duckdb 0.10.2-dev0.0 → 0.10.2-dev5.0
Sign up to get free protection for your applications and to get access to all the features.
- package/binding.gyp +22 -18
- package/binding.gyp.in +3 -0
- package/package.json +1 -1
- package/src/duckdb/extension/icu/icu-timezone.cpp +3 -1
- package/src/duckdb/extension/icu/icu_extension.cpp +6 -2
- package/src/duckdb/extension/json/buffered_json_reader.cpp +10 -3
- package/src/duckdb/extension/json/include/buffered_json_reader.hpp +2 -0
- package/src/duckdb/extension/json/include/json_scan.hpp +13 -7
- package/src/duckdb/extension/json/include/json_serializer.hpp +5 -4
- package/src/duckdb/extension/json/include/json_structure.hpp +3 -3
- package/src/duckdb/extension/json/json_functions/json_serialize_plan.cpp +15 -5
- package/src/duckdb/extension/json/json_functions/json_serialize_sql.cpp +15 -6
- package/src/duckdb/extension/json/json_functions/json_structure.cpp +21 -20
- package/src/duckdb/extension/json/json_functions/read_json.cpp +37 -3
- package/src/duckdb/extension/json/json_functions.cpp +7 -2
- package/src/duckdb/extension/json/json_scan.cpp +57 -33
- package/src/duckdb/extension/parquet/column_reader.cpp +12 -3
- package/src/duckdb/extension/parquet/column_writer.cpp +44 -7
- package/src/duckdb/extension/parquet/include/parquet_writer.hpp +5 -1
- package/src/duckdb/extension/parquet/parquet_extension.cpp +30 -3
- package/src/duckdb/extension/parquet/parquet_metadata.cpp +1 -1
- package/src/duckdb/extension/parquet/parquet_writer.cpp +4 -2
- package/src/duckdb/extension/parquet/zstd_file_system.cpp +1 -1
- package/src/duckdb/src/catalog/catalog.cpp +5 -1
- package/src/duckdb/src/catalog/catalog_entry/duck_schema_entry.cpp +21 -5
- package/src/duckdb/src/catalog/catalog_entry/duck_table_entry.cpp +8 -9
- package/src/duckdb/src/catalog/catalog_entry/index_catalog_entry.cpp +3 -7
- package/src/duckdb/src/catalog/catalog_entry/sequence_catalog_entry.cpp +1 -1
- package/src/duckdb/src/catalog/catalog_entry/table_catalog_entry.cpp +6 -7
- package/src/duckdb/src/catalog/catalog_entry.cpp +8 -0
- package/src/duckdb/src/catalog/catalog_search_path.cpp +5 -0
- package/src/duckdb/src/catalog/catalog_set.cpp +2 -2
- package/src/duckdb/src/catalog/default/default_functions.cpp +6 -6
- package/src/duckdb/src/catalog/default/default_schemas.cpp +1 -1
- package/src/duckdb/src/catalog/default/default_views.cpp +7 -7
- package/src/duckdb/src/catalog/dependency_catalog_set.cpp +2 -1
- package/src/duckdb/src/catalog/dependency_list.cpp +92 -8
- package/src/duckdb/src/catalog/dependency_manager.cpp +53 -68
- package/src/duckdb/src/catalog/duck_catalog.cpp +1 -1
- package/src/duckdb/src/common/adbc/adbc.cpp +287 -45
- package/src/duckdb/src/common/arrow/appender/union_data.cpp +2 -2
- package/src/duckdb/src/common/box_renderer.cpp +12 -12
- package/src/duckdb/src/common/crypto/md5.cpp +2 -1
- package/src/duckdb/src/common/enum_util.cpp +307 -1
- package/src/duckdb/src/common/enums/expression_type.cpp +4 -0
- package/src/duckdb/src/common/enums/optimizer_type.cpp +1 -1
- package/src/duckdb/src/common/file_system.cpp +60 -13
- package/src/duckdb/src/common/filename_pattern.cpp +13 -13
- package/src/duckdb/src/common/gzip_file_system.cpp +1 -1
- package/src/duckdb/src/common/http_state.cpp +1 -1
- package/src/duckdb/src/common/local_file_system.cpp +72 -71
- package/src/duckdb/src/common/multi_file_reader.cpp +48 -28
- package/src/duckdb/src/common/row_operations/row_matcher.cpp +2 -2
- package/src/duckdb/src/common/serializer/buffered_file_reader.cpp +13 -1
- package/src/duckdb/src/common/serializer/buffered_file_writer.cpp +32 -13
- package/src/duckdb/src/common/string_util.cpp +2 -3
- package/src/duckdb/src/common/tree_renderer.cpp +32 -67
- package/src/duckdb/src/common/types/bit.cpp +6 -6
- package/src/duckdb/src/common/types/data_chunk.cpp +2 -2
- package/src/duckdb/src/common/types/hash.cpp +6 -6
- package/src/duckdb/src/common/types/hyperloglog.cpp +2 -0
- package/src/duckdb/src/common/types/row/tuple_data_allocator.cpp +13 -0
- package/src/duckdb/src/common/types/row/tuple_data_layout.cpp +5 -7
- package/src/duckdb/src/common/types/uuid.cpp +1 -1
- package/src/duckdb/src/common/types/vector.cpp +22 -14
- package/src/duckdb/src/common/types.cpp +8 -1
- package/src/duckdb/src/common/vector_operations/comparison_operators.cpp +20 -18
- package/src/duckdb/src/common/vector_operations/generators.cpp +1 -1
- package/src/duckdb/src/common/vector_operations/is_distinct_from.cpp +267 -110
- package/src/duckdb/src/common/vector_operations/vector_hash.cpp +52 -23
- package/src/duckdb/src/common/virtual_file_system.cpp +33 -20
- package/src/duckdb/src/core_functions/aggregate/algebraic/avg.cpp +2 -2
- package/src/duckdb/src/core_functions/aggregate/distributive/minmax.cpp +3 -3
- package/src/duckdb/src/core_functions/aggregate/distributive/sum.cpp +31 -16
- package/src/duckdb/src/core_functions/aggregate/holistic/quantile.cpp +3 -0
- package/src/duckdb/src/core_functions/aggregate/nested/list.cpp +2 -0
- package/src/duckdb/src/core_functions/core_functions.cpp +1 -1
- package/src/duckdb/src/core_functions/function_list.cpp +2 -2
- package/src/duckdb/src/core_functions/scalar/date/time_bucket.cpp +1 -1
- package/src/duckdb/src/core_functions/scalar/generic/system_functions.cpp +46 -17
- package/src/duckdb/src/core_functions/scalar/list/array_slice.cpp +1 -1
- package/src/duckdb/src/core_functions/scalar/list/flatten.cpp +82 -45
- package/src/duckdb/src/core_functions/scalar/list/list_aggregates.cpp +2 -2
- package/src/duckdb/src/core_functions/scalar/list/list_sort.cpp +3 -0
- package/src/duckdb/src/core_functions/scalar/math/numeric.cpp +3 -2
- package/src/duckdb/src/core_functions/scalar/string/hex.cpp +2 -4
- package/src/duckdb/src/core_functions/scalar/string/repeat.cpp +12 -21
- package/src/duckdb/src/execution/column_binding_resolver.cpp +2 -10
- package/src/duckdb/src/execution/expression_executor/execute_comparison.cpp +133 -66
- package/src/duckdb/src/execution/expression_executor/execute_function.cpp +0 -2
- package/src/duckdb/src/execution/expression_executor.cpp +0 -4
- package/src/duckdb/src/execution/expression_executor_state.cpp +1 -1
- package/src/duckdb/src/execution/index/art/art.cpp +2 -2
- package/src/duckdb/src/execution/index/unknown_index.cpp +13 -13
- package/src/duckdb/src/execution/join_hashtable.cpp +1 -1
- package/src/duckdb/src/execution/nested_loop_join/nested_loop_join_mark.cpp +0 -1
- package/src/duckdb/src/execution/operator/csv_scanner/buffer_manager/csv_buffer.cpp +10 -7
- package/src/duckdb/src/execution/operator/csv_scanner/buffer_manager/csv_buffer_manager.cpp +32 -1
- package/src/duckdb/src/execution/operator/csv_scanner/buffer_manager/csv_file_handle.cpp +16 -2
- package/src/duckdb/src/execution/operator/csv_scanner/scanner/scanner_boundary.cpp +7 -7
- package/src/duckdb/src/execution/operator/csv_scanner/scanner/string_value_scanner.cpp +354 -159
- package/src/duckdb/src/execution/operator/csv_scanner/sniffer/csv_sniffer.cpp +11 -2
- package/src/duckdb/src/execution/operator/csv_scanner/sniffer/dialect_detection.cpp +22 -7
- package/src/duckdb/src/execution/operator/csv_scanner/sniffer/type_detection.cpp +1 -1
- package/src/duckdb/src/execution/operator/csv_scanner/state_machine/csv_state_machine_cache.cpp +10 -10
- package/src/duckdb/src/execution/operator/csv_scanner/table_function/csv_file_scanner.cpp +9 -18
- package/src/duckdb/src/execution/operator/csv_scanner/table_function/global_csv_state.cpp +180 -47
- package/src/duckdb/src/execution/operator/csv_scanner/util/csv_error.cpp +100 -58
- package/src/duckdb/src/execution/operator/csv_scanner/util/csv_reader_options.cpp +88 -21
- package/src/duckdb/src/execution/operator/helper/physical_vacuum.cpp +12 -13
- package/src/duckdb/src/execution/operator/join/physical_iejoin.cpp +2 -0
- package/src/duckdb/src/execution/operator/persistent/csv_rejects_table.cpp +118 -23
- package/src/duckdb/src/execution/operator/persistent/physical_batch_copy_to_file.cpp +2 -2
- package/src/duckdb/src/execution/operator/persistent/physical_batch_insert.cpp +14 -5
- package/src/duckdb/src/execution/operator/persistent/physical_copy_database.cpp +1 -1
- package/src/duckdb/src/execution/operator/persistent/physical_export.cpp +5 -5
- package/src/duckdb/src/execution/operator/schema/physical_attach.cpp +1 -1
- package/src/duckdb/src/execution/physical_plan/plan_simple.cpp +0 -9
- package/src/duckdb/src/execution/physical_plan/plan_vacuum.cpp +18 -0
- package/src/duckdb/src/execution/physical_plan_generator.cpp +5 -3
- package/src/duckdb/src/execution/radix_partitioned_hashtable.cpp +81 -106
- package/src/duckdb/src/execution/reservoir_sample.cpp +1 -1
- package/src/duckdb/src/execution/window_executor.cpp +48 -28
- package/src/duckdb/src/execution/window_segment_tree.cpp +20 -23
- package/src/duckdb/src/function/aggregate/sorted_aggregate_function.cpp +1 -1
- package/src/duckdb/src/function/cast/enum_casts.cpp +20 -55
- package/src/duckdb/src/function/cast/vector_cast_helpers.cpp +10 -9
- package/src/duckdb/src/function/cast_rules.cpp +9 -1
- package/src/duckdb/src/function/compression_config.cpp +1 -1
- package/src/duckdb/src/function/function_binder.cpp +45 -44
- package/src/duckdb/src/function/function_set.cpp +9 -9
- package/src/duckdb/src/function/pragma/pragma_queries.cpp +1 -2
- package/src/duckdb/src/function/scalar/compressed_materialization/compress_integral.cpp +21 -5
- package/src/duckdb/src/function/scalar/compressed_materialization/compress_string.cpp +2 -2
- package/src/duckdb/src/function/scalar/list/list_select.cpp +5 -2
- package/src/duckdb/src/function/scalar/list/list_zip.cpp +5 -4
- package/src/duckdb/src/function/scalar/operators/arithmetic.cpp +60 -32
- package/src/duckdb/src/function/scalar/sequence/nextval.cpp +1 -1
- package/src/duckdb/src/function/scalar/strftime_format.cpp +31 -25
- package/src/duckdb/src/function/scalar/string/caseconvert.cpp +6 -6
- package/src/duckdb/src/function/scalar/string/length.cpp +23 -2
- package/src/duckdb/src/function/scalar/string/like.cpp +1 -1
- package/src/duckdb/src/function/scalar/string/regexp/regexp_extract_all.cpp +1 -1
- package/src/duckdb/src/function/scalar/system/aggregate_export.cpp +3 -3
- package/src/duckdb/src/function/table/arrow/arrow_array_scan_state.cpp +1 -1
- package/src/duckdb/src/function/table/arrow.cpp +7 -1
- package/src/duckdb/src/function/table/copy_csv.cpp +17 -13
- package/src/duckdb/src/function/table/read_csv.cpp +52 -39
- package/src/duckdb/src/function/table/sniff_csv.cpp +7 -13
- package/src/duckdb/src/function/table/system/duckdb_constraints.cpp +1 -1
- package/src/duckdb/src/function/table/system/duckdb_databases.cpp +7 -1
- package/src/duckdb/src/function/table/system/duckdb_extensions.cpp +12 -2
- package/src/duckdb/src/function/table/system/duckdb_sequences.cpp +1 -2
- package/src/duckdb/src/function/table/system/duckdb_tables.cpp +2 -2
- package/src/duckdb/src/function/table/system/pragma_metadata_info.cpp +9 -2
