Suggestion: define `elfeed-score-load`

Question

Suggestion: define `elfeed-score-load`

firmart opened this issue 4 years ago · 5 comments

The function elfeed-score-enable installs (or not) the sort function and load the score file. It would be nice to separate the snippet which loads the score file to a separate function so that loading and unloading functions are both available. When I modified my score file, I used elfeed-score-enable to load it, but the denomination sounds unnatural.

Answer 1 · 2021-01-28T02:59:47.000Z

Do you mean like elfeed-score-load-score-file (bound to l in the keymap)? 😁

Answer 2 · 2021-01-28T07:56:36.000Z

Ah, yeah, my bad... IMO, it would be nice to reduce a bit of user interaction: i.e. let write/load functions don't prompt by default so that advanced users using multiple score files could do something like `C-u = l`. There is a slight inconvenient thing when I add new rules in the score file. I have to immediately load the score file, otherwise if I do rescore some entries with `= s`, it would overwrite silently my score file. So may it may be worthwhile to use a file watcher to monitor the score file and auto-reload it when is changed. Or, alternatively, load again the score file before write it. P.S. Thanks again for this wonderful package! I used to have my own scoring mechanism, but your package simplify greatly the task. Michael <notifications@github.com> writes:

…

Do you mean like `elfeed-score-load-score-file` (bound to `l` in the keymap)? 😁 -- You are receiving this because you authored the thread. Reply to this email directly or view it on GitHub: #8 (comment)

Answer 3 · 2021-01-29T01:54:43.000Z

I'd be uncomfortable just re-loading the thing unconditionally, but maybe check if the file is newer than the last read before scoring & warn? But in that case, it could be challenging to "merge" any in-memory changes with those that occurred on-disk. Hmmmm.... Oh! We write to disk after every scoring run, so it wouldn't be that bad. So yeah, maybe warn on score file changes since the last sync before scoring again?

Thanks again for this wonderful package!

You are most welcome; just glad someone's getting some use out of it. Thank you for all the bug reports & suggestions!

Answer 4 · 2021-01-29T03:21:58.000Z

maybe warn on score file changes since the last sync before scoring again?

Sounds good to me. Michael <notifications@github.com> writes:

…

I'd be uncomfortable just re-loading the thing unconditionally, but maybe check if the file is newer than the last read before scoring & warn? But in that case, it could be challenging to "merge" any in-memory changes with those that occurred on-disk. Hmmmm.... Oh! We write to disk after every scoring run, so it wouldn't be that bad. So yeah, maybe warn on score file changes since the last sync before scoring again? > Thanks again for this wonderful package! You are most welcome; just glad someone's getting some use out of it. Thank you for all the bug reports & suggestions! -- You are receiving this because you authored the thread. Reply to this email directly or view it on GitHub: #8 (comment)

Answer 5 · 2021-03-03T01:45:45.000Z

I'm going to close this in favor of issue #13, which I think captures the problem more succinctly -- tagged you there @firmart