Markdown

[cysouw]

I just found this repo after I finished my own approach to linguistic examples for Pandoc ... We should try to harmonise our markup-approaches!

best
Michael

https://github.com/cysouw/pandoc-ling

[parryc]

Hi @cysouw, I will first admit I have only a passing familiarity with Pandoc. That said, I am happy to harmonize what I have with your work. In your mind, what does that harmonization look like?

[cysouw]

I was mainly thinking about how people type the linguistic examples in markdown. Have a look at the (long) discussion on my readme about how I decided to format things. The reasons for my decisions were partly inspired by the available parsers in Pandoc, so I could build on that basis.

as for harmonisation, I was thinking whether we might get these input-formats closer together.

However, thinking about it: it might not be possible to completely merge them, and then it would possible be better to have them to be strongly different!

BTW: check out pandoc, it is really awesome...

[parryc]

What are you thoughts around adding gloss-specific formatting? Ex. what I have today as {b} or {m}. Perhaps that is out of the purview of pandoc. That is, it is up to the author to determine whether or not their downstream processors will support lower level control of how individual glosses appear, rather than something inherent to the structure.

That being said, it looks like the macrostructure of the gloss (e.g. the grammaticality judgement, label, etc.) could be incorporated into my repo. I would probably hew closely to what you have already designed, since your detailed readme has taken into account so much already. Not to mention it looks like you have generated an HTML version which will help me align the HTML format, as well.

[cysouw]

I treat {m} and {b} simply by using non-breaking spaces :-).

If you want to have multiple words combined, just insert non-breaking space instead of a normal space.

for empty positions in the gloss, the user can insert a tilde "~" (that is a convention found in many latex packages), but internally this tilde is simply replace by "space + non-breaking space + space", which automatically leads to an 'empty' space

as for the HTML: the whole principle of Pandoc is to allow conversion in many different formats. So yes, HTML is definitely part of the deal.

[parryc]

Got it, thanks for the explanation! The next time I sit down to update this (or if I start a new project that uses pandoc), I'll work to incorporate your schema. If you find a scenario where having a unified spec would be useful sooner rather than later, please let me know. Thanks again for bringing your project to my attention!