Add index and sequence check.

pull/15/head
Peter Hinch 2020-07-06 11:40:57 +01:00
rodzic 1281c08751
commit c0a4406710
2 zmienionych plików z 208 dodań i 76 usunięć

238
README.md
Wyświetl plik

@ -2,61 +2,82 @@
A place for assorted code ideas for MicroPython. Most are targeted at the
Pyboard variants.
# Installing MicroPython libraries
1. [Installation guides](./README.md#1-installation-guides)
1.1 [Installing MicroPython libraries](./README.md#11-installing-micropython-libraries)
1.2 [Fastbuild](./README.md#12-fastbuild) Build scripts and udev rules
1.3 [Installing PicoWeb](./README.md#13-installing-picoweb)
1.4 [Buildcheck](./README.md#14-buildcheck) Check firmware build date
2. [Hardware information and drivers](./README.md#2-hardware-information-and-drivers)
2.1 [ESP32](./README.md#21-esp32)
2.2 [SSD1306](./README.md#22-ssd1306) Write large fonts to the SSD1306
2.3 [Pyboard D](./README.md#23-pyboard-d) Assorted scraps of information
2.4 [DS3231 precision RTC](./README.md#24-ds3231-precision-rtc) Use cheap hardware to calibrate Pyboard RTC
3. [Essays](./README.md#3-essays) General thoughts
3.1 [Resilient](./README.md#31-resilient) A guide to writing resilient WiFi code
3.2 [Serialisation](./README.md#32-serialisation) MicroPython's four serialisation libraries
3.3 [Measurement of relative timing and phase of fast analog signals](./README.md#33-measurement-of-relative-timing-and-phase-of-fast-analog-signals)
4. [Code samples](./README.md#4-code-samples) Samples prefixed Pyboard are Pyboard specific
4.1 [Pyboard Mutex](./README.md#41-pyboard-mutex) Share data between threads and ISR's.
4.2 [Pyboard watchdog](./README.md#42-pyboard-watchdog) Access a Pyboard hardware WDT
4.3 [Software Watchdog](./README.md#43-software-watchdog) Cross-platform soft WDT
4.4 [Reverse](./README.md#44-reverse) Reversal algorithms for bytearrays
4.5 [Timed function](./README.md#45-timed-function)
4.6 [ESP8266 MQTT benchmark](./README.md#46-esp8266-mqtt-benchmark) Test performance of official library
4.7 [Rotary incremental encoder](./README.md#47-rotary-incremental-encoder) Fast, simple, proven algorithm
4.8 [A pseudo random number generator](./README.md#48-a-pseudo-random-number-generator)
4.9 [Verifying incrementing sequences](./README.md#49-verifying-incrementing-sequences)
4.10 [Bitmaps](./README.md#410-bitmaps) Non-allocating ways to access bitmaps
4.11 [Functors and singletons](./README.md#411-functors-and-singletons)
4.12 [A Pyboard power meter](./README.md#412-a-pyboard-power-meter)
# 1. Installation guides
## 1.1 Installing MicroPython libraries
This is more involved since the advent of the pycopy fork of MicroPython.
[This doc](./micropip/README.md) describes the issues and provides a utility
to simplify installation for users of official MicroPython firmware.
# Fastbuild
## 1.2 Fastbuild
Scripts for building MicroPython for various target hardware types and for
updating your local source. Now detects and builds for Pyboard D. See [docs](./fastbuild/README.md)
updating your local source. Now detects and builds for Pyboard D. See
[docs](./fastbuild/README.md)
# ESP32
## 1.3 Installing PicoWeb
Paul Sokolovsk's [PicoWeb](https://github.com/pfalcon/picoweb) requires his
fork of MicroPython.
Some time ago I was asked what was involved to install it on official firmware.
Changes were minor. However it should be stressed that while the version here
works, it is not up to date. See the [Easy installation](./PICOWEB.md) guide.
PR's with updated versions of PicoWeb are welcome.
## 1.4 Buildcheck
Raise an [exception](./buildcheck/buildcheck.py) if a firmware build is earlier
than a given date.
# 2. Hardware information and drivers
## 2.1 ESP32
Pinout diagram for the reference board with notes and warnings about reserved
pins etc. See [this doc](./ESP32/ESP32-Devkit-C-pinout.pdf).
# PicoWeb
[Easy installation](./PICOWEB.md) guide. Simplify installing this on
MicroPython hardware platforms under official MicroPython firmware.
# Serialisation
[A discussion](./SERIALISATION.md) of the need for serialisation and of the
relative characteristics of four libraries available to MicroPython. Includes a
tutorial on a Protocol Buffer library.
# SSD1306
## 2.2 SSD1306
A means of rendering multiple larger fonts to the SSD1306 OLED display. The
`Writer` class which performs this has been substantially improved and may now
be found as part of [this repository](https://github.com/peterhinch/micropython-font-to-py).
# mutex
## 2.3 Pyboard D
A class providing mutual exclusion enabling interrupt handlers and the main
program to access shared data in a manner which ensures data integrity.
Assorted [information](./pyboard_d/README.md) not yet in the official docs).
# watchdog
Access the simpler of the Pyboard's watchdog timers.
# software watchdog (soft_wdt)
A software watchdog timer with a fixed or variable timeout. Supports temporary
suspension and permanent cancellation. The latter can be useful when debugging
code to prevent a machine reboot when the application fails, terminates or is
interrupted with ctrl-c. See code and comments in swdt_tests.py.
# reverse
Fast reverse a bytearray in Arm Thumb assembler.
Python code to bit-reverse (fast-ish) 8, 16 and 32 bit words.
# DS3231
## 2.4 DS3231 precision RTC
This is a low cost precision battery backed real time clock (RTC) accurate to
+-2 minutes/year. Two drivers are provided, one portable across platforms and
@ -67,50 +88,90 @@ from the DS3231. Calibration to high precision may be achieved in five minutes.
The drivers are [documented here](./DS3231/README.md).
# Buildcheck
# 3. Essays
Raise an exception if a firmware build is earlier than a given date.
## 3.1 Resilient
# timed_function
A [guide](./resilient/README.md) to writing reliable ESP8266 networking code.
Probably applies to other WiFi connected MicroPython devices.
Time a function's execution using a decorator. Also a way to implement timeouts
using a closure.
## 3.2 Serialisation
# ESP8266 (MQTT benchmark)
[A discussion](./SERIALISATION.md) of the need for serialisation and of the
relative characteristics of four libraries available to MicroPython. Includes a
tutorial on a Protocol Buffer library.
benchmark.py Tests the performance of MQTT by periodically publishing while
subscribed to the same topic. Measures the round-trip delay. Adapt to suit your
## 3.3 Measurement of relative timing and phase of fast analog signals
This describes ways of using the Pyboard to perform precision measurements of
analog signals of up to around 50KHz. It is documented [here](./phase/README.md).
# 4. Code samples
## 4.1 Pyboard mutex
A [class](./mutex/README.md) providing mutual exclusion enabling hard interrupt
handlers and the main program to access shared data in a manner which ensures
data integrity.
