Skip to content

README.md

Latest commit

 

History

History
52 lines (36 loc) · 1.37 KB

README.md

File metadata and controls

52 lines (36 loc) · 1.37 KB
Raw

Pinkstack Realestate

Exploring ways of quick distributed scraping with the help of Akka.

Standalone mode

The easiest way to run the scraper with stand-alone mode is to use the neat CLI interface.

# Build "fat jar" with SBT
sbt assembly

# Run it with 
java -jar target/*/scraper.jar --categories prodaja --pages 10

By default, the scraper spits out JSON.

java -jar target/*/scraper.jar --categories prodaja --pages 2 | jq -R 'fromjson?'

So to make things bit easier for your eyes your can use jq to format or restructure output further for example to CSV.

java -jar target/*/scraper.jar --categories prodaja --pages 10 \
    | jq -R 'fromjson?' \
    | jq -r "([.refNumber, .title, .price, .location.latitude, .location.longitude]) | @csv" \
    > prodaja.csv

Adjusting parallelism and other fine application.conf switches can be easily done via loading of different configuration.

java -Dconfig.resource=quick.conf -jar target/*/scraper.jar --categories najem

Some configuration options can also be adjusted via environment variables i.e.

INITIAL_CATEGORIES=prodaja,najem
CATEGORY_PAGES_LIMIT=3

Authors