Fixed TOC nesting with non-sequential headings

[mossr]

When using \toc with a page that has nested headings that are not sequential, you currently get the following (which has h4 elements after h1 elements):

Before

After

With this PR, I fix it so you skip adding empty nested lists to the TOC:

Egregious MWE

Here's a MWE that's pretty egregious:

Before

After

MWE

Note, this with the following config.md

mintoclevel = 1
maxtoclevel = 7

@def title = "TOC Testing"

\toc

# H1

### H3

#### H4

## H2

## H2

##### H5

# H1

###### H6

## H2

## H2

#### H4

# H1

#### H4

#### H4

###### H6

#### H4

#### H4

[tlienart]

This is awesome, thanks a lot! code lgtm, will wait for CI to pass then do a patch release with this.

Ah it looks like the relevant test is failing, could you take a look?

[mossr]

Of course, I'll take a look

[mossr]

This test made me rethink how it should be handled. If we skip the indent entirely, then you may end up with headers that are actually nested under a top-level header but wouldn't appear as such in the TOC.

So I changed it so that the nesting is still added, but we hide the marker for the "skipped" sublists.

Test

old	new

Previous example

Notice now how all the h#'s line up properly.

[mossr]

Actually wait to merge this in. I think we can have the best of both worlds. Gimme a second

[mossr]

("a second == 4 hours later") I made the changes so both behaviors can co-exists. We only add empty nested items when appropriate. Here's what it would look like:

for something like

\toc
### Part of family 1: should not be nested `fd`

## Top-most ancestor of family 2
#### should be empty nested
### part of family 2

## Top-most ancestor of family 3
#### child1
#### child2
#### child3
#### child4

## Top-most ancestor of family 4
done.

I've updated the test to include this.

[tlienart]

Haha who would have thought that tocs are so interesting 😂 thanks a lot for all your work!

[mossr]

Of course, thank YOU for Franklin.jl! I love it so much that I redid my entire website using it

[yangcht]

Hej! I have been using \toc without problems but after recent updates. It starts to create errors like this. Do you have any idea how I can fix this? Or is this a bug? Thanks! @mossr

⋮ shutting down LiveServer… ✓
ERROR: ArgumentError: invalid Array dimensions
Stacktrace:
 [1] serve(fw::LiveServer.SimpleWatcher; host::String, port::Int64, dir::String, debug::Bool, verbose::Bool, coreloopfun::Franklin.var"#267#270"{…}, preprocess_request::typeof(identity), inject_browser_reload_script::Bool, launch_browser::Bool, allow_cors::Bool)
   @ LiveServer ~/.julia/packages/LiveServer/3Dvy0/src/server.jl:658
 [2] serve (repeats 2 times)
   @ ~/.julia/packages/LiveServer/3Dvy0/src/server.jl:578 [inlined]
 [3]
   @ Franklin ~/.julia/packages/Franklin/UskEZ/src/manager/franklin.jl:140
 [4] serve()
   @ Franklin ~/.julia/packages/Franklin/UskEZ/src/manager/franklin.jl:49
 [5] top-level scope
   @ REPL[3]:1

caused by: ArgumentError: invalid Array dimensions
Stacktrace:
  [1] Array
    @ Core ./boot.jl:475 [inlined]
  [2] Array
    @ Core ./boot.jl:484 [inlined]
  [3] fill
    @ Base ./array.jl:582 [inlined]
  [4] fill(v::Nothing, dims::Int64)
    @ Base ./array.jl:580
  [5] hfun_toc(params::Vector{String})
    @ Franklin ~/.julia/packages/Franklin/UskEZ/src/converter/html/functions.jl:244
  [6] top-level scope
    @ none:1
  [7] eval
    @ Core ./boot.jl:383 [inlined]
  [8] eval
    @ Franklin ~/.julia/packages/Franklin/UskEZ/src/Franklin.jl:1 [inlined]
  [9] convert_html_fblock(β::Franklin.HFun)
    @ Franklin ~/.julia/packages/Franklin/UskEZ/src/converter/html/functions.jl:18
 [10] process_html_qblocks(hs::String, qblocks::Vector{Franklin.AbstractBlock}, head::Int64, tail::Int64)
    @ Franklin ~/.julia/packages/Franklin/UskEZ/src/converter/html/html.jl:122
 [11] process_html_qblocks(hs::String, qblocks::Vector{Franklin.AbstractBlock})
    @ Franklin ~/.julia/packages/Franklin/UskEZ/src/converter/html/html.jl:102
 [12] convert_html(hs::String)
    @ Franklin ~/.julia/packages/Franklin/UskEZ/src/converter/html/html.jl:22
 [13] map
    @ Base ./tuple.jl:293 [inlined]
 [14] map
    @ Base ./tuple.jl:294 [inlined]
 [15] write_page(output_path::String, content::String; head::String, pgfoot::String, foot::String)
    @ Franklin ~/.julia/packages/Franklin/UskEZ/src/manager/write_page.jl:108
 [16] write_page
    @ Franklin ~/.julia/packages/Franklin/UskEZ/src/manager/write_page.jl:75 [inlined]
 [17] convert_and_write(root::String, file::String, head::String, pgfoot::String, foot::String, output_path::String)
    @ Franklin ~/.julia/packages/Franklin/UskEZ/src/manager/write_page.jl:215
 [18] process_file_err(case::Symbol, fpair::Pair{String, String}, head::String, pgfoot::String, foot::String, t::Float64)
    @ Franklin ~/.julia/packages/Franklin/UskEZ/src/manager/file_utils.jl:153
 [19] process_file(::Symbol, ::Pair{String, String}, ::String, ::Vararg{Any})
    @ Franklin ~/.julia/packages/Franklin/UskEZ/src/manager/file_utils.jl:104
 [20] fd_loop(cycle_counter::Int64, ::LiveServer.SimpleWatcher, watched_files::@NamedTuple{…})
    @ Franklin ~/.julia/packages/Franklin/UskEZ/src/manager/franklin.jl:408
 [21] #267
    @ Franklin ~/.julia/packages/Franklin/UskEZ/src/manager/franklin.jl:136 [inlined]
 [22] serve(fw::LiveServer.SimpleWatcher; host::String, port::Int64, dir::String, debug::Bool, verbose::Bool, coreloopfun::Franklin.var"#267#270"{…}, preprocess_request::typeof(identity), inject_browser_reload_script::Bool, launch_browser::Bool, allow_cors::Bool)
    @ LiveServer ~/.julia/packages/LiveServer/3Dvy0/src/server.jl:648
 [23] serve (repeats 2 times)
    @ ~/.julia/packages/LiveServer/3Dvy0/src/server.jl:578 [inlined]
 [24]
    @ Franklin ~/.julia/packages/Franklin/UskEZ/src/manager/franklin.jl:140
 [25] serve()
    @ Franklin ~/.julia/packages/Franklin/UskEZ/src/manager/franklin.jl:49
 [26] top-level scope
    @ REPL[3]:1
Some type information was truncated. Use `show(err)` to see complete types.