/translate-extract

Extract all Fitzgerald quotes from the metaphor analysis essay with chapter references.

Purpose

Create a structured registry of all direct Fitzgerald quotes used in analysis.md, enabling accurate lookup of official translations in the subsequent pipeline phase.

Input

• •mip-analysis/essay/analysis.md — Main essay containing embedded Fitzgerald quotes

Output

• •translations/quote-registry.json — Structured quote registry

Output Format

{
  "metadata": {
    "source_file": "mip-analysis/essay/analysis.md",
    "extraction_date": "YYYY-MM-DD",
    "quote_count": <number>
  },
  "quotes": [
    {
      "id": "Q001",
      "text": "exact quote text",
      "chapter": 1,
      "citation": "(Fitzgerald, Ch. 1)",
      "context": "surrounding sentence or phrase for disambiguation",
      "section": "essay section where quote appears"
    }
  ]
}

Extraction Rules

1. •Identify quotes: Look for text in double quotes followed by (Fitzgerald, Ch. X)
2. •Preserve exact text: Include all punctuation, ellipses, and formatting
3. •Record chapter: Parse the chapter number from the citation
4. •Capture context: Include 10-20 words before the quote for disambiguation
5. •Note section: Record which essay section (Introduction, DESIRE IS LIGHT, etc.)

Execution Steps

1. •Read mip-analysis/essay/analysis.md
2. •Parse all Fitzgerald quotes using pattern: "[^"]+" \(Fitzgerald, Ch\. \d\)
3. •For each quote:
   • •Extract exact text (including internal punctuation)
   • •Parse chapter number
   • •Capture surrounding context
   • •Identify essay section
4. •Generate sequential IDs (Q001, Q002, ...)
5. •Write structured JSON to translations/quote-registry.json

Verification

After extraction, report:

• •Total quote count
• •Distribution by chapter
• •Any quotes with ambiguous chapter references