docs/reference/glossary.rst: Add new terms and reduce complexity of old.

pull/1/head
Jim Mussared 2019-08-22 13:19:40 +10:00 committed by Damien George
parent e0befd9e04
commit a93495b66d
1 changed files with 160 additions and 115 deletions

View File

@ -4,152 +4,197 @@ Glossary
.. glossary::
baremetal
A system without a (full-fledged) OS, for example an
A system without a (full-fledged) operating system, for example an
:term:`MCU`-based system. When running on a baremetal system,
MicroPython effectively becomes its user-facing OS with a command
interpreter (REPL).
MicroPython effectively functions like a small operating system,
running user programs and providing a command interpreter
(:term:`REPL`).
buffer protocol
Any Python object that can be automatically converted into bytes, such
as ``bytes``, ``bytearray``, ``memoryview`` and ``str`` objects, which
all implement the "buffer protocol".
board
A PCB board. Oftentimes, the term is used to denote a particular
model of an :term:`MCU` system. Sometimes, it is used to actually
refer to :term:`MicroPython port` to a particular board (and then
may also refer to "boardless" ports like
:term:`Unix port <MicroPython Unix port>`).
Typically this refers to a printed circuit board (PCB) containing a
:term:`microcontroller <MCU>` and supporting components.
MicroPython firmware is typically provided per-board, as the firmware
contains both MCU-specific functionality but also board-level
functionality such as drivers or pin names.
bytecode
A compact representation of a Python program that generated by
compiling the Python source code. This is what the VM actually
executes. Bytecode is typically generated automatically at runtime and
is invisible to the user. Note that while :term:`CPython` and
MicroPython both use bytecode, the format is different. You can also
pre-compile source code offline using the :term:`cross-compiler`.
callee-owned tuple
A tuple returned by some builtin function/method, containing data
which is valid for a limited time, usually until next call to the
same function (or a group of related functions). After next call,
data in the tuple may be changed. This leads to the following
restriction on the usage of callee-owned tuples - references to
them cannot be stored. The only valid operation is extracting
values from them (including making a copy). Callee-owned tuples
is a MicroPython-specific construct (not available in the general
Python language), introduced for memory allocation optimization.
The idea is that callee-owned tuple is allocated once and stored
on the callee side. Subsequent calls don't require allocation,
allowing to return multiple values when allocation is not possible
(e.g. in interrupt context) or not desirable (because allocation
inherently leads to memory fragmentation). Note that callee-owned
tuples are effectively mutable tuples, making an exception to
Python's rule that tuples are immutable. (It may be interesting
why tuples were used for such a purpose then, instead of mutable
lists - the reason for that is that lists are mutable from user
application side too, so a user could do things to a callee-owned
list which the callee doesn't expect and could lead to problems;
a tuple is protected from this.)
This is a MicroPython-specific construct where, for efficiency
reasons, some built-in functions or methods may re-use the same
underlying tuple object to return data. This avoids having to allocate
a new tuple for every call, and reduces heap fragmentation. Programs
should not hold references to callee-owned tuples and instead only
extract data from them (or make a copy).
CircuitPython
A variant of MicroPython developed by `Adafruit Industries
<https://circuitpython.org>`_.
CPython
CPython is the reference implementation of Python programming
language, and the most well-known one, which most of the people
run. It is however one of many implementations (among which
Jython, IronPython, PyPy, and many more, including MicroPython).
As there is no formal specification of the Python language, only
CPython documentation, it is not always easy to draw a line
between Python the language and CPython its particular
implementation. This however leaves more freedom for other
implementations. For example, MicroPython does a lot of things
differently than CPython, while still aspiring to be a Python
language implementation.
CPython is the reference implementation of the Python programming
language, and the most well-known one. It is, however, one of many
implementations (including Jython, IronPython, PyPy, and MicroPython).
While MicroPython's implementation differs substantially from CPython,
it aims to maintain as much compatibility as possible.
cross-compiler
Also known as ``mpy-cross``. This tool runs on your PC and converts a
:term:`.py file` containing MicroPython code into a :term:`.mpy file`
containing MicroPython bytecode. This means it loads faster (the board
doesn't have to compile the code), and uses less space on flash (the
bytecode is more space efficient).
driver
A MicroPython library that implements support for a particular
component, such as a sensor or display.
FFI
Acronym for Foreign Function Interface. A mechanism used by the
:term:`MicroPython Unix port` to access operating system functionality.
