How to efficiently transfer bulk upload from an ESP8266 to ThingSpeak?

This link showed me how to successfully do a JSON bulk upload from an ESP8266 to ThingSpeak. I would like to post 60 values in one bulk upload to save on batteries. I know that it is possible to do a CSV bulk upload of data that does not include the "delta_t" or "field1" ascii text for every datapoint as with JSON. This repeated JSON text really bloats the data transfer (about 2000 characters for 60 values). I did successfully complete the Postman example provided in this link but I am unable to mimic this action in the code on my ESP8266. Does anybody know where example code exists - I've been looking for days? Sorry so cryptic but I'm trying to keep my query trim.

Answer 1

[Craig Larson]

Okay, I think I answered my question. Using this code example built for the Photon, I twisted it into minimal code for the ESP8266 just to demonstrate a bulk load of 6 data elements. Hard for me to belive, it works. From here I can build out the code necessary to build a larger string of 60 data elements. I hope that this can help others so am attaching the resulting code for others. Maybe I get an upvote to my current reputation of zero. The code is attached in the next post..

Answer 2

[Craig Larson]

Code edited and reattached.

Second edit: Attached (*V4) is a more fully developed CVS-like code that builds an demo array, assembles the data, and sends the data to Thingspeak. The demo array only has 6 data pairs (delta_t, and field1). The data_length is a compact 95 characters. When I expand this to my full 60 data pairs, the data_length should be 419 characters including the commas and a pipe separators. My JSON version equivalent had about 2000 characters. One of the things I did to keep this compact was to ommit the commas representing empty fields #2 through #8, and the location info. These commas were suggested in a Postman example found elsewhere of this site. Hope this helps somebody as it took me several days to build, and debug. .

[Christopher Stapels]

Thanks for taking the time to share. It is a difficult endpoint, but there is a lot of data to get into the right place. I think your work well help other users.

[Craig Larson]

Thanks Christopher!