How do I create this nested page structure?

[hamishtaplin]

I'm trying to figure out how to use the following structure of pages:

/projects/category1/index.md
/projects/category1/project1/index.md
/projects/category1/project2/index.md
/projects/category1/project3/index.md

/projects/category2/index.md
/projects/category2/project1/index.md
/projects/category2/project2/index.md
/projects/category2/project3/index.md

/projects/category3/index.md
/projects/category3/project1/index.md
/projects/category3/project2/index.md
/projects/category3/project3/index.md

Using this structure, I want to create 2 content "types", with associated templates

• category page (eg. /projects/category1/index.md)
• project page (eg. /projects/category1/project1/index.md)

I have gatsby-source-filesystem configured to pull all my pages in from a folder (content)

{
  resolve: `gatsby-source-filesystem`,
  options: {
     name: `pages`,
     path: `${__dirname}/content/`,
   },
},

In gatsby-node I need to write 2 queries that will match the category pages and the project pages so I can feed them to different templates via createPage to generate the pages.

I also need to create a few different types of navigation, eg.

All the category indexes:

/projects/category1
/projects/category2
/projects/category3

All the pages with the current category (via the category slug, eg. category1)

/projects/category1/project1/index.md
/projects/category1/project2/index.md
/projects/category1/project3/index.md

All the projects

/projects/category1/project1/
/projects/category1/project2/
/projects/category1/project3/
/projects/category2/project1/
/projects/category2/project2/
/projects/category2/project3/
/projects/category3/project1/
/projects/category3/project2/
/projects/category3/project3/

I'm pretty sure all of this can be done via a Regex on fileAbsolutePath. For example, this simply gets everything in the projects folder.

{
  allMarkdownRemark(filter: {fileAbsolutePath: {regex: "/projects/"}}) {
    nodes {
      id
      fields {
        slug
      }
    }
  }
}

What I have no idea how to do is to

• isolate the top-level category pages from their children
• given the slug of a category page (eg. category1) get all of it's children
• get everything apart from the category pages