Text To Image
Easily convert your text to grayscale images and vice versa.
With this tool you can encode text or plain text files to a grayscale image to be easily shared. Each pixel represents a single character’s decimal value. When decoding an image you can decode text straight to the console or to a plain text file.
Images use a PNG file extension. The image size will be automatically be set depending on the text length to be encoded.
Text should use a character encoding scheme using 8 bits or less until additional functionality is added to support UTF-8 (16 bit). If a character’s decimal value is greater than the limit (see below) the it will be divided by the limit and the new value used. When the character value equals the limit value the new value will be 1.
The limit value passed using either command line argument -l(–limit) specifies the decimal value limit for pixel values starting from 1. The default is 256 allowing for numbers from 0 to 255 (i.e. 8 bit pixels). If the limit value is greater than 8 bits then the value will still be wrapped around since the output image is grayscale.
Requirements
Python 3.x
Pillow (PIL fork)
Install
You can install text_to_image using pip or setup.py
Using pip
$> pip3 install text_to_imageUsing setup.py. First navigative to the root directory where setup.py is located then run
$> python3 setup.py install
How to Use
Once installed you can either directly use the encode.py and decode.py file or import and use the relevant functions.
Using the encode.py
$> python3 encode.py -t "Hello World!" image.png # encodes given text $> python3 encode.py -f my-text-file.txt image.png # encodes a text file $> python3 encode.py --help # for more information on argumentsUsing decode.py
$> python3 decode.py image.png # decodes a given image $> python3 decode.py -f my-text-file.txt image.png # decodes image.png to the given text file $> python3 decode.py --help # for more information on argumentsImporting
import text_to_image encoded_image_path = text_to_image.encode("Hello World!", "image.png") encoded_image_path = text_to_image.encode_file("input_text_file.txt", "output_image.png") decoded_text = text_to_image.decode("encoded_image.png") decoded_file_path = text_to_image.decode_to_file("encoded_image.png", "output_text_file.txt")
Tests
To run tests, navigate to the root directory where setup.py is located and run
$> python3-m unittest discover tests -v
Another image example:
TODO:
Add custom image sizes.
Expand pixel value to allow for UTF-8 characters.
Metadata
Release files for text-to-image 0.0.5
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| text_to_image-0.0.5.tar.gz | 6.3 kB | Without provenance | Details for text_to_image-0.0.5.tar.gz |
Release files / text_to_image-0.0.5.tar.gz
| Download URL | text_to_image-0.0.5.tar.gz |
|---|---|
| Size | 6.3 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
e5484c94ec6c2143e3aee2a7e839e6060dd8eee948470b786272f27e2ba78947
|
|
BLAKE2b-256 checksum How to use checksums |
1de2740b0270f1ca06ccdab62c599e0148732e16d056f42c7f8ea9d1d69d3412
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
Security for text-to-image 0.0.5
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.