why isn't php injected via the html parser

[Hippo0o]

PHP files are just html with php code blocks in it. So why does this parser work the other way around combining html around the php code blocks to inject html? Would'nt it make more sense if the html parser injected php?

I'm asking because i was investigating issues i have using neovim and files that mainly have html with some php code in it.

[calebdw]

PHP is a templating engine which can be used for any type of text file---the treesitter query injects html because that's the most common, but the node itself is called text in the grammar which is indicative of this. Additionally, it shouldn't be the html parser's responsibility to analyze .php files.

I'm asking because i was investigating issues i have using neovim and files that mainly have html with some php code in it.

What issues were you having?

[Hippo0o]

@calebdw i wrote about it here https://github.com/nvim-treesitter/nvim-treesitter/pull/4653#issuecomment-1593091425

The main issue seems to be that the @combined query will combine the ranges of the whole text node which collides with the ranges of the php nodes inbetween. And because of the ordering or smth the html ranges overshadow all inbetween.

[Hippo0o]

so the problem primarly lies with the neovim treesitter implementation of querying injections

[calebdw]

Hmm, I don't see any issues with the Neovim hightlighting?

so the problem primarly lies with the neovim treesitter implementation of querying injections

Sounds like this issue can be closed then

[Hippo0o]

Thanks for giving me an answer.