feat: support Spark-compatible `string_to_map` function#20120

Open

unknowntpo wants to merge 27 commits intoapache:mainfrom

unknowntpo:feat-string-to-map-fn

Commits on Feb 6, 2026

feat: add Spark-compatible string_to_map function
unknowntpo
committed
refactor: move string_to_map from string to map module

unknowntpo
and
claude
committed
refactor: move string_to_map test from spark/string to spark/map

unknowntpo
and
claude
committed
docs: add inline comments explaining string_to_map parsing logic

unknowntpo
and
claude
committed
fix: keep utils module private in map module

unknowntpo
and
claude
committed
fix lint
unknowntpo
committed
refine test cases
unknowntpo
committed
refactor: avoid heap allocation in string_to_map pairs iteration
unknowntpo
committed
refactor: avoid heap allocation in string_to_map kv splitting
unknowntpo
committed
refactor: rename get_scalar_string to extract_delimiter_from_string_array
unknowntpo
committed
refactor: improve extract_delimiter_from_string_array

unknowntpo
and
claude
committed
test: add multi-row sqllogictests for string_to_map

unknowntpo
and
claude
committed
refactor: use current_offset and add table-driven unit tests

unknowntpo
and
claude
committed
refactor: merge multi-row test into table-driven tests

unknowntpo
and
claude
committed
test: add duplicate key test case with LAST_WIN behavior

unknowntpo
and
claude
committed
refactor: add assert for args length in string_to_map

unknowntpo
and
claude
committed
refactor: put helper at bottom
unknowntpo
committed
style: run cargo fmt

unknowntpo
and
claude
committed
rename: string_to_map -> str_to_map (file rename only)

unknowntpo
and
claude
committed
refactor: rename SparkStringToMap to SparkStrToMap

unknowntpo
and
claude
committed
feat: use MapBuilder, EXCEPTION dedup policy, per-row delimiters

unknowntpo
and
claude
committed
refactor: use as_string_array helpers and support Utf8/LargeUtf8/Utf8View

unknowntpo
and
claude
committed
test: add per-row delimiter SLT test for str_to_map

unknowntpo
and
claude
committed
refactor: use NullBuffer::union for precomputed null handling in str_to_map

unknowntpo
and
claude
committed
refactor: reuse HashSet across rows to avoid per-row heap allocation
unknowntpo
committed
refactor: move Rust tests to SLT, extract delimiter constants, match Spark error message
unknowntpo
committed
chore: remove redundant duplicate key comment from SLT header
unknowntpo
committed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat: support Spark-compatible `string_to_map` function#20120

feat: support Spark-compatible `string_to_map` function#20120
unknowntpo wants to merge 27 commits intoapache:mainfrom
unknowntpo:feat-string-to-map-fn

Commits on Feb 6, 2026

feat: add Spark-compatible string_to_map function

refactor: move string_to_map from string to map module

refactor: move string_to_map test from spark/string to spark/map

docs: add inline comments explaining string_to_map parsing logic

fix: keep utils module private in map module

fix lint

refine test cases

refactor: avoid heap allocation in string_to_map pairs iteration

refactor: avoid heap allocation in string_to_map kv splitting

refactor: rename get_scalar_string to extract_delimiter_from_string_array

refactor: improve extract_delimiter_from_string_array

test: add multi-row sqllogictests for string_to_map

refactor: use current_offset and add table-driven unit tests

refactor: merge multi-row test into table-driven tests

test: add duplicate key test case with LAST_WIN behavior

refactor: add assert for args length in string_to_map

refactor: put helper at bottom

style: run cargo fmt

rename: string_to_map -> str_to_map (file rename only)

refactor: rename SparkStringToMap to SparkStrToMap

feat: use MapBuilder, EXCEPTION dedup policy, per-row delimiters

refactor: use as_string_array helpers and support Utf8/LargeUtf8/Utf8View

test: add per-row delimiter SLT test for str_to_map

refactor: use NullBuffer::union for precomputed null handling in str_to_map

refactor: reuse HashSet across rows to avoid per-row heap allocation

refactor: move Rust tests to SLT, extract delimiter constants, match Spark error message

chore: remove redundant duplicate key comment from SLT header