Open 8aad62bc-0568-48e0-8285-f8dc82dd5510 opened 3 years ago
Since this is an enhancement, it should only target 3.11.-- --Guido (mobile)
This does not apply to Windows (we don’t support curses there).
I’m trying to patch this bug.
Here are my current working questions:
What is the relationship between an fd (file descriptor) and a terminal? What software / hardware component goes to “fd 0” to receive input from it? Is there a GNU Screen command to receive stdin from “fd n”, fd 3 for example?
Looking at the source code:
def initscr():
import _curses, curses
# we call setupterm() here because it raises an error
# instead of calling exit() in error cases.
setupterm(term=_os.environ.get("TERM", "unknown"),
fd=_sys.__stdout__.fileno())
stdscr = _curses.initscr()
for key, value in _curses.__dict__.items():
if key[0:4] == 'ACS_' or key in ('LINES', 'COLS'):
setattr(curses, key, value)
return stdscr
why does initscr() begin by importing _curses and curses? Precompiled C curses and non-compiled C or is the second importing Python? How can a module be importing itself?
they call “setupterm”, a C curses function, because it raises an error if there’s a problem in case of just quitting. But how so, specifically? Do the errors get detected at stderr and then the terminal raises another error or something? I’m not clear on the details. Plus, why can they call this function without referring to the enclosing package, curses.setupterm? Is that a C thing that all functions are automatically added to the namespace?
Someone wrote that “initscr” actually calls “newterm”, in the code. So I guess I should look at the C newterm code, see how it works and see if a direct implementation is possible.
(I said in an email I would double post an email I sent but instead I’m posting more specific questions here related to that email. Also, I’m still studying an email Guido sent, so apologies for any redundant questions here.)
Thanks, Julius
I’m trying to patch this bug.
Here are my current working questions:
What is the relationship between an fd (file descriptor) and a terminal? What software / hardware component goes to “fd 0” to receive input from it? Is there a GNU Screen command to receive stdin from “fd n”, fd 3 for example?
Looking at the source code:
def initscr():
import _curses, curses
# we call setupterm() here because it raises an error
# instead of calling exit() in error cases.
setupterm(term=_os.environ.get("TERM", "unknown"),
fd=_sys.__stdout__.fileno())
stdscr = _curses.initscr()
for key, value in _curses.__dict__.items():
if key[0:4] == 'ACS_' or key in ('LINES', 'COLS'):
setattr(curses, key, value)
return stdscr
why does initscr() begin by importing _curses and curses? Precompiled C curses and non-compiled C or is the second importing Python? How can a module be importing itself?
they call “setupterm”, a C curses function, because it raises an error if there’s a problem in case of just quitting. But how so, specifically? Do the errors get detected at stderr and then the terminal raises another error or something? I’m not clear on the details. Plus, why can they call this function without referring to the enclosing package, curses.setupterm? Is that a C thing that all functions are automatically added to the namespace?
Someone wrote that “initscr” actually calls “newterm”, in the code. So I guess I should look at the C newterm code, see how it works and see if a direct implementation is possible.
(I said in an email I would double post an email I sent but instead I’m posting more specific questions here related to that email. Also, I’m still studying an email Guido sent, so apologies for any redundant questions here.)
Thanks, Julius
On Sun 5. Dec 2021 at 00:01, Éric Araujo \report@bugs.python.org\ wrote:
Change by Éric Araujo \merwok@netwok.org\:
---------- stage: -> needs patch versions: -Python 3.10, Python 3.6, Python 3.7, Python 3.8, Python 3.9
Python tracker \report@bugs.python.org\ \https://bugs.python.org/issue45934\
What is it about? The first message is by Guido, but not this message nor the title does not give me any idea what is it about.
Serhiy, this started with a post by Julius to the core-mentorship list. He wants to add newterm(), and found that an issue about that was just created by a different user (draganic1) -- apparently without a comment body, so it appears the first comment is mine (I didn't know that was possible in bpo).
Windows experts are welcome to take themselves off the nosy list.
Thank you Guido for clarification. I have found that post:
I agree that it is worth to implement newterm. In particular it would be useful in tests.
I’m currently planning on studying the C code for initscr and newterm so I can really understand how they work.
I’ll post any updates about this soon.
Thanks.
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: ```python assignee = 'https://github.com/serhiy-storchaka' closed_at = None created_at =
labels = ['type-feature', 'library', '3.11']
title = 'python curses newterm implementation'
updated_at =
user = 'https://bugs.python.org/draganic1'
```
bugs.python.org fields:
```python
activity =
actor = 'juliushamilton100'
assignee = 'serhiy.storchaka'
closed = False
closed_date = None
closer = None
components = ['Library (Lib)']
creation =
creator = 'draganic1'
dependencies = []
files = []
hgrepos = []
issue_num = 45934
keywords = []
message_count = 8.0
messages = ['407673', '407674', '407724', '407725', '407733', '407768', '407795', '407844']
nosy_count = 9.0
nosy_names = ['gvanrossum', 'paul.moore', 'tim.golden', 'zach.ware', 'serhiy.storchaka', 'steve.dower', 'Guido.van.Rossum', 'draganic1', 'juliushamilton100']
pr_nums = []
priority = 'normal'
resolution = None
stage = 'needs patch'
status = 'open'
superseder = None
type = 'enhancement'
url = 'https://bugs.python.org/issue45934'
versions = ['Python 3.11']
```