Cabret

Cabret is a yaml based static site generator, ideally with the same features as Hugo but with a simpler model. Here is a simple example of a Cabretfile.yaml

Build & Install

$ go build -v -o ./bin/cabret ./cmd/cabret
$ go install -v ./cmd/cabret

Introduction

build:
  - pipeline:
      - source: index.html
      - use: layout
        path: layouts/base.html
      - target: dist/index.html
  - pipeline:
      - source: posts/{{ .Id }}.md
      - plugin: markdown
      - use: layout
        path: layouts/base.html
      - target: dist/posts/{{ .Id }}/index.html

Cabret is based on the idea of a data pipeline. Along a pipeline flows a list of Content items that are just items with a data, mime type and metadata.

type Content struct {
    Type      string
    Metadata  map[string]any
    Data      []byte          
}

The Type is the mime type of the Data. The Metadata can be whatever the user wants and operations can populate it with special fields as they please. In some cases its useful to have Data be nil, when that happens the Type should be cabret.MetadataOnly (just the string "metadata-only")

Operations

Each pipeline is a list of operations, the first field in an operation should be one of source, use or target.

• source: <path pattern>

  A path pattern is a "glob pattern" with captures, some examples are

  posts/{id}.md 
  {{folders}}/{year}-{month}-{day}.json
  

  the only difference is that {{...}} captures can match slashes so can be used to match more than one "path part", while {...} can only match non slash characters.

  Load files matching the given pattern and populates the field .Metadata.MatchResult with the captures.

  Other supported forms are

  use: source
  path: <path pattern>
  

  use: source
  paths: 
    - <path pattern1>
    ...
    - <path patternN>
  

• target: <path template>

  A path template can contains {...} interpolations, the key inside the braces will be replaced using the .Metadata.MatchResult map. Double braces can also be used and are evaluated using Go text/template with context the whole item metadata.

  For each incoming item this will render the given path template with the item metadata and write the content to disk.

• use: <operation>

  This will apply the provided operations to the incoming items, for now available operations are

  • Template is a list operation that will concatenate all incoming items and render them using the specified template engine

    use: template
    engine: <name>
    

    for now engine can be one of

    • html — golang html template engine

    • text — golang text template engine

    • (TODO) handlebars — the famous Handlebars template engine

  • Layout is an item operation that works like the previous one as it will use a template engine to render the content of the output item. The main difference is that this will pass the previous item content in the template context in the .Content field.

    use: layout
    path: <glob pattern>
    

    use: layout
    paths: 
      - <glob pattern 1>
      ...
      - <glob pattern N>
    

    This operation will automatically choose between the go html or text template engines based on the first file name extension, TODO: make this configurable like the template operation)

    The template context is the incoming item metadata as well its data passed in the .Content variable.

    This is useful for wrapping partial html pages into full documents.

  • Markdown doesn't need any other options (TODO: add some options to configure goldmark)

    use: markdown
    

    This will render each incoming item using goldmark with some reasonable default options.

    By default this will also read YAML frontmatter.

  • Frontmatter as the previous one but just reads frontmatter.

    use: frontmatter
    

  • Categorize can be used in the following forms

    use: categorize
    key: <key>
    bind: <variable> # optional
    

    This will categorize all items based on the provided key, for example if key is tags then the incoming list of posts will be converted in a list of empty contents with a Category field telling the item's category and an Items field containing a list of posts with that tag.

    By default the output item category is placed into Category but this can be changed with the bind option.

  • Chunk can be used to paginate some items

    use: chunk
    size: <number>
    skip_remaining: <true or false> # optional, defaults to "false"
    

    This operation will group the incoming items in chunk with the provided size (except for the last one that can end up holding less items).

    The option skip_remaining will exclude the last chunk if it doesn't have exactly "size" items.

    The output items hold no data but have the following structure

    Metadata: {
        Index: /* 0-indexed current chunk number */,
        Total: /* total number of chunks */,
        Items: /* list of items of this chunk */,
    }
    

  • Slice extract a sub range of items

    use: slice
    from: <start> # inclusive, defaults to 0
    to: <end>     # exclusive, defaults to len(Items)
    

    Does what you expect to the incoming list of items.

  • Program is an item operation that will run a specified program passing the current item as stdin. The program output is then processed and becomes the new item data.

    use: program
    io: <stdin/stdout format> # optional, defaults to "raw"
    command: <shell command>  # required
    

    The io format specifies how the current item is passed to the given program

    • raw — will just pass the item .Data raw as stdin to the specified command.

    • json — will encode the whole item (also including type and metadata) as json ad pass that string as stdin to the specified command.

    TODO: Make the shell command be a go text/template string, for passing metadata directly thorough the command string.

  • (TODO) Sort sorts the incoming items using the provided key and direction (by default is ascending)

    use: sort
    direction: <ascending or descending>
    key: <key to use for sorting>
    

Architecture

• cabret.go — all abstract types and functions

• path/ — handles path patterns and templates

• config/ — contains config structure definitions and handles loading from YAML

• cmd/cabret/ — module that puts it all together in a CLI application

• operation/ — module containing all operations

  • operation/template/ — a small module that abstract the concept of a template, used by the layout operation.

• runner/ — a module that depends on config, parse and operations and evaluates operations.

• parse/ — handles the conversion of pipelines and lists of operations to core types.