Mirror upstream elastic/elasticsearch#134503 for AI review (snapshot of HEAD tree) #131
Conversation
BASE=62e53cd13c8846f0729390450a5cfbd0304e7f40 HEAD=6b2e950488ed317736d19f100fbcf58bb587fea4 Branch=main
WalkthroughAdds a new URL_ENCODE_COMPONENT function (docs, registry, capability, evaluator, tests), introduces UrlCodecUtils for RFC3986 encoding, refactors URL_ENCODE to use the utility and scratch buffers, clarifies URL_DECODE behavior (null + warning on failure), and updates/expands documentation, examples, JSON definitions, and test fixtures. Changes
Sequence Diagram(s)sequenceDiagram
autonumber
participant Client
participant ESQL as ESQL Engine
participant Fn as UrlEncode / UrlEncodeComponent
participant Util as UrlCodecUtils
participant Scratch as BreakingBytesRefBuilder
Client->>ESQL: EVAL URL_ENCODE(_)/URL_ENCODE_COMPONENT(_)
ESQL->>Fn: toEvaluator(...)
Fn->>Scratch: acquire (THREAD_LOCAL / Fixed)
Fn->>Util: urlEncode(val, scratch, plusForSpace)
Util-->>Fn: BytesRef (encoded or original)
Fn-->>ESQL: result BytesRef
ESQL-->>Client: encoded value
note over Fn,Util: plusForSpace=true for URL_ENCODE,<br/>false (%20) for URL_ENCODE_COMPONENT
sequenceDiagram
autonumber
participant Client
participant ESQL as ESQL Engine
participant Decode as UrlDecode
Client->>ESQL: EVAL URL_DECODE(_)
ESQL->>Decode: process(val)
alt Decoding succeeds
Decode-->>ESQL: decoded BytesRef
ESQL-->>Client: decoded value
else IllegalArgumentException
Decode-->>ESQL: null
ESQL-->>Client: null + Warning header
end
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60 minutes Poem
Pre-merge checks and finishing touches❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✨ Finishing touches
🧪 Generate unit tests (beta)
Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 3
📜 Review details
Configuration used: CodeRabbit UI
Review profile: ASSERTIVE
Plan: Pro
⛔ Files ignored due to path filters (4)
docs/reference/query-languages/esql/images/functions/url_encode_component.svgis excluded by!**/*.svgx-pack/plugin/esql/src/main/generated/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlDecodeEvaluator.javais excluded by!**/generated/**x-pack/plugin/esql/src/main/generated/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlEncodeComponentEvaluator.javais excluded by!**/generated/**x-pack/plugin/esql/src/main/generated/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlEncodeEvaluator.javais excluded by!**/generated/**
📒 Files selected for processing (31)
docs/reference/query-languages/esql/_snippets/functions/description/url_decode.md(1 hunks)docs/reference/query-languages/esql/_snippets/functions/description/url_encode.md(1 hunks)docs/reference/query-languages/esql/_snippets/functions/description/url_encode_component.md(1 hunks)docs/reference/query-languages/esql/_snippets/functions/examples/url_decode.md(1 hunks)docs/reference/query-languages/esql/_snippets/functions/examples/url_encode.md(1 hunks)docs/reference/query-languages/esql/_snippets/functions/examples/url_encode_component.md(1 hunks)docs/reference/query-languages/esql/_snippets/functions/layout/url_encode_component.md(1 hunks)docs/reference/query-languages/esql/_snippets/functions/parameters/url_decode.md(1 hunks)docs/reference/query-languages/esql/_snippets/functions/parameters/url_encode.md(1 hunks)docs/reference/query-languages/esql/_snippets/functions/parameters/url_encode_component.md(1 hunks)docs/reference/query-languages/esql/_snippets/functions/types/url_encode_component.md(1 hunks)docs/reference/query-languages/esql/kibana/definition/functions/url_decode.json(2 hunks)docs/reference/query-languages/esql/kibana/definition/functions/url_encode.json(2 hunks)docs/reference/query-languages/esql/kibana/definition/functions/url_encode_component.json(1 hunks)docs/reference/query-languages/esql/kibana/docs/functions/url_decode.md(1 hunks)docs/reference/query-languages/esql/kibana/docs/functions/url_encode.md(1 hunks)docs/reference/query-languages/esql/kibana/docs/functions/url_encode_component.md(1 hunks)x-pack/plugin/esql/qa/testFixtures/src/main/resources/string.csv-spec(5 hunks)x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/action/EsqlCapabilities.java(1 hunks)x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/expression/ExpressionWritables.java(2 hunks)x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/expression/function/EsqlFunctionRegistry.java(2 hunks)x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlDecode.java(2 hunks)x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlEncode.java(4 hunks)x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlEncodeComponent.java(1 hunks)x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/expression/function/scalar/util/UrlCodecUtils.java(1 hunks)x-pack/plugin/esql/src/test/java/org/elasticsearch/xpack/esql/expression/function/AbstractUrlEncodeDecodeTestCase.java(1 hunks)x-pack/plugin/esql/src/test/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlDecodeTests.java(1 hunks)x-pack/plugin/esql/src/test/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlEncodeComponentErrorTests.java(1 hunks)x-pack/plugin/esql/src/test/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlEncodeComponentSerializationTests.java(1 hunks)x-pack/plugin/esql/src/test/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlEncodeComponentTests.java(1 hunks)x-pack/plugin/esql/src/test/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlEncodeTests.java(1 hunks)
