Fix file upload progress

[JeromeDeLeon]

Based on the discussion from #1096

Thanks to @ladigitale for the advice. 👍

[codecov]

Codecov Report

Merging #1133 into master will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master    #1133   +/-   ##
=======================================
  Coverage   92.29%   92.30%           
=======================================
  Files          54       54           
  Lines        5230     5235    +5     
  Branches     1168     1169    +1     
=======================================
+ Hits         4827     4832    +5     
  Misses        403      403           

Impacted Files	Coverage Δ
src/RESTController.js	84.51% <100.00%> (+0.51%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 2081dc2...a3a7fb6. Read the comment docs.

[JeromeDeLeon]

I tested it without type and worked just fine. No duplication. Maybe the browser handles the calling of progress behind the scene if it should be calling progress for upload or progress for download.

[ladigitale]

Great thank you.

Could the type parameter be set at the end, because otherwise i think it could break code which is relying on previous behaviour. i mean code that expect that the progress value is at first place.
Wouldn't it be more "secure" to use this new parameter in the form of an object.
indeed, in the futur, if more data must be added it could be filled in this object seamlessly

[JeromeDeLeon]

I was thinking the same thing like putting loaded and total into an object allowing the future values to be wrapped into an object or putting type as an object at the end of the parameter but then again, the type could possibly remove if we could determine about the collision of upload progress and download progress. If you could test it on your end, that would be great help 👍

[dplewis]

Thanks for the fix. Nice code cleanup

[ladigitale]

I just found the time to test the pacth.
As expected at the end of the upload with this code you will yet have 2 consecutive callbacks of the same function with different meanings :

1 14535341 14535341 (meaning end of the upload)
1 165 165 (meaning end of the download sent as one chunk because of the low size of the response)

Thats why aditionnal imformations / api evolution would have been usefull.

however I really thank you for that work.

[ladigitale]

I hope I 'm not arrogant while writing. English is not my best part.
:)

[JeromeDeLeon]

Seems like type is really necessary there to distinguish the difference of the two (upload and download). I'll try to push another PR and let you know about it.