Skip to content

Conversation

@mandu
Copy link

@mandu mandu commented Dec 8, 2023

Local

Build

brew install gradle@7
./build-local.sh

Generate tiles

wget https://download.geofabrik.de/europe/united-kingdom/england/greater-london-latest.osm.pbf -O data/greater-london-latest.osm.pbf
./generate-tiles-local.sh

Docker

Build

./build-docker.sh

Generate tiles

wget https://download.geofabrik.de/europe/united-kingdom/england/greater-london-latest.osm.pbf -O data/greater-london-latest.osm.pbf
./generate-tiles-docker.sh

Copy link

@edmabon edmabon left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It builds and it works - I can't complain!

A couple of interesting things though.

  1. Compared to the binary you sent yesterday, this takes almost twice as long to build the same maps. The binary yesterday completely maxed out my CPU, this newer build doesn't.
  2. Compared to the binary from yesterday, the output is different with the same input map. I realise there are some changes to bring more tags in to the output (so would expect changes to the metadata files). But there are many changes to the reference and geometry files too. Is this because of the changes to the road classes that are being processed?

Is there any motivation to try and update the dependencies to more recent ones? Since this is just a one-off building process for the maps, I'd say no as we don't need it to be super optimised and up to date.

@mandu
Copy link
Author

mandu commented Dec 8, 2023

  1. Compared to the binary you sent yesterday, this takes almost twice as long to build the same maps. The binary yesterday completely maxed out my CPU, this newer build doesn't.

This is likely because of the xmx parameter (java max heap size) and I put the default to 2G.

  1. Compared to the binary from yesterday, the output is different with the same input map. I realise there are some changes to bring more tags in to the output (so would expect changes to the metadata files). But there are many changes to the reference and geometry files too. Is this because of the changes to the road classes that are being processed?

I made two PRs, here is the one that did some changes to the processing. I can merge this and rebase the other PR and if you are then able to check if we get same results?

@mandu
Copy link
Author

mandu commented Dec 8, 2023

Also notice that in the last 'force-push' I unified the behaviour of ./generate-tiles-local.sh and ./generate-tiles-docker.sh to write the tiles to out/ -folder

@mandu mandu merged commit 331a931 into master Dec 8, 2023
@mandu mandu deleted the gradle7-build branch December 8, 2023 11:00
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants