EloshonReference
All the Scraper's settings
One table per screen: the label shown in the app, the default value and the real effect. In brackets, the technical term when the label was renamed.
Checked against the software on November 1, 2026
Start: New scraping
| Setting | Default | Effect |
|---|---|---|
| Site address | Empty, required | Start page: home, category, product page or search page; the Scraper recognizes the kind of page. When empty, the start is refused with Enter a site address. |
| Source | Empty | Free name written in the store_slug column of every CSV row and shown as a label in Jobs. Empty: the column stays empty and Jobs shows the site's domain. |
| Default product type | auto | Written in the product_type column of every product. auto lets the Scraper fill it when it can, for example video for a video site. |
| Content type | Automatic | Products or Videos picks the reader. Automatic recognizes a shop or a video site from the first page and the sitemap. |
| Scraping mode | Standard scraping | Scraping via your browser reads the site in your Chrome or Edge, through the Eloshon Scraper extension. It needs the extension connected and this site allowed in the extension; under the option, the screen says what is missing. In this mode, the advanced options and the browser settings give way to Limits for this scraping: Max products, Listing pages, Product pages. |
| Max products | 1,000 | Reading stops at this number of products; the end screen says so and where to raise it. |
| Pages read, at most (max pages) | 150 | Most pages read to find the products, listing pages included. |
| Category levels (depth) | 3 | Levels of category links followed from the start page. Higher, the Scraper may find more products, but reads more pages. |
| Pages in parallel (parallel workers) | 4 | Product pages read at the same time. Higher may be faster, but the site may refuse sooner. The engine never reads more than 8 at once, even if you type more. |
| Browser when the page needs it (browser fallback) | On | Lets the app open its built-in browser when a page needs it. Off: no reading of pages that need JavaScript, no manual intervention, no stock by size or reviews loaded separately, no scrolling of lists. |
| Pages in the browser, at most | 40 | Most pages opened in the built-in browser, counted separately for finding the products and for reading product pages; reading a list may open more. Beyond it, pages stay as read without the browser. |
| Manual intervention (CAPTCHA) | On | On: the built-in browser opens in a visible window, keeps its profile from one scraping to the next and waits for you to pass a check. Off: the browser stays hidden, and a check page stops the scraping as on a blocked site. |
The five numbers start from the values in Settings. A number changed here does not touch the defaults; it stays in the form for the next starts, until the app is closed or until the next click on Save in Settings.
Defaults: Settings
These values apply to every scraping, and to a job resumed from Jobs. The four network settings are folded under Advanced.
| Setting | Default | Effect |
|---|---|---|
| Language | Français | Language of the whole interface. Saved on click. |
| Theme | System | System follows Windows' light or dark mode; Light and Dark fix it. Saved on click. |
| Maximum wait per page (timeout) | 20 seconds, typed in milliseconds | Longest wait for a page in simple reading; a slower page is requested again, within Retries. The time given to the built-in browser is worked out from this value. |
| Retries | 2 | New tries after a network error, a temporary refusal or a server error, with a wait that grows between two tries. |
| Pause between pages (rate limit) | 600 milliseconds | Shortest time between two requests to the same site in simple reading; the built-in browser, assisted unlock and the extension do not follow it. Longer: fewer refusals, but a slower scraping. |
| Pages in parallel (parallel workers) | 4 | Default of Pages in parallel, which can also be changed at each start. |
| Max products | 1,000 | Default of the same field in New scraping. |
| Pages read, at most (max pages) | 150 | Default of the same field in New scraping. |
| Category levels (depth) | 3 | Default of the same field in New scraping. |
| Pages in the browser, at most | 40 | Default of the same field in New scraping. |
| Specialized browser (nodriver) | off | Set from the Protection and intervention card. On, the Retry with the specialized browser button can drive your real Chrome, for free, after an analysis or at the end of a blocked or empty scraping. Needs Google Chrome, Python and nodriver. |
| License key | Empty | Needed for assisted unlock of protected sites, a paid service (pricing). Without a key, free local reading works. |
Site recipes (recipes.json) | — | Export recipes.json saves to a file the reading rules learned for each site. Import a file… first shows every change and keeps a copy of the current file; nothing changes before Apply the import. |
| Data folder | %APPDATA%\Eloshon Scraper\data | Read only. Jobs, CSV files, reports and the built-in browser's profile are kept there, never in the application folder. |
The numbers and the licence key are only kept after a click on Save. Leaving the screen, or changing the language or theme before saving, wipes the values you typed.
Site profiles: Sources
| Field or button | Default | Effect |
|---|---|---|
| Name | Empty, required | Name of the source; Use copies it into the Source field. |
| URL | Empty, required | Address copied into Site address. |
| Browser identity (User-Agent) | Empty | Identity announced to the site during a scraping, in simple reading as in the built-in browser. Empty, it is the built-in browser's, a recent Chrome, which is recommended: a robot identity gets more pages refused. Analysis, teaching, the specialized browser and the extension keep their own. |
| Default product type | Empty | Copied into the field of the same name; when empty, that field takes auto. |
| Use | — | Fills New scraping with the address, the name and the type, and gives it the source's identity, which is not shown there. |
| Delete | — | Asks for confirmation (Delete this source?) before deleting the source. Jobs already run stay. |
After Use, the source's identity is also used by the scrapings started afterwards from New scraping, without being shown there, until the next quick start from Dashboard.