## 4.2 Pyboard watchdog
[Access](./watchdog/wdog.py) the simpler of the Pyboard's watchdog timers.
## 4.3 Software watchdog
A [software watchdog](./soft_wdt/soft_wdt.py) timer with a fixed or variable
timeout. Supports temporary suspension and permanent cancellation. The latter
can be useful when debugging code to prevent a machine reboot when the
application fails, terminates or is interrupted with ctrl-c. See code and
comments in [the test script](./soft_wdt/swdt_tests.py).
## 4.4 Reverse
Fast [reverse](./reverse/reverse.py) a bytearray in Arm Thumb assembler.
Also includes cross-platform Python code to bit-reverse (fast-ish) 8, 16 and 32
bit words.
## 4.5 Timed function
Time a function's execution using a [decorator](./timed_function/timed_func.py)
and implement timeouts using a [closure](./timed_function/timeout.py).
## 4.6 ESP8266 MQTT benchmark
[This benchmark](./ESP8266/benchmark.py) tests the performance of MQTT by
periodically publishing while subscribed to the same topic. Measures the
round-trip delay. Uses the official `umqtt.simple` library. Adapt to suit your
server address and desired QOS (quality of service, 0 and 1 are supported).
After 100 messages reports maximum and minimum delays.
`conn.py` Connect in station mode using saved connection details where possible.
[This connect utility](./esp32/conn.py) connects in station mode using saved
connection details where possible.
# resilient
## 4.7 Rotary Incremental Encoder
A guide to writing reliable ESP8266 networking code. Probably applies to other
WiFi connected MicroPython devices, but reliable ones are thin on the ground.
Classes for handling incremental rotary position encoders. Note Pyboard timers
can do this in hardware. These samples cater for cases where that solution
can't be used. The [encoder_timed.py](./encoders/encoder_timed.py) sample
provides rate information by timing successive edges. In practice this is
likely to need filtering to reduce jitter caused by imperfections in the
encoder geometry.
# Rotary Incremental Encoder
Classes for handling incremental rotary position encoders. Note that the Pyboard
timers can do this in hardware. These samples cater for cases where that
solution can't be used. The `encoder_timed.py` sample provides rate information by
timing successive edges. In practice this is likely to need filtering to reduce
jitter caused by imperfections in the encoder geometry.
There are other algorithms but this is the simplest and fastest I've encountered.
There are other algorithms but this is the simplest and fastest I've
encountered.
These were written for encoders producing TTL outputs. For switches, adapt the
pull definition to provide a pull up or pull down as required.
The `encoder.portable.py` version should work on all MicroPython platforms.
Tested on ESP8266. Note that interrupt latency on the ESP8266 limits
performance. ESP32 has similar limitations.
The [encoder_portable.py](./encoders/encoder_portable) version should work on
all MicroPython platforms. Tested on ESP8266. Note that interrupt latency on
the ESP8266 limits performance. ESP32 has similar limitations.
# A pseudo random number generator
## 4.8 A pseudo random number generator
On the Pyboard V1.1, true random numbers may be generated rapidly with pyb.rng()
which uses a hardware random number generator on the microcontroller.
On the Pyboard V1.1, true random numbers may be generated rapidly with
`pyb.rng()` which uses a hardware random number generator on the
microcontroller.
There are two use cases for the pseudo random number generator. Firstly on
platforms lacking a hardware generator (e.g. the Pyboard Lite). And secondly
@ -119,28 +180,53 @@ number generator is seeded with an arbitrary initial value. On each call to the
function it will return a random number, but (given the same seed) the sequence
of numbers following initialisation will always be the same.
See the code for usage and timing documentation.
See [random.py](./random/random.py) for usage and timing documentation.
# Measurement of relative timing and phase of fast analog signals
## 4.9 Verifying incrementing sequences
This describes ways of using the Pyboard to perform precision measurements of
analog signals of up to around 50KHz. It is documented [here](./phase/README.md).
# bitmap: bool arrays and sets of integers
Classes for non-allocating handling of sets of small integers and small arrays
of booleans. Legal member values for a set and array index values are
constrained to 0 <= value <= max_value where max_value is a constructor arg.
The set and the array are different ways of viewing a bitmap implemented as a
bytearray: e.g. if max_value is 255 the bytearray occupies 32 bytes allocated by
the constructor.
## 4.10 Bitmaps
# Functors and singletons
A bitmap stored in a pre-allocated, fixed size bytearray may be viewed in two
ways:
1. As a set of positive integers whose values are constrained within limits.
2. As a fixed size one dimensional array of booleans.
These views provide a Pythonic interface while retaining the non-allocating
performance advantage relative to native sets and lists.
The file [bitmap.py](./bitmap/bitmap.py) offers classes supporting these views.
The constraint `0 <= value <= max_value` applies where `max_value` is a
constructor arg. The `max_value` arg defines the size of the underlying
bytearray. For example if `max_value` is 255, the bytearray will use 32 bytes.
The constraint applies to member values of a set, and to index values of a
boolean array.
These classes are lightweight. For example the `IntSet` class does not include
all the dunder (magic) methods required to match the native `set` class. These
may readily be added as required.
## 4.11 Functors and singletons
Two simple class decorators for objects useful in hardware interfacing.
Documented [here](./functor_singleton/README.md).
# A design for a hardware power meter
Singletons denote classes for which only a single instance will ever occur.
They are contentious in some circles, on the grounds that the single instance
guarantee may be violated in a specification change. They can be useful in
hardware contexts where a chip design is unlikely suddenly to change.
Singletons denoting hardware interfaces avoid globals and the need to pass
references around.
A functor is a class which is accessed via function call syntax. There is only
one instance, like a singleton. Initial access calls the constructor, with
subsequent accesses being via `__call__`. As an object it can retain state. As
an example, a functor might have a continuously running task: successive calls
modify the behaviour of the task.
# 4.12 A pyboard power meter
This uses a Pyboard to measure the power consumption of mains powered devices.
Unlike simple commercial devices it performs a true vector (phasor) measurement

