Input form layout: State and City are open above the Filters section instead of a collapsed "Place" section, and Max results per search is the last field before Filters. Same fields, same behavior.
- Search by form: type keywords (one search per line, as in Sympla's search box), pick a state and a city, a date range (a date or a span from today such as
30 days), categories, free or paid, event types and the sort (Sympla's order or soonest first). A place or a filter with no keyword runs one search over everything that matches. Pasted URLs still work beside the form.
- A city search returns only that city's events; Sympla itself lists other cities after them.
- A pasted keyword URL (
/eventos?s=...) now honours its own date (dt), category (cl), price (p) and type parameters; form filters merge into keyword URLs that lack them.
- Each event carries
search: the keyword, the filters label or the URL from the input that found it.
- The run's final status and
OUTPUT record list every search with the site's own count and what the run returned; a search bigger than Sympla's 1,000-result limit says so.
- Category, city + category and online pages follow Sympla's new collection pages (October 2026), which load their events from the search index; the old
/em-alta category URLs still work through Sympla's redirect.
startUrls is a plain list of URLs. The per-row maxEvents limit is gone: old rows are still accepted (the URL is kept, the limit ignored with a warning) and Max results per search (maxRequests) caps every search: each keyword, the filters-only search, each URL.- A run with a filter or a place but no keyword and no URL is allowed; a run with nothing at all ends with "Nothing to search".
- Run monitoring counts distinct users with a one-way hash of the Apify user ID; the ID itself is never stored. No input or output changes.
- When the built-in proxy refuses the connection, the run now fails at once with a clear message instead of retrying for minutes.
- The run no longer writes a
RUN_SUMMARY record to your key-value store or a "Run summary" line to the log.
- Built-in proxies moved to a new provider. No input or output changes.
- Rewritten from scratch: plain HTTP instead of a headless browser (faster, 256-512 MB).
- Bileto events (
bileto.sympla.com.br, most of Sympla's catalogue) are now scraped.
- New start URL forms: city and landing pages,
/eventos/gratis, keyword searches, legacy __id event links.
- Every event carries sessions and every ticket type with price, fee and availability.
- Pay-per-event pricing (
result-event-data).
- A run that finds nothing now says why. No matching events: the run succeeds with "No events found". Events Sympla changed in a way the scraper can no longer read: the run fails and the developer is notified automatically. Previously 1.x runs ended with 0 items and a generic message.
- Start URLs are checked before the run starts: an empty list or a non-Sympla link is rejected with a message instead of starting a run that can't find anything.
- Each start URL can set its own maximum number of events.
- New output schema (one
Event per item for both platforms).
- New input:
startUrls rows are {url, maxEvents}, and maxRequests ("Events per URL") is the limit for rows without their own. maxResults and proxyConfig are gone (proxies are built in).