- package/src/duckdb/src/function/table/system/pragma_table_info.cpp +10 -6
- package/src/duckdb/src/function/table/table_scan.cpp +1 -4
- package/src/duckdb/src/function/table/version/pragma_version.cpp +3 -3
- package/src/duckdb/src/include/duckdb/catalog/catalog.hpp +6 -2
- package/src/duckdb/src/include/duckdb/catalog/catalog_entry/duck_index_entry.hpp +1 -1
- package/src/duckdb/src/include/duckdb/catalog/catalog_entry/duck_schema_entry.hpp +1 -1
- package/src/duckdb/src/include/duckdb/catalog/catalog_entry/sequence_catalog_entry.hpp +1 -1
- package/src/duckdb/src/include/duckdb/catalog/catalog_entry.hpp +6 -1
- package/src/duckdb/src/include/duckdb/catalog/catalog_set.hpp +3 -3
- package/src/duckdb/src/include/duckdb/catalog/default/default_functions.hpp +3 -3
- package/src/duckdb/src/include/duckdb/catalog/dependency.hpp +26 -4
- package/src/duckdb/src/include/duckdb/catalog/dependency_list.hpp +39 -6
- package/src/duckdb/src/include/duckdb/catalog/dependency_manager.hpp +19 -14
- package/src/duckdb/src/include/duckdb/catalog/duck_catalog.hpp +1 -1
- package/src/duckdb/src/include/duckdb/catalog/standard_entry.hpp +4 -0
- package/src/duckdb/src/include/duckdb/common/allocator.hpp +3 -3
- package/src/duckdb/src/include/duckdb/common/arrow/arrow_wrapper.hpp +1 -1
- package/src/duckdb/src/include/duckdb/common/bit_utils.hpp +1 -1
- package/src/duckdb/src/include/duckdb/common/box_renderer.hpp +16 -16
- package/src/duckdb/src/include/duckdb/common/crypto/md5.hpp +0 -1
- package/src/duckdb/src/include/duckdb/common/enum_util.hpp +32 -0
- package/src/duckdb/src/include/duckdb/common/enums/expression_type.hpp +4 -2
- package/src/duckdb/src/include/duckdb/common/exception/binder_exception.hpp +14 -10
- package/src/duckdb/src/include/duckdb/common/exception/catalog_exception.hpp +4 -4
- package/src/duckdb/src/include/duckdb/common/exception/conversion_exception.hpp +6 -6
- package/src/duckdb/src/include/duckdb/common/exception/http_exception.hpp +3 -3
- package/src/duckdb/src/include/duckdb/common/exception/parser_exception.hpp +4 -4
- package/src/duckdb/src/include/duckdb/common/exception/transaction_exception.hpp +2 -2
- package/src/duckdb/src/include/duckdb/common/exception.hpp +57 -58
- package/src/duckdb/src/include/duckdb/common/exception_format_value.hpp +2 -2
- package/src/duckdb/src/include/duckdb/common/extra_type_info.hpp +2 -2
- package/src/duckdb/src/include/duckdb/common/file_open_flags.hpp +134 -0
- package/src/duckdb/src/include/duckdb/common/file_opener.hpp +9 -6
- package/src/duckdb/src/include/duckdb/common/file_system.hpp +35 -36
- package/src/duckdb/src/include/duckdb/common/filename_pattern.hpp +4 -6
- package/src/duckdb/src/include/duckdb/common/fixed_size_map.hpp +13 -13
- package/src/duckdb/src/include/duckdb/common/helper.hpp +42 -47
- package/src/duckdb/src/include/duckdb/common/http_state.hpp +1 -1
- package/src/duckdb/src/include/duckdb/common/hugeint.hpp +2 -2
- package/src/duckdb/src/include/duckdb/common/index_vector.hpp +10 -10
- package/src/duckdb/src/include/duckdb/common/local_file_system.hpp +13 -10
- package/src/duckdb/src/include/duckdb/common/memory_safety.hpp +3 -3
- package/src/duckdb/src/include/duckdb/common/multi_file_reader_options.hpp +1 -1
- package/src/duckdb/src/include/duckdb/common/opener_file_system.hpp +50 -24
- package/src/duckdb/src/include/duckdb/common/operator/abs.hpp +12 -4
- package/src/duckdb/src/include/duckdb/common/optional_ptr.hpp +5 -5
- package/src/duckdb/src/include/duckdb/common/pipe_file_system.hpp +1 -1
- package/src/duckdb/src/include/duckdb/common/platform.h +1 -1
- package/src/duckdb/src/include/duckdb/common/printer.hpp +5 -5
- package/src/duckdb/src/include/duckdb/common/profiler.hpp +2 -2
- package/src/duckdb/src/include/duckdb/common/progress_bar/display/terminal_progress_bar_display.hpp +5 -5
- package/src/duckdb/src/include/duckdb/common/random_engine.hpp +1 -1
- package/src/duckdb/src/include/duckdb/common/re2_regex.hpp +7 -7
- package/src/duckdb/src/include/duckdb/common/serializer/buffered_file_reader.hpp +3 -0
- package/src/duckdb/src/include/duckdb/common/serializer/buffered_file_writer.hpp +2 -2
- package/src/duckdb/src/include/duckdb/common/serializer/deserialization_data.hpp +17 -6
- package/src/duckdb/src/include/duckdb/common/serializer/deserializer.hpp +12 -0
- package/src/duckdb/src/include/duckdb/common/serializer/serialization_traits.hpp +14 -0
- package/src/duckdb/src/include/duckdb/common/serializer/serializer.hpp +16 -0
- package/src/duckdb/src/include/duckdb/common/sort/duckdb_pdqsort.hpp +3 -0
- package/src/duckdb/src/include/duckdb/common/sort/partition_state.hpp +2 -0
- package/src/duckdb/src/include/duckdb/common/string_util.hpp +4 -4
- package/src/duckdb/src/include/duckdb/common/tree_renderer.hpp +24 -25
- package/src/duckdb/src/include/duckdb/common/typedefs.hpp +5 -5
- package/src/duckdb/src/include/duckdb/common/types/cast_helpers.hpp +1 -2
- package/src/duckdb/src/include/duckdb/common/types/column/column_data_collection.hpp +10 -10
- package/src/duckdb/src/include/duckdb/common/types/column/column_data_collection_iterators.hpp +5 -5
- package/src/duckdb/src/include/duckdb/common/types/column/partitioned_column_data.hpp +1 -1
- package/src/duckdb/src/include/duckdb/common/types/constraint_conflict_info.hpp +1 -2
- package/src/duckdb/src/include/duckdb/common/types/data_chunk.hpp +2 -2
- package/src/duckdb/src/include/duckdb/common/types/date.hpp +1 -1
- package/src/duckdb/src/include/duckdb/common/types/hash.hpp +5 -5
- package/src/duckdb/src/include/duckdb/common/types/hyperloglog.hpp +1 -1
- package/src/duckdb/src/include/duckdb/common/types/interval.hpp +4 -4
- package/src/duckdb/src/include/duckdb/common/types/row/partitioned_tuple_data.hpp +1 -1
- package/src/duckdb/src/include/duckdb/common/types/row/tuple_data_layout.hpp +10 -3
- package/src/duckdb/src/include/duckdb/common/types/row/tuple_data_segment.hpp +1 -1
- package/src/duckdb/src/include/duckdb/common/types/selection_vector.hpp +12 -12
- package/src/duckdb/src/include/duckdb/common/types/string_heap.hpp +1 -1
- package/src/duckdb/src/include/duckdb/common/types/string_type.hpp +21 -20
- package/src/duckdb/src/include/duckdb/common/types/time.hpp +2 -2
- package/src/duckdb/src/include/duckdb/common/types/timestamp.hpp +3 -3
- package/src/duckdb/src/include/duckdb/common/types/uuid.hpp +2 -2
- package/src/duckdb/src/include/duckdb/common/types/value.hpp +2 -2
- package/src/duckdb/src/include/duckdb/common/types/vector.hpp +15 -7
- package/src/duckdb/src/include/duckdb/common/types/vector_buffer.hpp +1 -1
- package/src/duckdb/src/include/duckdb/common/types.hpp +14 -10
- package/src/duckdb/src/include/duckdb/common/uhugeint.hpp +2 -2
- package/src/duckdb/src/include/duckdb/common/union_by_name.hpp +1 -1
- package/src/duckdb/src/include/duckdb/common/unique_ptr.hpp +15 -14
- package/src/duckdb/src/include/duckdb/common/vector.hpp +21 -21
- package/src/duckdb/src/include/duckdb/common/vector_operations/aggregate_executor.hpp +2 -0
- package/src/duckdb/src/include/duckdb/common/vector_operations/general_cast.hpp +2 -1
- package/src/duckdb/src/include/duckdb/common/vector_operations/generic_executor.hpp +2 -3
- package/src/duckdb/src/include/duckdb/common/vector_operations/unary_executor.hpp +3 -2
- package/src/duckdb/src/include/duckdb/common/vector_operations/vector_operations.hpp +50 -32
- package/src/duckdb/src/include/duckdb/common/virtual_file_system.hpp +10 -11
- package/src/duckdb/src/include/duckdb/core_functions/aggregate/distributive_functions.hpp +1 -1
- package/src/duckdb/src/include/duckdb/core_functions/aggregate/sum_helpers.hpp +13 -1
- package/src/duckdb/src/include/duckdb/core_functions/function_list.hpp +1 -1
- package/src/duckdb/src/include/duckdb/core_functions/lambda_functions.hpp +2 -1
- package/src/duckdb/src/include/duckdb/execution/aggregate_hashtable.hpp +1 -1
- package/src/duckdb/src/include/duckdb/execution/column_binding_resolver.hpp +1 -1
- package/src/duckdb/src/include/duckdb/execution/expression_executor.hpp +1 -1
- package/src/duckdb/src/include/duckdb/execution/expression_executor_state.hpp +2 -5
- package/src/duckdb/src/include/duckdb/execution/merge_sort_tree.hpp +3 -3
- package/src/duckdb/src/include/duckdb/execution/operator/aggregate/aggregate_object.hpp +2 -2
- package/src/duckdb/src/include/duckdb/execution/operator/aggregate/distinct_aggregate_data.hpp +1 -1
- package/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/base_scanner.hpp +1 -1
- package/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/column_count_scanner.hpp +0 -3
- package/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/csv_buffer.hpp +9 -6
- package/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/csv_buffer_manager.hpp +3 -0
- package/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/csv_error.hpp +40 -22
- package/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/csv_file_handle.hpp +5 -0
- package/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/csv_file_scanner.hpp +1 -0
- package/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/csv_option.hpp +6 -4
- package/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/csv_reader_options.hpp +24 -10
- package/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/csv_state_machine.hpp +4 -0
- package/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/global_csv_state.hpp +3 -1
- package/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/skip_scanner.hpp +0 -3
- package/src/duckdb/src/include/duckdb/execution/operator/csv_scanner/string_value_scanner.hpp +58 -17
- package/src/duckdb/src/include/duckdb/execution/operator/helper/physical_batch_collector.hpp +1 -1
- package/src/duckdb/src/include/duckdb/execution/operator/helper/physical_explain_analyze.hpp +1 -1
- package/src/duckdb/src/include/duckdb/execution/operator/helper/physical_prepare.hpp +3 -3
- package/src/duckdb/src/include/duckdb/execution/operator/helper/physical_set.hpp +1 -1
- package/src/duckdb/src/include/duckdb/execution/operator/helper/physical_vacuum.hpp +4 -1
- package/src/duckdb/src/include/duckdb/execution/operator/persistent/csv_rejects_table.hpp +16 -5
- package/src/duckdb/src/include/duckdb/execution/physical_operator_states.hpp +6 -6
- package/src/duckdb/src/include/duckdb/execution/physical_plan_generator.hpp +2 -1
- package/src/duckdb/src/include/duckdb/execution/window_executor.hpp +1 -1
- package/src/duckdb/src/include/duckdb/execution/window_segment_tree.hpp +1 -1
- package/src/duckdb/src/include/duckdb/function/aggregate_function.hpp +1 -1
- package/src/duckdb/src/include/duckdb/function/cast/cast_function_set.hpp +5 -5
- package/src/duckdb/src/include/duckdb/function/cast/default_casts.hpp +4 -4
- package/src/duckdb/src/include/duckdb/function/cast/vector_cast_helpers.hpp +6 -6
- package/src/duckdb/src/include/duckdb/function/compression_function.hpp +4 -4
- package/src/duckdb/src/include/duckdb/function/copy_function.hpp +10 -13
- package/src/duckdb/src/include/duckdb/function/function.hpp +3 -3
