}

		// append @id column with value for auto-increment primary key
		// the @id value is correct, when: 1. without setting auto-increment primary key, 2. database AutoIncrementIncrement = 1
		switch values := db.Statement.Dest.(type) {
		case map[string]interface{}:
			values[pkFieldName] = insertID
		case *map[string]interface{}:
			(*values)[pkFieldName] = insertID
		case []map[string]interface{}, *[]map[string]interface{}:

/// info

A "bearer" token is not the only option.

But it's the best one for our use case.

And it might be the best for most use cases, unless you are an OAuth2 expert and know exactly why there's another option that better suits your needs.

In that case, **FastAPI** also provides you with the tools to build it.

    // Test #values()

    assertMultimapRemainsUnmodified(multimap, originalEntries);
    if (!multimap.isEmpty()) {
      Collection<V> values = multimap.asMap().entrySet().iterator().next().getValue();

      assertCollectionIsUnmodifiable(values, sampleValue);
    }

    // Test #entries()
    assertCollectionIsUnmodifiable(multimap.entries(), immutableEntry(sampleKey, sampleValue));

            DfsReferralData next = nextstart;

            if ( next != null ) {
                do {
                    if ( log.isDebugEnabled() ) {
                        log.debug("Next referral is " + next);
                    }
                    if ( res == null ) {
                        res = r.combine(next);
                    }
                    else {

### Zrozum problem

* Najpierw upewnij się, że **rozumiesz problem**, który próbuje rozwiązać pull request. Może być osadzony w większym kontekście w GitHubowej dyskusji lub problemie.

* Jest też duża szansa, że pull request nie jest konieczny, ponieważ problem można rozwiązać w **inny sposób**. Wtedy możesz to zasugerować lub o to zapytać.

### Nie martw się stylem

      return false;
    }

    Object target = thatIterator.next();
    E current = thisIterator.next();
    try {
      while (true) {
        int cmp = unsafeCompare(current, target);

        if (cmp < 0) {
          if (!thisIterator.hasNext()) {
            return false;
          }
          current = thisIterator.next();
        } else if (cmp == 0) {
          if (!thatIterator.hasNext()) {

                    // move on to next in r1
                    if (i1.hasNext()) {
                        res1 = i1.next();
                    } else {
                        done = true;
                    }
                }
            } else {
                // move on to next in r2
                if (i2.hasNext()) {
                    res2 = i2.next();
                } else {

# Benchmarks

Unabhängige TechEmpower-Benchmarks zeigen, **FastAPI**-Anwendungen, die unter Uvicorn ausgeführt werden, gehören zu <a href="https://www.techempower.com/benchmarks/#section=test&runid=7464e520-0dc2-473d-bd34-dbdfd7e85911&hw=ph&test=query&l=zijzen-7" class="external-link" target="_blank">den schnellsten existierenden Python-Frameworks</a>, nur Starlette und Uvicorn selbst (intern von FastAPI verwendet) sind schneller.

import static org.junit.Assert.assertThat;

import java.io.StringReader;
import java.util.List;

import org.junit.Test;

/**
 * @author wyukawa
 *
 */
public class IndexedIteratorTest {

    /**
     * @throws Exception
     */
    @Test
    public void test() throws Exception {
        final List<String> list = newArrayList();
        list.add("aaa");
        list.add("bbb");

	}
	body := string(slurp)
	for _, ref := range localLinks(body) {
		if *verbose {
			log.Printf("  links to %s", ref)
		}
		dest := *root + ref
		linkSources[dest] = append(linkSources[dest], url)
		crawl(dest, url)
	}
	for _, id := range pageIDs(body) {
		if *verbose {
			log.Printf(" url %s has #%s", url, id)
		}
		fragExists[urlFrag{url, id}] = true
	}
	return nil