[OSM-dev] Filename extensions for OSM files
scott at sacrosby.com
Wed May 11 23:54:13 BST 2011
On Wed, May 11, 2011 at 6:48 AM, Peter Körner <osm-lists at mazdermind.de> wrote:
> Am 11.05.2011 13:42, schrieb Frederik Ramm:
>> On 05/11/11 13:23, Peter Körner wrote:
>>> that way it is useful for a program to know, if a file fed into it via
>>> stdin will be processed fine right from the start. nobody likes
>>> processes that fail after 10 hours because of a missing
>>> --this-is-an-history-file flag.
>> Yeah, "fast failing" is probably the most useful effect of this. I've
>> written programs that relied on proper ordering and would either
>> explicitly fail or produce unexpected results at some time long into
>> processing the input; if there had been an option to make my program
>> fail unless you present it with sorted input, I might have used that.
> The PBF-Format has it, so the only question is: do we want to change the
> definition of the legacy xml format? Or would we rather just focus on our
> new bling-bling-pbf-format and just leave the xml stuff as it is?
The PBF format has the capability to indicate sort order, but that
isn't used at the moment.
There aren't many writers for the PBF format? AFAIK, there's only the
osmosis one and the splitter now has one. If PBF files are written by
being converted from XML files, then the XML file must have some means
to indicate that the data is sorted, and that indication must be
passed through the osmosis pipeline.
More information about the dev