Skip to content

Latest commit

 

History

History
62 lines (49 loc) · 2.08 KB

README.md

File metadata and controls

62 lines (49 loc) · 2.08 KB
Raw

python-zxing

A quick and dirty wrapper for the ZXing barcode library.

This is a hack subprocess control library that gives you a reasonable Python interface to the ZXing command line. ZXing will recognize and decode 1D and 2D barcodes in images, and return position information and decoded values. This will let you read barcodes from any images in Python.

If you need to threshold or filter your image prior to sending to ZXing, I recommend using functions from SimpleCV.

Usage

git clone https://github.com/zxing/zxing.git
cd zxing-master
mvn install
cd core
wget http://central.maven.org/maven2/com/google/zxing/core/2.2/core-2.2.jar
mv core-2.2.jar core.jar # Rename
mvn install
cd zxing-master/javase
wget http://central.maven.org/maven2/com/google/zxing/javase/2.2/javase-2.2.jar 
mv javase-2.2.jar javase.jar # Rename
mvn install
git clone git://github.com/oostendo/python-zxing.git

The library consists of two classes, BarCodeReader and BarCode. BarCode parses the output from ZXing's CommandLineRunner into a BarCode object which includes:

b = zxing.BarCode("""
file:default.png (format: FAKE_DATA, type: TEXT):
Raw result:
foo-bar|the bar of foo
Parsed result:
foo-bar 
the bar of foo
Also, there were 4 result points:
  Point 0: (24.0,18.0)
  Point 1: (21.0,196.0)
  Point 2: (201.0,198.0)
  Point 3: (205.23952,21.0)
""")

print b.format #FAKE_DATA
print b.raw #foo-bar|the bar of foo
print b.data #foo-bar\nthe bar of foo
print b.points #[(24.0, 18.0) ... ]

Initializing the barcode reader, you have to tell it where to find the ZXing core modules. It is the single parameter, which defaults to the parent directory.

reader = zxing.BarCodeReader("/var/opt/zxing")

barcode = reader.decode("/tmp/image.jpg")
(barcode1, barcode2) = reader.decode(["/tmp/1.png", "/tmp/2.png"])
code_list = reader.decode("/tmp/barcodes", True)

decode() takes an image path, directory, or list of images and has an optional parameter to use the "try_harder" option. If no barcode is found, it returns None objects.

Installation is manual.