<div dir="ltr">1. We may need to break backward compatibility at some point to improve the data structure. As long as we keep the same file format, we could provide a tool to build old databases out of new ones.<div><br></div>
<div>2. Good point and right, even now, cell_misc has subdirectories, not files. If we really want to have no limits on the number of maps, we could make the folder structure flat and have a fixed number of element directories and files inside them. I personally don't like such a limitation like max 32,000 maps. Maybe, moving/renaming vector/cell_misc element files to the mapset directory is a better option. mapset/element/mapname</div>
</div><div class="gmail_extra"><br><br><div class="gmail_quote">On Wed, May 14, 2014 at 5:18 PM, Glynn Clements <span dir="ltr"><<a href="mailto:glynn@gclements.plus.com" target="_blank">glynn@gclements.plus.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class=""><br>
Huidae Cho wrote:<br>
<br>
> I think in the long term it would be better to change the raster directory<br>
> structure such that raster and vector have the same mapset/(raster,<br>
> vector)/mapname/element path. Then we would be able to move some useful<br>
> element related functions from librast to libgis.<br>
<br>
</div>The main downsides are:<br>
<br>
1. You lose the ability to access the data with older versions of<br>
GRASS.<br>
<br>
2. You're more likely to run into limits on the number of maps within<br>
a mapset. IIRC, ext2/ext3 doesn't allow a directory to have more than<br>
32000 subdirectories, whereas there's no practical limit to the number<br>
of files. OTOH, given that raster maps invariably have a non-empty<br>
cell_misc subdirectory, this is probably a moot point.<br>
<span class="HOEnZb"><font color="#888888"><br>
--<br>
Glynn Clements <<a href="mailto:glynn@gclements.plus.com">glynn@gclements.plus.com</a>><br>
</font></span></blockquote></div><br></div>