🧰 Additional context used
🧬 Code graph analysis (9)
x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/expression/function/EsqlFunctionRegistry.java (1)
x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlEncodeComponent.java (1)
UrlEncodeComponent(35-99)
x-pack/plugin/esql/src/test/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlEncodeComponentErrorTests.java (2)
x-pack/plugin/esql-core/src/main/java/org/elasticsearch/xpack/esql/core/expression/Expression.java (1)
Expression(30-214)x-pack/plugin/esql/src/test/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlEncodeComponentTests.java (1)
UrlEncodeComponentTests(21-37)
x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/expression/function/scalar/util/UrlCodecUtils.java (1)
x-pack/plugin/esql/compute/src/main/java/org/elasticsearch/compute/operator/BreakingBytesRefBuilder.java (1)
BreakingBytesRefBuilder(21-209)
x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/expression/ExpressionWritables.java (1)
x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlEncodeComponent.java (1)
UrlEncodeComponent(35-99)
x-pack/plugin/esql/src/test/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlEncodeComponentTests.java (1)
x-pack/plugin/esql/src/test/java/org/elasticsearch/xpack/esql/expression/function/AbstractUrlEncodeDecodeTestCase.java (1)
AbstractUrlEncodeDecodeTestCase(30-269)
x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlEncode.java (2)
x-pack/plugin/esql/compute/src/main/java/org/elasticsearch/compute/operator/BreakingBytesRefBuilder.java (1)
BreakingBytesRefBuilder(21-209)x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/expression/function/scalar/util/UrlCodecUtils.java (1)
UrlCodecUtils(13-116)
x-pack/plugin/esql/src/test/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlEncodeComponentSerializationTests.java (1)
x-pack/plugin/esql-core/src/main/java/org/elasticsearch/xpack/esql/core/expression/Expression.java (1)
Expression(30-214)
x-pack/plugin/esql/src/test/java/org/elasticsearch/xpack/esql/expression/function/AbstractUrlEncodeDecodeTestCase.java (1)
x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/expression/function/scalar/util/UrlCodecUtils.java (1)
UrlCodecUtils(13-116)
x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlEncodeComponent.java (3)
x-pack/plugin/esql/compute/src/main/java/org/elasticsearch/compute/operator/BreakingBytesRefBuilder.java (1)
BreakingBytesRefBuilder(21-209)x-pack/plugin/esql/compute/src/main/java/org/elasticsearch/compute/operator/EvalOperator.java (1)
EvalOperator(23-270)x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/expression/function/scalar/util/UrlCodecUtils.java (1)
UrlCodecUtils(13-116)
🪛 markdownlint-cli2 (0.18.1)
docs/reference/query-languages/esql/_snippets/functions/types/url_encode_component.md
1-1: First line in a file should be a top-level heading
(MD041, first-line-heading, first-line-h1)
3-3: Emphasis used instead of a heading
(MD036, no-emphasis-as-heading)
docs/reference/query-languages/esql/_snippets/functions/parameters/url_encode_component.md
1-1: First line in a file should be a top-level heading
(MD041, first-line-heading, first-line-h1)
3-3: Emphasis used instead of a heading
(MD036, no-emphasis-as-heading)
docs/reference/query-languages/esql/_snippets/functions/layout/url_encode_component.md
1-1: First line in a file should be a top-level heading
(MD041, first-line-heading, first-line-h1)
3-3: Headings should be surrounded by blank lines
Expected: 1; Actual: 0; Below
(MD022, blanks-around-headings)
4-4: Fenced code blocks should be surrounded by blank lines
(MD031, blanks-around-fences)
9-9: Emphasis used instead of a heading
(MD036, no-emphasis-as-heading)
docs/reference/query-languages/esql/_snippets/functions/description/url_encode_component.md
1-1: First line in a file should be a top-level heading
(MD041, first-line-heading, first-line-h1)
3-3: Emphasis used instead of a heading
(MD036, no-emphasis-as-heading)
docs/reference/query-languages/esql/_snippets/functions/examples/url_encode_component.md
1-1: First line in a file should be a top-level heading
(MD041, first-line-heading, first-line-h1)
3-3: Emphasis used instead of a heading
(MD036, no-emphasis-as-heading)
🔇 Additional comments (28)
docs/reference/query-languages/esql/_snippets/functions/parameters/url_decode.md (1)
6-6: LGTM!The parameter description update improves consistency across the documentation by adding the article "The" and properly hyphenating "URL-encoded".
