Support for .zip files for MicroPython. Port of CPython zipfile.
A subset of .zip was standardized in ISO/IEC 21320–1:2015 "Document Container File: Core". Such files should work with this module. ZIP archives are used as a basis for many common files, such as EPUB, DOCX
micropython-zipfile was initially made for micropython-npyfile, to support Numpy .npz files (uses a ZIP archive). Which again was made for emlearn-micropython, a Machine Learning and Digital Signal Processing library for MicroPython.
Minimally useful.
- Supports read/write of ZIP archives, with or without compression
- Highly compatible with the CPython API
- Passes over 100 tests from the CPython test suite
- Tested on the Unix port and ESP32 port, running MicroPython 1.23
The majority of the code is copied from the CPython 3.12 implementation of zipfile
.
That code is, to the best of our understanding, under the PYTHON SOFTWARE FOUNDATION LICENSE VERSION 2.
All modifications/adaptations done in this project can be used under the same license.
This is an adapted implementation of the original zipfile module in CPython. With the exception of the Limitations documented below, it should have the same features.
This library uses the MicroPython deflate module to handle compression and decompression.
Therefore your MicroPython firmware build must include this module,
which may require you to enable the MICROPY_PY_DEFLATE
and MICROPY_PY_DEFLATE_COMPRESS
features.
If you get the error AttributeError: 'DeflateIO' object has no attribute 'write'
, you are missing MICROPY_PY_DEFLATE_COMPRESS
.
These limitations could be lifted if people contribute
- Only DEFLATE and STORED (uncompressed) is supported.
- LZMA compression not supported
- BZ2 compression not supported
PathLike
objects not supported.readline()
andreadlines()
on file objects not supported
These limitations are likely to be forever. This is because they are not that relevant in a microcontroller/embedded setting.
- PyFile objects not supported
- Usage as a command-line module not supported
- append mode (
'a'
option to ZipFile) not supported. - exclusive mode (
'x'
option to ZipFile) not supported. - Only
ascii
andutf-8
supported for metadata encoding
This package can be installed using mip.
For example:
mpremote mip install github:jonnor/micropython-zipfile
Or just copy the zipfile.py
file to your MicroPython device.
# Import the module
from zipfile import ZipFile, ZIP_DEFLATED, ZIP_STORED
# Create a .zip archive
# for compression, swap ZIP_STORED with ZIP_DEFLATED
path = 'myarchive.zip'
with ZipFile(path, 'w', ZIP_STORED) as archive:
contents = b'Hello World Hello World Hello World\n'
with archive.open('test.txt', 'w') as f:
f.write(contents)
# List the information from a .zip archive
print('\nListing information')
with ZipFile(path, 'r') as archive:
for info in archive.infolist():
print(info.filename)
print('\tSystem:\t\t' + str(info.create_system) + '(0 = Windows, 3 = Unix)')
print('\tZIP version:\t' + str(info.create_version))
print('\tCompressed:\t' + str(info.compress_size) + ' bytes')
print('\tUncompressed:\t' + str(info.file_size) + ' bytes')
# Read a file from inside .zip archive
print('\nReading file')
with ZipFile(path) as myzip:
with myzip.open('test.txt') as myfile:
print(myfile.read())
Also see examples. And consult tutorials and the API reference for the CPython zipfile module.
Contributions welcomed!
- maruno/mpy-blox zipfile.py. Minimal implementation of
zipfile.ZipFile
which supports only ureading. The functionread()
supports uncompressed or deflate compressed data. Not clear if ZIP64 is supported. Not separately installable.
Install the Unix/Windows port of MicroPython.
Install modules used by the tests
micropython -m mip install contextlib unittest os-path stat shutil
Run the basic test suite:
MICROPYPATH=.:$HOME/.micropython/lib micropython tests/test_zipfile.py
Run the full test suite with MicroPython:
MICROPYPATH=.:$HOME/.micropython/lib micropython -X heapsize=10M test_zipfile/test_core.py
The tests can also be ran under CPython
PYTHONPATH=./ python test_zipfile/test_core.py -v
MICROPYPATH=.:$HOME/.micropython/lib micropython tests/test_zipfile.py
Connect a MicroPython device via USB.
Copy over the data
mpremote cp zipfile.py :
mpremote run tests/test_zipfile.py