Doc Gen Unicode Diagnostic

run_shell
command: pandoc /tmp/document_source.md -o output.html 2>&1

run_shell
command: ls -lh output.* && file output.*

run_shell
command: [ -f output.pdf ] && echo "PDF created: $(wc -c < output.pdf) bytes" || echo "PDF MISSING"

# Generate Client Report in Multiple Formats

## Step 0: Pre-flight
run_shell
command: which pandoc xelatex || echo "Checking toolchain..."

## Step 1: Write markdown source
write_file
path: /tmp/client_report.md
content: |
  # Client Investment Report
  
  ## Executive Summary
  Portfolio performance shows strong returns — up 15% this quarter...
  
  ## Risk Analysis
  Key metrics: "Sharpe ratio" ✓ passed … continuing analysis

## Step 2: Sanitize for PDF (MANDATORY)
write_file
path: /tmp/client_report_sanitized.md
content: |
  # Client Investment Report
  
  ## Executive Summary
  Portfolio performance shows strong returns -- up 15% this quarter...
  
  ## Risk Analysis
  Key metrics: "Sharpe ratio" [x] passed ... continuing analysis

## Step 3: Convert to DOCX (original unicode OK)
run_shell
command: pandoc /tmp/client_report.md -o client_report.docx 2>&1

## Step 4: Convert to PDF (sanitized, xelatex first)
run_shell
command: pandoc /tmp/client_report_sanitized.md -o client_report.pdf --pdf-engine=xelatex 2>&1

## Step 5: Convert to HTML (original unicode OK)
run_shell
command: pandoc /tmp/client_report.md -o client_report.html 2>&1

## Step 6: Verify
run_shell
command: ls -lh client_report.* && file client_report.*

If error contains "xelatex not found" or "LaTeX error":
  → Try next engine: --pdf-engine=pdflatex or --pdf-engine=wkhtmltopdf

If error contains "encoding" or "UTF-8":
  → Unicode not properly sanitized; re-check Step 2
  → Add -f markdown+utf8 to pandoc command

If error contains "template" or "class":
  → LaTeX template issue; try --pdf-engine=wkhtmltopdf

If error is "unknown error" (no stderr captured):
  → Re-run with 2>&1 to capture actual error message
  → Check if pandoc is installed: which pandoc

If wkhtmltopdf fails:
  → Install: apt-get install wkhtmltopdf
  → Or use Python alternative: reportlab or fpdf2

run_shell
command: python3 -c "
from fpdf import FPDF
pdf = FPDF()
pdf.add_page()
pdf.set_font('Arial', '', 12)
pdf.cell(0, 10, 'Document Title')
pdf.output('output.pdf')
"

run_shell
command: python3 -c "
from reportlab.lib.pagesizes import letter
from reportlab.pdfgen import canvas
c = canvas.Canvas('output.pdf', pagesize=letter)
c.drawString(100, 750, 'Document Title')
c.save()
"

# Markdown to Word (Unicode-safe)
pandoc input.md -o output.docx

# Markdown to PDF with xelatex (BEST for Unicode)
pandoc input.md -o output.pdf --pdf-engine=xelatex

# Markdown to PDF with pdflatex (requires sanitization)
pandoc input.md -o output.pdf --pdf-engine=pdflatex

# Markdown to PDF with wkhtmltopdf (HTML-based, good fallback)
pandoc input.md -o output.pdf --pdf-engine=wkhtmltopdf

# Markdown to HTML
pandoc input.md -o output.html

# With metadata
pandoc input.md -o output.pdf --metadata title="Document Title"

# Force UTF-8 encoding
pandoc -f markdown+utf8 input.md -o output.pdf