Re: Extending timeclock.el

[John Task]

Eli Zaretskii <eliz@gnu.org> writes:

> The way we extend existing packages is by adding code to the existing
> files, not by starting new FOO-x.el files.  This is a natural way of
> extending them, even if the extension adds something that is only
> loosely coupled with the previous functionalities.

Noted, thanks.

> I'm not sure this is being used.  Do you see any code that logs "b"
> lines in the file?

Interesting.  I looked at the code and it seems that nothing logs "b"
lines in the file.  Maybe it was removed?  Anyway, I think I can
safely ignore it then.

> I'm guessing that this means you move the log file to some archive,
> e.g., when a year ends or something.

Yes, that makes sense.  Though archiving log files doesn't seem too
statistic-friendly...  Maybe I'll have to warn about that somewhere.

> You don't have to design and implement anything that didn't exist
> unless you think it's a must-to-have feature.  We can always wait for
> users to request such a feature in some distant future.  What I
> described was just a rationale for having "O", and my point was that
> allowing people to say the project is finished is useful even if for
> now this is largely unused.

OK.

> As I said, the timelog file could announce what kind it is of.

Yes, that would be a solution.  I implemented a safe comment string
already, so something like #+FORMAT: should do the trick.

> You can build tests using the ert package.  We use it in our test
> suite (under the test/ subdirectory of the Emacs source tree).

Good.  I'll look into that.