# Changelog of Auto-Data.net Scraper: Car Specs by Brand and Model (`mrbridge/auto-data-scraper`) Actor

- **URL**: https://apify.com/mrbridge/auto-data-scraper/changelog.md
- **Full Actor documentation**: https://apify.com/mrbridge/auto-data-scraper.md

## Changelog

### 0.1.3 (2026-10-21)

- The full spec sheet (`rawSpecs`) of a variant is now read with a single HTML parse per page.
- Status messages are coalesced: a run no longer sends one status update per variant, so far fewer API calls are made and the "Setting status message timed out" warnings go away.
- Default memory raised to 1,024 MB after measurement, one Actor Start event unchanged.

### 0.1.2 (2026-10-21)

- Variant pages are now cached between runs in a key-value store named `auto-data-html-cache` in your Apify account, for 30 days. A later run on the same models does not fetch them again, and fewer warnings are logged.
- Inverted production years (`yearMin` greater than `yearMax`) are refused, and the input now raises a warning for options that cannot take effect.
- The weight-to-power ratio of hybrid variants is now computed on the system power.
- The default memory is now 512 MB, down from 2048 MB, because a run peaks far below it. A run also starts a single Actor Start event instead of two.
- The README documents the known limits of the columns: row key, ranges, empty `powerKw`, gross battery capacity, electric consumption, sparse old pages and production years copied as printed.
- The Apify links of the README no longer carry the `fpr` parameter.
- README: ranges wording, scrapedAt and cache notes

### 0.1.0 (2026-10-19)

First version of Auto Data Scraper, built from the auto-data.net part of Car Data Scraper.

- Reads car technical specifications from auto-data.net only, for a whole brand or a list of models, one row per engine variant across every generation.
- Flat summary columns for power, torque, 0-100 km/h, top speed, dimensions, weight, fuel economy, drivetrain, and for electric and hybrid variants system power, gross battery capacity and electric range.
- New `rawSpecs` array on every data row: the full spec sheet of the variant page, in page order.
- New `url` and `autodataId` columns. The columns `sources`, `conflicts`, `dedupeConfidence`, `sourcesDetail`, `flags`, `quarterMileS` and `peakG` of Car Data Scraper are not produced, because there is a single source. Every shared column keeps its name, type and value.
- Five dataset views: `overview`, `performance`, `detailed`, `fullSpecSheet` and `errors`.
- Pay-per-event pricing: one Car variant event per data row, error rows never charged, plus the standard Actor Start event.
- The source options of the multi-source Actor (`sources`, `extraSources`, `sourcePriority`, `conflictThresholdPct`) are ignored when sent through the API, with one warning per key.