-
Notifications
You must be signed in to change notification settings - Fork 379
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[1.0][Config] remove path option. #366
Conversation
The question is why? Why the path has been removed? I don't want to have hardcoded filtername in the cached image path:
|
if you still need this feature you can rule it by configuring several resolvers: https://gist.github.com/makasim/9802395 |
@makasim yeah, I have answered in the email. Let's bring it here. I had this config in 0.*: http://pastebin.com/y4KGrBhv Now, using your example, I should do it like this: http://pastebin.com/ZXtwYMPK With |
Also, there is a lot of confusion with |
i don't understand this enough to give a real opinion. i don't know how the bundle works exactly - would it help to allow to embed resolver config into the filter config? or is there something else that could make this more elegant? his example indeed looks problematic... |
Basically, I would like bundle users to have full control on path where cached image is being stored.
Well, I think it is important to have flexibility in where do you store your generated content. For someone, who won't configure that, will use default stuff. But someone who want to have control - can't ;) I think we need to collect more opinions as well. |
@ASKozienko wdyt? |
@ASKozienko so basically it's not 1.0 than until you implement something you mentioned with 'id', right?
If so, how does the users specify different cached location per filter in one resolver? |
@svscorp just don't care where cache is |
@ASKozienko who doesn't? I do care, for an instance :) |
@svscorp |
i see why from the implementation we do not want to mix formats and caching. but i think the urls are important, as they face outside. what if there would be a configuration shortcut for the use case of svscorp? |
Yes! It is the thing I am missing since It is a first time I see feature remove (without providing new implementation) in tag increment, that's why I am protecting it here :) Thanks for fast responses guys, btw! |
yeah, or fork the project. But I would like to stay with original project better ;) |
@svscorp you can name the filter like
instead of
|
your statement is try to 1.x and any next releases but for 0.x. Quote from semantic versioning site. which we try to follow.
|
Yeah, I know. It's more like a hacky, naming the filter as a partial path :) Anyway, I understand I shouldn't refer to 0.* in 1.* discussion. But could you implement path configuration in 1.*? |
@svscorp there were several requests on same topic. I will implement when I have time not high priority though. |
@makasim I can do that. Just need to pick up the right strategy you guys will accept. |
@svscorp solution does not have to couple resolvers and filters, I am thinking of path generation strategy. The strategy maybe configured in several ways. Later inject to resolvers. I see here some problems, for example how to do the strategy which would satisfy all resolvers? Personally I don`t come with any clean solution yet. Second do we really need such overhead, for generated content? |
[1.0][Config] remove path option.
I merge this because I see some interest in ability to customize cached paths. So this would be done later |
👍 |
No description provided.