Skip to content

Optimized text for full unicode and some escape sequences (#129169) #7

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
MitchLewis930 wants to merge 1 commit into
base: pr_017_before
Choose a base branch
from
Open

Optimized text for full unicode and some escape sequences (#129169) #7

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
82 changes: 67 additions & 15 deletions ...t/impl/src/main/java/org/elasticsearch/xcontent/provider/json/ESUTF8StreamJsonParser.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,9 +21,14 @@

import java.io.IOException;
import java.io.InputStream;
import java.util.ArrayList;
import java.util.List;

public class ESUTF8StreamJsonParser extends UTF8StreamJsonParser {
protected int stringEnd = -1;
protected int stringLength;

private final List<Integer> backslashes = new ArrayList<>();
Copy link

Copilot AI Jan 31, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Using an ArrayList as an instance field that's cleared and reused could lead to memory retention if large strings with many backslashes are parsed. Consider using a smaller initial capacity or switching to a primitive int array with manual size tracking to reduce allocation overhead and memory footprint for the common case of few or no backslashes.

Suggested change
private final List<Integer> backslashes = new ArrayList<>();
private final List<Integer> backslashes = new ArrayList<>(4);

Copilot uses AI. Check for mistakes.

public ESUTF8StreamJsonParser(
IOContext ctxt,
Expand All @@ -43,15 +48,12 @@ public ESUTF8StreamJsonParser(
/**
* Method that will try to get underlying UTF-8 encoded bytes of the current string token.
* This is only a best-effort attempt; if there is some reason the bytes cannot be retrieved, this method will return null.
* Currently, this is only implemented for ascii-only strings that do not contain escaped characters.
*/
public Text getValueAsText() throws IOException {
if (_currToken == JsonToken.VALUE_STRING && _tokenIncomplete) {
if (stringEnd > 0) {
final int len = stringEnd - 1 - _inputPtr;
// For now, we can use `len` for `stringLength` because we only support ascii-encoded unescaped strings,
// which means each character uses exactly 1 byte.
return new Text(new XContentString.UTF8Bytes(_inputBuffer, _inputPtr, len), len);
return new Text(new XContentString.UTF8Bytes(_inputBuffer, _inputPtr, len), stringLength);
}
return _finishAndReturnText();
}
Expand All @@ -69,21 +71,71 @@ protected Text _finishAndReturnText() throws IOException {
final int[] codes = INPUT_CODES_UTF8;
final int max = _inputEnd;
final byte[] inputBuffer = _inputBuffer;
while (ptr < max) {
stringLength = 0;
backslashes.clear();

loop: while (true) {
if (ptr >= max) {
return null;
}
int c = inputBuffer[ptr] & 0xFF;
if (codes[c] != 0) {
if (c == INT_QUOTE) {
stringEnd = ptr + 1;
final int len = ptr - startPtr;
// For now, we can use `len` for `stringLength` because we only support ascii-encoded unescaped strings,
// which means each character uses exactly 1 byte.
return new Text(new XContentString.UTF8Bytes(inputBuffer, startPtr, len), len);
switch (codes[c]) {
case 0 -> {
++ptr;
++stringLength;
}
case 1 -> {
if (c == INT_QUOTE) {
// End of the string
break loop;
}
assert c == INT_BACKSLASH;
backslashes.add(ptr);
++ptr;
if (ptr >= max) {
// Backslash at end of file
return null;
}
c = inputBuffer[ptr] & 0xFF;
if (c == '"' || c == '/' || c == '\\') {
ptr += 1;
stringLength += 1;
} else {
// Any other escaped sequence requires replacing the sequence with
// a new character, which we don't support in the optimized path
return null;
}
}
case 2, 3, 4 -> {
int bytesToSkip = codes[c];
if (ptr + bytesToSkip > max) {
return null;
}
ptr += bytesToSkip;
++stringLength;
}
default -> {
return null;
}
return null;
}
++ptr;
}
return null;

stringEnd = ptr + 1;
if (backslashes.isEmpty()) {
return new Text(new XContentString.UTF8Bytes(inputBuffer, startPtr, ptr - startPtr), stringLength);
} else {
byte[] buff = new byte[ptr - startPtr - backslashes.size()];
int copyPtr = startPtr;
int destPtr = 0;
for (Integer backslash : backslashes) {
int length = backslash - copyPtr;
System.arraycopy(inputBuffer, copyPtr, buff, destPtr, length);
destPtr += length;
copyPtr = backslash + 1;
}
System.arraycopy(inputBuffer, copyPtr, buff, destPtr, ptr - copyPtr);
return new Text(new XContentString.UTF8Bytes(buff), stringLength);
}
}

