[OSM-dev] Osmosis Replication Statistics

[Brett Henderson]

Tom Hughes wrote:
> In message <46D40241.1030906 at bretth.com>
>         Brett Henderson <brett at bretth.com> wrote:
>
>   
>> Next up I may try to generate a delete changeset for the problematic 
>> original tiger data.  I don't know how this should applied to the 
>> database but it is the same problem the new TIGER import and AND data 
>> are currently facing.
>>     
>
> I'm perfectly capable of adding the history records for it, but so
> far I have avoided doing so because it will mean adding millions of
> extra records to the database.
>
> Tom
>   
That's fair enough.  I'm trying to figure out what to focus on next.  
Currently on my list of potentials are:
* An API download task.  Marcus Wolschon has already written this, I 
just have to make a couple of minor changes before applying it into the 
codebase.
* An API upload task.  This will duplicate the functionality of 
Martijn's bulk_import.pl so has slipped in priority.  Once the kinks 
have been ironed out in that tool I'll re-use the ideas.
* Create a new task for dumping "current" tables.  This might be useful 
in the short term to speed up planet creation.
* Complete the changeset application code in --write-mysql-change and 
add "placeholder" functionality to allow new entities to be created.
* Add referential integrity checking to database writing code.
* Add a task for reading changes from JOSM style change files.
* A new cron or daemon run osmosis-extract application for producing 
regular changesets (ie. daily snapshots or perhaps even less).
* Add configurable logging (eg. -quiet or -verbose type switches) to 
customise osmosis output.
* Add progress reporting features to osmosis tasks so you can tell where 
it's up to.
* Enhance pipe connection code to be smarter in how it auto-connects tasks.
* Re-think command line, it's fairly verbose at the moment.

Please feel free to provide suggestions on what is required or would be 
helpful.

I'll add the download task next because that already exists.  After that 
I haven't decided.  Of most interest to me is an automated extract 
mechanism (ie. "osmosis-extract") but that may take some time.  Other 
features such as configurable logging or progress reporting might be 
more useful today.

Cheers,
Brett