- package/src/duckdb/src/include/duckdb/function/function_binder.hpp +24 -23
- package/src/duckdb/src/include/duckdb/function/function_serialization.hpp +1 -1
- package/src/duckdb/src/include/duckdb/function/function_set.hpp +1 -1
- package/src/duckdb/src/include/duckdb/function/pragma_function.hpp +1 -1
- package/src/duckdb/src/include/duckdb/function/scalar/regexp.hpp +3 -3
- package/src/duckdb/src/include/duckdb/function/scalar/strftime_format.hpp +7 -7
- package/src/duckdb/src/include/duckdb/function/scalar/string_functions.hpp +3 -3
- package/src/duckdb/src/include/duckdb/function/scalar_function.hpp +6 -6
- package/src/duckdb/src/include/duckdb/function/table/arrow/arrow_duck_schema.hpp +4 -4
- package/src/duckdb/src/include/duckdb/function/table/arrow.hpp +2 -2
- package/src/duckdb/src/include/duckdb/function/table/read_csv.hpp +1 -3
- package/src/duckdb/src/include/duckdb/function/table/table_scan.hpp +1 -1
- package/src/duckdb/src/include/duckdb/function/table_function.hpp +12 -11
- package/src/duckdb/src/include/duckdb/function/udf_function.hpp +66 -60
- package/src/duckdb/src/include/duckdb/main/appender.hpp +6 -6
- package/src/duckdb/src/include/duckdb/main/attached_database.hpp +3 -1
- package/src/duckdb/src/include/duckdb/main/buffered_data/buffered_data.hpp +2 -2
- package/src/duckdb/src/include/duckdb/main/buffered_data/simple_buffered_data.hpp +1 -1
- package/src/duckdb/src/include/duckdb/main/capi/capi_internal.hpp +2 -2
- package/src/duckdb/src/include/duckdb/main/chunk_scan_state/query_result.hpp +2 -2
- package/src/duckdb/src/include/duckdb/main/client_context.hpp +1 -1
- package/src/duckdb/src/include/duckdb/main/client_context_file_opener.hpp +3 -2
- package/src/duckdb/src/include/duckdb/main/client_data.hpp +1 -0
- package/src/duckdb/src/include/duckdb/main/client_properties.hpp +2 -1
- package/src/duckdb/src/include/duckdb/main/config.hpp +14 -5
- package/src/duckdb/src/include/duckdb/main/connection.hpp +27 -26
- package/src/duckdb/src/include/duckdb/main/database.hpp +19 -3
- package/src/duckdb/src/include/duckdb/main/database_file_opener.hpp +58 -0
- package/src/duckdb/src/include/duckdb/main/database_path_and_type.hpp +2 -2
- package/src/duckdb/src/include/duckdb/main/error_manager.hpp +6 -6
- package/src/duckdb/src/include/duckdb/main/extension_entries.hpp +8 -0
- package/src/duckdb/src/include/duckdb/main/extension_helper.hpp +1 -0
- package/src/duckdb/src/include/duckdb/main/external_dependencies.hpp +2 -1
- package/src/duckdb/src/include/duckdb/main/prepared_statement.hpp +8 -8
- package/src/duckdb/src/include/duckdb/main/query_profiler.hpp +2 -59
- package/src/duckdb/src/include/duckdb/main/query_result.hpp +3 -3
- package/src/duckdb/src/include/duckdb/main/relation/query_relation.hpp +1 -1
- package/src/duckdb/src/include/duckdb/main/relation.hpp +1 -1
- package/src/duckdb/src/include/duckdb/main/secret/secret.hpp +11 -7
- package/src/duckdb/src/include/duckdb/main/secret/secret_manager.hpp +2 -2
- package/src/duckdb/src/include/duckdb/main/secret/secret_storage.hpp +4 -4
- package/src/duckdb/src/include/duckdb/main/settings.hpp +78 -70
- package/src/duckdb/src/include/duckdb/optimizer/column_lifetime_analyzer.hpp +45 -0
- package/src/duckdb/src/include/duckdb/optimizer/compressed_materialization.hpp +11 -13
- package/src/duckdb/src/include/duckdb/optimizer/join_order/cardinality_estimator.hpp +2 -2
- package/src/duckdb/src/include/duckdb/optimizer/join_order/cost_model.hpp +1 -1
- package/src/duckdb/src/include/duckdb/optimizer/join_order/join_node.hpp +1 -1
- package/src/duckdb/src/include/duckdb/optimizer/join_order/query_graph.hpp +1 -1
- package/src/duckdb/src/include/duckdb/optimizer/join_order/query_graph_manager.hpp +1 -1
- package/src/duckdb/src/include/duckdb/optimizer/join_order/relation_manager.hpp +1 -1
- package/src/duckdb/src/include/duckdb/optimizer/matcher/expression_matcher.hpp +7 -7
- package/src/duckdb/src/include/duckdb/optimizer/matcher/function_matcher.hpp +7 -7
- package/src/duckdb/src/include/duckdb/optimizer/matcher/type_matcher.hpp +1 -1
- package/src/duckdb/src/include/duckdb/optimizer/optimizer.hpp +3 -0
- package/src/duckdb/src/include/duckdb/optimizer/statistics_propagator.hpp +32 -30
- package/src/duckdb/src/include/duckdb/optimizer/unnest_rewriter.hpp +1 -1
- package/src/duckdb/src/include/duckdb/parallel/event.hpp +1 -1
- package/src/duckdb/src/include/duckdb/parallel/executor_task.hpp +1 -1
- package/src/duckdb/src/include/duckdb/parallel/interrupt.hpp +2 -2
- package/src/duckdb/src/include/duckdb/parallel/pipeline_event.hpp +1 -1
- package/src/duckdb/src/include/duckdb/parser/column_list.hpp +4 -4
- package/src/duckdb/src/include/duckdb/parser/constraints/unique_constraint.hpp +49 -8
- package/src/duckdb/src/include/duckdb/parser/expression/bound_expression.hpp +1 -1
- package/src/duckdb/src/include/duckdb/parser/expression/positional_reference_expression.hpp +1 -1
- package/src/duckdb/src/include/duckdb/parser/expression/star_expression.hpp +1 -1
- package/src/duckdb/src/include/duckdb/parser/parsed_data/alter_scalar_function_info.hpp +1 -1
- package/src/duckdb/src/include/duckdb/parser/parsed_data/alter_table_function_info.hpp +1 -1
- package/src/duckdb/src/include/duckdb/parser/parsed_data/alter_table_info.hpp +1 -1
- package/src/duckdb/src/include/duckdb/parser/parsed_data/copy_database_info.hpp +40 -0
- package/src/duckdb/src/include/duckdb/parser/parsed_data/create_function_info.hpp +2 -1
- package/src/duckdb/src/include/duckdb/parser/parsed_data/create_index_info.hpp +1 -0
- package/src/duckdb/src/include/duckdb/parser/parsed_data/create_info.hpp +2 -1
- package/src/duckdb/src/include/duckdb/parser/parsed_data/create_macro_info.hpp +1 -1
- package/src/duckdb/src/include/duckdb/parser/parsed_data/create_pragma_function_info.hpp +1 -1
- package/src/duckdb/src/include/duckdb/parser/parsed_data/create_secret_info.hpp +2 -2
- package/src/duckdb/src/include/duckdb/parser/parsed_data/extra_drop_info.hpp +3 -3
- package/src/duckdb/src/include/duckdb/parser/parsed_data/parse_info.hpp +3 -2
- package/src/duckdb/src/include/duckdb/parser/parsed_data/vacuum_info.hpp +2 -5
- package/src/duckdb/src/include/duckdb/parser/parser.hpp +1 -1
- package/src/duckdb/src/include/duckdb/parser/parser_extension.hpp +3 -3
- package/src/duckdb/src/include/duckdb/parser/query_error_context.hpp +1 -1
- package/src/duckdb/src/include/duckdb/parser/transformer.hpp +3 -5
- package/src/duckdb/src/include/duckdb/planner/bind_context.hpp +2 -2
- package/src/duckdb/src/include/duckdb/planner/binder.hpp +16 -9
- package/src/duckdb/src/include/duckdb/planner/expression/bound_case_expression.hpp +1 -1
- package/src/duckdb/src/include/duckdb/planner/expression/bound_cast_expression.hpp +1 -1
- package/src/duckdb/src/include/duckdb/planner/expression/bound_default_expression.hpp +1 -1
- package/src/duckdb/src/include/duckdb/planner/expression/bound_expanded_expression.hpp +34 -0
- package/src/duckdb/src/include/duckdb/planner/expression/bound_subquery_expression.hpp +1 -1
- package/src/duckdb/src/include/duckdb/planner/expression.hpp +1 -1
- package/src/duckdb/src/include/duckdb/planner/expression_binder/base_select_binder.hpp +5 -8
- package/src/duckdb/src/include/duckdb/planner/expression_binder/column_alias_binder.hpp +3 -4
- package/src/duckdb/src/include/duckdb/planner/expression_binder/group_binder.hpp +3 -2
- package/src/duckdb/src/include/duckdb/planner/expression_binder/having_binder.hpp +3 -5
- package/src/duckdb/src/include/duckdb/planner/expression_binder/order_binder.hpp +6 -14
- package/src/duckdb/src/include/duckdb/planner/expression_binder/qualify_binder.hpp +3 -6
- package/src/duckdb/src/include/duckdb/planner/expression_binder/select_bind_state.hpp +52 -0
- package/src/duckdb/src/include/duckdb/planner/expression_binder/select_binder.hpp +4 -10
- package/src/duckdb/src/include/duckdb/planner/expression_binder/table_function_binder.hpp +5 -2
- package/src/duckdb/src/include/duckdb/planner/expression_binder.hpp +1 -0
- package/src/duckdb/src/include/duckdb/planner/expression_iterator.hpp +2 -0
- package/src/duckdb/src/include/duckdb/planner/filter/conjunction_filter.hpp +3 -6
- package/src/duckdb/src/include/duckdb/planner/logical_operator.hpp +2 -0
- package/src/duckdb/src/include/duckdb/planner/logical_tokens.hpp +1 -0
- package/src/duckdb/src/include/duckdb/planner/operator/list.hpp +1 -0
- package/src/duckdb/src/include/duckdb/planner/operator/logical_copy_database.hpp +4 -12
- package/src/duckdb/src/include/duckdb/planner/operator/logical_create_secret.hpp +1 -1
- package/src/duckdb/src/include/duckdb/planner/operator/logical_cteref.hpp +2 -2
- package/src/duckdb/src/include/duckdb/planner/operator/logical_delim_get.hpp +1 -1
- package/src/duckdb/src/include/duckdb/planner/operator/logical_dummy_scan.hpp +1 -1
- package/src/duckdb/src/include/duckdb/planner/operator/logical_explain.hpp +1 -1
- package/src/duckdb/src/include/duckdb/planner/operator/logical_export.hpp +2 -2
- package/src/duckdb/src/include/duckdb/planner/operator/logical_expression_get.hpp +2 -2
- package/src/duckdb/src/include/duckdb/planner/operator/logical_extension_operator.hpp +2 -2
- package/src/duckdb/src/include/duckdb/planner/operator/logical_materialized_cte.hpp +2 -2
- package/src/duckdb/src/include/duckdb/planner/operator/logical_pragma.hpp +1 -1
- package/src/duckdb/src/include/duckdb/planner/operator/logical_prepare.hpp +3 -2
- package/src/duckdb/src/include/duckdb/planner/operator/logical_recursive_cte.hpp +3 -3
- package/src/duckdb/src/include/duckdb/planner/operator/logical_reset.hpp +1 -1
- package/src/duckdb/src/include/duckdb/planner/operator/logical_set.hpp +2 -1
- package/src/duckdb/src/include/duckdb/planner/operator/logical_vacuum.hpp +52 -0
- package/src/duckdb/src/include/duckdb/planner/operator_extension.hpp +1 -1
- package/src/duckdb/src/include/duckdb/planner/parsed_data/bound_create_table_info.hpp +2 -2
- package/src/duckdb/src/include/duckdb/planner/query_node/bound_select_node.hpp +5 -4
- package/src/duckdb/src/include/duckdb/planner/subquery/flatten_dependent_join.hpp +2 -2
- package/src/duckdb/src/include/duckdb/storage/arena_allocator.hpp +4 -1
- package/src/duckdb/src/include/duckdb/storage/buffer/block_handle.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/buffer_manager.hpp +3 -3
- package/src/duckdb/src/include/duckdb/storage/checkpoint/row_group_writer.hpp +2 -3
- package/src/duckdb/src/include/duckdb/storage/checkpoint/string_checkpoint_state.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/checkpoint/table_data_writer.hpp +3 -3
- package/src/duckdb/src/include/duckdb/storage/checkpoint_manager.hpp +5 -5
- package/src/duckdb/src/include/duckdb/storage/compression/alp/alp_analyze.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/compression/alp/alp_compress.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/compression/alp/alp_fetch.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/compression/alp/alp_scan.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/compression/alprd/algorithm/alprd.hpp +3 -3
