- ConnectionReleased
 - CallEnd
 
## Cache directory

The cache directory must be exclusively owned by a single instance.

Deleting the cache when it is no longer needed can be done.  However this may delete the purpose of the cache
which is designed to persist between app restarts.

```kotlin
cache.delete()
```
 
## Pruning the Cache

Pruning the entire Cache to clear space temporarily can be done using evictAll.

          .url(url)
          .post(true)
          .build(),
      )
    runBatch(badProviders, names)
    println("cached first run\n****************\n")
    names = listOf("google.com", "graph.facebook.com")
    bootstrapClient =
      bootstrapClient.newBuilder()
        .cache(dnsCache)
        .build()
    dnsProviders =
      providers(
        client = bootstrapClient,
        http2Only = true,

 */

package org.jetbrains.kotlin.analysis.api.fir.scopes

import org.jetbrains.kotlin.analysis.api.fir.KtFirAnalysisSession
import org.jetbrains.kotlin.analysis.api.fir.utils.cached
import org.jetbrains.kotlin.analysis.api.lifetime.withValidityAssertion
import org.jetbrains.kotlin.analysis.api.scopes.KtScopeNameFilter
import org.jetbrains.kotlin.analysis.api.symbols.KtConstructorSymbol

Some requests will have a cached response. When this cached response isn’t fresh, OkHttp can do a _conditional GET_ to download an updated response if it’s newer than what’s cached. This requires headers like `If-Modified-Since` and `If-None-Match` to be added.

## Rewriting Responses

 * Fix: Persist proper `Content-Encoding` header to cache for GZip responses.
 * Fix: Eliminate rare race condition in SPDY streams that would prevent connection reuse.
 * Fix: Change HTTP date formats to UTC to conform to RFC2616 section 3.3.
 * Fix: Support SPDY header blocks with trailing bytes.
 * Fix: Allow `;` as separator for `Cache-Control` header.

```Python hl_lines="22-24"
{!../../../docs_src/extending_openapi/tutorial001.py!}
```

### Cache the OpenAPI schema

You can use the property `.openapi_schema` as a "cache", to store your generated schema.

That way, your application won't have to generate the schema every time a user opens your API docs.

It will be generated only once, and then the same cached schema will be used for the next requests.

```Python hl_lines="13-14  25-26"

    open fun headers(headers: Headers) = commonHeaders(headers)

    /**
     * Sets this request's `Cache-Control` header, replacing any cache control headers already
     * present. If [cacheControl] doesn't define any directives, this clears this request's
     * cache-control headers.
     */
    open fun cacheControl(cacheControl: CacheControl): Builder = commonCacheControl(cacheControl)

# Sourcing this enables Bazel remote cache (public, read-only)
# The cache configs are different for MacOS and Linux
if [[ $(uname -s) == "Darwin" ]]; then
  TFCI_BAZEL_COMMON_ARGS="$TFCI_BAZEL_COMMON_ARGS --config tf_public_macos_cache"
else
  TFCI_BAZEL_COMMON_ARGS="$TFCI_BAZEL_COMMON_ARGS --config tf_public_cache"

    a bug where a shared connection pool did not guarantee shared connections in some cases.
 *  Fix: Prefer the server's response body on all conditional cache misses. Previously we would
    return the cached response's body if it had a newer `Last-Modified` date.
 *  Fix: Update the stored timestamp on conditional cache hits.
 *  New: Optimized HTTP/2 request header encoding. More headers are HPACK-encoded and string
    literals are now Huffman-encoded.

    server!

 *  **Cache improvements.** This release fixes some severe cache problems
    including a bug where the cache could be corrupted upon certain access
    patterns. We also fixed a bug where the cache was being cleared due to a
    corrupted journal. We've added APIs to configure a request's `Cache-Control`
    headers, and to manually clear the cache.