Long road to 4.0

[ikalnytskyi]

• [ ] frontmatter processor
  • [x] use --- for YAML format (hugo/zola style)
  • [x] use +++ for TOML format (hugo/zola style)
  • [x] autodetect format
  • [x] allow delimiter overwrite
  • [ ] auto detect YAML/TOML by delimiter
• [x] commonmark processor
  • [x] replace misletoe with markdown-it-py
  • [x] support optional markdown-it-py extensions
• [ ] items refactoring
  • [ ] website item should be set on the application level
  • [ ] generators should be using the item class from application using some factory/clone function
  • [ ] items must be regular dictionaries
  • [ ] items should get a helper method to return URL (absolute and with hash trick)
• [ ] jinja processor
  • [ ] add filter/function to introduce "hashed" assets to overcome caching issue
• [ ] command processor
  • [ ] a processor that can invoke any command and acts as a passthrough
  • [ ] or maybe the result is used as a content? :thinking:
• [ ] image processor
  • [ ] Use Pillow to resize or chance format
• [ ] git processor
  • [ ] get updated_at and created_at values from git
  • [ ] get author from git
• [ ] some processor to extract published date
• [ ] Some general blog processor that
  • [ ] Takes source directory as input
  • [ ] Takes output directory as input
  • [ ] Sets feed generation
  • [ ] Sets renderer engines
  • [ ] Uses frontmatter by default
  • [ ] Generates sitemap
  • [ ] Contains NO LOGIC, only invokes other processors
• [ ] Use dataclasses instead of mappings for the document models inside pipeline
• [ ] Set document model class in the configuration class (to support various use cases)
• [ ] Ensure that basic blog workflows is super simple and does not require dozens processors