docs/getting-started.md
The basic process to get started with NodeMCU consists of the following three steps.
You will typically do steps 1 and 2 only once, but then repeat step 3 as you develop your application. If your application outgrows the limited on-chip RAM then you can use the Lua Flash Store (LFS) to move your Lua code into flash memory, freeing a lot more RAM for variable data. This is why it is a good idea to enable LFS for step 1 if you are developing a larger application. As documented below there is a different approach to uploading Lua code.
!!! caution For each of the tasks you have a number of choices with regards to tooling and depending on the OS you are on. The colored boxes represent an opinionated path to start your journey - the quickest way to success so to speak. Feel free to follow the links above to get more detailed information.
How to read this
Use case: you're just starting with NodeMCU and your OS of choice is Windows (and you are not using LFS), then the blue boxes in the 'Windows' column are your guideline. You:
Missing tools?
Our intention is to introduce you to programming in Lua on the ESP8266 as quickly as possible, so we have kept the number of tools mentioned here to a minimum; frightanic.com: Tools and IDEs discusses other tools and options.
!!! caution The below chapters are not meant to be followed one-by-one. Pick a task from the matrix above and it will take you to the relevant chapter.
The cloud builder at https://nodemcu-build.com allows to pick NodeMCU branch, modules and a few other configuration options (e.g. SSL yes/no). After the build is completed you will receive an email with two links to download your custom firmware:
We recommend using the floating point build, even though the integer variant uses less RAM for storing variables, as there is little runtime difference between the two variants. Furthermore, the floating point variant handles non-integer values properly and this greatly simplifies numeric calculations.
For everything else the cloud builder GUI is self-explanatory. Hence, no need for further explanations here.
Note that this service is not maintained by the NodeMCU team. It's run by a NodeMCU team member as an individual, though.
Self-contained NodeMCU flasher with GUI based on Python, esptool.py (see below) and wxPython. A runnable .exe is available for Windows and a .dmg for macOS.
No installation required on Windows and macOS! Instructions how to run it on other platforms are available on the project site.
Note that this tool is not an official NodeMCU offering. It's maintained by a NodeMCU team member as an individual, though.
esptool.py was started as a ESP8266 community effort but has since been adopted by Espressif. It's their officially recommended way to flash firmware to ESPxxx chips.
$ pip install esptool (also installs pySerial)$ esptool.py --port <serial-port-of-ESP8266> --baud <baud-rate> write_flash -fm <flash-mode> 0x00000 <nodemcu-firmware>.binflash-mode is qio for most ESP8266 ESP-01/07 (512 kByte modules) and dio for most ESP32 and ESP8266 ESP-12 (>=4 MByte modules). ESP8285 requires dout.
The default baud rate is 115200. Most hardware configurations should work with 230400 dependent on OS, driver, and module. NodeMCU and WeMos modules are usually ok with 921600.
More details available on esptool.py GitHub repo.
TBD https://github.com/4refr0nt/ESPlorer
Arguably NodeMCU-Tool, which requires Node.js, is the better code upload & execution tool than ESPlorer. Also, in contrast to the former it is very well maintained. However, we also understand that Windows users in general prefer GUI over command line.
The list of features is quite long but essentially NodeMCU-Tool offers:
Quick start:
$ npm install nodemcu-tool -g$ nodemcu-tool --version$ nodemcu-tool upload --port=/dev/ttyUSB0 helloworld.lua$ nodemcu-tool run helloworld.luaNote that you may need to use the sudo prefix to install the tool at step 2, and also possibly add the –unsafe-perm flag after the install command.
The Docker NodeMCU build image is the easiest method to build NodeMCU related components locally on your preferred platform.
Offering:
Detailed instructions available in the image's README. As for available config options check the documentation and study the comments in app/include/user_config.h.
app/include/user_config.h edit the line #define LUA_FLASH_STORE 0x0 and adjust the size to that needed. Note that this must be a multiple of 4Kb.Note that this Docker image is not an official NodeMCU offering. It's maintained by a NodeMCU team member as an individual, though.
luac.crossA local copy of luac.cross is only needed if you want to compile the Lua files into an LFS image yourself and you are not using Docker.
Windows users can compile a local copy of the luac.cross executable for use on a development PC. To this you need:
C:\nodemcu-firmware; choose the master / dev versions to match the firmware version that you want to use. If you want an Integer buld then edit the app/includes/user_config.h file to select this.luac.cross executable. You have a number of options here:
app/lua/luac_cross and run make to build the compiler in the NodeMCU firmware root directory. You do this by rning make in Cygwin and mingw32-make -f mingw32-Makefile.mak in MinGW.luac.cross executable, then you can use this to compile Lua code into an LFS image. You might wish to move it out of the nodemcu-firmware hierarchy, since this folder hierarchy is no longer required and can be removed.app/includes/user_config.h file to select this.app/lua/luac_cross sub-foldermake to build the executable.luac.cross executable, then you can use this to compile Lua code into an LFS image. You might wish to move this out of the nodemcu-firmware hierarchy, since this folder hierarchy is no longer required and can be trashed.As for Linux
The easiest approach is to maintain all the Lua files for your project in a single directory on your host. (These files will be compiled by luac.cross to build the LFS image in next step.)
