forked from D-Net/dnet-hadoop
115 lines
3.7 KiB
Java
115 lines
3.7 KiB
Java
|
|
package eu.dnetlib.dhp.bmuse.bioschema;
|
|
|
|
import static eu.dnetlib.dhp.bmuse.utils.SparkSessionSupport.runWithSparkSession;
|
|
|
|
import java.util.Objects;
|
|
import java.util.Optional;
|
|
import java.util.function.Function;
|
|
import java.util.stream.Stream;
|
|
|
|
import org.apache.commons.io.IOUtils;
|
|
import org.apache.hadoop.fs.Path;
|
|
import org.apache.hadoop.io.SequenceFile;
|
|
import org.apache.hadoop.io.Text;
|
|
import org.apache.hadoop.io.compress.GzipCodec;
|
|
import org.apache.spark.SparkConf;
|
|
import org.apache.spark.api.java.JavaSparkContext;
|
|
import org.apache.spark.sql.SparkSession;
|
|
import org.apache.spark.util.LongAccumulator;
|
|
import org.jsoup.nodes.Element;
|
|
import org.jsoup.select.Elements;
|
|
import org.slf4j.Logger;
|
|
import org.slf4j.LoggerFactory;
|
|
|
|
import eu.dnetlib.dhp.bmuse.utils.ArgumentApplicationParser;
|
|
import eu.dnetlib.dhp.bmuse.utils.BMUSEScraper;
|
|
import eu.dnetlib.dhp.bmuse.utils.FunctionalInterfaceSupport;
|
|
import eu.dnetlib.dhp.bmuse.utils.UrlParser;
|
|
|
|
public class SparkScraper {
|
|
|
|
static Logger logger = LoggerFactory.getLogger(SparkScraper.class);
|
|
|
|
public static void main(String[] args) throws Exception {
|
|
|
|
final ArgumentApplicationParser parser = new ArgumentApplicationParser(
|
|
IOUtils
|
|
.toString(
|
|
SparkScraper.class
|
|
.getResourceAsStream(
|
|
"/eu/dnetlib/dhp/bmuse/bioschema/generate_dataset.json")));
|
|
parser.parseArgument(args);
|
|
Boolean isSparkSessionManaged = Optional
|
|
.ofNullable(parser.get("isSparkSessionManaged"))
|
|
.map(Boolean::valueOf)
|
|
.orElse(Boolean.TRUE);
|
|
|
|
final String nameNode = parser.get("nameNode");
|
|
final String workingPath = parser.get("workingPath");
|
|
final String rdfOutput = parser.get("rdfOutput");
|
|
final String sitemapUrl = parser.get("sitemapUrl");
|
|
final String sitemapURLKey = parser.get("sitemapURLKey");
|
|
final String dynamic = parser.get("dynamic");
|
|
final String maxScrapedPages = parser.get("maxScrapedPages");
|
|
Boolean dynamicValue = true;
|
|
if (Objects.nonNull(dynamic)) {
|
|
dynamicValue = Boolean.parseBoolean(dynamic);
|
|
}
|
|
final boolean scrapingType = dynamicValue.booleanValue();
|
|
|
|
SparkConf conf = new SparkConf();
|
|
runWithSparkSession(
|
|
conf,
|
|
isSparkSessionManaged,
|
|
spark -> {
|
|
final LongAccumulator scraped = spark.sparkContext().longAccumulator("scraped");
|
|
final LongAccumulator errors = spark.sparkContext().longAccumulator("errors");
|
|
|
|
JavaSparkContext sc = JavaSparkContext.fromSparkContext(spark.sparkContext());
|
|
|
|
System.setProperty("webdriver.chrome.whitelistedIps", "");
|
|
|
|
BMUSEScraper scraper = new BMUSEScraper();
|
|
String url = sitemapUrl.toLowerCase();
|
|
Elements urls = UrlParser.getSitemapList(url, sitemapURLKey);
|
|
long total = urls.size();
|
|
|
|
Path output = new Path(
|
|
nameNode
|
|
.concat(workingPath)
|
|
.concat(rdfOutput));
|
|
try (SequenceFile.Writer writer = SequenceFile
|
|
.createWriter(
|
|
sc.hadoopConfiguration(),
|
|
SequenceFile.Writer.file(output),
|
|
SequenceFile.Writer.keyClass(Text.class),
|
|
SequenceFile.Writer.valueClass(Text.class),
|
|
SequenceFile.Writer.compression(SequenceFile.CompressionType.BLOCK, new GzipCodec()))) {
|
|
Stream<Element> urlStream = null;
|
|
if (Objects.nonNull(maxScrapedPages)) {
|
|
urlStream = urls.stream().limit(Long.parseLong(maxScrapedPages));
|
|
} else {
|
|
urlStream = urls.stream();
|
|
}
|
|
urlStream.forEach(u -> {
|
|
try {
|
|
final Text key = new Text(u.text());
|
|
final Text value = new Text(scraper.scrapeUrl(u.text(), scrapingType));
|
|
writer.append(key, value);
|
|
scraped.add(1l);
|
|
} catch (Exception e) {
|
|
logger.error(u.text(), e);
|
|
errors.add(1l);
|
|
}
|
|
});
|
|
}
|
|
|
|
logger
|
|
.info(
|
|
"Total pages to scrape: " + total + " Scraped: " + scraped.value() +
|
|
" Errors: " + errors.value());
|
|
});
|
|
}
|
|
}
|