Multi export

[sashadev-sky]

Fixes #270 (<=== Add issue number here)
/ PR at #310

Also:

• Bug: Multiple image select with boxZoom #186 / Leaflet incremental update #235
• multiple image collection actions (like export) #29
• addToolbarItem() to enable custom features #62
• Fix toolbar undefined Fix toolbar undefined #295 / issue at Issue with "LeafletToolbar" when using "leaflet.distordableimage" #273
• Work towards Separate each tool into its own file, starting with "toggleExport" #161

Make sure these boxes are checked before your pull request (PR) is ready to be reviewed and merged. Thanks!

• PR is descriptively titled 📑 and links the original issue above 🔗
• tests pass -- look for a green checkbox ✔️ a few minutes after opening your PR -- or run tests locally with grunt
• code is in uniquely-named feature branch and has no merge conflicts 📁
• screenshots/GIFs are attached 📎 in case of UI updates
• @mention the original creator of the issue in a comment below for help or for a review

We're happy to help you get this ready -- don't be afraid to ask for help, and don't be discouraged if your tests fail at first!

If tests do fail, click on the red X to learn why by reading the logs.

Please be sure you've reviewed our contribution guidelines at https://publiclab.org/contributing-to-public-lab-software

Thanks!

=================================

Changes:

• Removed: unused: "strict" from JSHINT gruntfile
• L.Map.BoxSelector is fixed (Multi-select with shift-drag)
• src/edit/tools folder with DistortableImage.ControlBar.js and DistortableImage.PopupBar.js, each corresponding to the new class names for the 2 types of toolbars and EditAction.js the classname for the class we extend to create new actions.
• leaflet-toolbar upgraded to "0.4.0-alpha.2" and added as a dependency in package.json
• Two multi-image interface tools: Exports, Deletes
• Toolbar API: removeTool, addTool, hasTool for the DistortableCollection and DistortableImage.Edit class - allows changing the toolbar actions in runtime
  • Did not use the implementation made by @rexagod in Add (reveal) functionality for more options in toolbar (also _removeTool) #155 but will circle back around later to review it and compare carefully
• DistortableCollection and DistortableImageOverlay instantiation updates: new action option allows overwriting the default toolbar actions for respective toolbars initially if don't want to use the runtime API above
• DistortableCollection instantiation updates: images are now only added to the collection class, not the map directly.
• _updateCorners and _updateCorner() are now setCorners and setCorner() but not ready to be documented yet - need overlay.fire("update") in them to work sep. but I don't want to change where we currently fire it in our code bc of the bugs with rotation, scaling ,etc.
• Deletion of unnecessary committed leaflet.distortableimage.js.map file two months ago
• Testing updates to match changes to DistortableCollection class

========================

Pending:

• loader UI: considering for now just adding animated SVG loader icon to replace Export icon during load
• Multiple Exports still not working
• make a DistortableCollection.Edit class so the Tool API will work the same on it as on DistortableImage.Edit - then update docs

[jywarren]

I think we need to re-order the corners we get from this section:

Leaflet.DistortableImage/src/DistortableImageOverlay.js

Lines 100 to 109 in 83e3371

	this._corners = [
	map.containerPointToLatLng(center.subtract(offset)),
	map.containerPointToLatLng(
	center.add(L.point(offset.x, -offset.y))
	),
	map.containerPointToLatLng(
	center.add(L.point(-offset.x, offset.y))
	),
	map.containerPointToLatLng(center.add(offset))
	];

[jywarren]

Yes, i'm going to bump version number so it'll get picked up by MapKnitter, then we can come back to UI. Thanks!

[jywarren]

OK, i'm adding a note to the README that this library does NW, NE, SW, SE corners!

[jywarren]

Done and pushing to my gh-pages to test: http://jywarren.github.io/Leaflet.DistortableImage/examples/select

[jywarren]

Darn! I'm still getting convert-im6.q16: invalid argument for option Perspective : 'Unsolvable Matrix' @ error/distort.c/GenerateCoefficients/837. -- why? let me double-check the order...

[jywarren]

[{"lat":51.51300227287522,"lng":-0.12977341306395831},
{"lat":51.53297533262061,"lng":-0.11981705320067705},
{"lat":51.53297533262061,"lng":-0.07981995237059891},
{"lat":51.50869919147977,"lng":-0.07673346321098508}

That's: SW NW NE SE -- that should be correct... in the submitted order... let me check where they're coming out on the cloud logs...

[jywarren]

0,0 0,3573     NW to SW
521,0 0,0      NE to NW
521,348 0,0      SE to NW
0,348 0,4343      SW to SW

...roughly. Main problem is that points 2 and 3 of 4 are both being sent to 0,0!