@Override
Expand Down
129 changes: 101 additions & 28 deletions ...l/src/test/java/org/elasticsearch/xcontent/provider/json/ESUTF8StreamJsonParserTests.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,16 +13,14 @@
import com.fasterxml.jackson.core.JsonParser;
import com.fasterxml.jackson.core.JsonToken;

import org.elasticsearch.common.Strings;
import org.elasticsearch.core.CheckedConsumer;
import org.elasticsearch.test.ESTestCase;
import org.elasticsearch.xcontent.XContentBuilder;
import org.elasticsearch.xcontent.XContentString;
import org.elasticsearch.xcontent.json.JsonXContent;
import org.hamcrest.Matchers;

import java.io.IOException;
import java.nio.charset.StandardCharsets;
import java.util.Locale;

public class ESUTF8StreamJsonParserTests extends ESTestCase {

Expand All @@ -45,11 +43,13 @@ public void testGetValueAsText() throws IOException {
assertThat(parser.nextFieldName(), Matchers.equalTo("foo"));
assertThat(parser.nextValue(), Matchers.equalTo(JsonToken.VALUE_STRING));

var textRef = parser.getValueAsText().bytes();
assertThat(textRef, Matchers.notNullValue());
assertThat(textRef.offset(), Matchers.equalTo(9));
assertThat(textRef.offset() + textRef.length(), Matchers.equalTo(12));
assertTextRef(textRef, "bar");
var text = parser.getValueAsText();
assertThat(text, Matchers.notNullValue());

var bytes = text.bytes();
assertThat(bytes.offset(), Matchers.equalTo(9));
assertThat(bytes.offset() + bytes.length(), Matchers.equalTo(12));
assertTextRef(bytes, "bar");

assertThat(parser.getValueAsString(), Matchers.equalTo("bar"));
assertThat(parser.getValueAsText(), Matchers.nullValue());
Expand All @@ -62,17 +62,36 @@ public void testGetValueAsText() throws IOException {
assertThat(parser.nextFieldName(), Matchers.equalTo("foo"));
assertThat(parser.nextValue(), Matchers.equalTo(JsonToken.VALUE_STRING));

var text = parser.getValueAsText();
assertThat(text, Matchers.notNullValue());
assertTextRef(text.bytes(), "bar\"baz\"");
});

testParseJson("{\"foo\": \"b\\u00e5r\"}", parser -> {
assertThat(parser.nextToken(), Matchers.equalTo(JsonToken.START_OBJECT));
assertThat(parser.nextFieldName(), Matchers.equalTo("foo"));
assertThat(parser.nextValue(), Matchers.equalTo(JsonToken.VALUE_STRING));

assertThat(parser.getValueAsText(), Matchers.nullValue());
assertThat(parser.getValueAsString(), Matchers.equalTo("bar\"baz\""));
assertThat(parser.getValueAsString(), Matchers.equalTo("bår"));
});

testParseJson("{\"foo\": \"bår\"}", parser -> {
assertThat(parser.nextToken(), Matchers.equalTo(JsonToken.START_OBJECT));
assertThat(parser.nextFieldName(), Matchers.equalTo("foo"));
assertThat(parser.nextValue(), Matchers.equalTo(JsonToken.VALUE_STRING));

assertThat(parser.getValueAsText(), Matchers.nullValue());
var text = parser.getValueAsText();
assertThat(text, Matchers.notNullValue());

var bytes = text.bytes();
assertThat(bytes.offset(), Matchers.equalTo(9));
assertThat(bytes.offset() + bytes.length(), Matchers.equalTo(13));
assertTextRef(bytes, "bår");

assertThat(parser.getValueAsString(), Matchers.equalTo("bår"));

assertThat(parser.nextToken(), Matchers.equalTo(JsonToken.END_OBJECT));
});

testParseJson("{\"foo\": [\"lorem\", \"ipsum\", \"dolor\"]}", parser -> {
Expand Down Expand Up @@ -112,43 +131,97 @@ public void testGetValueAsText() throws IOException {
});
}

private boolean validForTextRef(String value) {
for (char c : value.toCharArray()) {
if (c == '"') {
return false;
private record TestInput(String input, String result, boolean supportsOptimized) {}

private static final TestInput[] ESCAPE_SEQUENCES = {
new TestInput("\\b", "\b", false),
new TestInput("\\t", "\t", false),
new TestInput("\\n", "\n", false),
new TestInput("\\f", "\f", false),
new TestInput("\\r", "\r", false),
new TestInput("\\\"", "\"", true),
new TestInput("\\/", "/", true),
new TestInput("\\\\", "\\", true) };

private int randomCodepoint(boolean includeAscii) {
while (true) {
char val = Character.toChars(randomInt(0xFFFF))[0];
Copy link

