ccorn

A better name might be `pos_tuple`. This is inspired by the internal name `range_access::pos` that `get_cursor` delegates the work to, and by the fact that the zip cursor (but not...

> I understand the general idea, but how does this work within the views' ecosystem? It extends usabilty of the views ecosystem by keeping relevant state information accessible. > This...

> The problem remains, that you can create arbitrary complicated tuple and variant structures, i.e. std::tuple, this is more or less solved by `base()` you can recursively access that structure...

To conclude: `base()` would enable inspection of pipeline predecessors. But it is not optimized for that; it needlessly requires copyable iterators instead of deliberately returning a reference to const *Base*...

I have not tried to build TFF on my G5 yet, but I have experienced lots of problems with macports builds on my G5 (`+universal` for archs `ppc` and `ppc64`)...

@kencu: Thanks for this! Just looking at the contents of the island patch makes it look definitely push-worthy to me.

This seems to apply only to links augmented with `target="_blank"`. In my experiments, if I open a non-augmented link in a new tab via context menu, `window.opener` is `null`. In...

The roughly equivalent input ~~~ kramdown item : ^ para ~~~ does not have that problem.

Using `irb`, I find that `test.md` gets parsed to ~~~ irb => #, :location=>1, :options=>{}, :abbrev_defs=>{}, :abbrev_attr=>{}} [1, :ial=>nil} [2, :raw_text=>"item"} [nil}>]>, 2} [, 3} [3}>]>]>]>]> ~~~ (note the `,`)...

Hmm, the same issue occurs with other list types: ~~~ shell $ kramdown