- package/src/duckdb/src/include/duckdb/storage/compression/alprd/alprd_analyze.hpp +2 -2
- package/src/duckdb/src/include/duckdb/storage/compression/alprd/alprd_compress.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/compression/alprd/alprd_fetch.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/compression/alprd/alprd_scan.hpp +3 -3
- package/src/duckdb/src/include/duckdb/storage/compression/chimp/algorithm/flag_buffer.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/compression/chimp/algorithm/leading_zero_buffer.hpp +3 -3
- package/src/duckdb/src/include/duckdb/storage/compression/chimp/chimp.hpp +3 -3
- package/src/duckdb/src/include/duckdb/storage/compression/chimp/chimp_analyze.hpp +0 -1
- package/src/duckdb/src/include/duckdb/storage/compression/chimp/chimp_fetch.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/compression/chimp/chimp_scan.hpp +4 -4
- package/src/duckdb/src/include/duckdb/storage/compression/patas/patas.hpp +4 -4
- package/src/duckdb/src/include/duckdb/storage/compression/patas/patas_analyze.hpp +0 -1
- package/src/duckdb/src/include/duckdb/storage/compression/patas/patas_fetch.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/compression/patas/patas_scan.hpp +3 -3
- package/src/duckdb/src/include/duckdb/storage/data_pointer.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/index.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/magic_bytes.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/object_cache.hpp +3 -3
- package/src/duckdb/src/include/duckdb/storage/optimistic_data_writer.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/partial_block_manager.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/single_file_block_manager.hpp +4 -1
- package/src/duckdb/src/include/duckdb/storage/standard_buffer_manager.hpp +24 -24
- package/src/duckdb/src/include/duckdb/storage/statistics/numeric_stats_union.hpp +3 -3
- package/src/duckdb/src/include/duckdb/storage/statistics/segment_statistics.hpp +2 -2
- package/src/duckdb/src/include/duckdb/storage/storage_info.hpp +2 -2
- package/src/duckdb/src/include/duckdb/storage/storage_manager.hpp +4 -4
- package/src/duckdb/src/include/duckdb/storage/string_uncompressed.hpp +9 -6
- package/src/duckdb/src/include/duckdb/storage/table/append_state.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/table/chunk_info.hpp +2 -2
- package/src/duckdb/src/include/duckdb/storage/table/column_checkpoint_state.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/table/column_data.hpp +10 -1
- package/src/duckdb/src/include/duckdb/storage/table/data_table_info.hpp +3 -2
- package/src/duckdb/src/include/duckdb/storage/table/row_group.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/table/row_group_segment_tree.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/table/row_version_manager.hpp +2 -1
- package/src/duckdb/src/include/duckdb/storage/table/scan_state.hpp +3 -3
- package/src/duckdb/src/include/duckdb/storage/table/segment_lock.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/table/segment_tree.hpp +4 -4
- package/src/duckdb/src/include/duckdb/storage/table/standard_column_data.hpp +2 -0
- package/src/duckdb/src/include/duckdb/storage/table/table_index_list.hpp +3 -2
- package/src/duckdb/src/include/duckdb/storage/table/table_statistics.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/table/update_segment.hpp +1 -1
- package/src/duckdb/src/include/duckdb/storage/table_storage_info.hpp +2 -1
- package/src/duckdb/src/include/duckdb/storage/write_ahead_log.hpp +1 -1
- package/src/duckdb/src/include/duckdb/transaction/delete_info.hpp +20 -1
- package/src/duckdb/src/include/duckdb/transaction/transaction.hpp +1 -1
- package/src/duckdb/src/include/duckdb/transaction/transaction_data.hpp +1 -1
- package/src/duckdb/src/include/duckdb/transaction/undo_buffer.hpp +1 -1
- package/src/duckdb/src/include/duckdb/transaction/update_info.hpp +1 -1
- package/src/duckdb/src/main/appender.cpp +1 -1
- package/src/duckdb/src/main/attached_database.cpp +11 -3
- package/src/duckdb/src/main/capi/arrow-c.cpp +6 -2
- package/src/duckdb/src/main/capi/cast/utils-c.cpp +1 -1
- package/src/duckdb/src/main/capi/duckdb-c.cpp +1 -1
- package/src/duckdb/src/main/capi/pending-c.cpp +1 -1
- package/src/duckdb/src/main/capi/prepared-c.cpp +2 -2
- package/src/duckdb/src/main/capi/result-c.cpp +4 -4
- package/src/duckdb/src/main/client_context.cpp +4 -12
- package/src/duckdb/src/main/client_context_file_opener.cpp +15 -3
- package/src/duckdb/src/main/client_data.cpp +5 -0
- package/src/duckdb/src/main/config.cpp +82 -82
- package/src/duckdb/src/main/database.cpp +31 -7
- package/src/duckdb/src/main/database_manager.cpp +3 -2
- package/src/duckdb/src/main/database_path_and_type.cpp +4 -4
- package/src/duckdb/src/main/error_manager.cpp +1 -1
- package/src/duckdb/src/main/extension/extension_alias.cpp +9 -9
- package/src/duckdb/src/main/extension/extension_helper.cpp +10 -5
- package/src/duckdb/src/main/extension/extension_install.cpp +1 -1
- package/src/duckdb/src/main/extension/extension_load.cpp +111 -37
- package/src/duckdb/src/main/query_profiler.cpp +1 -118
- package/src/duckdb/src/main/secret/secret_manager.cpp +1 -2
- package/src/duckdb/src/main/secret/secret_storage.cpp +1 -1
- package/src/duckdb/src/main/settings/settings.cpp +81 -65
- package/src/duckdb/src/optimizer/column_lifetime_analyzer.cpp +8 -1
- package/src/duckdb/src/optimizer/compressed_materialization/compress_aggregate.cpp +3 -0
- package/src/duckdb/src/optimizer/compressed_materialization.cpp +26 -28
- package/src/duckdb/src/optimizer/cse_optimizer.cpp +5 -5
- package/src/duckdb/src/optimizer/join_order/cardinality_estimator.cpp +10 -6
- package/src/duckdb/src/optimizer/optimizer.cpp +14 -17
- package/src/duckdb/src/optimizer/pushdown/pushdown_cross_product.cpp +16 -5
- package/src/duckdb/src/optimizer/pushdown/pushdown_inner_join.cpp +7 -4
- package/src/duckdb/src/optimizer/pushdown/pushdown_set_operation.cpp +2 -2
- package/src/duckdb/src/optimizer/remove_unused_columns.cpp +3 -3
- package/src/duckdb/src/optimizer/rule/ordered_aggregate_optimizer.cpp +2 -2
- package/src/duckdb/src/optimizer/statistics/expression/propagate_aggregate.cpp +1 -1
- package/src/duckdb/src/optimizer/statistics/expression/propagate_between.cpp +8 -8
- package/src/duckdb/src/optimizer/statistics/expression/propagate_case.cpp +1 -1
- package/src/duckdb/src/optimizer/statistics/expression/propagate_cast.cpp +1 -1
- package/src/duckdb/src/optimizer/statistics/expression/propagate_columnref.cpp +1 -1
- package/src/duckdb/src/optimizer/statistics/expression/propagate_comparison.cpp +7 -7
- package/src/duckdb/src/optimizer/statistics/expression/propagate_conjunction.cpp +6 -6
- package/src/duckdb/src/optimizer/statistics/expression/propagate_constant.cpp +1 -1
- package/src/duckdb/src/optimizer/statistics/expression/propagate_function.cpp +2 -2
- package/src/duckdb/src/optimizer/statistics/expression/propagate_operator.cpp +10 -10
- package/src/duckdb/src/optimizer/statistics/operator/propagate_aggregate.cpp +1 -1
- package/src/duckdb/src/optimizer/statistics/operator/propagate_cross_product.cpp +1 -1
- package/src/duckdb/src/optimizer/statistics/operator/propagate_filter.cpp +14 -7
- package/src/duckdb/src/optimizer/statistics/operator/propagate_get.cpp +2 -2
- package/src/duckdb/src/optimizer/statistics/operator/propagate_join.cpp +9 -9
- package/src/duckdb/src/optimizer/statistics/operator/propagate_limit.cpp +1 -1
- package/src/duckdb/src/optimizer/statistics/operator/propagate_order.cpp +1 -1
- package/src/duckdb/src/optimizer/statistics/operator/propagate_projection.cpp +2 -2
- package/src/duckdb/src/optimizer/statistics/operator/propagate_set_operation.cpp +1 -1
- package/src/duckdb/src/optimizer/statistics/operator/propagate_window.cpp +1 -1
- package/src/duckdb/src/optimizer/statistics_propagator.cpp +39 -18
- package/src/duckdb/src/parallel/pipeline_finish_event.cpp +1 -1
- package/src/duckdb/src/parallel/task_scheduler.cpp +8 -1
- package/src/duckdb/src/parser/constraints/unique_constraint.cpp +4 -2
- package/src/duckdb/src/parser/parsed_data/create_index_info.cpp +64 -0
- package/src/duckdb/src/parser/parsed_data/vacuum_info.cpp +1 -0
- package/src/duckdb/src/parser/transform/expression/transform_function.cpp +13 -5
- package/src/duckdb/src/parser/transform/expression/transform_subquery.cpp +24 -0
- package/src/duckdb/src/parser/transform/helpers/transform_cte.cpp +1 -8
- package/src/duckdb/src/parser/transform/statement/transform_create_function.cpp +2 -1
- package/src/duckdb/src/parser/transform/statement/transform_create_index.cpp +3 -0
- package/src/duckdb/src/parser/transform/statement/transform_delete.cpp +2 -6
- package/src/duckdb/src/parser/transform/statement/transform_insert.cpp +2 -6
- package/src/duckdb/src/parser/transform/statement/transform_pivot_stmt.cpp +2 -6
- package/src/duckdb/src/parser/transform/statement/transform_pragma.cpp +1 -1
- package/src/duckdb/src/parser/transform/statement/transform_select.cpp +5 -2
- package/src/duckdb/src/parser/transform/statement/transform_select_node.cpp +2 -8
- package/src/duckdb/src/parser/transform/statement/transform_update.cpp +1 -6
- package/src/duckdb/src/parser/transform/statement/transform_use.cpp +3 -2
- package/src/duckdb/src/parser/transformer.cpp +14 -2
- package/src/duckdb/src/planner/binder/expression/bind_aggregate_expression.cpp +40 -17
- package/src/duckdb/src/planner/binder/expression/bind_unnest_expression.cpp +8 -3
- package/src/duckdb/src/planner/binder/expression/bind_window_expression.cpp +32 -30
- package/src/duckdb/src/planner/binder/query_node/bind_cte_node.cpp +28 -17
- package/src/duckdb/src/planner/binder/query_node/bind_select_node.cpp +146 -101
- package/src/duckdb/src/planner/binder/query_node/bind_setop_node.cpp +28 -26
- package/src/duckdb/src/planner/binder/query_node/plan_cte_node.cpp +29 -0
- package/src/duckdb/src/planner/binder/query_node/plan_subquery.cpp +10 -10
- package/src/duckdb/src/planner/binder/statement/bind_copy.cpp +3 -2
- package/src/duckdb/src/planner/binder/statement/bind_copy_database.cpp +23 -28
- package/src/duckdb/src/planner/binder/statement/bind_create.cpp +11 -12
- package/src/duckdb/src/planner/binder/statement/bind_create_table.cpp +9 -11
- package/src/duckdb/src/planner/binder/statement/bind_insert.cpp +10 -13
- package/src/duckdb/src/planner/binder/statement/bind_pragma.cpp +3 -3
- package/src/duckdb/src/planner/binder/statement/bind_set.cpp +3 -0
- package/src/duckdb/src/planner/binder/statement/bind_vacuum.cpp +66 -65
- package/src/duckdb/src/planner/binder/tableref/bind_basetableref.cpp +54 -46
- package/src/duckdb/src/planner/binder/tableref/bind_pivot.cpp +3 -0
- package/src/duckdb/src/planner/binder/tableref/bind_subqueryref.cpp +1 -1
- package/src/duckdb/src/planner/binder/tableref/bind_table_function.cpp +5 -5
- package/src/duckdb/src/planner/binder.cpp +78 -6