This is not available on :term:`baremetal` ports.
filesystem
Most MicroPython ports and boards provide a filesystem stored in flash
that is available to user code via the standard Python file APIs such
as ``open()``. Some boards also make this internal filesystem
accessible to the host via USB mass-storage.
frozen module
A Python module that has been cross compiled and bundled into the
firmware image. This reduces RAM requirements as the code is executed
directly from flash.
Garbage Collector
A background process that runs in Python (and MicroPython) to reclaim
unused memory in the :term:`heap`.
GPIO
General-purpose input/output. The simplest means to control
electrical signals. With GPIO, user can configure hardware
signal pin to be either input or output, and set or get
its digital signal value (logical "0" or "1"). MicroPython
abstracts GPIO access using :class:`machine.Pin` and :class:`machine.Signal`
General-purpose input/output. The simplest means to control electrical
signals (commonly referred to as "pins") on a microcontroller. GPIO
typically allows pins to be either input or output, and to set or get
their digital value (logical "0" or "1"). MicroPython abstracts GPIO
access using the :class:`machine.Pin` and :class:`machine.Signal`
classes.
GPIO port
A group of :term:`GPIO` pins, usually based on hardware
properties of these pins (e.g. controllable by the same
register).
A group of :term:`GPIO` pins, usually based on hardware properties of
these pins (e.g. controllable by the same register).
heap
A region of RAM where MicroPython stores dynamic data. It is managed
automatically by the :term:`Garbage Collector`. Different MCUs and
boards have vastly different amounts of RAM available for the heap, so
this will affect how complex your program can be.
interned string
A string referenced by its (unique) identity rather than its
address. Interned strings are thus can be quickly compared just
by their identifiers, instead of comparing by content. The
drawbacks of interned strings are that interning operation takes
time (proportional to the number of existing interned strings,
i.e. becoming slower and slower over time) and that the space
used for interned strings is not reclaimable. String interning
is done automatically by MicroPython compiler and runtimer when
it's either required by the implementation (e.g. function keyword
arguments are represented by interned string id's) or deemed
beneficial (e.g. for short enough strings, which have a chance
to be repeated, and thus interning them would save memory on
copies). Most of string and I/O operations don't produce interned
strings due to drawbacks described above.
An optimisation used by MicroPython to improve the efficiency of
working with strings. An interned string is referenced by its (unique)
identity rather than its address and can therefore be quickly compared
just by its identifier. It also means that identical strings can be
de-duplicated in memory. String interning is almost always invisible to
the user.
MCU
Microcontroller. Microcontrollers usually have much less resources
than a full-fledged computing system, but smaller, cheaper and
than a desktop, laptop, or phone, but are smaller, cheaper and
require much less power. MicroPython is designed to be small and
optimized enough to run on an average modern microcontroller.
micropython-lib
MicroPython is (usually) distributed as a single executable/binary
file with just few builtin modules. There is no extensive standard
library comparable with :term:`CPython`. Instead, there is a related, but
separate project
`micropython-lib <https://github.com/micropython/micropython-lib>`_
which provides implementations for many modules from CPython's
standard library. However, large subset of these modules require
POSIX-like environment (Linux, FreeBSD, MacOS, etc.; Windows may be
partially supported), and thus would work or make sense only with
`MicroPython Unix port`. Some subset of modules is however usable
for `baremetal` ports too.
library comparable with :term:`CPython`'s. Instead, there is a related,
but separate project `micropython-lib
<https://github.com/micropython/micropython-lib>`_ which provides
implementations for many modules from CPython's standard library.
Unlike monolithic :term:`CPython` stdlib, micropython-lib modules
are intended to be installed individually - either using manual
copying or using :term:`upip`.
Some of the modules are are implemented in pure Python, and are able to
be used on all ports. However, the majority of these modules use
:term:`FFI` to access operating system functionality, and as such can
only be used on the :term:`MicroPython Unix port` (with limited support
for Windows).
Unlike the :term:`CPython` stdlib, micropython-lib modules are
intended to be installed individually - either using manual copying or
using :term:`upip`.
MicroPython port
MicroPython supports different :term:`boards <board>`, RTOSes,
and OSes, and can be relatively easily adapted to new systems.
MicroPython with support for a particular system is called a
"port" to that system. Different ports may have widely different
functionality. This documentation is intended to be a reference
of the generic APIs available across different ports ("MicroPython
core"). Note that some ports may still omit some APIs described
here (e.g. due to resource constraints). Any such differences,
and port-specific extensions beyond MicroPython core functionality,
would be described in the separate port-specific documentation.
