![](https://secure.gravatar.com/avatar/ab1880b4dd926314fbb103b666d1d0bb.jpg?s=120&d=mm&r=g)
Hi everyone, I'm going to be leading a class on Python at the University of California, Berkeley next semester (starting in January). I'm interested in using actual bugs in Python as exercises for the class, the goal being 1) to give the students something "real" to work with, and 2) to (hopefully) contribute back anything we manage to put together to the Python dev team. What I'm looking for are some oustanding bugs which the more experienced devs feel would be straightforward to implement (which might not have been implemented yet because they're not very interesting, or there are simply too many other things to do right now). I'd appreciate any and all suggestions any of you might be able to provide. I think that Python will have a better chance at wider adoption if universities would provide courses for the language. I'm running this class partly because I'd like to give back to the community which produced a language I've enjoyed using, and any assistance would be greatly appreciated. If you feel this is off topic, please feel free to reply to me off-list. Thanks, Randy Chung
![](https://secure.gravatar.com/avatar/d6b9415353e04ffa6de5a8f3aaea0553.jpg?s=120&d=mm&r=g)
"Randy Chung" <aoshi@OCF.Berkeley.EDU> wrote in message news:41C02165.2010000@ocf.berkeley.edu...
Hi everyone,
I'm going to be leading a class on Python at the University of California, Berkeley next semester (starting in January).
Great.
I'm interested in using actual bugs in Python as exercises
Please consider including review of existing patches. Besides being useful, it will also teach students how to submit good patches of their own. Your primary choice is whether to work on changes in the C code for the interpreter and builtin modules or changes in the Python code in standard library modules. The former would be more appropriate for a class on application programming in C, so I suggest the latter for learning Python programming. Besides which, improving the library will probably be a focus of the just-starting 2.5 cycle. You can access existing patch and bug lists via http://sourceforge.net/tracker/?group_id=5470 For each, you can narrow list by selecting, for instance, Category 'Python Library'. Terry J. Reedy
![](https://secure.gravatar.com/avatar/0be5431acfe4c377fe26256e4adf68ca.jpg?s=120&d=mm&r=g)
Terry Reedy wrote:
"Randy Chung" <aoshi@OCF.Berkeley.EDU> wrote in message news:41C02165.2010000@ocf.berkeley.edu... [SNIP]
I'm interested in using actual bugs in Python as exercises
Please consider including review of existing patches. Besides being useful, it will also teach students how to submit good patches of their own.
To go along with this suggestion, see http://www.python.org/dev/dev_intro.html for the basic overview of what a review tends to consist of. -Brett
![](https://secure.gravatar.com/avatar/c7b89689e0799cf497f55e0f9c1260d7.jpg?s=120&d=mm&r=g)
On Dec 15, 2004, at 2:58 PM, Terry Reedy wrote:
Your primary choice is whether to work on changes in the C code for the interpreter and builtin modules or changes in the Python code in standard library modules.
The PyPy folks might be useful too given that they are writing Python in Python. http://codespeak.net/pypy/ -a
![](https://secure.gravatar.com/avatar/3acb8bae5a2b5a28f6fe522a4ea9b873.jpg?s=120&d=mm&r=g)
Randy Chung wrote:
I'm going to be leading a class on Python at the University of California, Berkeley next semester (starting in January). I'm interested in using actual bugs in Python as exercises for the class, the goal being 1) to give the students something "real" to work with, and 2) to (hopefully) contribute back anything we manage to put together to the Python dev team.
Welcome to the club! I'm just running a class on development processes in open source software (i.e. with a somewhat different scope), where students already have fixed bugs in Mozilla and PHP - unfortunately, none of them were interested in Python. Nevertheless, this is quite fun for both the students, and myself - especially when the Mozilla guys do a review after 3h, the super review after 36h, and explain that they cannot commit the fix because the code is frozen (and then do after two weeks, well before the presentation in the class).
What I'm looking for are some oustanding bugs which the more experienced devs feel would be straightforward to implement
I'd look in the one-to-two-year old range of bugs. One risk in such a course is that some of the developers fixes the bug while your student is working on it, which is discouraging. How many bugs do you need? The *really* easy ones are documentation bugs, but I expect that students don't see this as a sufficient challenge. There are currently roughly 100 documentation bugs open. You can further filter out all assigned bugs; the unassigned ones are likely resting because nobody cares (yet). Looking at unassigned bugs starting from offset 300, I think all of the following bugs might be suitable: - 840065 - 839151 - 837242 (really trivial) - 828743 - 813986
If you feel this is off topic, please feel free to reply to me off-list.
This is clearly python-dev relevant. Regards, Martin
participants (5)
-
"Martin v. Löwis"
-
Andrew P. Lentvorski, Jr.
-
Brett C.
-
Randy Chung
-
Terry Reedy