The current extractor looks for a couple of meta tags to extract information from, but these are only the few common one's also used by search engines. It'd be nice if we could pass more data that won't be visible to users. Of course, best option would be if we could pass the data from astro component itself - perhaps getImageUrl could take params that would be passed to lib? Or it itself could somehow generate the image, making it available in dev mode?
Though even just reading all meta tags, having an option to add a function to extract information, or reading a specific tag (e.g. <script type="takumi-data">) would be helpful.
Currently, the data needs to be gotten externally by url/path, but I don't see a way to access astro data like content stores to get the data from there.
edit: I just noticed that the parsed document is being passed, so this is possible to work around, but still would be nice to have a more ergonomic solution (and also have this be more visible in documentation)
The current extractor looks for a couple of meta tags to extract information from, but these are only the few common one's also used by search engines. It'd be nice if we could pass more data that won't be visible to users. Of course, best option would be if we could pass the data from astro component itself - perhaps getImageUrl could take params that would be passed to lib? Or it itself could somehow generate the image, making it available in dev mode?
Though even just reading all meta tags, having an option to add a function to extract information, or reading a specific tag (e.g.
<script type="takumi-data">) would be helpful.Currently, the data needs to be gotten externally by url/path, but I don't see a way to access astro data like content stores to get the data from there.
edit: I just noticed that the parsed document is being passed, so this is possible to work around, but still would be nice to have a more ergonomic solution (and also have this be more visible in documentation)