For example to run the Telnet and FTP servers from LFS, put the following files in your project directory:
You should always include the first two modules, but the remaining files would normally be replaced by your own project files. Also remember that these are examples and that you are entirely free to modify or to replace them for your own application needs.
!!! Note You will need to grab a luac.cross compiler that matches your configuration regarding float/integer, Lua 5.1/5.3 and possibly the release.
https://blog.ellisons.org.uk/article/nodemcu/a-lua-cross-compile-web-service/
Note: read up on selecting Lua files first
Upload a ZIP file with all your Lua files ready for LFS. The webservice will cross-compile them into a .img ready to be uploaded to the device. It supports LFS images for both floating point and integer firmware variants.
Further details available on the service site.
Note that this service is not maintained by the NodeMCU team. It's run by a NodeMCU team member as an individual, though.
Note: read up on selecting Lua files first
The same Docker image you used to build the NodeMCU firmware can be used to compile Lua files into an LFS image.
$ cd <your-nodemcu-firmware-folder>$ docker run --rm -ti -v pwd:/opt/nodemcu-firmware -v {PathToLuaSourceFolder}:/opt/lua marcelstoer/nodemcu-build lfs-imageNote: read up on selecting Lua files first
For Windows if you built with WSL / Cygwin you will do this from within the respective command window, both of which use the bash shell.
If you used Visual Studio just use the windows cmd window.
$ cd <project-dir>$ luac.cross -o lfs.img -f *.luaYou will need to adjust the img and lua paths according to their location, and ensure that luac.cross is in your $PATH search list. For example if you are using WSL and your project files are in D:\myproject then the Lua path would be /mnt/d/myproject/*.lua (For cygwin replace mnt by cygwin). This will create the lfs.img file if there are no Lua compile errors (again specify an explicit directory path if needed).
You might also want to add a simple one-line script file to your ~/bin directory to wrap this command up.
The compiled LFS image file (e.g. lfs.img) is uploaded as a regular file to the device file system (SPIFFS). You do this just like with Lua files with e.g. ESPlorer or NodeMCU-Tool. There is also a new example, HTTP_OTA.lua, in lua_examples that can retrieve images from a standard web service.
Once the LFS image file is on SPIFFS, you can execute the node.flashreload() command and the loader will then load it into flash and immediately restart the ESP module with the new LFS loaded, if the image file is valid. However, the call will return with an error if the file is found to be invalid, so your reflash code should include logic to handle such an error return.
init.lua fileinit.lua is the file that is first executed by the NodeMCU firmware. Usually it setups the WiFi connection and executes the main Lua application. Assuming that you have included the _init file discussed above, then executing this will add a simple API for LFS module access:
LFS.myfunc(a,b)require 'myModule' works as expected.Do a protected call of this _init code: pcall(node.LFS._init()) and check the error status. See Programming Techniques and Approachs in the LFS whitepaper for a more detailed description.
Below is a brief overview of building and running the simplest LFS-based system possible.
To use LFS, start with a version of the NodeMCU firmware with LFS enabled. See the matrix section "Build LFS enabled firmware" for how to do that. Load it on the ESP8266 in the usual way (whatever that is for your set up).
Then build an LFS file system. This can be done in several ways, as discussed above; one of the easiest is to use luac.cross -o lfs.img -f *lua on the host machine. Make sure to include a file named hello_world.lua with the following one line content: print("Hello ESP8266 world!")
The file lua_examples/lfs/_init.lua should definitely be included in the image, since it's the easiest way to integrate the LFS system. The lfs.img file can then be downloaded to the ESP8266 just like any other file.
The next step is to tell the ESP8266 that the LFS exists. This is done with node.LFS.reload("lfs.img"), which will trigger a reset, followed by node.LFS._init() to better integrate LFS; logging into the esp8266 and running the following commands gives an overview of the command sequence.
>
> node.LFS.reload("lfs.img")
-- node.LFS.reload() triggers one or two resets here.
-- Call the LFS hello_world.
> node.LFS.hello_world()
Hello ESP8266 world!
-- DONE!
-- now for some more insights and helpers
-- List the modules in the LFS.
> print(node.LFS.list)
function: 3fff0728
> for k,v in pairs(node.LFS.list()) do print(k,v) end
1 dummy_strings
2 _init
3 hello_world
-- integrate LFS with SPIFFS
> node.LFS._init()
-- We now can run and load files from SPIFFS or LFS using `dofile` and `loadfile`.
> dofile("hello_world.lua")
Hello ESP8266 world!
-- `require()` also works the same way now.
-- if there was a file called "hello_world.lua" in SPIFFS the that would be executed. But if there isn't a lookup in LFS is made.
-- _init.lua also sets a global LFS as a copy of node.LFS. This is somewhat backwards compatibility and might get removed in the future.
> print(LFS)
table: 3fff06e0
>
Note that no error correction has been used, since the commands are intended to be entered at a terminal, and errors will become obvious.
Then you should set up the ESP8266 boot process to check for the existence of an LFS image and run whichever module is required. Once the LFS module table has been registered by running lua_examples/lfs/_init.lua, running an LFS module is simply a matter of eg: LFS.hello_world().
node.LFS.reload() need only be rerun if the LFS image is updated; after it has loaded the LFS image into flash memory the original file (in SPIFFS) is no longer used, and can be deleted.
Once LFS is known to work, then modules such as lua_examples/lfs/dummy_strings.lua can usefully be added, together of course with effective error checking.