Alibaba Scraper & Product Spec Sheet

Scrapes product listings from Alibaba search / category pages (same window.__page__data_sse*._offer_list engine as the base Alibaba Scraper), then — for each product row — opens that product's own detail page and reads its embedded window.detailData JSON to add the full specification table, the colour/size option matrix with stock, packed size & weight, and certificate records. Built for sellers, dropshippers and sourcing agents who need to publish an Alibaba listing on their own store or quote freight without emailing the supplier.

What it scrapes

• Search results — Alibaba search and category browse pages (identical to the base actor).
• Product detail page — one extra GET per product to https://www.alibaba.com/product-detail/x_{productId}.html (or the card's own productUrl when present) to pull the supplier's own spec sheet, variant matrix, packaging and certificates.

Features

• Bulk URLs — process many search URLs in one run (unchanged from the base).
• Max Items — total product rows across all URLs (unchanged from the base).
• Live dataset — rows appear in the output table as each page finishes (unchanged from the base).
• 📋 Full spec sheet (includeSpecSheet, default on) — every attribute the supplier published: material, place of origin, model number, fabric weight, feature tags, and the "Key attributes" rows.
• 🎨 Colour & size options (includeVariantOptions, default on) — every colour/size/style axis the listing offers, with swatch images and the stock Alibaba reports per combination.
• 📦 Packed size & weight (includePackagingAndWeight, default on) — single package L×W×H (cm), gross weight (kg), packed volume, and packaging properties — the numbers a freight forwarder asks for.
• 🏅 Product certificates (includeCertificates, default on) — certificate name, number, issuing authority, validity window and scan images (CE, RoHS, FCC, GCC, Oeko-Tex, UKCA, …).
• Optional Apify Proxy — unchanged from the base; Residential is recommended for both the search and detail lanes (Alibaba blocks datacenter IPs on search and rate-limits detail pages after a handful of requests).

Cost: every toggle above spends one extra page request per product. Turn all four off to run at the base actor's exact cost — the run then skips the detail fetch entirely and still returns the 38 base fields plus the new fields in their empty/null shape.

Input

Field	Type	Description
urls	array	Alibaba search/category URLs (unchanged from the base).
maxItems	integer	Total product rows across all URLs, 1–100000, default 10 (unchanged from the base).
proxyConfiguration	object	Apify Proxy; Residential recommended (unchanged from the base).
includeSpecSheet	boolean	Add the full attribute/specification table. Default true.
includeVariantOptions	boolean	Add the colour/size/style option matrix with stock. Default true.
includePackagingAndWeight	boolean	Add packed size, gross weight, packed volume. Default true.
includeCertificates	boolean	Add certificate name/number/issuer/validity. Default true.

Output

Every dataset row carries the base's 38 fields (title, productUrl, price, badges, companyName, productId, …) unchanged, plus:

Field	Type	Notes
detailUrl	string	The product-detail URL that was (or would be) fetched.
detailFetched	boolean	true only if the detail page was successfully fetched and parsed.
specifications	array	{ name, value, attrNameId } — deduped union of the supplier's basic, industry and other attribute tables.
keyAttributes	array	{ name, value } — the "Key attributes" rows on their own.
variantAxes	array	{ axisId, axisName, type, values: [{ id, name, colorHex, image }] }.
variantCombinations	array	{ skuId, options: [{ axisName, valueName }], stock, stockIsUnlimited, dispatchFrom, storeCode, unitPriceUsd }. unitPriceUsd is only non-null on overseas-warehouse listings — a normal China-dispatch listing has no per-variant price anywhere in the page, only a product-level price.
variantCount	integer	len(variantCombinations) — 0 on listings with no options.
packaging	object	{ packageSize, packageLength, packageWidth, packageHeight, unitWeightKg, unitVolume, packagingProperties }. packageSize is the supplier's raw LxWxH string; the split numeric fields are null when it doesn't parse as exactly three numbers.
certificates	array	{ name, number, type, ownerType, issuingAuthority, validPeriod, iconUrl, imageList, aiSummary }. Empty on listings with no certificates — not every product carries one.
categoryPath	array	Breadcrumb category names, e.g. ["Apparel & Accessories", "Men's Clothing", "Men's Jackets"].
categoryIds	object	{ productCategoryId, productLeafId, firstLevelCateId }.
rating	object or null	{ scope: "product"|"store", averageStar, totalReviewCount } — Alibaba publishes either a product-level or a store-level rating, never both, so scope tells you which one you got.

When a detail page can't be fetched (rate-limited, blocked, or the toggle is off), the row still contains the base's 38 fields with real values — only the additions above default to null/[]/0 with detailFetched: false. A row is never dropped for a missing detail page.

Local run

cd alibaba-scraper-and-product-spec-sheet
pip install -r requirements.txt

Create storage/key_value_stores/default/INPUT.json (Apify CLI) or use the Console. Example:

{
  "urls": ["https://www.alibaba.com/trade/search?keywords=jacket&page=1"],
  "maxItems": 5,
  "proxyConfiguration": { "useApifyProxy": true, "apifyProxyGroups": ["RESIDENTIAL"] },
  "includeSpecSheet": true,
  "includeVariantOptions": true,
  "includePackagingAndWeight": true,
  "includeCertificates": true
}

$apify run

Legal

Data is taken from publicly visible pages only. You are responsible for compliance with Alibaba's terms and applicable laws.