Repository navigation
Fixed block ordering code in compiler.pyassem #46724
Description
| BPO | 2472 |
|---|---|
| Nosy | @nascheme, @pitrou |
| Files |
Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.
Show more details
GitHub fields:
assignee = 'https://github.com/nascheme'
closed_at = <Date 2009-02-06.21:11:25.677>
created_at = <Date 2008-03-24.12:09:38.554>
labels = ['type-bug', 'library']
title = 'Fixed block ordering code in compiler.pyassem'
updated_at = <Date 2009-02-06.22:26:41.855>
user = 'https://github.com/pitrou'bugs.python.org fields:
activity = <Date 2009-02-06.22:26:41.855>
actor = 'pitrou'
assignee = 'nascheme'
closed = True
closed_date = <Date 2009-02-06.21:11:25.677>
closer = 'nascheme'
components = ['Library (Lib)']
creation = <Date 2008-03-24.12:09:38.554>
creator = 'pitrou'
dependencies = []
files = ['9839']
hgrepos = []
issue_num = 2472
keywords = ['patch']
message_count = 5.0
messages = ['64410', '64412', '64473', '81300', '81312']
nosy_count = 2.0
nosy_names = ['nascheme', 'pitrou']
pr_nums = []
priority = 'normal'
resolution = 'accepted'
stage = 'resolved'
status = 'closed'
superseder = None
type = 'behavior'
url = 'https://bugs.python.org/issue2472'
versions = ['Python 2.6']
Activity
pitrou commentedon Mar 24, 2008
This is a rewrite of the block ordering code in the compiler package
(specifically, the flowgraph part). The previous code was littered with
self-admitted "hacks", "fixups" and "XXX" :-) They are all removed and
replaced with a clean
order_blocksfunction which does the rightthing from the start.
The patch also replaces a wrong startBlock() with a nextBlock() in
compiler.pycodegen (startBlock can only be used when the previous block
does an unconditional transfer to another one, otherwise the two
adjacent blocks may not be emitted in order).
I've run test_compiler a couple of times, and tested execution of
several functions. They all run fine. Unless someone has specific
reasons to reject the patch, I'd recommend applying it even if not many
people use the compiler package :) I needed the fixes for my work on bpo-2459.
pitrou commentedon Mar 24, 2008
By enabling TEST_ALL I've just run
test_compiler.CompilerTest.testCompileLibraryagainst the wholestdlib, and there were no errors. It's a good sign :-)
pitrou commentedon Mar 25, 2008
Neal, I don't have sufficient permissions to assign bugs to anybody, but
here you are in the nosy list. :)
nascheme commentedon Feb 6, 2009
Thanks Antoine. For some reason I don't think I ever got an email about
this issue. I did some further cleanups and optimizations. Committed
as SVN rev 69373.
Lib/compiler is still in need of some fixing since it doesn't handle
decorators and some other new language features. I'll try to keep at it.
pitrou commentedon Feb 6, 2009
Thanks! If I knew you hadn't received any notification I would have
tried to ping you... I'm glad this is finally accepted :)