import __hello__ doesn't print any message!!!

That was changed in [a2213060d9dd], see issue #1414.

What’s import __hello__?  I don’t have it in any version from 2.4 to 3.2.

Okay, it doesn’t work with -m __hello__, but using -c "import __hello__" I can see the message in all versions.

On Fri, Mar 25, 2011 at 5:52 PM, Éric Araujo <report@bugs.python.org> wrote:
> Okay, it doesn’t work with -m __hello__, but using -c "import __hello__" I can see the message in all versions.

Can you elaborate on that? I.e. the versions and the actual message.
The current bytecode of the __hello__ module is (default branch):

  1           0 LOAD_CONST               1 (True)
              3 STORE_NAME               1 (initialized)
              6 LOAD_CONST               0 (None)
              9 RETURN_VALUE

I'd be a bit surprised if that really prints a message.

I can see the message (“Hello world...”) in 2.4 to 2.7, but actually not in 3.x.

Why is this still open, and more importantly, why is it a release blocker?

> Why is this still open

Because import __hello__ doesn't print hello.

> and more importantly, why is it a release blocker?

Because it is a very important issue!

A patch against 3.1. The new bytecode is now:

  1           0 LOAD_CONST               2 (True) 
              3 STORE_NAME               1 (initialized) 

  2           6 LOAD_NAME                2 (print) 
              9 LOAD_CONST               0 ('Hello world...') 
             12 CALL_FUNCTION            1 
             15 POP_TOP              
             16 LOAD_CONST               1 (None) 
             19 RETURN_VALUE

+print("Hello world...")

I would prefer a more positive "Hello World!", but thanks for the patch.

New changeset 44fd95cead7b by Victor Stinner in branch '3.1':
Issue #11614: import __hello__ prints "Hello World!". Patch written by Andreas
http://hg.python.org/cpython/rev/44fd95cead7b

New changeset 2c5736e8a221 by Victor Stinner in branch '3.2':
(Merge 3.1) Issue #11614: import __hello__ prints "Hello World!". Patch written
http://hg.python.org/cpython/rev/2c5736e8a221

New changeset 79c49d2fc768 by Victor Stinner in branch 'default':
(Merge 3.2) Issue #11614: import __hello__ prints "Hello World!". Patch written
http://hg.python.org/cpython/rev/79c49d2fc768

Thanks Andreas for your fix! Python 3 is now better than Python 2: it prints "Hello World!" instead of "Hello World...".

New changeset 9e1d7cd15d2e by Victor Stinner in branch '3.1':
Issue #11614: Fix importlib tests for the new __hello__ module
http://hg.python.org/cpython/rev/9e1d7cd15d2e

New changeset 6119f6cc9b2d by Victor Stinner in branch '3.2':
(Merge 3.1) Issue #11614: Fix importlib tests for the new __hello__ module
http://hg.python.org/cpython/rev/6119f6cc9b2d

New changeset b61a1cd4943b by Victor Stinner in branch 'default':
(Merge 3.2) Issue #11614: Fix importlib tests for the new __hello__ module
http://hg.python.org/cpython/rev/b61a1cd4943b

New changeset 9a4d4506680a by Victor Stinner in branch '3.1':
Issue #11614: Fix test_linecache, cjkencodings_test.py doesn't exist anymore
http://hg.python.org/cpython/rev/9a4d4506680a

New changeset 43cbfacae463 by Victor Stinner in branch '3.2':
(Merge 3.1) Issue #11614: Fix test_linecache, cjkencodings_test.py doesn't
http://hg.python.org/cpython/rev/43cbfacae463

New changeset 06473da99270 by Victor Stinner in branch 'default':
(Merge 3.2) Issue #11614: Fix test_linecache, cjkencodings_test.py doesn't
http://hg.python.org/cpython/rev/06473da99270

> New changeset 9a4d4506680a by Victor Stinner in branch '3.1':
> Issue #11614: Fix test_linecache, cjkencodings_test.py doesn't exist anymore

Oh, these commits are related to #12057, not this one.

New changeset 47bc3509f76b by Victor Stinner in branch '2.7':
Issue #11614: Fix test_linecache, cjkencodings_test.py doesn't exist anymore
http://hg.python.org/cpython/rev/47bc3509f76b

The byte code could be further optimized (because this is such a speed-critical module! :)):

  1           0 LOAD_CONST               2 (True) 
              3 STORE_NAME               1 (initialized) 

  2           6 LOAD_NAME                2 (print) 
              9 LOAD_CONST               0 ('Hello world...') 
             12 CALL_FUNCTION            1 
             15 RETURN_VALUE

> The byte code could be further optimized (because this is such a speed-critical module! :)): (...)

You propose to replace

              8 CALL_FUNCTION            1
             10 POP_TOP
             12 LOAD_CONST               2 (None)
             14 RETURN_VALUE

with:

             12 CALL_FUNCTION            1 
             15 RETURN_VALUE

It changes the semantics of Python. Technically, you *can* override the print() builtin function:

vstinner@apu$ ./python
Python 3.8.0a2+ (heads/master:dc078947a5, Mar  7 2019, 12:23:23) 
>>> import builtins
>>> def mock(*args, **kw): return 3
... 
>>> builtins.print=mock
>>> print("hello")
3
>>> import __phello__
>>> # doesn't print anything
... 

It would be possible if you ensure that print() isn't replaced.

Longer explanation:
https://fatoptimizer.readthedocs.io/en/latest/semantics.html

To use more efficient bytecode without modying the Python semantics, you need to deoptimize if print() is replaced. I implemented that in my old FAT Python project :-)
https://fatoptimizer.readthedocs.io/en/latest/optimizations.html#call-pure

--

I would be more interested by a tool to update/regenerate M___hello__ in Python/frozen.c.

> It changes the semantics of Python.

When would the return value actually matter? Would it affect the value of the variable `__hello__`?