Inscrap

Inscrap is a small command-line tool for exporting product catalogs from InSales stores. It reads product URLs from a sitemap.xml, fetches product data through the store's JSON API, and writes the result to a file.

It is useful when you need to quickly collect product titles, descriptions, prices, stock levels, images, and variants without browsing the catalog manually.

Русская версия

Quick start

Requirements: Python 3.13 or newer and, preferably, uv.

Run Inscrap directly from a cloned repository:

uv run inscrap run https://shop.example.com -o products.xlsx

You can pass a sitemap URL instead of the store URL:

uv run inscrap run https://shop.example.com/sitemap.xml

Without -o, the result is saved as products.json in the current directory.

Installation

Install Inscrap as a global uv tool to use it from any directory:

uv tool install .

Check the installation:

inscrap run --help

You can also build and install a wheel:

uv build
uv tool install dist/insales_scraper-1.1.0-py3-none-any.whl

Or install a wheel from a specific path:

uv tool install /path/to/insales_scraper-1.0-py3-none-any.whl

Remove the global installation with:

uv tool uninstall insales-scraper

Usage

After installation, the command is:

inscrap run URL [OPTIONS]

Choose the output format by changing the file extension:

inscrap run https://shop.example.com -o products.json
inscrap run https://shop.example.com -o products.csv
inscrap run https://shop.example.com -o products.xlsx
inscrap run https://shop.example.com -o products.txt

The output directory must already exist:

mkdir -p data
inscrap run https://shop.example.com -o data/catalog.xlsx

Request options

Option	Description	Default
-o, --output	Output path. The format is selected by the .json, .csv, .xlsx, or .txt extension.	products.json
-c, --concurrency	Number of concurrent requests, from 1 to 50.	5
-r, --retries	Number of retry attempts after a request error.	5
-t, --transport	HTTP transport: httpx or curl_cffi.	httpx
-f, --fatalist	Stop the export after the first product request error.	disabled
-p, --proxy	Proxy URL used for HTTP requests.	not set

For a large catalog, increase concurrency gradually so you do not put unnecessary load on the store:

inscrap run https://shop.example.com -c 10 -r 3 -o catalog.csv

List the available HTTP transports with:

inscrap transport

Exported data

Each product includes its ID, title, description, URL, availability, images, and variants. Each variant includes its own ID, title, SKU, barcode, availability, quantity, current price, and old price.

Format	Data layout
JSON	The complete nested structure with products and their variants.
CSV / XLSX	One row per variant. Product fields are repeated for each of its variants.
TXT	A text representation of the scraped products.

How it works

1. Inscrap opens the provided sitemap.xml. If you pass a store URL, it adds /sitemap.xml.
2. It keeps sitemap links that contain /product/.
3. It requests the JSON version of each product page by adding .json to the product URL.
4. It saves successfully loaded products to the selected file. By default, products with errors are skipped, and their number is included in the final statistics.

Built with

Typer and Rich power the CLI. HTTP requests use httpx or curl-cffi, and exports use polars and xlsxwriter.