[jywarren]

Oh gosh, you're right! The exporter service uses lon -- ok, changing!

[sashadev-sky]

@jywarren Ok great! I had on my todo list to add to the README a blurb about coordinate order because at one point we had them backwards!

[jywarren]

Hooray! There we go!

https://mapknitter-exports-warps.storage.googleapis.com/1562027947/1562027947.jpg

{"status_url":"https://mapknitter-exports-warps.storage.googleapis.com/1562027947/status.json","status":"complete","tms":"public/tms/1562027947/","geotiff":"public/warps/1562027947/1562027947.tif","zip":"public/tms/1562027947.zip","jpg":"public/warps/1562027947/1562027947.jpg","export_id":1562027947,"user_id":null,"size":"78.9256MB","width":"7388","height":"2670","start_time":"2019-07-02 00:39:06 +0000","run_time":null,"gem_version":"1.0.9","cm_per_pixel":100.0}

There are some timing issues... or something... for example I'm still getting responses like this as well:

{"status_url":null,"status":"starting","tms":false,"geotiff":false,"zip":false,"jpg":false,"export_id":1562027947,"user_id":null,"size":null,"width":null,"height":null,"start_time":"2019-07-02 00:39:06 +0000","run_time":null,"gem_version":"1.0.9","cm_per_pixel":null}

[sashadev-sky]

@jywarren It is working! But I am stuck on warping 1/1 still. Is this because I'm not on gh-pages?

[jywarren]

It's a big export... just give it a few more minutes?

[jywarren]

Mine has some caching issues. If i call:

$.ajax('http://export.mapknitter.org/id/1562027947/status.json').done(function(r){console.log(r)})

I get status: 'starting' even though I've /also/ received a complete response earlier. I keep getting started and so it won't kill our interval timer...

Can't seem to bust that cached response with this either:

$.ajax('http://export.mapknitter.org/id/1562027947/status.json?_=12421', {cache: false}).done(function(r){console.log(r)})

OK. We basically got there. Let's see if we can fix this cache issue and tomorrow we can merge this and publish it.

[jywarren]

Ah! OK! So -- here it is -- if we try to bust the cache by appending a timestamp, like:

$.ajax('http://export.mapknitter.org/id/1562027947/status.json?123').done(function(r){console.log(r)})

It will strip the ?123 when it redirects to https://mapknitter-exports-warps.storage.googleapis.com/1562027947/status.json

But if we directly call the final URL with a ?123 appended to bust /that/ cache, it works!

$.ajax('https://mapknitter-exports-warps.storage.googleapis.com/1562027947/status.json?123').done(function(r){console.log(r)})

So, we really do have to get this segment of code working:

Leaflet.DistortableImage/src/DistortableCollection.js

Lines 353 to 359 in 2d7ad24

	// this may be overridden to update the UI to show export progress or completion
	function _defaultUpdater(data) {
	// optimization: fetch status directly from google storage:
	if (data.hasOwnProperty('status_url') && statusUrl !== data.status_url && data.status_url.match('.json')) { statusUrl = data.status_url; } if (data.status === "complete") {
	clearInterval(updateInterval);
	alert("Export complete. " + data.jpg);
	}

Where it can detect the "final" address.

[jywarren]

Actually I think I can fix it here too:

https://github.com/publiclab/mapknitter-exporter-sinatra/blob/4620dd9d28edae612fe59ee30a1ed303f8e47289/app.rb#L51-L59

OK - i did this in that project. I'm going to merge now! Let's do all this follow-up too!

[jywarren]

So, prompt, spinner, and then MK integration!

[jywarren]

Exciting!

[jywarren]

and -- we should rejigger the section of code above that tries to match "complete" and the new JSON address. I think if we fix it, things will run a lot smoother!

[jywarren]

I published this to NPM!

[sashadev-sky]

@jywarren should I try to connect it in MK first?

[sashadev-sky]

Ill open a new issue for the next 3 parts u just listed!

[jywarren]

I'm calling it a night! I'll check in in the morning, thanks Sasha!

…

On Mon, Jul 1, 2019, 9:12 PM Sasha Boginsky ***@***.***> wrote: Ill open a new issue for the next 3 parts u just listed! — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#314?email_source=notifications&email_token=AAAF6J4LZWICOIY4UNVJFBLP5KTMJA5CNFSM4H4LIEJKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODY7YKIQ#issuecomment-507479330>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAAF6J7WFWRZOMVIMZC56NLP5KTMJANCNFSM4H4LIEJA> .