Gist 731c0162a924e371a567e83d242d7fa2

100%

Gist 731c0162a924e371a567e83d242d7fa2

✓ Published0🌍 Public

Last edited Aug 22, 2024

Created on Aug 22, 2024

This example shows a Python utility that generates `.pyi` stub files from source code by parsing and transforming Abstract Syntax Trees. The tool processes functions, classes, and decorators—specifically handling `@patch` and `@delegates`—ellipsizing bodies and cleaning annotations based on local type definitions. It uses the `ast` module for parsing, `inspect.signature` for extracting function parameters, and `fastcore` utilities like `@patch` and `delegates`. The script reads a Python file, walks the tree to collect class names, imports the module dynamically, and rewrites the tree before writing the stubbed output.

AI-generated description

Similar vizzes

Loading thumbnail…

Gist b33154065c8c297e8bb1ee38a500a7f6

This example shows how to convert Python type annotations into a structured prompt that guides users through solving problems with the fastcore library’s py2pyi module. It uses plain text instructions and analogies to explain the module’s role in generating type stubs. The visualization relies on a text-based conversational interface, not graphical rendering, to break down tasks into beginner-friendly steps.

深深度碎片

64% match

Loading thumbnail…

Gist 87f69df227eafef3f4fc3774184404e2

This example shows a collection of Python metaclasses and decorators that manipulate class creation, initialization, and function signatures. The code defines metaclasses like `FixSigMeta`, which adjusts `__signature__` based on overridden `__init__` methods, and `PrePostInitMeta`, which calls optional `__pre_init__` and `__post_init__` hooks. Decorators such as `delegates` and `use_kwargs_dict` rewrite `**kwargs` parameters by borrowing signatures from other functions or classes. The implementation relies on the `inspect` module to read and replace `Parameter` objects, with no external data source or rendering involved.

深深度碎片

57% match

Loading thumbnail…

A small script that warns about functions that are not called from within the code (fork from https://gist.github.com/Herst/287aa233c0acba6ed2bc2d093dec786d ported to Python 2)

This example demonstrates a Python script that identifies functions defined in source code but never called elsewhere in the same set of files. It shows how static analysis can flag potentially dead code for cleanup. The implementation uses Python's `ast` module and a custom `NodeVisitor` subclass to traverse abstract syntax trees, collecting both function definitions and references across multiple modules. A separate, simpler version relies on regular expressions via `re.findall` to match definitions and calls. Both approaches compare the set of defined names against referenced names, excluding patterns like `visit_*` or names starting with `test_`, and print the file and line number of each unused function.

HHerst

53% match

Loading thumbnail…

Gist ffc170d3bf3e7ca6ced3f5585f3f97ef

This notebook explores the internal structure of the fastai library by tracing the import chain triggered by `from fastai.vision.all import *`. It shows the file listings and source code of key modules like `vision/all.py`, `basics.py`, and `callback/all.py`, revealing the hierarchical organization of the framework. Using shell commands (`ls`, `less`) within Jupyter cells, it documents how the vision module re-exports components from data, augment, learner, and callback subpackages, providing a code-level walkthrough of fastai's architecture.

深深度碎片

51% match

Loading thumbnail…

Gist d68534c6cd5f4fe95e426fb37c694be4

This example presents a Markdown-rendered Python cheatsheet originally authored by Mosh Hamedani, converted and annotated by IoTeacher. It displays code snippets covering variables, input handling, string operations, arithmetic, control flow, loops, lists, tuples, and dictionaries. The visualization is plain text with syntax highlighting applied through Markdown formatting. It does not use JavaScript libraries, datasets, or interactive rendering; instead, it relies on static code blocks and headings to organize the tutorial content.

IIoTeacher

49% match

Loading thumbnail…

Convert jupyter notebooks to python scripts

This example demonstrates a Python script that converts Jupyter notebook files (`.ipynb`) into plain Python scripts (`.py`). It shows how to parse notebook JSON structures and extract code cells while preserving cell boundaries as comments. The script reads the input file using Python's built-in `json` library and writes the output with a standard shebang line. It handles both notebook format versions 4 and earlier, iterating through `cells` or `worksheets[0].cells` accordingly. The code relies on straightforward file I/O operations and conditional logic to accommodate schema differences, making it a practical utility for version control or code review workflows.

CChristosT

49% match

Loading thumbnail…

A "Best of the Best Practices" (BOBP) guide to developing in Python.

This example presents a Markdown-based coding guide that distills Python best practices into a concise reference. It shows a structured document with sections on values, general development guidelines, and specific style rules for naming, indentation, imports, documentation, and line lengths. The guide uses code snippets and direct quotes from PEP 8, PEP 20, and PEP 257 to illustrate conventions like `lower_case_with_underscores` for variables and `CapWords` for classes. The visualization relies solely on the plain text formatting of the source file, with no external libraries or data processing.

AAbsarF

48% match

Loading thumbnail…

Gist bcaa860506520cefb711aa4a9f885c12

This example assembles a large sprite sheet of data visualization badges, arranging hundreds of PNG icons into a single 8192-pixel square grid. It shows the process of converting individual SVG badge files into a unified texture atlas for efficient loading in a projector tool. The script uses Python’s Pillow library (`Image.open`, `resize`, `paste`) to read, scale, and tile each icon, while calculating the grid dimensions with `math.sqrt`. It sorts the source files from a local SVG directory, converts them to PNGs, and saves the final composite to `sprite_image.png`.

11wheel

47% match