MicroPython supports different :term:`boards <board>`, RTOSes, and
OSes, and can be relatively easily adapted to new systems. MicroPython
with support for a particular system is called a "port" to that
system. Different ports may have widely different functionality. This
documentation is intended to be a reference of the generic APIs
available across different ports ("MicroPython core"). Note that some
ports may still omit some APIs described here (e.g. due to resource
constraints). Any such differences, and port-specific extensions
beyond the MicroPython core functionality, would be described in the
separate port-specific documentation.
MicroPython Unix port
Unix port is one of the major :term:`MicroPython ports <MicroPython port>`.
It is intended to run on POSIX-compatible operating systems, like
Linux, MacOS, FreeBSD, Solaris, etc. It also serves as the basis
of Windows port. The importance of Unix port lies in the fact
that while there are many different :term:`boards <board>`, so
two random users unlikely have the same board, almost all modern
OSes have some level of POSIX compatibility, so Unix port serves
as a kind of "common ground" to which any user can have access.
So, Unix port is used for initial prototyping, different kinds
of testing, development of machine-independent features, etc.
All users of MicroPython, even those which are interested only
in running MicroPython on :term:`MCU` systems, are recommended
to be familiar with Unix (or Windows) port, as it is important
productivity helper and a part of normal MicroPython workflow.
The unix port is one of the major :term:`MicroPython ports
<MicroPython port>`. It is intended to run on POSIX-compatible
operating systems, like Linux, MacOS, FreeBSD, Solaris, etc. It also
serves as the basis of Windows port. The Unix port is very useful for
quick development and testing of the MicroPython language and
machine-independent features. It can also function in a similar way to
:term:`CPython`'s ``python`` executable.
.mpy file
The output of the :term:`cross-compiler`. A compiled form of a
:term:`.py file` that contains MicroPython bytecode instead of Python
source code.
native
Usually refers to "native code", i.e. machine code for the target
microcontroller (such as ARM Thumb, Xtensa, x86/x64). The ``@native``
decorator can be applied to a MicroPython function to generate native
code instead of bytecode for that function, which will likely be
faster but use more RAM.
port
Either :term:`MicroPython port` or :term:`GPIO port`. If not clear
from context, it's recommended to use full specification like one
of the above.
Usually short for :term:`MicroPython port`, but could also refer to
:term:`GPIO port`.
.py file
A file containing Python source code.
REPL
An acronym for "Read, Eval, Print, Loop". This is the interactive
Python prompt, useful for debugging or testing short snippets of code.
Most MicroPython boards make a REPL available over a UART, and this is
typically accessible on a host PC via USB.
stream
Also known as a "file-like object". An object which provides sequential
read-write access to the underlying data. A stream object implements
a corresponding interface, which consists of methods like ``read()``,
``write()``, ``readinto()``, ``seek()``, ``flush()``, ``close()``, etc.
A stream is an important concept in MicroPython, many I/O objects
implement the stream interface, and thus can be used consistently and
interchangeably in different contexts. For more information on
streams in MicroPython, see `uio` module.
Also known as a "file-like object". An Python object which provides
sequential read-write access to the underlying data. A stream object
implements a corresponding interface, which consists of methods like
``read()``, ``write()``, ``readinto()``, ``seek()``, ``flush()``,
``close()``, etc. A stream is an important concept in MicroPython;
many I/O objects implement the stream interface, and thus can be used
consistently and interchangeably in different contexts. For more
information on streams in MicroPython, see the `uio` module.
UART
Acronym for "Universal Asynchronous Receiver/Transmitter". This is a
peripheral that sends data over a pair of pins (TX & RX). Many boards
include a way to make at least one of the UARTs available to a host PC
as a serial port over USB.
upip
(Literally, "micro pip"). A package manage for MicroPython, inspired
by :term:`CPython`'s pip, but much smaller and with reduced functionality.
upip runs both on :term:`Unix port <MicroPython Unix port>` and on
:term:`baremetal` ports (those which offer filesystem and networking
support).
(Literally, "micro pip"). A package manager for MicroPython, inspired
by :term:`CPython`'s pip, but much smaller and with reduced
functionality.
upip runs both on the :term:`Unix port <MicroPython Unix port>` and on
:term:`baremetal` ports which offer filesystem and networking support.