Repository navigation
use bytes for code objects #45363
Description
| BPO | 1022 |
|---|---|
| Nosy | @gvanrossum, @brettcannon |
| 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/gvanrossum'
closed_at = <Date 2007-10-24.20:02:52.703>
created_at = <Date 2007-08-25.17:47:08.019>
labels = ['interpreter-core']
title = 'use bytes for code objects'
updated_at = <Date 2007-10-24.20:02:52.702>
user = 'https://bugs.python.org/nnorwitz'bugs.python.org fields:
activity = <Date 2007-10-24.20:02:52.702>
actor = 'gvanrossum'
assignee = 'gvanrossum'
closed = True
closed_date = <Date 2007-10-24.20:02:52.703>
closer = 'gvanrossum'
components = ['Interpreter Core']
creation = <Date 2007-08-25.17:47:08.019>
creator = 'nnorwitz'
dependencies = []
files = ['8317']
hgrepos = []
issue_num = 1022
keywords = ['patch']
message_count = 4.0
messages = ['55290', '55436', '55597', '56727']
nosy_count = 3.0
nosy_names = ['gvanrossum', 'nnorwitz', 'brett.cannon']
pr_nums = []
priority = 'low'
resolution = 'rejected'
stage = None
status = 'closed'
superseder = None
type = None
url = 'https://bugs.python.org/issue1022'
versions = ['Python 3.0']
Activity
nnorwitz commentedon Aug 25, 2007
This patch has a hack to read in marshaled code objects. Bytes can't be
marshaled currently (they are stored as strings). So when reading a
marshaled code object, the hack converts them to bytes.
All tests except test_modulefinder pass.
The issue is that since bytes are mutable, they aren't suitable for code
objects since a user could change them at any time.
gvanrossum commentedon Aug 29, 2007
I'll defer thinking about this until post-a1.
brettcannon commentedon Sep 3, 2007
I have security/stability issues with code objects having mutable
bytecode. This would allow someone to possibly crash the interpreter
with crappy bytecode. I also have a worry that someone might come up
with some clever way of causing different code to execute through
manipulating what LOAD_GLOBALS or STORE_GLOBALS accesses/sets.
gvanrossum commentedon Oct 24, 2007
We won't be doing it this way.