspension points at compile-time, and a new
keyword makes that easy.
keyword makes that easy. The CPython referrence implementation also
exploits it heavily, to detect which functions *are* generator-
functions (although a new keyword in place of "def" would solve that
for CPython -- but people asking the "why a new keyword?" question
don't want any new keyword).
Q: Then why not some other special syntax without a new keyword? For
example, one of these instead of "yield 3":
return 3 and continue
return and continue 3
return generating 3
continue return 3
return >> , 3
from generator return 3
return >> 3
return << 3
>> 3
<< 3
A: Did I miss one <wink>? Out of hundreds of messages, I counted two
suggesting such an alternative, and extracted the above from them.
It would be nice not to need a new keyword, but nicer to make yield
very clear -- I don't want to have to *deduce* that a yield is
occurring from making sense of a previous senseless sequence of
keywords or operators. Still, if this attracts enough interest,
proponents should settle on a single consensus suggestion, and Guido
will Pronounce on it.
Q. Why allow "return" at all? Why not force termination to be spelled
"raise StopIteration"?
@ -323,6 +396,9 @@ Q & A
these mechanisms for advanced users. That's not an argument for
forcing everyone to work at that level, though. "return" means "I'm
done" in any kind of function, and that's easy to explain and to use.
Note that "return" isn't always equivalent to "raise StopIteration"
in try/except construct, either (see the "Specification: Return"
section).
Q. Then why not allow an expression on "return" too?