- package/src/duckdb/src/planner/expression/bound_cast_expression.cpp +11 -3
- package/src/duckdb/src/planner/expression/bound_expanded_expression.cpp +22 -0
- package/src/duckdb/src/planner/expression_binder/base_select_binder.cpp +11 -58
- package/src/duckdb/src/planner/expression_binder/column_alias_binder.cpp +5 -8
- package/src/duckdb/src/planner/expression_binder/group_binder.cpp +5 -4
- package/src/duckdb/src/planner/expression_binder/having_binder.cpp +5 -19
- package/src/duckdb/src/planner/expression_binder/lateral_binder.cpp +8 -8
- package/src/duckdb/src/planner/expression_binder/order_binder.cpp +42 -26
- package/src/duckdb/src/planner/expression_binder/qualify_binder.cpp +3 -22
- package/src/duckdb/src/planner/expression_binder/select_bind_state.cpp +52 -0
- package/src/duckdb/src/planner/expression_binder/select_binder.cpp +43 -5
- package/src/duckdb/src/planner/expression_binder/table_function_binder.cpp +19 -7
- package/src/duckdb/src/planner/logical_operator.cpp +20 -3
- package/src/duckdb/src/planner/operator/logical_copy_database.cpp +4 -14
- package/src/duckdb/src/planner/operator/logical_delete.cpp +1 -1
- package/src/duckdb/src/planner/operator/logical_get.cpp +1 -1
- package/src/duckdb/src/planner/operator/logical_insert.cpp +1 -1
- package/src/duckdb/src/planner/operator/logical_update.cpp +1 -1
- package/src/duckdb/src/planner/operator/logical_vacuum.cpp +65 -0
- package/src/duckdb/src/planner/planner.cpp +4 -4
- package/src/duckdb/src/planner/subquery/flatten_dependent_join.cpp +26 -27
- package/src/duckdb/src/storage/arena_allocator.cpp +9 -0
- package/src/duckdb/src/storage/buffer/buffer_pool_reservation.cpp +1 -1
- package/src/duckdb/src/storage/buffer_manager.cpp +2 -10
- package/src/duckdb/src/storage/checkpoint/table_data_writer.cpp +2 -3
- package/src/duckdb/src/storage/checkpoint_manager.cpp +15 -8
- package/src/duckdb/src/storage/compression/bitpacking.cpp +6 -1
- package/src/duckdb/src/storage/compression/string_uncompressed.cpp +1 -1
- package/src/duckdb/src/storage/data_pointer.cpp +1 -1
- package/src/duckdb/src/storage/data_table.cpp +18 -7
- package/src/duckdb/src/storage/local_storage.cpp +8 -5
- package/src/duckdb/src/storage/magic_bytes.cpp +6 -5
- package/src/duckdb/src/storage/partial_block_manager.cpp +1 -1
- package/src/duckdb/src/storage/serialization/serialize_logical_operator.cpp +20 -9
- package/src/duckdb/src/storage/serialization/serialize_nodes.cpp +12 -10
- package/src/duckdb/src/storage/serialization/serialize_parse_info.cpp +23 -0
- package/src/duckdb/src/storage/single_file_block_manager.cpp +46 -19
- package/src/duckdb/src/storage/standard_buffer_manager.cpp +21 -5
- package/src/duckdb/src/storage/statistics/array_stats.cpp +1 -1
- package/src/duckdb/src/storage/statistics/base_statistics.cpp +2 -3
- package/src/duckdb/src/storage/statistics/list_stats.cpp +1 -1
- package/src/duckdb/src/storage/statistics/struct_stats.cpp +1 -1
- package/src/duckdb/src/storage/storage_info.cpp +19 -19
- package/src/duckdb/src/storage/storage_manager.cpp +18 -13
- package/src/duckdb/src/storage/table/chunk_info.cpp +11 -3
- package/src/duckdb/src/storage/table/column_data.cpp +88 -66
- package/src/duckdb/src/storage/table/row_group.cpp +7 -7
- package/src/duckdb/src/storage/table/row_version_manager.cpp +2 -2
- package/src/duckdb/src/storage/table/standard_column_data.cpp +4 -0
- package/src/duckdb/src/storage/table/update_segment.cpp +3 -1
- package/src/duckdb/src/storage/table_index_list.cpp +6 -1
- package/src/duckdb/src/storage/temporary_file_manager.cpp +1 -1
- package/src/duckdb/src/storage/wal_replay.cpp +8 -7
- package/src/duckdb/src/storage/write_ahead_log.cpp +3 -4
- package/src/duckdb/src/transaction/cleanup_state.cpp +10 -3
- package/src/duckdb/src/transaction/commit_state.cpp +11 -4
- package/src/duckdb/src/transaction/duck_transaction.cpp +23 -3
- package/src/duckdb/src/transaction/rollback_state.cpp +1 -1
- package/src/duckdb/src/transaction/transaction_context.cpp +1 -1
- package/src/duckdb/src/transaction/undo_buffer.cpp +3 -1
- package/src/duckdb/third_party/fmt/include/fmt/core.h +0 -5
- package/src/duckdb/third_party/fsst/fsst.h +1 -1
- package/src/duckdb/third_party/fsst/libfsst.cpp +1 -140
- package/src/duckdb/third_party/fsst/libfsst.hpp +0 -13
- package/src/duckdb/third_party/hyperloglog/hyperloglog.hpp +4 -0
- package/src/duckdb/third_party/libpg_query/include/nodes/parsenodes.hpp +8 -1
- package/src/duckdb/third_party/libpg_query/include/parser/gram.hpp +2 -1
- package/src/duckdb/third_party/libpg_query/src_backend_parser_gram.cpp +15401 -15354
- package/src/duckdb/third_party/libpg_query/src_backend_parser_scan.cpp +299 -538
- package/src/duckdb/third_party/lz4/lz4.cpp +2605 -0
- package/src/duckdb/third_party/lz4/lz4.hpp +843 -0
- package/src/duckdb/third_party/parquet/parquet_types.cpp +3 -0
- package/src/duckdb/third_party/parquet/parquet_types.h +2 -1
- package/src/duckdb/third_party/re2/re2/bitmap256.cc +44 -0
- package/src/duckdb/third_party/re2/re2/bitmap256.h +3 -35
- package/src/duckdb/third_party/re2/re2/bitstate.cc +31 -24
- package/src/duckdb/third_party/re2/re2/compile.cc +146 -164
- package/src/duckdb/third_party/re2/re2/dfa.cc +174 -181
- package/src/duckdb/third_party/re2/re2/filtered_re2.cc +19 -3
- package/src/duckdb/third_party/re2/re2/filtered_re2.h +27 -23
- package/src/duckdb/third_party/re2/re2/mimics_pcre.cc +21 -11
- package/src/duckdb/third_party/re2/re2/nfa.cc +91 -131
- package/src/duckdb/third_party/re2/re2/onepass.cc +11 -10
- package/src/duckdb/third_party/re2/re2/parse.cc +171 -154
- package/src/duckdb/third_party/re2/re2/perl_groups.cc +35 -35
- package/src/duckdb/third_party/re2/re2/pod_array.h +55 -0
- package/src/duckdb/third_party/re2/re2/prefilter.cc +40 -40
- package/src/duckdb/third_party/re2/re2/prefilter.h +24 -2
- package/src/duckdb/third_party/re2/re2/prefilter_tree.cc +70 -84
- package/src/duckdb/third_party/re2/re2/prefilter_tree.h +5 -4
- package/src/duckdb/third_party/re2/re2/prog.cc +315 -58
- package/src/duckdb/third_party/re2/re2/prog.h +77 -44
- package/src/duckdb/third_party/re2/re2/re2.cc +333 -221
- package/src/duckdb/third_party/re2/re2/re2.h +277 -201
- package/src/duckdb/third_party/re2/re2/regexp.cc +137 -105
- package/src/duckdb/third_party/re2/re2/regexp.h +45 -40
- package/src/duckdb/third_party/re2/re2/set.cc +40 -17
- package/src/duckdb/third_party/re2/re2/set.h +11 -6
- package/src/duckdb/third_party/re2/re2/simplify.cc +50 -41
- package/src/duckdb/third_party/re2/re2/sparse_array.h +392 -0
- package/src/duckdb/third_party/re2/re2/sparse_set.h +264 -0
- package/src/duckdb/third_party/re2/re2/stringpiece.cc +1 -1
- package/src/duckdb/third_party/re2/re2/stringpiece.h +11 -8
- package/src/duckdb/third_party/re2/re2/tostring.cc +8 -6
- package/src/duckdb/third_party/re2/re2/unicode_casefold.cc +39 -10
- package/src/duckdb/third_party/re2/re2/unicode_casefold.h +1 -1
- package/src/duckdb/third_party/re2/re2/unicode_groups.cc +5019 -4566
- package/src/duckdb/third_party/re2/re2/unicode_groups.h +1 -1
- package/src/duckdb/third_party/re2/re2/walker-inl.h +21 -20
- package/src/duckdb/third_party/re2/util/logging.h +14 -18
- package/src/duckdb/third_party/re2/util/mix.h +4 -4
- package/src/duckdb/third_party/re2/util/mutex.h +48 -15
- package/src/duckdb/third_party/re2/util/rune.cc +5 -5
- package/src/duckdb/third_party/re2/util/strutil.cc +1 -16
- package/src/duckdb/third_party/re2/util/strutil.h +1 -3
- package/src/duckdb/third_party/re2/util/utf.h +1 -1
- package/src/duckdb/third_party/re2/util/util.h +9 -1
- package/src/duckdb/third_party/utf8proc/include/utf8proc_wrapper.hpp +2 -0
- package/src/duckdb/third_party/utf8proc/utf8proc_wrapper.cpp +36 -1
- package/src/duckdb/ub_src_common.cpp +0 -2
- package/src/duckdb/ub_src_execution_physical_plan.cpp +2 -0
- package/src/duckdb/ub_src_planner_expression.cpp +2 -0
- package/src/duckdb/ub_src_planner_expression_binder.cpp +2 -0
- package/src/duckdb/ub_src_planner_operator.cpp +2 -0
- package/test/jsdoc.test.ts +3 -0
@@ -0,0 +1,2605 @@
|
|
1
|
+
/*
|
2
|
+
LZ4 - Fast LZ compression algorithm
|
3
|
+
Copyright (C) 2011-2020, Yann Collet.
|
4
|
+
|
5
|
+
BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
|
6
|
+
|
7
|
+
Redistribution and use in source and binary forms, with or without
|
8
|
+
modification, are permitted provided that the following conditions are
|
9
|
+
met:
|
10
|
+
|
11
|
+
* Redistributions of source code must retain the above copyright
|
12
|
+
notice, this list of conditions and the following disclaimer.
|
13
|
+
* Redistributions in binary form must reproduce the above
|
14
|
+
copyright notice, this list of conditions and the following disclaimer
|
15
|
+
in the documentation and/or other materials provided with the
|
16
|
+
distribution.
|
17
|
+
|
18
|
+
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
|
19
|
+
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
|
20
|
+
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
|
21
|
+
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
|
22
|
+
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
|
23
|
+
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
|
24
|
+
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
|
25
|
+
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
|
26
|
+
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
27
|
+
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
28
|
+
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
29
|
+
|
30
|
+
You can contact the author at :
|
31
|
+
- LZ4 homepage : http://www.lz4.org
|
32
|
+
- LZ4 source repository : https://github.com/lz4/lz4
|
33
|
+
*/
|
34
|
+
|
35
|
+
/*-************************************
|
36
|
+
* Tuning parameters
|
37
|
+
**************************************/
|
38
|
+
/*
|
39
|
+
* LZ4_HEAPMODE :
|
40
|
+
* Select how default compression functions will allocate memory for their hash table,
|
41
|
+
* in memory stack (0:default, fastest), or in memory heap (1:requires malloc()).
|
42
|
+
*/
|
43
|
+
#ifndef LZ4_HEAPMODE
|
44
|
+
# define LZ4_HEAPMODE 0
|
45
|
+
#endif
|
46
|
+
|
47
|
+
/*
|
48
|
+
* LZ4_ACCELERATION_DEFAULT :
|
49
|
+
* Select "acceleration" for LZ4_compress_fast() when parameter value <= 0
|
50
|
+
*/
|
51
|
+
#define LZ4_ACCELERATION_DEFAULT 1
|
52
|
+
/*
|
53
|
+
* LZ4_ACCELERATION_MAX :
|
54
|
+
* Any "acceleration" value higher than this threshold
|
55
|
+
* get treated as LZ4_ACCELERATION_MAX instead (fix #876)
|
56
|
+
*/
|
57
|
+
#define LZ4_ACCELERATION_MAX 65537
|
58
|
+
|
59
|
+
|
60
|
+
/*-************************************
|
61
|
+
* CPU Feature Detection
|
62
|
+
**************************************/
|
63
|
+
/* LZ4_FORCE_MEMORY_ACCESS
|
64
|
+
* By default, access to unaligned memory is controlled by `memcpy()`, which is safe and portable.
|
65
|
+
* Unfortunately, on some target/compiler combinations, the generated assembly is sub-optimal.
|
66
|
+
* The below switch allow to select different access method for improved performance.