Copilot AI Jan 31, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Character.toChars() returns an array that could contain two elements for supplementary characters (codepoints > 0xFFFF), but the code only takes the first element. Since randomInt(0xFFFF) limits values to the BMP, this works, but the method name 'randomCodepoint' suggests it should support all codepoints. Either rename the method to clarify it only handles BMP characters, or extend the range and handle supplementary characters properly.

Suggested change
char val = Character.toChars(randomInt(0xFFFF))[0];
char val = (char) randomInt(0xFFFF);

Copilot uses AI. Check for mistakes.
if (val <= 0x7f && includeAscii == false) {
continue;
}
if (c == '\\') {
return false;
if (val >= Character.MIN_SURROGATE && val <= Character.MAX_SURROGATE) {
continue;
}
if ((int) c < 32 || (int) c >= 128) {
return false;
return val;
}
}

private TestInput buildRandomInput(int length) {
StringBuilder input = new StringBuilder(length);
StringBuilder result = new StringBuilder(length);
boolean forceSupportOptimized = randomBoolean();
boolean doesSupportOptimized = true;
for (int i = 0; i < length; ++i) {
if (forceSupportOptimized == false && randomBoolean()) {
switch (randomInt(9)) {
Copy link

Copilot AI Jan 31, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The magic number 9 in the switch statement is unclear. This should be documented or replaced with a named constant that explains why 10 possible values (0-9) are used, where case 0 handles escape sequences, case 1 handles unicode escapes, and default handles unicode characters. The probability distribution (1/10 for escapes, 1/10 for unicode escapes, 8/10 for unicode chars) appears intentional but is not obvious.

Copilot uses AI. Check for mistakes.
case 0 -> {
var escape = randomFrom(ESCAPE_SEQUENCES);
input.append(escape.input());
result.append(escape.result());
doesSupportOptimized = doesSupportOptimized && escape.supportsOptimized();
}
case 1 -> {
int value = randomCodepoint(true);
input.append(String.format(Locale.ENGLISH, "\\u%04x", value));
result.append(Character.toChars(value));
doesSupportOptimized = false;
}
default -> {
var value = Character.toChars(randomCodepoint(false));
input.append(value);
result.append(value);
}
}
} else {
var value = randomAlphanumericOfLength(1);
input.append(value);
result.append(value);
}
}
return true;
return new TestInput(input.toString(), result.toString(), doesSupportOptimized);
}

public void testGetValueRandomized() throws IOException {
XContentBuilder jsonBuilder = JsonXContent.contentBuilder().startObject();
StringBuilder inputBuilder = new StringBuilder();
inputBuilder.append('{');

final int numKeys = 128;
String[] keys = new String[numKeys];
String[] values = new String[numKeys];
TestInput[] inputs = new TestInput[numKeys];
for (int i = 0; i < numKeys; i++) {
String currKey = randomAlphanumericOfLength(6);
String currVal = randomUnicodeOfLengthBetween(0, 512);
jsonBuilder.field(currKey, currVal);
var currVal = buildRandomInput(randomInt(512));
inputBuilder.append('"');
inputBuilder.append(currKey);
inputBuilder.append("\":\"");
inputBuilder.append(currVal.input());
inputBuilder.append('"');
if (i < numKeys - 1) {
inputBuilder.append(',');
}
keys[i] = currKey;
values[i] = currVal;
inputs[i] = currVal;
}

jsonBuilder.endObject();
testParseJson(Strings.toString(jsonBuilder), parser -> {
inputBuilder.append('}');
testParseJson(inputBuilder.toString(), parser -> {
assertThat(parser.nextToken(), Matchers.equalTo(JsonToken.START_OBJECT));
for (int i = 0; i < numKeys; i++) {
assertThat(parser.nextFieldName(), Matchers.equalTo(keys[i]));
assertThat(parser.nextValue(), Matchers.equalTo(JsonToken.VALUE_STRING));

String currVal = values[i];
if (validForTextRef(currVal)) {
String currVal = inputs[i].result();
if (inputs[i].supportsOptimized()) {
assertTextRef(parser.getValueAsText().bytes(), currVal);
} else {
assertThat(parser.getValueAsText(), Matchers.nullValue());
Expand Down