docs/reference/query-languages/esql/_snippets/functions/examples/url_encode_component.md (1)
1-13: LGTM!The example correctly demonstrates URL_ENCODE_COMPONENT behavior, where all characters are percent-encoded except alphanumerics,
.,-,_, and~. The output correctly shows even protocol separators (://) and slashes are encoded, which is appropriate for component-level encoding.Note: The static analysis hints about markdown formatting are false positives for auto-generated documentation files.
x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/expression/ExpressionWritables.java (2)
44-44: LGTM!The import is correctly added for the new UrlEncodeComponent class.
235-235: LGTM!The UrlEncodeComponent.ENTRY registration follows the correct pattern and is appropriately positioned alongside the related UrlEncode entry.
docs/reference/query-languages/esql/_snippets/functions/parameters/url_encode.md (1)
6-6: LGTM!The parameter description update improves consistency by adding the article "The", matching the style used in related documentation.
docs/reference/query-languages/esql/_snippets/functions/types/url_encode_component.md (1)
1-9: LGTM!The supported types table correctly shows that both
keywordandtextinputs returnkeywordoutput, which is consistent with other string conversion functions.Note: The static analysis hints about markdown formatting are false positives for auto-generated documentation files.
x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/expression/function/EsqlFunctionRegistry.java (2)
99-99: LGTM!The import is correctly added for the new UrlEncodeComponent class.
545-545: LGTM!The UrlEncodeComponent registration in snapshot functions follows the correct pattern and is appropriately positioned alongside the related url_encode and url_decode functions.
x-pack/plugin/esql/src/test/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlEncodeTests.java (1)
29-29: LGTM!The refactoring from a boolean parameter to
PercentCodecTestType.ENCODEimproves code clarity and type safety. Using an enum makes the test parameterization more maintainable and extensible for different encoding/decoding types.docs/reference/query-languages/esql/_snippets/functions/parameters/url_encode_component.md (1)
1-7: LGTM!The parameter documentation correctly describes the string parameter for URL_ENCODE_COMPONENT and is consistent with the documentation style used for related functions.
Note: The static analysis hints about markdown formatting are false positives for auto-generated documentation files.
docs/reference/query-languages/esql/_snippets/functions/description/url_encode.md (1)
5-5: LGTM! Documentation clarifies encoding behavior.The updated description explicitly specifies which characters are not percent-encoded and clarifies that spaces are encoded as
+. This aligns with application/x-www-form-urlencoded encoding standards.docs/reference/query-languages/esql/kibana/docs/functions/url_encode_component.md (1)
1-9: LGTM! Clear documentation for RFC3986 component encoding.The documentation correctly distinguishes URL_ENCODE_COMPONENT from URL_ENCODE by using
%20for spaces instead of+. This follows RFC3986 percent-encoding for URI components, which is appropriate for encoding individual query parameters or path segments.docs/reference/query-languages/esql/_snippets/functions/layout/url_encode_component.md (1)
1-27: LGTM! Standard documentation layout for new function.This is auto-generated documentation scaffolding for the URL_ENCODE_COMPONENT function. The structure follows the established pattern for ESQL function documentation.
x-pack/plugin/esql/src/test/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlDecodeTests.java (1)
29-29: LGTM! Improved test parameterization.The change from boolean to enum-based parameterization (
PercentCodecTestType.DECODE) improves code clarity and makes the test infrastructure more extensible for different codec types.docs/reference/query-languages/esql/kibana/docs/functions/url_decode.md (1)
4-8: LGTM! Documentation accurately reflects error handling.The updated description correctly documents that URL_DECODE returns
nulland adds a warning header when decoding fails. This aligns with the implementation change in UrlDecode.java that addswarnExceptions = { IllegalArgumentException.class }to handle invalid input gracefully.x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlDecode.java (3)
47-47: LGTM! Description updated to match documented behavior.The function description now accurately reflects that URL_DECODE returns
nulland adds a warning header when decoding fails, which aligns with thewarnExceptionsannotation added below.