|
67
|
+
* Method 0 (default) : use `memcpy()`. Safe and portable.
|
68
|
+
* Method 1 : `__packed` statement. It depends on compiler extension (ie, not portable).
|
69
|
+
* This method is safe if your compiler supports it, and *generally* as fast or faster than `memcpy`.
|
70
|
+
* Method 2 : direct access. This method is portable but violate C standard.
|
71
|
+
* It can generate buggy code on targets which assembly generation depends on alignment.
|
72
|
+
* But in some circumstances, it's the only known way to get the most performance (ie GCC + ARMv6)
|
73
|
+
* See https://fastcompression.blogspot.fr/2015/08/accessing-unaligned-memory.html for details.
|
74
|
+
* Prefer these methods in priority order (0 > 1 > 2)
|
75
|
+
*/
|
76
|
+
#ifndef LZ4_FORCE_MEMORY_ACCESS /* can be defined externally */
|
77
|
+
# if defined(__GNUC__) && \
|
78
|
+
( defined(__ARM_ARCH_6__) || defined(__ARM_ARCH_6J__) || defined(__ARM_ARCH_6K__) \
|
79
|
+
|| defined(__ARM_ARCH_6Z__) || defined(__ARM_ARCH_6ZK__) || defined(__ARM_ARCH_6T2__) )
|
80
|
+
# define LZ4_FORCE_MEMORY_ACCESS 2
|
81
|
+
# elif (defined(__INTEL_COMPILER) && !defined(_WIN32)) || defined(__GNUC__)
|
82
|
+
# define LZ4_FORCE_MEMORY_ACCESS 1
|
83
|
+
# endif
|
84
|
+
#endif
|
85
|
+
|
86
|
+
/*
|
87
|
+
* LZ4_FORCE_SW_BITCOUNT
|
88
|
+
* Define this parameter if your target system or compiler does not support hardware bit count
|
89
|
+
*/
|
90
|
+
#if defined(_MSC_VER) && defined(_WIN32_WCE) /* Visual Studio for WinCE doesn't support Hardware bit count */
|
91
|
+
# undef LZ4_FORCE_SW_BITCOUNT /* avoid double def */
|
92
|
+
# define LZ4_FORCE_SW_BITCOUNT
|
93
|
+
#endif
|
94
|
+
|
95
|
+
|
96
|
+
|
97
|
+
/*-************************************
|
98
|
+
* Dependency
|
99
|
+
**************************************/
|
100
|
+
/*
|
101
|
+
* LZ4_SRC_INCLUDED:
|
102
|
+
* Amalgamation flag, whether lz4.c is included
|
103
|
+
*/
|
104
|
+
#ifndef LZ4_SRC_INCLUDED
|
105
|
+
# define LZ4_SRC_INCLUDED 1
|
106
|
+
#endif
|
107
|
+
|
108
|
+
#ifndef LZ4_STATIC_LINKING_ONLY
|
109
|
+
#define LZ4_STATIC_LINKING_ONLY
|
110
|
+
#endif
|
111
|
+
|
112
|
+
#ifndef LZ4_DISABLE_DEPRECATE_WARNINGS
|
113
|
+
#define LZ4_DISABLE_DEPRECATE_WARNINGS /* due to LZ4_decompress_safe_withPrefix64k */
|
114
|
+
#endif
|
115
|
+
|
116
|
+
#define LZ4_STATIC_LINKING_ONLY /* LZ4_DISTANCE_MAX */
|
117
|
+
#include <stdlib.h>
|
118
|
+
|
119
|
+
#include "lz4.hpp"
|
120
|
+
/* see also "memory routines" below */
|
121
|
+
|
122
|
+
|
123
|
+
/*-************************************
|
124
|
+
* Compiler Options
|
125
|
+
**************************************/
|
126
|
+
|
127
|
+
#define LZ4_FORCE_INLINE static
|
128
|
+
#define LZ4_FORCE_O2
|
129
|
+
|
130
|
+
#if (defined(__GNUC__) && (__GNUC__ >= 3)) || (defined(__INTEL_COMPILER) && (__INTEL_COMPILER >= 800)) || defined(__clang__)
|
131
|
+
# define expect(expr,value) (__builtin_expect ((expr),(value)) )
|
132
|
+
#else
|
133
|
+
# define expect(expr,value) (expr)
|
134
|
+
#endif
|
135
|
+
|
136
|
+
#ifndef likely
|
137
|
+
#define likely(expr) expect((expr) != 0, 1)
|
138
|
+
#endif
|
139
|
+
#ifndef unlikely
|
140
|
+
#define unlikely(expr) expect((expr) != 0, 0)
|
141
|
+
#endif
|
142
|
+
|
143
|
+
/* Should the alignment test prove unreliable, for some reason,
|
144
|
+
* it can be disabled by setting LZ4_ALIGN_TEST to 0 */
|
145
|
+
#ifndef LZ4_ALIGN_TEST /* can be externally provided */
|
146
|
+
# define LZ4_ALIGN_TEST 1
|
147
|
+
#endif
|
148
|
+
|
149
|
+
namespace duckdb_lz4 {
|
150
|
+
/*-************************************
|
151
|
+
* Memory routines
|
152
|
+
**************************************/
|
153
|
+
|
154
|
+
/*! LZ4_STATIC_LINKING_ONLY_DISABLE_MEMORY_ALLOCATION :
|
155
|
+
* Disable relatively high-level LZ4/HC functions that use dynamic memory
|
156
|
+
* allocation functions (malloc(), calloc(), free()).
|
157
|
+
*
|
158
|
+
* Note that this is a compile-time switch. And since it disables
|
159
|
+
* public/stable LZ4 v1 API functions, we don't recommend using this
|
160
|
+
* symbol to generate a library for distribution.
|
161
|
+
*
|
162
|
+
* The following public functions are removed when this symbol is defined.
|
163
|
+
* - lz4 : LZ4_createStream, LZ4_freeStream,
|
164
|
+
* LZ4_createStreamDecode, LZ4_freeStreamDecode, LZ4_create (deprecated)
|
165
|
+
* - lz4hc : LZ4_createStreamHC, LZ4_freeStreamHC,
|
166
|
+
* LZ4_createHC (deprecated), LZ4_freeHC (deprecated)
|
167
|
+
* - lz4frame, lz4file : All LZ4F_* functions
|
168
|
+
*/
|
169
|
+
#if defined(LZ4_STATIC_LINKING_ONLY_DISABLE_MEMORY_ALLOCATION)
|
170
|
+
# define ALLOC(s) lz4_error_memory_allocation_is_disabled
|
171
|
+
# define ALLOC_AND_ZERO(s) lz4_error_memory_allocation_is_disabled
|
172
|
+
# define FREEMEM(p) lz4_error_memory_allocation_is_disabled
|
173
|
+
#elif defined(LZ4_USER_MEMORY_FUNCTIONS)
|
174
|
+
/* memory management functions can be customized by user project.
|
175
|
+
* Below functions must exist somewhere in the Project
|
176
|
+
* and be available at link time */
|
177
|
+
void* LZ4_malloc(size_t s);
|
178
|
+
void* LZ4_calloc(size_t n, size_t s);
|
179
|
+
void LZ4_free(void* p);
|
180
|
+
# define ALLOC(s) LZ4_malloc(s)
|
181
|
+
# define ALLOC_AND_ZERO(s) LZ4_calloc(1,s)
|
182
|
+
# define FREEMEM(p) LZ4_free(p)
|
183
|
+
#else
|
184
|
+
# include <stdlib.h> /* malloc, calloc, free */
|
185
|
+
# define ALLOC(s) malloc(s)
|
186
|
+
# define ALLOC_AND_ZERO(s) calloc(1,s)
|
187
|
+
# define FREEMEM(p) free(p)
|
188
|
+
#endif
|
189
|
+
|
190
|
+
#if ! LZ4_FREESTANDING
|
191
|
+
# include <string.h> /* memset, memcpy */
|
192
|
+
#endif
|
193
|
+
#if !defined(LZ4_memset)
|
194
|
+
# define LZ4_memset(p,v,s) memset((p),(v),(s))
|
195
|
+
#endif
|
196
|
+
#define MEM_INIT(p,v,s) LZ4_memset((p),(v),(s))
|
197
|
+
|
198
|
+
|
199
|
+
/*-************************************
|
200
|
+
* Common Constants
|
201
|
+
**************************************/
|
202
|
+
#define MINMATCH 4
|
203
|
+
|
204
|
+
#define WILDCOPYLENGTH 8
|
205
|
+
#define LASTLITERALS 5 /* see ../doc/lz4_Block_format.md#parsing-restrictions */
|
206
|
+
#define MFLIMIT 12 /* see ../doc/lz4_Block_format.md#parsing-restrictions */
|
207
|
+
#define MATCH_SAFEGUARD_DISTANCE ((2*WILDCOPYLENGTH) - MINMATCH) /* ensure it's possible to write 2 x wildcopyLength without overflowing output buffer */
|
208
|
+
#define FASTLOOP_SAFE_DISTANCE 64
|
209
|
+
static const int LZ4_minLength = (MFLIMIT+1);
|
210
|
+
|
211
|
+
#define KB *(1 <<10)
|
212
|
+
#define MB *(1 <<20)
|
213
|
+
#define GB *(1U<<30)
|
214
|
+
|
215
|
+
#define LZ4_DISTANCE_ABSOLUTE_MAX 65535
|
216
|
+
#if (LZ4_DISTANCE_MAX > LZ4_DISTANCE_ABSOLUTE_MAX) /* max supported by LZ4 format */
|
217
|
+
# error "LZ4_DISTANCE_MAX is too big : must be <= 65535"
|
218
|
+
#endif
|
219
|
+
|
220
|
+
#define ML_BITS 4
|
221
|
+
#define ML_MASK ((1U<<ML_BITS)-1)
|
222
|
+
#define RUN_BITS (8-ML_BITS)
|
223
|
+
#define RUN_MASK ((1U<<RUN_BITS)-1)
|
224
|
+
|
225
|
+
|
226
|
+
/*-************************************
|
227
|
+
* Error detection
|
228
|
+
**************************************/
|
229
|
+
#if defined(LZ4_DEBUG) && (LZ4_DEBUG>=1)
|
230
|
+
# include <assert.h>
|
231
|
+
#else
|
232
|
+
# ifndef assert
|
233
|
+
# define assert(condition) ((void)0)
|
234
|
+
# endif
|
235
|
+
#endif
|
236
|
+
|
237
|
+
#define LZ4_STATIC_ASSERT(c) { enum { LZ4_static_assert = 1/(int)(!!(c)) }; } /* use after variable declarations */
|
238
|
+
|
239
|
+
#if defined(LZ4_DEBUG) && (LZ4_DEBUG>=2)
|
240
|
+
# include <stdio.h>
|
241
|
+
static int g_debuglog_enable = 1;
|
242
|
+
# define DEBUGLOG(l, ...) { \
|
243
|
+
if ((g_debuglog_enable) && (l<=LZ4_DEBUG)) { \
|
244
|
+
fprintf(stderr, __FILE__ ": "); \
|
245
|
+
fprintf(stderr, __VA_ARGS__); \
|
246
|
+
fprintf(stderr, " \n"); \
|
247
|
+
} }
|
248
|
+
#else
|
249
|
+
# define DEBUGLOG(l, ...) {} /* disabled */
|
250
|
+
#endif
|
251
|
+
|
252
|
+
static int LZ4_isAligned(const void* ptr, size_t alignment)
|
253
|
+
{
|
254
|
+
return ((size_t)ptr & (alignment -1)) == 0;
|
255
|
+
}
|
256
|
+
|
257
|
+
|
258
|
+
/*-************************************
|
259
|
+
* Types
|
260
|
+
**************************************/
|
261
|
+
#include <limits.h>
|
262
|
+
#if defined(__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */)
|
263
|
+
# include <stdint.h>
|
264
|
+
typedef uint8_t BYTE;
|
265
|
+
typedef uint16_t U16;
|
266
|
+
typedef uint32_t U32;
|
267
|
+
typedef int32_t S32;
|
268
|
+
typedef uint64_t U64;
|
269
|
+
typedef uintptr_t uptrval;
|
270
|
+
#else
|
271
|
+
# if UINT_MAX != 4294967295UL
|
272
|
+
# error "LZ4 code (when not C++ or C99) assumes that sizeof(int) == 4"
|
273
|
+
# endif
|
274
|
+
typedef unsigned char BYTE;
|
275
|
+
typedef unsigned short U16;
|
276
|
+
typedef unsigned int U32;
|
277
|
+
typedef signed int S32;
|
278
|
+
typedef unsigned long long U64;
|
279
|
+
typedef size_t uptrval; /* generally true, except OpenVMS-64 */
|
280
|
+
#endif
|
281
|
+
|
282
|
+
#if defined(__x86_64__)
|
283
|
+
typedef U64 reg_t; /* 64-bits in x32 mode */
|
284
|
+
#else
|
285
|
+
typedef size_t reg_t; /* 32-bits in x32 mode */
|
286
|
+
#endif
|
287
|
+
|
288
|
+
typedef enum {
|
289
|
+
notLimited = 0,
|
290
|
+
limitedOutput = 1,
|
291
|
+
fillOutput = 2
|
292
|
+
} limitedOutput_directive;
|
293
|
+
|
294
|
+
|
295
|
+
/*-************************************
|
296
|
+
* Reading and writing into memory
|
297
|
+
**************************************/
|
298
|
+
|
299
|
+
/**
|
300
|
+
* LZ4 relies on memcpy with a constant size being inlined. In freestanding
|
301
|
+
* environments, the compiler can't assume the implementation of memcpy() is
|
302
|
+
* standard compliant, so it can't apply its specialized memcpy() inlining
|
303
|
+
* logic. When possible, use __builtin_memcpy() to tell the compiler to analyze
|
304
|
+
* memcpy() as if it were standard compliant, so it can inline it in freestanding
|
305
|
+
* environments. This is needed when decompressing the Linux Kernel, for example.
