host maps on github

[jvonau]

Related iiab/iiab-admin-console#483 (comment) "I propose that we put the map-catalog.json on githubusercontent"
Needs cities1000.sqlite and map-catalog.json added to maps and regions.json corrected as per https://github.com/jvonau/maps/pull/1/files
Would need the last commit rolled back before committing to master, it's provided for testing convenience of the new github location scheme until the above changes are implemented in maps.
Tested on 22.04 VM

[jvonau]

iiab/maps#54

[tim-moody]

I thought that our standard was to use the vars in a role for testing, but to make the var in vars/default_vars.yml the definitive value. Is the intent that this is not a user configurable value? I assume it will not be visible to the Admin Console install.

[holta]

I thought that our standard was to use the vars in a role for testing, but to make the var in vars/default_vars.yml the definitive value. Is the intent that this is not a user configurable value? I assume it will not be visible to the Admin Console install.

Sounds good to me!

[jvonau]

I thought that our standard was to use the vars in a role for testing, but to make the var in vars/default_vars.yml the definitive value.

That has not changed, but #3165 didn't take care of default_vars, used defaults/main.yml to point to the testing repo that is still used today. Thought that value would of changed back to iiab/maps once iiab/maps#45 came alive but I'm not the one who approves PRs

Is the intent that this is not a user configurable value? I assume it will not be visible to the Admin Console install.

Any entries in defaults/main.yml or default_vars.yml are still user configurable via local_vars.yml but may not have entries present to do so relying on the default value that is defined.

[holta]

Either way.

Certainly as @tim-moody suggests, if values are changed very rarely every few years, they should likely be in vars/default_vars.yml primarily.

[jvonau]

Guess to sum this up 'iiab_map_url' is going away, 'osm_repo_url' and 'maps_branch' are taking its place becoming the authoritative source pointer once iiab/maps have the required files added. Now 'map_catalog_url' is not quite accurate it's more 'map_mbtiles_url', I can change that while the hood is up or defer that to Adam. Just need to add one more commit to enable default_vars.yml over defaults/main.yml or I can roll back the last commit for the same effect before merging.

[holta]

Either is fine, Thanks Jerry!

[tim-moody]

should the names osm_repo and map_branch sync up, like osm_repo_branch
or even just allow for the branch as part of the url

[georgejhunt]

My strategy is to test the values in osm-vector-maps/defaults/main.yml, which are used for testing massive changes to the map system. And then change the variable name in the roles/install.yml to iiab_map_url (which is set in iiab/vars/default_vars.yml) as a separate, and later PR.

So I'm not in favor of iiab_map_url going away.

[tim-moody]

@georgejhunt I think it's your call.

[jvonau]

iiab_map_url is currently used to retrieve 2 files from http://download.iiab.io where do you propose these files be served from?

[jvonau]

should the names osm_repo and map_branch sync up, like osm_repo_branch or even just allow for the branch as part of the url

That discussion should of been part of PR 3165, I just went with what is already presently defined.

[jvonau]

#3195

[holta]

@georgejhunt I think it's your call.

Thanks @georgejhunt if you can make a recommendation — so @jvonau can refine this PR per your recommendation?

[jvonau]

iiab/maps#55 (comment)

[georgejhunt]

I recommend that you take me out of the loop. Considering my diminished capacity, I'm not good at analyzing the changes in https://github.com/iiab/iiab/pull/3192/files, and figuring why the position of installing packages was moved up in the playlist. Which changes matter, and which do not? What are the unintended I think this is a good time to let Jerry and Adam take over the development of OSM. I think master still functions, and as Jerry says, right now we are using a development branch, which needs to be swapped. I have never been very good at the back and forth interactive stuff. I don't monitor our skype channel. I'm finding that if I don't enjoy it, I put it off. I've been putting off IIAB a lot recently.

…

On Tue, Apr 26, 2022 at 10:39 AM A Holt ***@***.***> wrote: @georgejhunt <https://github.com/georgejhunt> I think it's your call. Thanks @georgejhunt <https://github.com/georgejhunt> if you can make a recommendation — so @jvonau <https://github.com/jvonau> can refine this PR per your recommendation? — Reply to this email directly, view it on GitHub <#3192 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAOTQHGO5W37GAQ6SDRLT2DVHAS4XANCNFSM5UKWRSGA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[holta]

good time to let Jerry and Adam take over the development of OSM

@georgejhunt you made all this map/satellite work happen (none of which could have happened without you!) and I apologize I've been quite busy this week with 3 unrelated hardware failures.

But yes we will help package this up as best we can in coming days — calling you when we get stuck.

Thanks to @jvonau's initiative overnight and @tim-moody who's helping a ton here — truly a joint effort and tremendous joint accomplishment with the potential for Many Thousands of kids + teachers to use this over coming years!

[holta]

New PR:

• PR remove maps regions.json iiab-admin-console#484

Thanks everyone for these discussions, helping to clarify:

• GitHub host maps#54
• Garbage in what should be a source file resources/regions.json & discussion of cities1000.sqlite maps#55