53-53: LGTM! Refined parameter description.The parameter description is now more concise and grammatically correct: "The URL-encoded string to decode."
86-86: LGTM! Correct error handling implementation.The
warnExceptions = { IllegalArgumentException.class }annotation correctly implements the documented behavior.URLDecoder.decode()throwsIllegalArgumentExceptionfor invalid escape sequences (e.g.,%ZZ, incomplete sequences), and this annotation will convert those exceptions into anullreturn value with a warning header in the response.docs/reference/query-languages/esql/kibana/docs/functions/url_encode.md (1)
4-7: LGTM! Consistent documentation updates.The description clarifies the URL_ENCODE encoding rules, explicitly stating which characters are not percent-encoded and that spaces are encoded as
+. The updated example is clear and demonstrates the function usage effectively.docs/reference/query-languages/esql/_snippets/functions/description/url_decode.md (1)
5-5: LGTM! Documentation consistently updated.The description update matches the changes in other documentation files and accurately describes the error handling behavior (null return with warning header on decode failure).
docs/reference/query-languages/esql/_snippets/functions/description/url_encode_component.md (1)
1-6: LGTM!The description accurately documents URL_ENCODE_COMPONENT behavior. The markdown linting issues flagged by static analysis are expected for auto-generated documentation files and can be safely ignored.
docs/reference/query-languages/esql/_snippets/functions/examples/url_decode.md (1)
6-12: LGTM!The simplified example improves clarity while maintaining accurate demonstration of URL_DECODE functionality.
x-pack/plugin/esql/src/test/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlEncodeComponentErrorTests.java (1)
22-37: LGTM!The test class follows the established pattern for error testing in ESQL functions, properly integrating with the test infrastructure and correctly handling type error messages.
docs/reference/query-languages/esql/_snippets/functions/examples/url_encode.md (1)
6-11: LGTM!The updated example effectively demonstrates URL_ENCODE behavior, showing both space encoding (as +) and percent-encoding of special characters.
x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/action/EsqlCapabilities.java (1)
1463-1466: LGTM!The capability addition follows the established pattern and is properly snapshot-gated, consistent with the existing URL_ENCODE capability.
docs/reference/query-languages/esql/kibana/definition/functions/url_encode_component.json (1)
1-37: LGTM!The function definition is well-structured with appropriate signatures, examples, and metadata. The encoding behavior is correctly documented.
x-pack/plugin/esql/src/test/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlEncodeComponentSerializationTests.java (1)
14-20: LGTM!The serialization test follows the standard pattern for ESQL scalar function serialization testing, properly extending the base class and implementing the required factory method.
x-pack/plugin/esql/src/test/java/org/elasticsearch/xpack/esql/expression/function/scalar/convert/UrlEncodeComponentTests.java (1)
21-37: LGTM!The test class properly integrates with the URL encoding/decoding test infrastructure, using the ENCODE_COMPONENT test type to ensure component-specific encoding rules are tested correctly.
| "name" : "string", | ||
| "type" : "keyword", | ||
| "optional" : false, | ||
| "description" : "The URL to encode." |
There was a problem hiding this comment.
🧹 Nitpick | 🔵 Trivial
Consider clarifying the parameter description.
The parameter description says "The URL to encode" but should be "The URL component to encode" or "The string to encode" to better reflect that this function encodes URL components (parts of a URL), not complete URLs. This distinction matters because URL_ENCODE_COMPONENT uses different encoding rules than URL_ENCODE (e.g., spaces become %20 instead of +).
Apply this diff to improve clarity:
- "description" : "The URL to encode."
+ "description" : "The URL component to encode."Note: The same change should be applied to line 25.
📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| "description" : "The URL to encode." | |
| "description" : "The URL component to encode." |
🤖 Prompt for AI Agents
In
docs/reference/query-languages/esql/kibana/definition/functions/url_encode_component.json
around lines 13 and 25, the parameter description "The URL to encode." is
misleading; change it to "The URL component to encode." or "The string to
encode." to reflect that this function encodes URL components (not full URLs)
and uses component-specific encoding rules; update both occurrences to the
clearer wording.
| | EVAL a = TRIM(URL_DECODE(first_name)) | ||
| | EVAL b = URL_DECODE(TO_LOWER(first_name)) | ||
| | EVAL c = URL_DECODE(TO_LOWER(first_name)) | ||
| | KEEP a,b; |
There was a problem hiding this comment.