Wyświetl plik

@ -0,0 +1,46 @@
# check_mid.py Check a sequence of incrementing message ID's.
# Released under the MIT licence. See LICENSE.
# Copyright (C) Peter Hinch 2020
# For use in test scripts: message ID's increment without bound rather
# than modulo N. Assumes message ID's start with 0 or 1.
# Missing and duplicate message counter. Handles out-of-order messages.
# Out of order messages will initially be missing to arrive later.
# The most recent n message ID's are therefore not checked. If a
# message is missing after n have been received, it is assumed lost.
class CheckMid:
def __init__(self, buff=5):
self._buff = buff
self._mids = set()
self.miss = 0 # Count missing message ID's
self.dupe = 0 # Duplicates
self.oord = 0 # Received out of order
self.bcnt = 0 # Reboot count
def __call__(self, mid):
mids = self._mids
if mid <= 1 and len(mids) > 1: # Target has rebooted
self._mids.clear()
self.miss = 0
self.dupe = 0
self.oord = 0
self.bcnt += 1
if mid in mids:
self.dupe += 1
elif mids and mid < max(mids):
self.oord += 1
mids.add(mid)
if len(mids) > self._buff:
oldest = min(mids)
mids.remove(oldest)
self.miss += min(mids) - oldest - 1
# Usage/demo
#cm = CheckMid()
#s1 = (1,2,3,4,5,8,9,10,11,12,13,17,17,16,18,19,20,21,22,23,24,29,28,27,26,30,31,32,33,34,35,36)
#for x in s1:
#cm(x)
#print(cm.dupe, cm.miss, cm.oord, cm.bcnt)