Replies: 2 comments 2 replies
-
I thought quite a lot about using a patch methodology. The issue is two fold. The first is combinatorics. If a file is marked as EJS that means there are at least two different possible versions of that file depending on the features requested by the user for the application. Most often code router versus file router. The other issue is logic. In the case of the base files, in particular the file where the providers are defined, we need to layer all the providers requested by the user into that file. To accomplish these with patches would mean having all the possible combination of providers. |
Beta Was this translation helpful? Give feedback.
-
FWIW, I really considered this. Particularly for adding stuff in after the app has been built (a feature that the current alpha release supports). And I just couldn't get it to work. |
Beta Was this translation helpful? Give feedback.
-
@jherr @tannerlinsley
I think ejs is very hard to maintain, I tried to contribute to ejs, I have been always confused, because of the syntax and the inconsistent sytax highlighting, just unreadable code.
Then I was reminded of pnpm patch, applies a batch to your node_modules based on a diff, that is similar to git diff.
💡Why not use git to apply add ones?
Want to add Tanstack query just apply this patch
here is an example of a patch, generate with
git diff HEAD^ HEAD > changes.patch
the I can just do
git apply changes.patch
.I am just throwing an idea, but it think it is worth exploring
Beta Was this translation helpful? Give feedback.
All reactions