yashdw – Data Wrangling Practical 1

A small Python package holding solutions to the 10 questions of Data Wrangling Practical 1, runnable from Jupyter with yashdw.q1() … yashdw.q10() or the shortcut yashdw.1 … yashdw.10.

The 10 questions (from the practical file)

The full question text of each is in its function docstring: yashdw.question(3).

Notes on the code: Q.4 uses the array exactly as the question lists it ([12,15,18,20,22,24,26,28,30,32]). Your original notebook was missing the 28. Q.6 is random, so its numbers change on every run.

Usage

import yashdw

yashdw.q1()        # plain function call - works everywhere
yashdw.1           # shortcut (see below)
yashdw.run(5)      # same as q5()
yashdw.question(8) # print the question text
yashdw.show(8)     # print the source code of solution 8
yashdw.insert(8)   # put solution 8's code into a NEW editable cell

From a terminal (no Jupyter needed): python -m yashdw 3 or python -m yashdw all.

How the yashdw.1 shortcut works (and its limits)

• yashdw.1 is not valid Python: the tokenizer reads .1 as a number, and names cannot start with a digit. A real attribute is impossible.
• So import yashdw registers an IPython input transformer. Before a cell runs, it rewrites a line that is exactly yashdw.1 (1–10) into yashdw.q1().
• It executes the solution. It does NOT insert code into your cell. The cell keeps showing yashdw.1, and the output appears below it.
• To get editable code, use yashdw.insert(1). It uses IPython's set_next_input to create a new cell below containing the code. That works in IPython and in Jupyter Notebook/JupyterLab, but I could not test it on your frontend. Writing into the same cell would need a Jupyter frontend extension, which this package does not include.
• The shortcut only works when the line is by itself (yashdw.1 or yashdw.1;) in IPython/Jupyter. In plain python scripts or the standard python shell use yashdw.q1().
• If the shortcut is not active, run %load_ext yashdw or yashdw.enable_shortcut().

Installation

Requirements: Python 3.8+, numpy, pandas (and jupyter/ipython for notebooks). Installing needs internet unless you prepared the wheels beforehand (next section).

Windows (Command Prompt / PowerShell)

cd path\to\yashdw_module
py -m pip install -r requirements.txt
py -m pip install .

Linux / macOS

cd path/to/yashdw_module
python3 -m pip install -r requirements.txt
python3 -m pip install .

Android / Termux

pkg update
pkg install python python-numpy python-pandas
pip install ipython notebook      # Jupyter on Termux can be difficult; ipython alone is fine
cd yashdw_module && pip install .

Termux packages numpy/pandas through pkg; building them with pip is slow and often fails. Jupyter Notebook on Termux can need extra build tools. ipython also supports yashdw.1.

Jupyter

Install into the same Python that runs Jupyter. From inside a notebook:

%pip install /full/path/to/yashdw_module

Then import yashdw.

Without installing at all

Put the yashdw/ folder next to your notebook (or add its parent to sys.path). It only needs numpy and pandas to be installed already.

Offline use (prepare while online)

Offline works only for things already downloaded or installed.

1. While online (the five-minute window), in the project folder:

   python install.py --wheels      # downloads numpy, pandas, etc. wheels into ./wheelhouse
   python install.py               # installs everything now
   

   Pip stores wheels for your OS and Python version only. Prepare on the same machine/Python you will use.
2. Later, offline:

   python install.py --offline     # installs from ./wheelhouse, no internet
   python install.py --check       # runs all 10 solutions
   

   If the packages are already installed (step 1, second command), you need nothing more: import yashdw works offline.

Also prepare beforehand: confirm jupyter notebook starts, and run python install.py --check.

Optional: publishing to PyPI (not done)

This package has not been published. The ZIP works locally without PyPI. If you ever want to publish: pip install build twine, python -m build, then twine upload dist/* with your own PyPI account. The name yashdw may already be taken.

Files

yashdw/__init__.py     shortcut, helpers, IPython transformer
yashdw/solutions.py    q1() ... q10()
yashdw/__main__.py     python -m yashdw N
install.py             install / offline helper
examples/practical_examples.ipynb