Recent revs have added print statements that trace pyzo's import and startup logic. This has put to rest a major concern, namely that importing pyzo or one of its packages or modules might *do too much*. We don't want pyzo imports *themselves *to interfere with integrating pyzo into Leo. Happily, this does not appear to be a problem:
1. Leo can probably import pyzo modules in almost any order. 2. Importing pyzo.core.pyzologging starts pyzo's logging, but that is benign. Leo can use or ignore the pyzo logger. 3. Using the ShellStackWidget starts up a Python kernel, which takes considerable time. However, the file browser demo does *not* start any kernel. Starting kernels does *not *happen automatically when pyzo starts up. In a real emergency, Leo could use import hooks. But that almost certainly won't be needed. *Summary* Pyzo's import structure is benign: - Importing the top-level pyzo module starts the pyzo logger. Leo can use or ignore the pyzo logger as desired. - Pyzo kernels do *not *automatically start during startup. This is crucial. After importing pyzo modules, Leo will be free to monkey patch any function, class or method as desired. In particular, shut*down* issues will be much easier to handle than startup issues. In short, Leo almost certainly can use pyzo code without any serious problems whatever. This is a big step forward. Edward -- You received this message because you are subscribed to the Google Groups "leo-editor" 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]. Visit this group at https://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout.
