Apify scraping tutorials

{
  "jobUrlsOrIds": [],
  "starterSearchQuery": "data engineer",
  "starterSearchLocation": "London",
  "maxResults": 1
}

1. 00:00.000 - Your first useful Apify scrape

   Let's turn one public LinkedIn job into a useful record. We'll run the Neuton Job Details Actor once, inspect the result, then export a checked CSV.

2. 00:10.024 - Open the Actor. Check Pricing.

   Open the Actor on Apify and read the current Pricing tab. You need an Apify account, but no LinkedIn login. Set a charge limit before starting.

3. 00:19.429 - Paste a one-row starter

   In Input, open the JSON editor. Leave job URLs empty, enter data engineer and London, and set max results to one. The Actor finds a current public listing at run time.

4. 00:31.373 - Start once. Inspect Dataset.

   Select Start once. When the run finishes, open Dataset. A succeeded run can still have zero useful rows, so check the data itself.

5. 00:40.715 - Verify the row against its source

   Check job ID and URL, then title, company, location, description, and application path against the public listing. The row on screen is illustrative. Missing salary stays null.

6. 00:52.424 - Read RUN_SUMMARY

   If the Dataset is empty, open Storage, the default key-value store, and Run Summary. Check its failed inputs and rows saved. The Log has further diagnostics.

7. 01:03.024 - Export only after review

   Once the row passes review, export Dataset as CSV. Open the file and verify job ID and source URL before sending it to a spreadsheet.

8. 01:13.176 - One input. One checked row. One export.

   That's the workflow: one bounded input, one checked row, one export. Links to the Actor and full guide are below. Review live pricing before your own run.

{
  "queries": [],
  "dois": ["10.1038/s41586-020-2649-2"],
  "maxResults": 1
}

1. 00:00.000 - Your first paper-metadata scrape

   Already have a DOI? Use it to request one source-linked bibliographic record. We'll run Neuton's Crossref Works Actor, inspect the Dataset, and export a checked CSV.

2. 00:11.176 - Use Crossref for DOI metadata

   Crossref offers publisher-deposited DOI metadata: titles, authors, journals, dates and more when available. This Actor returns metadata, not the paper PDF or full text.

3. 00:23.483 - Look up one DOI

   On Input, open JSON. Leave queries empty and enter the single DOI shown here. Max results is per search query; it is not a global charge cap.

4. 00:34.083 - Check Pricing. Start once.

   Before Start, read the live Pricing tab and set a run charge limit. Run once, then open Dataset. A succeeded run may still return no usable row.

5. 00:44.299 - Check the DOI-backed row

   Check DOI and URL first. Then verify title, authors, journal, publisher and publication date against the DOI landing page. Missing abstracts can be normal; don't invent precision.

6. 00:57.053 - If empty, open RUN_SUMMARY

   If empty, open Storage and Run Summary. It lists saved rows and failed DOI lookups separately. Read the Log for the same run before trying again.

7. 01:07.013 - Export a checked record

   Export the checked row as CSV, preserving DOI and URL. If you need citation graphs or open-access indicators, use OpenAlex separately and verify its identifiers. Metadata never replaces reading the paper.