Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.lang.python > #95306
| Date | 2015-08-12 20:44 +0200 |
|---|---|
| From | "Sven R. Kunze" <srkunze@mail.de> |
| Subject | Re: Hooking Mechanism when Entering and Leaving a Try Block |
| References | <55CA6D7F.1060705@mail.de> <CALwzidkBoDo1TZj2CWfimYtagtArN9r9dvw1Ke9HqcthdipV4Q@mail.gmail.com> <55CB6EAC.5040004@mail.de> <CAPTjJmoqg9sA-QB3RoRPQFx1osOYyBzff-UX8f2zrzxWpt7sYQ@mail.gmail.com> |
| Newsgroups | comp.lang.python |
| Message-ID | <mailman.132.1439405104.3627.python-list@python.org> (permalink) |
On 12.08.2015 18:11, Chris Angelico wrote: > On Thu, Aug 13, 2015 at 2:05 AM, Sven R. Kunze <srkunze@mail.de> wrote: >> Unfortunately, no. :( >> >> It should work out of the box with no "let me replace all my try-except >> statements in my 10 million line code base". > (Please don't top-post.) Is this some guideline? I actually quite dislike pick somebody's mail to pieces. It actually pulls things out of context. But if this is a rule for this, so be it. > Sounds to me like you want some sort of AST transform, possibly in an > import hook. Check out something like MacroPy for an idea of how > powerful this sort of thing can be. Sounds like I MacroPy would enable me to basically insert a function call before and after each try: block at import time. Is that correct so far? That sounds not so bad at all. However, if that only works due to importing it is not a solution. I need to make sure I catch all try: blocks, the current stack is in (and is about to step into). Ah yes, and it should work with Python 3 as well. Regards, Sven
Back to comp.lang.python | Previous | Next | Find similar | Unroll thread
Re: Hooking Mechanism when Entering and Leaving a Try Block "Sven R. Kunze" <srkunze@mail.de> - 2015-08-12 20:44 +0200
csiph-web