|
306
|
+
*/
|
307
|
+
#if !defined(LZ4_memcpy)
|
308
|
+
# if defined(__GNUC__) && (__GNUC__ >= 4)
|
309
|
+
# define LZ4_memcpy(dst, src, size) __builtin_memcpy(dst, src, size)
|
310
|
+
# else
|
311
|
+
# define LZ4_memcpy(dst, src, size) memcpy(dst, src, size)
|
312
|
+
# endif
|
313
|
+
#endif
|
314
|
+
|
315
|
+
#if !defined(LZ4_memmove)
|
316
|
+
# if defined(__GNUC__) && (__GNUC__ >= 4)
|
317
|
+
# define LZ4_memmove __builtin_memmove
|
318
|
+
# else
|
319
|
+
# define LZ4_memmove memmove
|
320
|
+
# endif
|
321
|
+
#endif
|
322
|
+
|
323
|
+
static unsigned LZ4_isLittleEndian(void)
|
324
|
+
{
|
325
|
+
const union { U32 u; BYTE c[4]; } one = { 1 }; /* don't use static : performance detrimental */
|
326
|
+
return one.c[0];
|
327
|
+
}
|
328
|
+
|
329
|
+
|
330
|
+
#if defined(LZ4_FORCE_MEMORY_ACCESS) && (LZ4_FORCE_MEMORY_ACCESS==2)
|
331
|
+
/* lie to the compiler about data alignment; use with caution */
|
332
|
+
|
333
|
+
static U16 LZ4_read16(const void* memPtr) { return *(const U16*) memPtr; }
|
334
|
+
static U32 LZ4_read32(const void* memPtr) { return *(const U32*) memPtr; }
|
335
|
+
static reg_t LZ4_read_ARCH(const void* memPtr) { return *(const reg_t*) memPtr; }
|
336
|
+
|
337
|
+
static void LZ4_write16(void* memPtr, U16 value) { *(U16*)memPtr = value; }
|
338
|
+
static void LZ4_write32(void* memPtr, U32 value) { *(U32*)memPtr = value; }
|
339
|
+
|
340
|
+
#elif defined(LZ4_FORCE_MEMORY_ACCESS) && (LZ4_FORCE_MEMORY_ACCESS==1)
|
341
|
+
|
342
|
+
/* __pack instructions are safer, but compiler specific, hence potentially problematic for some compilers */
|
343
|
+
/* currently only defined for gcc and icc */
|
344
|
+
typedef union { U16 u16; U32 u32; reg_t uArch; } __attribute__((packed)) LZ4_unalign;
|
345
|
+
|
346
|
+
static U16 LZ4_read16(const void* ptr) { return ((const LZ4_unalign*)ptr)->u16; }
|
347
|
+
static U32 LZ4_read32(const void* ptr) { return ((const LZ4_unalign*)ptr)->u32; }
|
348
|
+
static reg_t LZ4_read_ARCH(const void* ptr) { return ((const LZ4_unalign*)ptr)->uArch; }
|
349
|
+
|
350
|
+
static void LZ4_write16(void* memPtr, U16 value) { ((LZ4_unalign*)memPtr)->u16 = value; }
|
351
|
+
static void LZ4_write32(void* memPtr, U32 value) { ((LZ4_unalign*)memPtr)->u32 = value; }
|
352
|
+
|
353
|
+
#else /* safe and portable access using memcpy() */
|
354
|
+
|
355
|
+
static U16 LZ4_read16(const void* memPtr)
|
356
|
+
{
|
357
|
+
U16 val; LZ4_memcpy(&val, memPtr, sizeof(val)); return val;
|
358
|
+
}
|
359
|
+
|
360
|
+
static U32 LZ4_read32(const void* memPtr)
|
361
|
+
{
|
362
|
+
U32 val; LZ4_memcpy(&val, memPtr, sizeof(val)); return val;
|
363
|
+
}
|
364
|
+
|
365
|
+
static reg_t LZ4_read_ARCH(const void* memPtr)
|
366
|
+
{
|
367
|
+
reg_t val; LZ4_memcpy(&val, memPtr, sizeof(val)); return val;
|
368
|
+
}
|
369
|
+
|
370
|
+
static void LZ4_write16(void* memPtr, U16 value)
|
371
|
+
{
|
372
|
+
LZ4_memcpy(memPtr, &value, sizeof(value));
|
373
|
+
}
|
374
|
+
|
375
|
+
static void LZ4_write32(void* memPtr, U32 value)
|
376
|
+
{
|
377
|
+
LZ4_memcpy(memPtr, &value, sizeof(value));
|
378
|
+
}
|
379
|
+
|
380
|
+
#endif /* LZ4_FORCE_MEMORY_ACCESS */
|
381
|
+
|
382
|
+
|
383
|
+
static U16 LZ4_readLE16(const void* memPtr)
|
384
|
+
{
|
385
|
+
if (LZ4_isLittleEndian()) {
|
386
|
+
return LZ4_read16(memPtr);
|
387
|
+
} else {
|
388
|
+
const BYTE* p = (const BYTE*)memPtr;
|
389
|
+
return (U16)((U16)p[0] + (p[1]<<8));
|
390
|
+
}
|
391
|
+
}
|
392
|
+
|
393
|
+
static void LZ4_writeLE16(void* memPtr, U16 value)
|
394
|
+
{
|
395
|
+
if (LZ4_isLittleEndian()) {
|
396
|
+
LZ4_write16(memPtr, value);
|
397
|
+
} else {
|
398
|
+
BYTE* p = (BYTE*)memPtr;
|
399
|
+
p[0] = (BYTE) value;
|
400
|
+
p[1] = (BYTE)(value>>8);
|
401
|
+
}
|
402
|
+
}
|
403
|
+
|
404
|
+
/* customized variant of memcpy, which can overwrite up to 8 bytes beyond dstEnd */
|
405
|
+
LZ4_FORCE_INLINE
|
406
|
+
void LZ4_wildCopy8(void* dstPtr, const void* srcPtr, void* dstEnd)
|
407
|
+
{
|
408
|
+
BYTE* d = (BYTE*)dstPtr;
|
409
|
+
const BYTE* s = (const BYTE*)srcPtr;
|
410
|
+
BYTE* const e = (BYTE*)dstEnd;
|
411
|
+
|
412
|
+
do { LZ4_memcpy(d,s,8); d+=8; s+=8; } while (d<e);
|
413
|
+
}
|
414
|
+
|
415
|
+
static const unsigned inc32table[8] = {0, 1, 2, 1, 0, 4, 4, 4};
|
416
|
+
static const int dec64table[8] = {0, 0, 0, -1, -4, 1, 2, 3};
|
417
|
+
|
418
|
+
|
419
|
+
#ifndef LZ4_FAST_DEC_LOOP
|
420
|
+
# if defined __i386__ || defined _M_IX86 || defined __x86_64__ || defined _M_X64
|
421
|
+
# define LZ4_FAST_DEC_LOOP 1
|
422
|
+
# elif defined(__aarch64__) && defined(__APPLE__)
|
423
|
+
# define LZ4_FAST_DEC_LOOP 1
|
424
|
+
# elif defined(__aarch64__) && !defined(__clang__)
|
425
|
+
/* On non-Apple aarch64, we disable this optimization for clang because
|
426
|
+
* on certain mobile chipsets, performance is reduced with clang. For
|
427
|
+
* more information refer to https://github.com/lz4/lz4/pull/707 */
|
428
|
+
# define LZ4_FAST_DEC_LOOP 1
|
429
|
+
# else
|
430
|
+
# define LZ4_FAST_DEC_LOOP 0
|
431
|
+
# endif
|
432
|
+
#endif
|
433
|
+
|
434
|
+
#if LZ4_FAST_DEC_LOOP
|
435
|
+
|
436
|
+
LZ4_FORCE_INLINE void
|
437
|
+
LZ4_memcpy_using_offset_base(BYTE* dstPtr, const BYTE* srcPtr, BYTE* dstEnd, const size_t offset)
|
438
|
+
{
|
439
|
+
assert(srcPtr + offset == dstPtr);
|
440
|
+
if (offset < 8) {
|
441
|
+
LZ4_write32(dstPtr, 0); /* silence an msan warning when offset==0 */
|
442
|
+
dstPtr[0] = srcPtr[0];
|
443
|
+
dstPtr[1] = srcPtr[1];
|
444
|
+
dstPtr[2] = srcPtr[2];
|
445
|
+
dstPtr[3] = srcPtr[3];
|
446
|
+
srcPtr += inc32table[offset];
|
447
|
+
LZ4_memcpy(dstPtr+4, srcPtr, 4);
|
448
|
+
srcPtr -= dec64table[offset];
|
449
|
+
dstPtr += 8;
|
450
|
+
} else {
|
451
|
+
LZ4_memcpy(dstPtr, srcPtr, 8);
|
452
|
+
dstPtr += 8;
|
453
|
+
srcPtr += 8;
|
454
|
+
}
|
455
|
+
|
456
|
+
LZ4_wildCopy8(dstPtr, srcPtr, dstEnd);
|
457
|
+
}
|
458
|
+
|
459
|
+
/* customized variant of memcpy, which can overwrite up to 32 bytes beyond dstEnd
|
460
|
+
* this version copies two times 16 bytes (instead of one time 32 bytes)
|
461
|
+
* because it must be compatible with offsets >= 16. */
|
462
|
+
LZ4_FORCE_INLINE void
|
463
|
+
LZ4_wildCopy32(void* dstPtr, const void* srcPtr, void* dstEnd)
|
464
|
+
{
|
465
|
+
BYTE* d = (BYTE*)dstPtr;
|
466
|
+
const BYTE* s = (const BYTE*)srcPtr;
|
467
|
+
BYTE* const e = (BYTE*)dstEnd;
|
468
|
+
|
469
|
+
do { LZ4_memcpy(d,s,16); LZ4_memcpy(d+16,s+16,16); d+=32; s+=32; } while (d<e);
|
470
|
+
}
|
471
|
+
|
472
|
+
/* LZ4_memcpy_using_offset() presumes :
|
473
|
+
* - dstEnd >= dstPtr + MINMATCH
|
474
|
+
* - there is at least 8 bytes available to write after dstEnd */
|
475
|
+
LZ4_FORCE_INLINE void
|
476
|
+
LZ4_memcpy_using_offset(BYTE* dstPtr, const BYTE* srcPtr, BYTE* dstEnd, const size_t offset)
|
477
|
+
{
|
478
|
+
BYTE v[8];
|
479
|
+
|
480
|
+
assert(dstEnd >= dstPtr + MINMATCH);
|
481
|
+
|
482
|
+
switch(offset) {
|
483
|
+
case 1:
|
484
|
+
MEM_INIT(v, *srcPtr, 8);
|
485
|
+
break;
|
486
|
+
case 2:
|
487
|
+
LZ4_memcpy(v, srcPtr, 2);
|
488
|
+
LZ4_memcpy(&v[2], srcPtr, 2);
|
489
|
+
#if defined(_MSC_VER) && (_MSC_VER <= 1933) /* MSVC 2022 ver 17.3 or earlier */
|
490
|
+
# pragma warning(push)
|
491
|
+
# pragma warning(disable : 6385) /* warning C6385: Reading invalid data from 'v'. */
|
492
|
+
#endif
|
493
|
+
LZ4_memcpy(&v[4], v, 4);
|
494
|
+
#if defined(_MSC_VER) && (_MSC_VER <= 1933) /* MSVC 2022 ver 17.3 or earlier */
|
495
|
+
# pragma warning(pop)
|
496
|
+
#endif
|
497
|
+
break;
|
498
|
+
case 4:
|
499
|
+
LZ4_memcpy(v, srcPtr, 4);
|
500
|
+
LZ4_memcpy(&v[4], srcPtr, 4);
|
501
|
+
break;
|
502
|
+
default:
|
503
|
+
LZ4_memcpy_using_offset_base(dstPtr, srcPtr, dstEnd, offset);
|
504
|
+
return;
|
505
|
+
}
|
506
|
+
|
507
|
+
LZ4_memcpy(dstPtr, v, 8);
|
508
|
+
dstPtr += 8;
|
509
|
+
while (dstPtr < dstEnd) {
|
510
|
+
LZ4_memcpy(dstPtr, v, 8);
|
511
|
+
dstPtr += 8;
|
512
|
+
}
|
513
|
+
}
|
514
|
+
#endif
|
515
|
+
|
516
|
+
|
517
|
+
/*-************************************
|
518
|
+
* Common functions
|
519
|
+
**************************************/
|
520
|
+
static unsigned LZ4_NbCommonBytes (reg_t val)
|
521
|
+
{
|
522
|
+
assert(val != 0);
|
523
|
+
if (LZ4_isLittleEndian()) {
|
524
|
+
if (sizeof(val) == 8) {
|
525
|
+
const U64 m = 0x0101010101010101ULL;
|
526
|
+
val ^= val - 1;
|
527
|
+
return (unsigned)(((U64)((val & (m - 1)) * m)) >> 56);
|
528
|
+
} else /* 32 bits */ {
