EloshonReference
The platforms and sites the Scraper already knows
Each platform the Scraper recognises was worked on site by site: real shops opened one by one, to understand where each one keeps the price, the variants or the barcodes, then a method written and checked on real pages before it was released. The result: on these platforms, the Scraper knows where to look from the first page, without asking you anything.
Checked against the software on November 1, 2026
In short
For the platforms on this page, the Scraper already has its method: you paste the address, and it reads. For a site it does not know yet, it searches on its own, and you can show it what is missing in one click. The shop names quoted here belong to their owners; Eloshon has no connection with them.
Before you read a site
Check the site's terms of use and the rights to its content (text, photos, trademarks) before you read or reuse it. Each user remains responsible for how they use the data they read. This note is general information, not legal advice.
Recipe, method: what do these words mean?
Every online shop stores its information in its own way. Some publish the list of their products in a hidden file, others slip the price and the variants into invisible code, and others only show them once the page is displayed. A method is the path to follow for a whole platform (all Shopify shops, for example). A recipe is the path for one single site. When the Scraper knows them, it goes straight to the information: reading is faster, more complete, and makes fewer mistakes.
Work done site by site
- Real shops. Each method, except Magento's, starts from real shops, studied one by one: where the price is, where the variants are, where the barcode is, how to go to the next page.
- Checks. Each method is checked on real pages before it is released (except Magento's), then protected by automatic tests, on pages built with the same shape as the real ones. For Shopify, WooCommerce, PrestaShop 1.6, nopCommerce and CS-Cart, one or two real saved pages are also read again at each test.
- Safeguards. When a method works out a value, it compares it with what the page shows. For example, on PrestaShop, the calculated prices of the combinations (PrestaShop's word for variants) are only kept if the price of the combination shown by default matches the price on the page; otherwise, none of them is kept.
- Your own recipes. What you teach the Scraper about a site (with Teach this page in the app, or with 🎯 in the extension) is kept, and will be used again at the next scraping of that site.
Recognised platforms
The “Since” column gives the version of the app that brings the method. Your version is written at the bottom of the app's left bar (“v0.5.…”).
| Platform | Since | What the Scraper reads | Limits |
|---|---|---|---|
| Shopify | 0.4.6 | The list of a shop's products (and, since 0.5.58, of a single collection), in one go, without opening each product page: title, description, price, crossed-out price, availability, reference, brand, images, type, weight. On a product page (since 0.5.61): each variant with its reference, its price, its availability and its barcode. | The public list does not give the barcodes. If the shop stops the reading in the middle of the list, the Scraper says so: the list may be incomplete. |
| WooCommerce | 0.4.6 | The list of a shop's products (and, since 0.5.58, of a single category): name, short description, price, crossed-out price, reference, gallery, tags, categories, specifications. On a product page (since 0.5.61): each variation with its reference, its price and its availability. | No barcode. No price per variation in the list (it comes from the product pages). |
| Magento 2 | 0.5.61 | Category pages: name, reference, price, crossed-out price, stock, image, short description, options. | Not the whole shop, only its categories. Method written from Magento's documentation, not yet checked on a real shop. |
| PrestaShop 1.6 (and 1.5) | 0.5.62 (1.5 template: 0.5.65) | Each combination: its option, its reference, its EAN code, its stock, its price including all taxes (kept only if the calculated price of the default combination matches the price shown) and its crossed-out price. Also the gallery and the short description. | PrestaShop 1.7 and later versions do not have their method yet: they are read like an unknown site. |
| nopCommerce | 0.5.65 | Price, price before discount, full description, packs with their price, large images, categories, and every page of a category. | The reference and the barcode of each pack are not read. |
| CS-Cart | 0.5.65 | Price, crossed-out price, reference, description, gallery, breadcrumb trail (the path of categories above the product), and the hidden table of variations (code, option, stock, price). | The crossed-out price of the other variations is not read. |
| osCommerce, xt:Commerce, modified | 0.5.65 | The addresses of the products and categories, the name and the brand, the breadcrumb trail, the gallery. | Paid accessories offered with the product are not read as options. |
| JTL-Shop | 0.5.70 | Platform recognised. Its product pages are read through their product data (reference, brand, price); its options, with their price and their item number, are pointed out with 🎯. | The barcode and the photo of each option, on their separate page, are not read. |
| BigCommerce | 0.5.72 | Reference, barcode, the manufacturer's real reference, stock, price, most precise category, table of specifications with its sections and its PDF documents. A missing weight is never written as 0. | Accessories offered with the product are not read. |
Whatever the platform, the Scraper also reads the product data that many sites hide in their pages for Google (JSON-LD and microdata: name, brand, reference, barcode, price, stock, images, breadcrumb trail), as well as the product lists of sites built with Next.js or Nuxt.
Sites with their own recipe
Two sites have a recipe that comes with the app. For them, the Scraper uses the site's own product search directly, instead of reading the pages one by one: it is much faster.
| Site | Pages read | What is read | Checked on | Limits |
|---|---|---|---|---|
| Colruyt (colruyt.be) | Categories | Name, price, reference, image, address of the product page, brand, category, pack size. | 09/10/2026: one category announced 178 products; the job read 30 of them, its limit. | The prices are those of a single store (by default, the one in Halle). No barcode. |
| Lidl (lidl.be) | Special offer pages and departments | Title, price, crossed-out price, currency, reference, image, address, brand, description, category, pack size (taken from the base price, for example “500 g”, for food). | 25/09/2026: 114 products from a special offer page; 26/09/2026: 122 products from a department, over several pages. | lidl.be only. Lidl's numbers are not barcodes. A special offer page only exists for as long as the offer lasts. |
Other sites already read during our tests
These sites have no recipe of their own: they are read by the methods above. The results come from our tests; a site can change at any time.
- MediaMarkt (mediamarkt.be): on 20 and 21/10/2026, a list of televisions, 30 products out of 30 accepted, with their barcode and their brand, thanks to the product data hidden in the pages.
- Allbirds (allbirdsbenelux.nl, Shopify): on 27 and 28/09/2026, 88 products accepted.
- Foot Locker (footlocker.be): on 30/10/2026, a product page read with its 17 sizes, their barcodes and their stock.
- Electrodepot (electrodepot.be, Magento): on 21/10/2026, a list of televisions, 963 products accepted.
- Levi’s (levi.com): on 19/09/2026, a category of men’s clothing, 463 products accepted.
- Brico (brico.be): on 08/10/2026, a category of garden furniture, 282 products accepted.
- Nike (nike.com): on 07/10/2026, a sale page, 200 products accepted.
- Puma (eu.puma.com, Salesforce Commerce Cloud): on 07/10/2026, a men’s list, 100 products accepted.
- Delhaize (delhaize.be): on 12/10/2026, a category of beers and spirits, 97 products accepted.
- Hydrobuilder (hydrobuilder.com, Shopify): on 01/11/2026, a collection of LED grow lights, 10 products accepted.
- Growers House (growershouse.com, Shopify): on 31/10/2026, 10 products accepted.
What about a site that is not on this list?
- The Scraper searches on its own: it recognises product cards, prices and next pages on many shops. Start with Analyze a site before reading it.
- If it misses a field, show it the field: Teach this page in the app, or 🎯 in the extension (Point out a price or a field in the extension). It will remember it for that site.
- Some sites, often those of big brands, refuse any automatic reading, even with a recipe: the Scraper says so, and suggests the possible ways forward (Why a site refuses to be read).