Re: lua-factory: the analysis of the changes proposed in #753141
- From: George Sedov <radist morse gmail com>
- To: Victor Toso <lists victortoso com>
- Cc: "grilo-list gnome org" <grilo-list gnome org>
- Subject: Re: lua-factory: the analysis of the changes proposed in #753141
- Date: Fri, 30 Oct 2015 12:14:36 +0300
Hi,
So, this changes a lot. What I like least is that you are
splitting
functions provided by grl-lua-library. I do like grl as a library
-
globally accessible.
aside of `callback` which is the only one I think it makes sense
to
keep as an argument here.
Why is callback so different from options in your opinion? They are
both context-dependent in the very same way. They both do not make
any
sense outside of their respective scopes for which they were
created.
And they both do not belong to the globally accessible library,
IMO.
Callback makes more sense to me to have as parameters then the
options
as one could simply say that callback per se is not part of Grl
library
- Plugin developer should keep reference to the callback (as userdata
on async functions) in order to finish a operation later on
As functions, get_media_keys or get_requested_keys should always be
reacheable when a operation is ongoing IMHO.
After three or four fetch() the user might want to get those keys to
only write the keys that were requested from application or only
write the keys that has no value yet. Passing a table of utility
functions around for that does not seem right to me.
Still do not see how it is different from the callback. Callback
finishes the operation, so the user should pass it as a userdata even
after 3 or 4 or whatever fetch()'s, just like options(), if he wants
some info regarding the operation. These two functions are
complimenting each other in a way: you call options() when you want to
get some info about the operation, and you call callback() when you
want to return some info: media or (I hope in the future it'll be
implemented) error.
I might prefer to have an actual GrlMedia as parameter and maybe the
operations options as argument (as table) as well. I think this would
be enough to remove context-dependend functions -- aside of callback.
Funny you should say that, because that's exactly how I ended up
implementing it. :) Well, at least for the media I did.
<snip>
grl_source_search (query, media, options, callback)
local userdata = {}
if options.count < 5 then
grl.callback("fail!")
callback()
end
-- save what you want here
userdata.media = media
userdata.options = options.req_keys
userdata.callback = callback
grl.fetch("something", fetch_done, userdata)
</snip>
What do you think?
search() is a bad example, since it doesn't have access to any kind of
media at all, but here's how the browse() now looks in vk.lua (you can
see the code in #744238)
function grl_source_browse (media, options, callback)
local skip = options ("skip")
local count = options ("count")
if (not media) or (not media.id) then
### root folder ###
else
local media_id = split (media.id, "_")
### work further with media-id ###
And now you want to make the same thing for the options() as well. Yes,
it will make the API clearer, but I wonder if there will be some
performance implications. Here's the list of keys we provide in
get_options:
"type" (like I don't know what function I'm currently writing)
"count"
"skip"
"flags"
"key-filter" (with an additional argument "key name")
"range-filter" (same as previous)
"operation-id"
"media-id" "query" "search" (mutually exclusive, and, IMO, redundant)
"requested-keys"
As you can see, the "key-filter" and "range-filter" are the biggest
problem. Unfortunately, I'm not quite sure how filtering works. Is it a
set of additional restrictions aimed to narrow the results? Then we
probably should provide all of them that are listed in the respective
***_list() functions. How much of them can there be? How much time it
will take to retrieve all of them for every operation?
Also, about the "flags". Right now it returns an int, which is pretty
useless unless you decode it. But even if we return a set of flags
(namely "normal" "full" "idle_loop" and "fast_only") what exactly lua
plugin can use it for? "normal" is a default, when nothing else is on.
"full" means that grilo should use all the plugins available in order
to retrieve the needed info. Pointless in lua. "idle_loop" means (and I
quote) "use idle loop to relay results". Whatever this means, it has
nothing to do with lua. And finally "fast_only". And while yes, lua
could use this info, it is redundant since we already have a list of
requested-keys. And if we provided the list of slow-keys correctly,
then we should be fine.
Cheers,
George
[
Date Prev][Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]