My first impression is that I don't think this has anything to do with upgrading. If you're using a browser that makes 2 parallel requests to the endpoint that triggers pdb then you will run into really weird issues where you have two pdbs running at once. However, this has pretty much always been the case so you're probably aware of it.
Are you using pserve with or without `--reload`? On Wed, Mar 15, 2017 at 12:34 PM, Lele Gaifax <[email protected]> wrote: > Hi all, > > I've spent a couple of hours trying to understand an issue I'm facing with > Pyramid 1.8.3 when I enter into an interactive pdb session. It's not clear > what happens, and even less what's causing it: I tried hard to distill a > reproducible recipe, or at least to narrow the context, but without luck. > > I will try to summarize, sorry for the length and bear with me please... > > It all started with the goal of "modernizing" the dependencies of an > application[1], in particular upgrading it from Pyramid 1.6 to 1.8 and from > SQLAlchemy 1.0 to 1.1. This took very little effort[4], all tests pass, and > everything seems working as expected. > > Given that yesterday I noticed a minor i18n issue on the production > instance, > and verified that it's still present in my unreleased/modernized version, I > tried to spot the glitch using current sources, adding a pdb.set_trace() > call at the right point and started my local instance as usual. When the > execution enters into pdb however it is immediately clear that something > strange is happening with the in/out streams: the usual pdb interactive > session begins with something like (“@” is the cursor position) > > > /path/to/file.py(123)name_of_function() > -> statement_at_line_123_of_file.py > (Pdb) @ > > where you can enter pdb commands. In particular, I can enter into a full > interactive interpreter, something like > > (Pdb) interact > *interactive* > >>> import sys > >>> print(sys.version_info) > sys.version_info(major=3, minor=6, micro=1, releaselevel='candidate', > serial=1) > >>> @ > > With Pyramid 1.8.3 instead the “(Pdb)” prompt is not at the right place > and I > may[2] see the following > > > /home/lele/wip/sol3/src/sol/models/competitor.py(184)caption() > -> if self.player1 is None: > (Pdb) > /home/lele/wip/sol3/src/sol/models/competitor.py(184)caption() > -> if self.player1 is None: > @ > > Effectively, it looks like *two* different interleaved sessions are going > on > together, see: > > interact > *interactive* > (Pdb) l > 179 if player_caption is None: > 180 player_caption = lambda player, h, l, cc: > player.caption(html=h, localized=l, > 181 > css_class=cc) > >>> 182 > 183 import pdb; pdb.set_trace() > 184 -> if self.player1 is None: > 185 return _('Player NOT assigned yet!', just_subst=not > localized) > 186 else: > 187 captions = [player_caption(self.player1, html, > localized, css_class)] > 188 if self.player2 is not None: > 189 captions.append(player_caption(self.player2, > html, localized, css_class)) > @ > > At that point I hit the Enter key once and I get the (Pdb) prompt, another > Enter repeats the "l" command I entered above and the I get the usual ">>>" > Python prompt: > > (Pdb) > 190 if self.player3 is not None: > 191 captions.append(player_caption(self.player3, > html, localized, css_class)) > 192 if self.player4 is not None: > 193 captions.append(player_caption(self.player4, > html, localized, css_class)) > 194 return njoin(captions, localized=localized) > 195 > 196 description = property(caption) > 197 > 198 @property > 199 def player1FullName(self): > 200 "Full name of the first player" > >>> @ > > The "interact" here is just an example to show a different prompt, but the > very same confusion happens trying normal interactive pdb commands. Again, > the > feeling is that I have *two* interactive sessions running at the same time, > contending my inputs and interleaving their output. > > If I switch back to the master branch[3], reinstall its requirements.txt > (thus > running with Pyramid 1.6.1, waitress 0.9.0 and no hupper) and repeat the > session, none of the above non-sense happens! > > I then created a "starter" Pyramid project from scratch, but could not > reproduce the issue in any way... > > I have obviously read the changelogs of Pyramid and waitress, but did not > spot anything that seems related. > > Any hint on how to dig further, or shedding some light, will be highly > appreciated! > > Thanks in advance, > ciao, lele. > > [1] https://bitbucket.org/lele/sol > [2] “may”, because it's non-deterministic > [3] almost the same sources, different pinned requirements > [4] the only change was related to the dropped logging_file_config() > function > -- > nickname: Lele Gaifax | Quando vivrò di quello che ho pensato ieri > real: Emanuele Gaifas | comincerò ad aver paura di chi mi copia. > [email protected] | -- Fortunato Depero, 1929. > > -- > You received this message because you are subscribed to the Google Groups > "pylons-discuss" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > To view this discussion on the web visit https://groups.google.com/d/ > msgid/pylons-discuss/87efxyigvr.fsf%40metapensiero.it. > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "pylons-discuss" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/pylons-discuss/CAKdhhwH2L3ByRDmmnqiL5tQzoDwACkrOzt3GPgt%2BU7MgwqVszw%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
