new_database

Python exam snippet bank — V2

[!summary] What changed in V2 V2 keeps the final exam-focused dataset as the base and adds the strongest missing material from the older dataset in the same scheme: new snippets for syntax fragments, aliasing/copies, numeric corner cases, loop control/iterators, pandas missing-value alignment, and OOP inheritance, plus several targeted add-on pieces in existing snippets.

At a glance

• topics: 8
• subtopics: 23
• snippets: 51
• pieces: 163
• presets: 5
• question bank coverage checked against: 168 questions
• trap slugs: 103

Package layout

python_exam_snippet_bank_v2/
├── README.md
├── db/
│   └── snippet_bank.sqlite
├── content/
│   └── <topic>/<subtopic>/<snippet-slug>/
│       ├── README.md
│       ├── 01_<piece-slug>.md
│       └── ...
├── exports/
│   ├── topics.tsv
│   ├── subtopics.tsv
│   ├── snippets.tsv
│   ├── pieces.tsv
│   ├── question_taxonomy.tsv
│   ├── question_stress_test.tsv
│   ├── presets.tsv
│   ├── preset_items.tsv
│   ├── navigation_plan.tsv
│   └── ...
└── docs/
    ├── SNIPPETS_CATALOG.md
    ├── PRESETS.md
    ├── NAVIGATION_PLAN.md
    ├── STRESS_TEST_REPORT.md
    ├── LEGACY_INTEGRATION.md
    ├── CHANGELOG.md
    ├── SCHEMA.md
    └── HANDOFF.md

Recommended entry points

• start with docs/PRESETS.md if you want the quickest overview
• open docs/LEGACY_INTEGRATION.md to see what the V2 pass imported from the old dataset
• open docs/SNIPPETS_CATALOG.md for the full human-readable bank
• use db/snippet_bank.sqlite or the TSV exports for implementation work

Notes

[!tip] Stress-test status The original final-bank dataset already cleared the threshold for all 168 bank questions.
V2 adds backup/reference coverage and a few trap clarifiers, so the stress-test export is carried forward unchanged as the baseline coverage record.

[!info] Why SQLite + Markdown Metadata and relationships stay normalized in SQLite/TSV, while multi-line piece bodies live as standalone Markdown files so the frontend does not have to parse giant escaped JSON strings.