🧹 Nitpick | 🔵 Trivial
Drop unused duplicate URL_DECODE evaluation.
The new EVAL c = URL_DECODE(TO_LOWER(first_name)) duplicates the previous line and c is never kept. Please drop it to avoid needless work and future confusion.
-| EVAL c = URL_DECODE(TO_LOWER(first_name))📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| | EVAL a = TRIM(URL_DECODE(first_name)) | |
| | EVAL b = URL_DECODE(TO_LOWER(first_name)) | |
| | EVAL c = URL_DECODE(TO_LOWER(first_name)) | |
| | KEEP a,b; | |
| | EVAL a = TRIM(URL_DECODE(first_name)) | |
| | EVAL b = URL_DECODE(TO_LOWER(first_name)) | |
| | KEEP a,b; |
🤖 Prompt for AI Agents
In x-pack/plugin/esql/qa/testFixtures/src/main/resources/string.csv-spec around
lines 2639 to 2642, there is a duplicate unused evaluation: remove the line
"EVAL c = URL_DECODE(TO_LOWER(first_name))" so you only perform the needed
evaluations for a and b and avoid computing an unused variable; keep the
remaining lines as-is (EVAL a = TRIM(URL_DECODE(first_name)), EVAL b =
URL_DECODE(TO_LOWER(first_name)), KEEP a,b;).
| scratch.grow(size); | ||
| scratch.clear(); | ||
|
|
||
| int lo = val.offset; | ||
| int hi = val.offset + val.length; | ||
|
|
||
| for (int i = lo; i < hi; ++i) { | ||
| byte b = val.bytes[i]; | ||
| char c = (char) (b & 0xFF); | ||
|
|
||
| if (plusForSpace && c == ' ') { | ||
| scratch.append((byte) '+'); | ||
| continue; | ||
| } | ||
|
|
||
| if (isRfc3986Safe(c)) { | ||
| scratch.append(b); | ||
| continue; | ||
| } | ||
|
|
||
| // every encoded byte is represented by 3 chars: %XY | ||
| scratch.append((byte) '%'); | ||
|
|
||
| // the X in %XY is the hex value for the high nibble | ||
| scratch.append((byte) HEX_DIGITS[(c >> 4) & 0x0F]); | ||
|
|
||
| // the Y in %XY is the hex value for the low nibble | ||
| scratch.append((byte) HEX_DIGITS[c & 0x0F]); | ||
| } |
There was a problem hiding this comment.
Fix aliasing bug when input already uses the scratch buffer.
If a caller feeds the result of a previous urlEncode call back into this method with the same scratch (e.g. re‑encoding a component), scratch.clear() zeros the shared BytesRef’s length before we capture the bounds. The loop then sees hi == lo and returns an empty string. Repro:
BreakingBytesRefBuilder scratch = new BreakingBytesRefBuilder(breaker, "test");
BytesRef once = UrlCodecUtils.urlEncode(new BytesRef("foo bar"), scratch, true);
BytesRef twice = UrlCodecUtils.urlEncode(once, scratch, true); // returns length 0Capture lo/hi before clearing (or clone the input) so aliasing stays safe.
- scratch.grow(size);
- scratch.clear();
-
- int lo = val.offset;
- int hi = val.offset + val.length;
+ scratch.grow(size);
+
+ int lo = val.offset;
+ int hi = val.offset + val.length;
+
+ scratch.clear();🤖 Prompt for AI Agents
In
x-pack/plugin/esql/src/main/java/org/elasticsearch/xpack/esql/expression/function/scalar/util/UrlCodecUtils.java
around lines 27 to 55, the method clears the shared scratch buffer before
capturing the input bounds which causes an aliasing bug when the caller passes
the same BytesRef instance that backs scratch; capture the input bounds and/or
clone the input bytes before calling scratch.clear() so the loop iterates over
the original data (e.g., compute lo and hi or copy val =
BytesRef.deepCopyOf(val) prior to scratch.clear()), then proceed with encoding
into scratch.
1 participant
Single commit with tree=6b2e950488ed317736d19f100fbcf58bb587fea4^{tree}, parent=62e53cd13c8846f0729390450a5cfbd0304e7f40. Exact snapshot of upstream PR head. No conflict resolution attempted.
Summary by CodeRabbit