On Tue, Oct 18, 2016 at 8:58 AM, 'Terry Brown' via leo-editor < [email protected]> wrote:
> No. Commit hooks are always for systems with git. > > What I meant was that it would be confusing to sometimes report, at the > top of the log pane, the preceding commit hash, and sometimes report the > current commit hash depending on whether .git was available. > I now see that this is primarily a "fit and finish" issue. Leo's signon should follow accepted practice by giving build info, including date and hash. So I'm now convinced that git hooks should stay. We can debate how often signon info helps devs fix bugs, but that's a secondary consideration. Technical point, my recollection when the system was first developed > was that best practice was to use bash for hooks. Of course anyone > developing Leo has Python on their system somewhere, but not > necessarily on the path, whereas git, even on Windows, executes hooks > in bash. > The hook can add to sys.path as needed--it's running from a known place, namely leo-editor/.git. More importantly, the hook should generate a .json file compatible with those created and read by leoVersion.py. Calling leoVersion.create_commit_timestamp_json is the proper way to do this. I'm going to create the new hooks next, before b1 goes out the door. 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.
