Re: [O] Printable calendar?

[Peter Davis]

On Fri, May 30, 2014 at 01:20:21PM -0400, Nick Dokos wrote:
> Peter Davis <address@hidden> writes:
> 
> > Ok, I was able to get the column rules I want. (See below)
> >
> > I'm still puzzled by the right/left alignment. In the org buffer the
> > columns appear correctly aligned, but in HTML output, the left (Sun)
> > and right (Sat) columns are right-aligned,
> > while all the others are left-aligned.
> >
> > Clues?
> >
> 
> You can force the misbehaving columns to behave - more or less: the M
> value on the 17th will cause problems (btw, I prefer
> to have a non-exported "zeroth" column for things like / and !
> that are basically table metadata - see (info "(org) Advanced features")
> for details):
> 

Thanks, Nick. That worked beautifully (once I noticed that May has 31 days 
instead of 30 twice.)

> The misbehaviour is caused by the heuristic used in
> org-export-table-cell-alignment:
> 
> ,----
> | Return alignment as specified by the last alignment cookie in the
> | same column as TABLE-CELL.  If no such cookie is found, a default
> | alignment value will be deduced from fraction of numbers in the
> | column (see `org-table-number-fraction' for more information).
> `----
> 
> You can play around with org-table-number-fraction (default: 0.5) to
> change the behaviour. A value of 0.25 will right-align them all, whereas
> a value of 0.75 will left-align them all. But I wouldn't want to bet my
> life on that: it depends on the contents of the table so it seems like a
> fragile solution at best.
> 
> BTW, the 0.25 and 0.75 values above are purely trial-and-error (actually
> derived from the smallest ratio I found edebugging over the columns:
> 6/21).
> 
> Nick
> 
> Footnotes:
> 
> [fn:1] The heuristic counts empty cells as numbers if the non-empty row
>        above it is a number, so for the first column for example, there
>        are 21 cells and 11 of them are "numbers".
>        

Interesting, but it sounds to me like org is trying to be too smart for its own 
good here, making inferences based on the contents of the
cells. I'll stick with the '<l>' notation, which seems more reliable.

Thank you!

-pd


-- 
----
Peter Davis
The Tech Curmudgeon
www.techcurmudgeon.com