Closed zackad closed 4 months ago
From dependencies graph above, we need to pull this packages:
I have to ask, is the purpose of this repo to have a continued development on a prettier plugin for melody? or was the original project an "almost there" solution for formatting twig templates and you want to use it as a starting point?
If it's the latter option, and you're looking to create a prettier plugin specifically for twig, and melody support isn't really critical, I have some thoughts.
I would have to say that the most salvageable parts of the original project would have to be the printer and the tests.
I do wonder if a project such as Twig lexer could potentially be a more suitable starting point for returning AST.
Not having worked within the melody framework I am just speculating, but it seems like each template is designed to have a .melody.twig file sitting alongside a .js file. The reason I believe that this is an important point is that the .melody.twig templates aren't supposed to have any kind of data manipulation going on, hence the absence of arrow functions in the parser.
Happy to move this to a discussion or discuss further on discord, but personally what I am looking for is a prettier plugin for twig files that makes no assumptions about the "flavor" of twig that your project is running, would you be happy for this project to go in that direction?
To be honest, I don't even know how this project even works. All I want is a tool that would reformat twig in a consistent and sensible way. Unfortunately this is the only tools I found so far that fulfill my need.
I have to ask, is the purpose of this repo to have a continued development on a prettier plugin for melody? or was the original project an "almost there" solution for formatting twig templates and you want to use it as a starting point?
Personally I want to drop all melody
stuff from this project and focus only on twig template. But at the same time I don't want to discredit the original project/author.
I would have to say that the most salvageable parts of the original project would have to be the printer and the tests.
I do wonder if a project such as Twig lexer could potentially be a more suitable starting point for returning AST.
parser, lexer, AST and such is a foreign concept to me. I have no idea how they work together. That's why there's not much activity on this project since I'm still learning.
By the way, are you interested to become maintainer of this project? Seems like you are more familiar on this topic than I do. Your contribution is greatly appreciated.
Happy to move this to a discussion or discuss further on discord, but personally what I am looking for is a prettier plugin for twig files that makes no assumptions about the "flavor" of twig that your project is running, would you be happy for this project to go in that direction?
I prefer discussing in the open forum like this, so everyone can participate or atleast see what has been discussed without creating yet another account.
Don't get too carried away this is all pretty foreign to me too. I have had about 3 shots at building a robust formatter for twig so far, but each time it's been too much to do on my own so happy to work through it with you. If you'd like me to help maintain the project I would be happy to, otherwise also happy to submit some PRs where I can.
Of couse I'm gonna stick around to make sure that this tool work properly since I'm use it daily. For now I've invite you as collaborator.
Don't feel burdened by this role. I just want to make sure that when I'm busy with other thing, there's someone who can takeover managing PR.
Thanks mate. If we are going down the path of targeting this directly at twig rather than melody. How do you feel about renaming the package to avoid confusion. I would propose keeping some attribution in the README regarding the origins of a lot of the code so that the original author is still attributed.
I would probably aim for a 1.0.0 release perhaps you would be able to setup a milestone and maybe a discussion and we could discuss in there what that could look like
Agree with package renaming, which means we need to publish as separate package. AFAIK we can't rename published package on npm. I'm not sure if we should publish the new package name as-is right now or wait until release 1.0.
I've setup milestone and enable discussion. But I'm not sure if it would be any different that using github issue. Maybe I'm not aware of different use case, never use it.
This project depends on several
melody-*
packages that hasn't been updated for awhile. I don't think that forking and re-publishing those packages is not an ideal solution. Let's just import it directly into this project.