[OSM-dev] Planetfile with ChangeSet

m2huber at hsr.ch m2huber at hsr.ch
Wed Jun 3 13:07:10 BST 2009


Hi @ all

Thanks for your Answer.

The md5 checksum is correct, either.

But with bzcat and the direct piping into Osmosis worked well.

We tried on our project to read the bz2 File directly, which didn't work with the same exception as in osmosis. With the unzipped file it worked perfectly.

It's a pitty, that the direct read of the bz2 File doesn't work.. does anyone has an Idea with that? 
It would be nice if that would work as well. During development we used the extracts from download.geofabrik.de and never had any problems reading the bz2 Files...

Thanks anyway

Mike

-----Ursprüngliche Nachricht-----
Von: Frederik Ramm [mailto:frederik at remote.org] 
Gesendet: Dienstag, 2. Juni 2009 17:14
An: Huber Michael (m2huber at hsr.ch)
Cc: dev at openstreetmap.org
Betreff: Re: AW: [OSM-dev] Planetfile with ChangeSet

Hi,

m2huber at hsr.ch wrote:
> As we firstly tried to Run the program with the newest planet file we discovered a problem with the changesets. After your mail I also tried the import the latest planet with osmosis and it didn't work either. I don't know what exactly the problem is, but I guess that there could be a Problem with the formattation of the XML File?

The XML file looks fine to me. Your Osmosis complains about an error in 
line 4319 of planet-090520. I have:

% bzcat planet-090520.osm.bz2 |head -4319|tail -1
   <changeset id="4316" created_at="2007-04-01T13:41:04Z" 
closed_at="2007-04-01T17:28:47Z" open="false" min_lon="-7.0227251" 
min_lat="53.1442826" max_lon="-6.1289473" max_lat="54.0376818" 
user="mackerski" uid="6367"/>

Which looks perfectly all right! May I suggest:

1. Take the md5 sum of your planet file and compare with the one on the 
download server (to check if the file has downloaded correctly).

2. Do you read the bz2 file directly from your app, or do you uncompress 
it first? Because there have been issues with compressed files before. 
If you try to use bzcat to pipe the file into osmosis (then use 
"file=-"), does that perhaps work better?

Bye
Frederik




More information about the dev mailing list