Markdown Discuss - Feb 2009 - RFC: Markdown Table Syntax

On Feb 26, 2009, at 9:51 AM, I wrote:
> I can see that really opened a can of worms here. Sorry about that.  
> I'll have to move up my plan to formally propose a table syntax. The  
> comment that started this sub-thread was a throwaway line, really. I  
> probably should have omitted it.
And then, on Feb 26, 2009, at 3:28 PM, John Gruber wrote:
> 3. A hypothetical official table syntax for Markdown will almost  
> certainly look very much, if not exactly, like Michel's table syntax  
> in PHP Markdown Extra.
Well, perhaps it's foolish of me, but since I'd already spent several  
hours writing it up before I saw that post, I went ahead and finished  
my proposal as an informal request for comments, which I've [blogged  
here][]:

[blogged here]
http://www.justatheory.com/computers/markup/markdown-table-rfc.html

The high-level overview is this:

* I looked at prior art from database clients and MultiMarkdown
* I identified a few things I liked in the DB clients that aren't
   in MultiMarkdown (continuing lines, implicit cell alignment)
* I offered some refinements of the MultiMarkdown syntax

Those refinements, in a nutshell, are, simply:

* Implicit cell alignment using space characters, rather than
   the explicit formatting hints in the header lines
* Cell content continuation using : for succeeding lines
* Stricter use of space, for proper alignment in plain text (which
   all of the MultiMarkdown examples I?ve seen tend to do anyway)
* Allow + to separate columns in the header-demarking lines
* A table does not have to start right at the beginning of a line

Many if not all of these refinements can preserve compatibility with  
what's already there, although such was not my goal. Overall, what I  
was concentrating on was how I'd like tables to look in plain text,  
given how often I've created them in code documentation, and how often  
I look at them in terminal-based database clients. I like what I've  
come up with, drawing as it does on solid prior art, and would  
appreciate those who take an interest in such arcana taking a look.

Again, I appreciate how friendly and willing the folks have been on  
this list, especially to a presumptuous n00b like myself, and hope to  
continue to contribute to that for a long time.

Best,

David

Le 2009-02-26 ? 19:54, David E. Wheeler a ?crit :
> <http://www.justatheory.com/computers/markup/markdown-table-rfc.html>
>
> [...]
>
> Those refinements, in a nutshell, are, simply:
>
> * Implicit cell alignment using space characters, rather than
>  the explicit formatting hints in the header lines
> * Cell content continuation using : for succeeding lines
> * Stricter use of space, for proper alignment in plain text (which
>  all of the MultiMarkdown examples I?ve seen tend to do anyway)
> * Allow + to separate columns in the header-demarking lines
> * A table does not have to start right at the beginning of a line
Perhaps you should mention that you're now forcing a table cells to be  
properly aligned using a monospace font. This in fact makes the syntax  
impossible to use in a proportional font context.

While it's true that many people use a monospace font and that in a  
text editor you can change the font, I'd like to mention that Markdown  
is also used for writing comments and blog posts in textareas not  
formatted with a monospace font, and for which it may not be easy to  
change the font. Forcing proper alignment makes the table syntax  
unusable in these situations.

Now, given that cell continuation using colons relies on that proper  
alignment with a monospace font feature (or else you risk mistaking  
colons in the text for column separators), I don't find that syntax  
very satisfying.

Allowing `+` as column separator in the header underline looks like a  
good idea though.


-- 
Michel Fortin
michel.fortin at michelf.com
http://michelf.com/

On 26 Feb 2009, at 7:54 PM, David E. Wheeler wrote:
>> 3. A hypothetical official table syntax for Markdown will almost  
>> certainly look very much, if not exactly, like Michel's table  
>> syntax in PHP Markdown Extra.
>
> Well, perhaps it's foolish of me, but since I'd already spent  
> several hours writing it up before I saw that post, I went ahead and  
> finished my proposal as an informal request for comments, which I've  
> [blogged here][]:
I never say never. Well, usually.

-J.G.