-
Notifications
You must be signed in to change notification settings - Fork 0
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Visualise networks on map? #6
Comments
@duncandewhurst @lgs85 (cc @odscjames) This is now on the dev branch, this URL maybe change in the new future. I've just set it up with basic circle markers for nodes and a colour for spans. I also, relatively randomly, added some data in a popup for each feature. We can customise all of this as appropriate. It also uses OSM base layers. This is under an acceptable use policy. We meet the requirements provided we don't err into 'heavy use', I'd guess that OFDS won't, but I could be wrong (and I'd need to check that actually constitutes heavy use). There are other options if we think this is likely, e.g. MapBox, Bing Maps (using a plugin), Esri ArcGIS (official plugin), MapQuest (official plugins) and Here Maps. Examples here: http://leaflet-extras.github.io/leaflet-providers/preview/ |
Looks good! A few questions and suggestions:
|
Most of this is for @Lathrisk to tackle but:
All of them. The JSON->GeoJSON conversion just puts them all together in the 2 files (nodes & spans). If we want the networks split on the map we'll need to add something to that conversion in ODFSKit first. It won't be hard, it just needs prioritising against other work.
@Lathrisk we can chat about this one, it might be that the ODFSKit conversion would be a good place to produce a meta file with field use info? |
I think that's fine. Let's definitely colour the spans and nodes by network by default. |
Ok - @Lathrisk isn't available for a few days but I'll make these decisions: We don't need to make separate GeoJSON files per network then, and "Let's definitely colour the spans and nodes by network by default." can be handled by the same mechanism as colouring it by other fields. However it sounds like having meta information available on what fields are used in the GeoJSON would be very handy, so I'll do that. Thoughts: .phase.name, .status.name, etc, anything .name - could 2 items have different id's but very similar names or even exactly the same names? There isn't a rule for unique names. Should these ones actually set colours by the id, but show name or some kind of "name (id)" to user? |
👍
In case it's relevant or reusable, Lib CoVE calculates field coverage since OpenDataServices/lib-cove#89.
Good point. I've opened an issue on making names unique: Open-Telecoms-Data/open-fibre-data-standard#172. I don't think we want to show ids to users. Setting colours by id but showing names sounds like a reasonable solution. I've added a legend to the list of suggestions in #6 (comment) |
Yes - I've just looked at that code as part of doing additional fields - that's why I specially picked this point up :-) |
Helpful for Open-Telecoms-Data/cove-ofds#6 Also fix: bug in get_json() that meant it could not be called twice
Will be adding more to meta later, like errors on conversion. Added field coverage because thought would be helpful for Open-Telecoms-Data/cove-ofds#6 but actually I added to wrong conversion! Leaving as may be useful and that is simpler than removing. Also fix: bug in get_json() that meant it could not be called twice
I've added the following to the list of requirements in #6 (comment):
The current feedback to users is contradictory: |
It's possible for the GeoJSON to not actually have any geography elements. In this case, don't show the map. I can add something to Libcoveofds so that the meta file has a flag for this. In this case, keep the visualisation box and have a message to use "your data has no lat/lngs etc" and put in warning colour - @duncandewhurst will write exact message. |
@duncandewhurst @odscjames @codemacabre |
I've mapped (no pun intended) out the possible scenarios in the table below - let me know if I missed anything. What are the possible causes of the Ideally, the JSON-> GeoJSON conversion script should only produce valid GeoJSON data. Otherwise, if there is an issue with the JSON data that means the GeoJSON would be invalid, then the conversion should fail and report the issue in the data conversion box. If a user uploads GeoJSON files, are they passed directly to the map or are they first converted to JSON and then back to GeoJSON? If it's the former, then we'll need error handling in the map box too (3rd scenario below).
Note that I updated the successful message slightly to mention coordinate ordering. |
Ideally, but there will be some creative work in the lib to write lots of bad test data then make sure the lib handles it. Open-Telecoms-Data/lib-cove-ofds#27 So for now there could be dodgy GeoJson that crashes our map JS Lib, I think?
The later, and this was something I meant to raise as a general discussion when things are less hectic. #36 for later. |
All outstanding comments in this issue are resolved, additional feedback now in : #51 |
(Possible overlap with #5 . Also see Open-Telecoms-Data/open-fibre-data-standard#124 )
The text was updated successfully, but these errors were encountered: