Skip to content

Commit

Permalink
feat(event): event 정보 크롤러 구현
Browse files Browse the repository at this point in the history
  • Loading branch information
jacobhboy committed Jan 29, 2024
1 parent 3293481 commit 43e1586
Show file tree
Hide file tree
Showing 2 changed files with 108 additions and 0 deletions.
Original file line number Diff line number Diff line change
@@ -0,0 +1,11 @@
package com.sickgyun.server.event.exception;

import org.springframework.http.HttpStatus;

import com.sickgyun.server.common.exception.SickgyunException;

public class EventUrlConnectingError extends SickgyunException {
public EventUrlConnectingError() {
super(HttpStatus.INTERNAL_SERVER_ERROR, "이벤트 정보 크롤링 도중 서버 연결에 실패하였습니다.");
}
}
Original file line number Diff line number Diff line change
@@ -0,0 +1,97 @@
package com.sickgyun.server.event.scheduler;

import java.io.IOException;
import java.time.YearMonth;
import java.util.ArrayList;
import java.util.Collection;
import java.util.List;
import java.util.stream.Stream;

import org.jsoup.Jsoup;
import org.jsoup.nodes.Document;
import org.jsoup.nodes.Element;
import org.jsoup.select.Elements;
import org.springframework.beans.factory.annotation.Value;
import org.springframework.scheduling.annotation.Scheduled;
import org.springframework.stereotype.Component;
import org.springframework.transaction.annotation.Transactional;

import com.sickgyun.server.event.domain.Event;
import com.sickgyun.server.event.domain.repository.EventRepository;
import com.sickgyun.server.event.exception.EventUrlConnectingError;

import lombok.RequiredArgsConstructor;

@Component
@RequiredArgsConstructor
public class EventScheduler {

@Value("${event.url}")
private String url;
private final EventRepository eventRepository;

@Scheduled(cron = "0 0 5 1 * ?") // 매달 1일 5시
@Transactional
public void getReqruitInformation() {
YearMonth now = YearMonth.now();
YearMonth nextMonth = now.plusMonths(1);

List<Event> thisMonthEvent = getEventsByMonth(now);
List<Event> nextMonthEvent = getEventsByMonth(nextMonth);

List<Event> allEvents = Stream.of(thisMonthEvent, nextMonthEvent)
.flatMap(Collection::stream)
.toList();

eventRepository.deleteAll();
eventRepository.saveAll(allEvents);
}

private List<Event> getEventsByMonth(YearMonth date) {
Document document = connectToServer(date);
Elements rawEvents = getRawEvents(document);
return getAllEvents(rawEvents, date);
}

private static List<Event> getAllEvents(Elements rawEvents, YearMonth yearMonth) {
List<Event> events = new ArrayList<>();

for (Element rawReqruit : rawEvents) {
String imageSrc = "https://dev-event.vercel.app/" + rawReqruit.select("img").get(2).attr("src");
String name = rawReqruit.getElementsByClass("Item_item__content__title___fPQa").text();
String host = rawReqruit.getElementsByClass("Item_host__zNXMy").text();
String date = rawReqruit.getElementsByClass("Item_date__kVMJZ").text();
String hashtags = rawReqruit.getElementsByClass("Item_tags___ujeV").text()
.replace(" ", "")
.replace("#", " #")
.substring(1);

events.add(
new Event(
imageSrc,
name,
host,
date,
hashtags,
yearMonth
)
);

}
return events;
}

private static Elements getRawEvents(Document document) {
return document.getElementsByClass("Item_item__86e_I");
}

private Document connectToServer(YearMonth date) {
Document document;
try {
document = Jsoup.connect(String.format(url, date.getYear(), date.getMonth().getValue())).get();
} catch (IOException e) {
throw new EventUrlConnectingError();
}
return document;
}
}

0 comments on commit 43e1586

Please sign in to comment.