Heading-section path for each top-level block

Usage

ast_sections(x, ...)

Arguments

x

A pandoc document, a pandoc_blocks wrapper, or a plain list of top-level blocks.

Value

A list with one element per top-level block, each a named character vector c(h1, h2, h3, h4, h5, h6) with NA where no heading at that level is in scope.

Details

Walks the top-level block stream of a pandoc document and, for each block, reports the chain of enclosing pandoc_header titles. Because the Pandoc block list is flat (headings are siblings of the content they introduce, not parents of it), this is the piece of structure that q2r does not otherwise expose. It is the analog of parsermd's rmd_node_sections() and the basis for select_section().

Each block is assigned a length-6 named character vector (h1..h6). A header is considered part of the section it opens, so a level-2 header ## Methods carries h2 = "Methods" itself and resets any deeper (h3..h6) entries. Heading titles are the flattened text of the header (ast_text()).

Examples

if (FALSE) { # \dontrun{
doc = parse_qmd("# A\n\nintro\n\n## B\n\nbody\n")
ast_sections(doc)
} # }