Where can I post a new debugger?

F

Francesco

I have written a new debugger.
I think people will find it useful.
Where can I post it for people to download?

Thanks,
Nir

Hello Nir,

You could create a soureforge project (?)!
 
I

Irmen de Jong

Ziaran said:
I have written a new debugger.
I think people will find it useful.
Where can I post it for people to download?

I suppose you don't have a web page?
You can just zip it together and upload it to a web page.
Or if you're serious about it, create a Sourceforge project
and put it there ;-)

--Irmen
 
L

Lothar Scholz

Ziaran _ said:
I have written a new debugger.
I think people will find it useful.
Where can I post it for people to download?

What is so different from the existing ones ?
 
?

=?iso-8859-1?Q?Fran=E7ois?= Pinard

Lothar Scholz said:
What is so different from the existing ones ?
[Ziaran]
http://ziaran.freehosting.net/

An URL is hardly an explanation, and alone, hardly an English sentence.
People should not have to start a browser merely to figure out an email.

You think people will find your new debugger useful? Good for you!
Now, if you want us to think the same, you should tell us a little more.
Once interested, then we may start a browser :).
 
Z

Ziaran

François,

I did not mean to piss you off.

I like URLs, they are so easy to click, and there is always the possibility
for pleasant surprises inside.



Allow me to correct myself. I hope people will find my pdb debugger
modification useful.



I started learning Python 3 weeks ago, and was trying to implement an
academic exercise for that purpose, but alas I had a bug in my
implementation, and trying to debug it I soon found out that Idle does not
break on breakpoints that are crossed by non-main threads. I tried pdb too
and found that for different reasons it was not suitable for that task
either. And then two nights of web searching did not bring me any closer to
a life-saving debugger (and believe me that if I ever learned anything in
the respectable software engineering company I work for, it is web surfing.
I might say I am one of the world-wide-best in that area). So, anyway, I
decided do modify pdb for my needs.



I am fond of source-level text debuggers.

They are usually much more powerful than GUI based debuggers, and I saw that
pdb has some great features.



I think my modification takes pdb closer to its potential, however modest it
may be. It helped me find my bug in no time, and I know Python much better
than when I started, too.



For a somewhat more technical explanation, let me quote from the __doc__



"A remote debugger for Python 2.3.3



This is a modification of pdb.py to allow remoting of the debugger IO to a
seperate server process, and support convenient debugging of multiple
threads.



New debugger commands:

ts (Thread Set) - Set debugger focus to a specific thread, or list

active threads.

tc (Thread Command) - Send command to a specific thread.

bn (Break Now) - Break all threads immediately, or as soon as they

return to execute python code.

save - Save breakpoints of the current debug session.

load - Load breakpoints of the current debug session.



Other improvements:

Post mortem on unhandled exceptions.

Debugger starts in right frame.



To debug a thread (except the main thread):



1. Place the statement 'rpdb.set_trace()' as the first statement in the
run() method of your thread. This statement will be ignored if the script
was not invoked by the debugger, so it can remain in the code.



2. When execution reaches that statement the debugger will break and the
debugger-server will report the new thread."



And here is the link again; hoping this time people in need will believe it
worth while to click:

http://ziaran.freehosting.net



Yours,

Nir (Ziaran)



Lothar Scholz said:
What is so different from the existing ones ?
[Ziaran]
http://ziaran.freehosting.net/

An URL is hardly an explanation, and alone, hardly an English sentence.
People should not have to start a browser merely to figure out an email.

You think people will find your new debugger useful? Good for you!
Now, if you want us to think the same, you should tell us a little more.
Once interested, then we may start a browser :).
 
M

Michel Claveau/Hamster

Bonsoir !

Je comprend votre point de vue.
Cependant, une adress URL ne me gène pas (avis purement personnel).

Bonne nuit.
 
J

Jeremy Fincher

François Pinard said:
Lothar Scholz said:
What is so different from the existing ones ?
[Ziaran]
http://ziaran.freehosting.net/

An URL is hardly an explanation, and alone, hardly an English sentence.
People should not have to start a browser merely to figure out an email.

You think people will find your new debugger useful? Good for you!
Now, if you want us to think the same, you should tell us a little more.
Once interested, then we may start a browser :).

I think a URL, when linked to an explanatory page (such as the one he
linked to), is a great explanation. Why should he have to say the
same thing twice (on usenet and then again on the webpage) because
you're too lazy to click a link?

Jeremy
 
?

=?iso-8859-1?Q?Fran=E7ois?= Pinard

[Jeremy Fincher]
I think a URL, when linked to an explanatory page (such as the one he
linked to), is a great explanation. Why should he have to say the
same thing twice (on usenet and then again on the webpage) because
you're too lazy to click a link?

Every single day, I receive hundreds of suggestions to visit various
URLs. Spambayes is a great tool for filtering most of these out. I've
been on mailing lists where people had this habit of sending messages
holding URLs, probably with the feeling of distributing pieces of cake.
Most of the times, in my opinion, they were not even worth following.

It is not a question of being lazy, but to make better use of our time,
a resource which is much in demand for many of us. I rather think the
laziness is on the shoulders of the guys who publish these unexplained
URLs, as they would rather have hundreds of people starting a browser,
than providing a short explanation themselves.

It happens that maintainers broadcast announcement for new releases,
describing solved bugs in detail, telling where the archive or the Web
site is, while failing to provide one or two sentences describing what
the package does. This is usually an oversight much more than laziness
or bad will, as these maintainers are working so deep in their own
packages, it is so evident for them, that they forget that it may not be
on the mind of each and every reader of the mailing list.

P.S. - Here is another issue, only slightly related. There is a very
sad trend I observe growing since many years, in which maintainers tend
to over-praise their own time, and more or less despise the time of
their possible contributors. They ask users to first spouse their work
methods, each maintainer its own set, of course. This has the effect of
creating religions around packages. Some maintainers even like that :).
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Members online

No members online now.

Forum statistics

Threads
473,755
Messages
2,569,536
Members
45,011
Latest member
AjaUqq1950

Latest Threads

Top