ReceiptExtract - OCR, Photo/PDF to CSV — Receipt提取 - OCR, Photo/PDF to CSV

Receipts

提取 transaction data from receipt images or PDFs with Receipt提取.

Keep the 工作流 simple: locate the API 令牌, 上传 one receipt file (or a directory for bulk mode), inspect the JSON, then present either raw JSON or a 清理ed summary. Prefer the bundled 辅助工具 script for repeatable usage.

Quick 工作流

Identify the 输入 file

Accept common image 格式化s (.jpg, .jpeg, .png, .网页p) and PDFs. If the file came from chat, use the attached local path.

Locate the API 令牌

设置 RECEIPT提取_API_令牌 in your 环境 before 运行ning commands. Do not paste the 令牌 back into chat.

Call the 上传 端点

端点: POST https://www.receipt提取.com/API/receipt/上传 Auth header: Authorization: Bearer <令牌> Multipart form field: file

解析 the 响应

成功 shape typically includes: 成功 data.merchant data.date data.items[] data.tax data.total data.correctness检查 data.taxBreakdown[] credit信息 savedReceiptId

Present the 结果

For humans: summarize merchant, date, items, tax, total, and any anomalies. For integrations: return raw JSON or convert to CSV. Preferred command

Use the 辅助工具 script:

导出 RECEIPT提取_API_令牌="your-令牌" python3 scripts/提取_receipt.py /path/to/receipt.png

Optional flags:

python3 scripts/提取_receipt.py /path/to/receipt.pdf --格式化 summary python3 scripts/提取_receipt.py /path/to/receipt.jpg --格式化 csv python3 scripts/提取_receipt.py --输入-dir /path/to/receipts --格式化 summary python3 scripts/提取_receipt.py --输入-dir /path/to/receipts --recursive --格式化 json

Bulk processing

Use --输入-dir to process multiple receipts in one 运行. The 辅助工具 script 发送s one API 请求 per file and continues even if some files fAIl.

Supported file types: .jpg, .jpeg, .png, .网页p, .pdf Use --recursive to include nested folders Exit code is non-zero when one or more files fAIl Each receipt consumes credits independently Fallback command

Use curl when the 辅助工具 script is unnecessary:

curl -sS -X POST "https://www.receipt提取.com/API/receipt/上传" \ -H "Authorization: Bearer $RECEIPT提取_API_令牌" \ -F "file=@/path/to/receipt.png"

输出 handling JSON

Prefer JSON when the user wants the full 提取ed payload or when another 工具 will consume the 结果. In bulk mode, JSON includes processed, succeeded, fAIled, and per-file 结果s.

Summary

In bulk mode, summary prints one 状态 line per file followed by total counts.

Use a concise 格式化 like:

Merchant: Walmart Date: 2023-06-09 Total: 76.37 Tax: 8.18 Items:

• BEDDING — 39.97
• STEAMER — 27.97

CSV

When the user asks for CSV, 输出 line-item rows with these columns when avAIlable:

source_file (bulk mode) merchant date description quantity total_price item_tax sku receipt_tax receipt_total saved_receipt_id http_状态 (bulk mode) 成功 (bulk mode) error (bulk mode) Error handling

Interpret common 失败s like this:

400 — invalid 输入, missing file, unsupported type, or file too large 401 — missing/invalid 令牌 402 — insufficient credits 429 — rate limited; retry with backoff 500 — server error; safe to retry carefully

If the 响应 is malformed or 成功 is false:

show the error plAInly do not invent 提取ed data mention likely causes if obvious (bad 令牌, no credits, unsupported file) Practical notes Treat the API 结果 as the source of truth, but sanity-检查 obvious issues. Flag suspicious 输出 instead of silently "fixing" it. Example: Canadian receipt with tax currency labeled USD. correctness检查: true is a useful confidence 签名al, not a guarantee. Preserve the original file path and savedReceiptId when useful for 追踪ability. In bulk mode, keep one 请求 per file and preserve each source file path for 追踪ability. Security Keep the 令牌 out of chat replies. Prefer 环境 variables or secret 管理器s over embedding 令牌s in scripts. Do not commit 令牌s, raw headers, or secret-bearing examples into git. Resources 辅助工具 script: scripts/提取_receipt.py API reference notes: references/API.md