Closed davidex720 closed 5 years ago
You're welcome to make pull request to implement this, shouldn't be too complicated since we have all necessary parts already in the code:
senddata.cpp
which buffers all outgoing messages. This makes it easy to implement a bigger buffer. Just increase size of RTOS queue and store it whereever. T-Beam has 4 MB PSRAM, this should be sufficienct for thousands of messages, no need for microSD)timekeeper.cpp
which tries to keep system time synched with an external timesource, e.g. GPS. Using this version you can just take systime to timestamp messages.very cool! the idea to use a microSD is coming from the need to validate the package sniffing, a lot easier on a microSD than on PSRAM very cool the gps time implementation, I'll have a look at that
What do you mean by validate the package sniffing ? The intention of paxcounter is to have a privacy respecting simple counter, not to have an offline wireshark device for collecting sniffed data.
i was thinking about tests
a lot easier on a microSD than on PSRAM
I don't see that it makes a big difference how and where to store the data.
@davidex720 You just need to rewrite the mallocator function to store data where and how you want it, e.g. on a filesystem.
ok, thank you, i'll do it
@DavideNesi If you do it, would be nice to see your pull request here after.
@davidex720 for SD card usage with ESP32, see here:
https://github.com/espressif/arduino-esp32/issues/1190#issuecomment-470721024
Using tbeam it could be really helpful to have the opportunity to store messages that were not successfully sent. Doing so it could be possible to make something like a buffer of older message that the device could send once it reconnects to a gateway. Using GPS lock the tbeam could store the messages with both time and gps coordinate