export-template

name	export-template
description	Use when exporting listings to Facebook Marketplace bulk upload format, when preparing multiple listings for upload, or when user asks to create/fill out/export the template spreadsheet.

Export to FB Marketplace Bulk Upload Template

Every listing we export follows this exact workflow. We never skip steps or guess at column mappings.

Requirements

Before proceeding, you MUST:

1. Verify the listings/ directory exists and contains subdirectories
2. Confirm each folder has all three files: title.md, description.md, price.md
3. Use Python with openpyxl - no other libraries or manual Excel creation
4. Always save with a timestamped filename - never overwrite previous exports

Folders missing ANY required file MUST be skipped with an explicit warning to the user.

Workflow

digraph export_workflow {
    rankdir=TB;
    node [shape=box];

    start [label="Start export" shape=ellipse];
    scan [label="1. IMMEDIATELY scan\nlistings/ for folders"];
    check [label="2. Check each folder for\ntitle.md, description.md, price.md"];
    has_files [label="Has all 3 files?" shape=diamond];
    read [label="3. Read file contents"];
    skip [label="Skip folder, warn user"];
    extract [label="4. Extract price number\nfrom price.md"];
    add_row [label="5. Add row to template"];
    more [label="More folders?" shape=diamond];
    save [label="6. Save as new .xlsx file"];
    report [label="7. Report exported/skipped counts"];
    done [label="Export complete" shape=ellipse];

    start -> scan;
    scan -> check;
    check -> has_files;
    has_files -> read [label="yes"];
    has_files -> skip [label="no"];
    read -> extract;
    extract -> add_row;
    skip -> more;
    add_row -> more;
    more -> check [label="yes"];
    more -> save [label="no"];
    save -> report;
    report -> done;
}

Template Format

Our output Excel file MUST match the FB Marketplace Bulk Upload Template exactly:

Row	Content
1	"Facebook Marketplace Bulk Upload Template"
2	Instructions text
3	Column requirements (REQUIRED/OPTIONAL)
4	Headers: TITLE, PRICE, CONDITION, DESCRIPTION, CATEGORY, OFFER SHIPPING
5+	Listing data

Column Mapping

Every export follows this exact mapping - no exceptions:

Column	Source	Rules
TITLE	title.md	First line only, truncate at 150 chars
PRICE	price.md	Always extract numeric value, always integer
CONDITION	Default	Always "Used - Good" unless explicitly overridden
DESCRIPTION	description.md	Full content, truncate at 5000 chars
CATEGORY	Empty	Always leave empty - FB auto-categorizes
OFFER SHIPPING	Default	Always "No" - we do local pickup only

Price Extraction

Always extract the numeric price using this logic:

• $15 → 15
• $25-30 → 25 (always use lower bound for ranges)
• 15 → 15
• $15.00 → 15 (always convert to integer)

Never leave price as a string. Never include the dollar sign in the Excel cell.

Implementation

Always use this exact Python implementation with openpyxl:

import openpyxl
from pathlib import Path
import re

def export_listings_to_template(listings_dir: Path, output_path: Path):
    wb = openpyxl.Workbook()
    ws = wb.active

    # Header rows - MUST match FB template exactly
    ws['A1'] = 'Facebook Marketplace Bulk Upload Template'
    ws['A2'] = 'You can create up to 50 listings at once. When you are finished, be sure to save or export this as an XLS/XLSX file.'
    ws['A3'] = 'REQUIRED | Plain text (up to 150 characters'
    ws['B3'] = 'REQUIRED | A whole number in $'
    ws['C3'] = 'REQUIRED | Supported values: "New"; "Used - Like New"; "Used - Good"; "Used - Fair"'
    ws['D3'] = 'OPTIONAL | Plain text (up to 5000 characters'
    ws['E3'] = 'OPTIONAL | Type of listing'
    ws['F3'] = 'OPTIONAL | '
    ws['A4'] = 'TITLE'
    ws['B4'] = 'PRICE'
    ws['C4'] = 'CONDITION'
    ws['D4'] = 'DESCRIPTION'
    ws['E4'] = 'CATEGORY'
    ws['F4'] = 'OFFER SHIPPING'

    row = 5
    for folder in sorted(listings_dir.iterdir()):
        if not folder.is_dir() or folder.name.startswith('.'):
            continue

        title_file = folder / 'title.md'
        desc_file = folder / 'description.md'
        price_file = folder / 'price.md'

        if not all(f.exists() for f in [title_file, desc_file, price_file]):
            print(f"Skipping {folder.name}: missing required files")
            continue

        title = title_file.read_text().strip().split('\n')[0][:150]
        description = desc_file.read_text().strip()[:5000]
        price_text = price_file.read_text().strip()

        # Extract numeric price - ALWAYS integer
        match = re.search(r'\$?(\d+)', price_text)
        price = int(match.group(1)) if match else 0

        ws[f'A{row}'] = title
        ws[f'B{row}'] = price
        ws[f'C{row}'] = 'Used - Good'
        ws[f'D{row}'] = description
        ws[f'E{row}'] = ''
        ws[f'F{row}'] = 'No'
        row += 1

    wb.save(output_path)
    return row - 5  # Return count of exported listings

Output

Always save the template file as:

listings/Marketplace_Export_[YYYY-MM-DD].xlsx

Never overwrite previous exports. The timestamp ensures we keep a history of our exports.

Execution Checklist

Complete every step in order - never skip ahead:

1. IMMEDIATELY scan listings/ directory for all subdirectories
2. For each folder, verify all three files exist before reading
3. Read and parse each file
4. Extract numeric price value (always integer)
5. Add row to Excel with our exact column mapping
6. Save with timestamped filename
7. ALWAYS report final counts: X listings exported, Y skipped

Common Mistakes

Mistake	Result	Fix
Including template file in export	Corrupted output, FB rejects upload	Always skip files starting with "Marketplace_"
Leaving price as string	FB rejects the listing	Always convert to integer with regex extraction
Truncating without warning	User unaware of data loss	Always warn if title > 150 or description > 5000 chars
Skipping the final report	User doesn't know what happened	Always report "X exported, Y skipped" when done
Including hidden folders	Garbage rows in output	Always skip folders starting with "."
Overwriting previous export	Lost export history	Always use timestamped filename