Current example in the README is not much of use

[kaushalyap]

Other

Current example on README create index generated pages by path, this is less useful, why?

• Users does not want to search queries that match the slug, they want the pages where content matches with their queries.

I know what is shown in README is a dummy example, but it is worth to have real life example like product search. Example using Medusa would be great.

[bidoubiwa]

Hello @kaushalyap ! I agree with your comment, we should definitely provide a better example in the README! Initially we thought about creating an example following this tutorial, where the data inside the markdown would be indexed in Meilisearch.

If you are willing to create the example it would be awesome 🔥 No problem if you don't :)

[kaushalyap]

Dear @bidoubiwa

Initially we thought about creating an example following this tutorial

That example is better and simpler than pulling data from Medusa, when I figured things out and have enough time I'll give Medusa example a try.

According to my current thinking, the approach would be to generate pages in gatsby and then pull matching product information from medusa using their gatsby plugin and transforming page path and product information to generate the search index.

While there is Meilisearch plugin which automatically generate index with products but we do not have access to slug to have a complete index, while we could update the already created index in front-end using the admin key, this seems like redundant work and security risk. What do you think?

[oluademola]

Hello @kaushalyap

Thank you very much for your suggestion. I am not sure I completely understand your example with Medusa, and it is something we have not considered. However, this being a potential example in the README, the intention would be to create something clear and simple as the previously mentioned tutorial. Let me know your thoughts and if I'm missing anything.

Thanks again.

[kaushalyap]

@oluademola I agree it should be simple and understandable by majority. So I would suggest going with a markdown page indexing example.