|
529
|
+
const U32 m = 0x01010101;
|
530
|
+
return (unsigned)((((val - 1) ^ val) & (m - 1)) * m) >> 24;
|
531
|
+
}
|
532
|
+
} else /* Big Endian CPU */ {
|
533
|
+
assert(false);
|
534
|
+
return 0;
|
535
|
+
}
|
536
|
+
}
|
537
|
+
|
538
|
+
|
539
|
+
#define STEPSIZE sizeof(reg_t)
|
540
|
+
LZ4_FORCE_INLINE
|
541
|
+
unsigned LZ4_count(const BYTE* pIn, const BYTE* pMatch, const BYTE* pInLimit)
|
542
|
+
{
|
543
|
+
const BYTE* const pStart = pIn;
|
544
|
+
|
545
|
+
if (likely(pIn < pInLimit-(STEPSIZE-1))) {
|
546
|
+
reg_t const diff = LZ4_read_ARCH(pMatch) ^ LZ4_read_ARCH(pIn);
|
547
|
+
if (!diff) {
|
548
|
+
pIn+=STEPSIZE; pMatch+=STEPSIZE;
|
549
|
+
} else {
|
550
|
+
return LZ4_NbCommonBytes(diff);
|
551
|
+
} }
|
552
|
+
|
553
|
+
while (likely(pIn < pInLimit-(STEPSIZE-1))) {
|
554
|
+
reg_t const diff = LZ4_read_ARCH(pMatch) ^ LZ4_read_ARCH(pIn);
|
555
|
+
if (!diff) { pIn+=STEPSIZE; pMatch+=STEPSIZE; continue; }
|
556
|
+
pIn += LZ4_NbCommonBytes(diff);
|
557
|
+
return (unsigned)(pIn - pStart);
|
558
|
+
}
|
559
|
+
|
560
|
+
if ((STEPSIZE==8) && (pIn<(pInLimit-3)) && (LZ4_read32(pMatch) == LZ4_read32(pIn))) { pIn+=4; pMatch+=4; }
|
561
|
+
if ((pIn<(pInLimit-1)) && (LZ4_read16(pMatch) == LZ4_read16(pIn))) { pIn+=2; pMatch+=2; }
|
562
|
+
if ((pIn<pInLimit) && (*pMatch == *pIn)) pIn++;
|
563
|
+
return (unsigned)(pIn - pStart);
|
564
|
+
}
|
565
|
+
|
566
|
+
|
567
|
+
#ifndef LZ4_COMMONDEFS_ONLY
|
568
|
+
/*-************************************
|
569
|
+
* Local Constants
|
570
|
+
**************************************/
|
571
|
+
static const int LZ4_64Klimit = ((64 KB) + (MFLIMIT-1));
|
572
|
+
static const U32 LZ4_skipTrigger = 6; /* Increase this value ==> compression run slower on incompressible data */
|
573
|
+
|
574
|
+
|
575
|
+
/*-************************************
|
576
|
+
* Local Structures and types
|
577
|
+
**************************************/
|
578
|
+
typedef enum { clearedTable = 0, byPtr, byU32, byU16 } tableType_t;
|
579
|
+
|
580
|
+
/**
|
581
|
+
* This enum distinguishes several different modes of accessing previous
|
582
|
+
* content in the stream.
|
583
|
+
*
|
584
|
+
* - noDict : There is no preceding content.
|
585
|
+
* - withPrefix64k : Table entries up to ctx->dictSize before the current blob
|
586
|
+
* blob being compressed are valid and refer to the preceding
|
587
|
+
* content (of length ctx->dictSize), which is available
|
588
|
+
* contiguously preceding in memory the content currently
|
589
|
+
* being compressed.
|
590
|
+
* - usingExtDict : Like withPrefix64k, but the preceding content is somewhere
|
591
|
+
* else in memory, starting at ctx->dictionary with length
|
592
|
+
* ctx->dictSize.
|
593
|
+
* - usingDictCtx : Everything concerning the preceding content is
|
594
|
+
* in a separate context, pointed to by ctx->dictCtx.
|
595
|
+
* ctx->dictionary, ctx->dictSize, and table entries
|
596
|
+
* in the current context that refer to positions
|
597
|
+
* preceding the beginning of the current compression are
|
598
|
+
* ignored. Instead, ctx->dictCtx->dictionary and ctx->dictCtx
|
599
|
+
* ->dictSize describe the location and size of the preceding
|
600
|
+
* content, and matches are found by looking in the ctx
|
601
|
+
* ->dictCtx->hashTable.
|
602
|
+
*/
|
603
|
+
typedef enum { noDict = 0, withPrefix64k, usingExtDict, usingDictCtx } dict_directive;
|
604
|
+
typedef enum { noDictIssue = 0, dictSmall } dictIssue_directive;
|
605
|
+
|
606
|
+
|
607
|
+
/*-************************************
|
608
|
+
* Local Utils
|
609
|
+
**************************************/
|
610
|
+
int LZ4_versionNumber (void) { return LZ4_VERSION_NUMBER; }
|
611
|
+
const char* LZ4_versionString(void) { return LZ4_VERSION_STRING; }
|
612
|
+
int LZ4_compressBound(int isize) { return LZ4_COMPRESSBOUND(isize); }
|
613
|
+
int LZ4_sizeofState(void) { return sizeof(LZ4_stream_t); }
|
614
|
+
|
615
|
+
|
616
|
+
/*-****************************************
|
617
|
+
* Internal Definitions, used only in Tests
|
618
|
+
*******************************************/
|
619
|
+
//#if defined (__cplusplus)
|
620
|
+
//extern "C" {
|
621
|
+
//#endif
|
622
|
+
|
623
|
+
int LZ4_compress_forceExtDict (LZ4_stream_t* LZ4_dict, const char* source, char* dest, int srcSize);
|
624
|
+
|
625
|
+
int LZ4_decompress_safe_forceExtDict(const char* source, char* dest,
|
626
|
+
int compressedSize, int maxOutputSize,
|
627
|
+
const void* dictStart, size_t dictSize);
|
628
|
+
int LZ4_decompress_safe_partial_forceExtDict(const char* source, char* dest,
|
629
|
+
int compressedSize, int targetOutputSize, int dstCapacity,
|
630
|
+
const void* dictStart, size_t dictSize);
|
631
|
+
//#if defined (__cplusplus)
|
632
|
+
//}
|
633
|
+
//#endif
|
634
|
+
|
635
|
+
/*-******************************
|
636
|
+
* Compression functions
|
637
|
+
********************************/
|
638
|
+
LZ4_FORCE_INLINE U32 LZ4_hash4(U32 sequence, tableType_t const tableType)
|
639
|
+
{
|
640
|
+
if (tableType == byU16)
|
641
|
+
return ((sequence * 2654435761U) >> ((MINMATCH*8)-(LZ4_HASHLOG+1)));
|
642
|
+
else
|
643
|
+
return ((sequence * 2654435761U) >> ((MINMATCH*8)-LZ4_HASHLOG));
|
644
|
+
}
|
645
|
+
|
646
|
+
LZ4_FORCE_INLINE U32 LZ4_hash5(U64 sequence, tableType_t const tableType)
|
647
|
+
{
|
648
|
+
const U32 hashLog = (tableType == byU16) ? LZ4_HASHLOG+1 : LZ4_HASHLOG;
|
649
|
+
if (LZ4_isLittleEndian()) {
|
650
|
+
const U64 prime5bytes = 889523592379ULL;
|
651
|
+
return (U32)(((sequence << 24) * prime5bytes) >> (64 - hashLog));
|
652
|
+
} else {
|
653
|
+
const U64 prime8bytes = 11400714785074694791ULL;
|
654
|
+
return (U32)(((sequence >> 24) * prime8bytes) >> (64 - hashLog));
|
655
|
+
}
|
656
|
+
}
|
657
|
+
|
658
|
+
LZ4_FORCE_INLINE U32 LZ4_hashPosition(const void* const p, tableType_t const tableType)
|
659
|
+
{
|
660
|
+
if ((sizeof(reg_t)==8) && (tableType != byU16)) return LZ4_hash5(LZ4_read_ARCH(p), tableType);
|
661
|
+
return LZ4_hash4(LZ4_read32(p), tableType);
|
662
|
+
}
|
663
|
+
|
664
|
+
LZ4_FORCE_INLINE void LZ4_clearHash(U32 h, void* tableBase, tableType_t const tableType)
|
665
|
+
{
|
666
|
+
switch (tableType)
|
667
|
+
{
|
668
|
+
default: /* fallthrough */
|
669
|
+
case clearedTable: { /* illegal! */ assert(0); return; }
|
670
|
+
case byPtr: { const BYTE** hashTable = (const BYTE**)tableBase; hashTable[h] = NULL; return; }
|
671
|
+
case byU32: { U32* hashTable = (U32*) tableBase; hashTable[h] = 0; return; }
|
672
|
+
case byU16: { U16* hashTable = (U16*) tableBase; hashTable[h] = 0; return; }
|
673
|
+
}
|
674
|
+
}
|
675
|
+
|
676
|
+
LZ4_FORCE_INLINE void LZ4_putIndexOnHash(U32 idx, U32 h, void* tableBase, tableType_t const tableType)
|
677
|
+
{
|
678
|
+
switch (tableType)
|
679
|
+
{
|
680
|
+
default: /* fallthrough */
|
681
|
+
case clearedTable: /* fallthrough */
|
682
|
+
case byPtr: { /* illegal! */ assert(0); return; }
|
683
|
+
case byU32: { U32* hashTable = (U32*) tableBase; hashTable[h] = idx; return; }
|
684
|
+
case byU16: { U16* hashTable = (U16*) tableBase; assert(idx < 65536); hashTable[h] = (U16)idx; return; }
|
685
|
+
}
|
686
|
+
}
|
687
|
+
|
688
|
+
LZ4_FORCE_INLINE void LZ4_putPositionOnHash(const BYTE* p, U32 h,
|
689
|
+
void* tableBase, tableType_t const tableType,
|
690
|
+
const BYTE* srcBase)
|
691
|
+
{
|
692
|
+
switch (tableType)
|
693
|
+
{
|
694
|
+
case clearedTable: { /* illegal! */ assert(0); return; }
|
695
|
+
case byPtr: { const BYTE** hashTable = (const BYTE**)tableBase; hashTable[h] = p; return; }
|
696
|
+
case byU32: { U32* hashTable = (U32*) tableBase; hashTable[h] = (U32)(p-srcBase); return; }
|
697
|
+
case byU16: { U16* hashTable = (U16*) tableBase; hashTable[h] = (U16)(p-srcBase); return; }
|
698
|
+
}
|
699
|
+
}
|
700
|
+
|
701
|
+
LZ4_FORCE_INLINE void LZ4_putPosition(const BYTE* p, void* tableBase, tableType_t tableType, const BYTE* srcBase)
|
702
|
+
{
|
703
|
+
U32 const h = LZ4_hashPosition(p, tableType);
|
704
|
+
LZ4_putPositionOnHash(p, h, tableBase, tableType, srcBase);
|
705
|
+
}
|
706
|
+
|
707
|
+
/* LZ4_getIndexOnHash() :
|
708
|
+
* Index of match position registered in hash table.
|
709
|
+
* hash position must be calculated by using base+index, or dictBase+index.
|
710
|
+
* Assumption 1 : only valid if tableType == byU32 or byU16.
|
711
|
+
* Assumption 2 : h is presumed valid (within limits of hash table)
|
712
|
+
*/
|
713
|
+
LZ4_FORCE_INLINE U32 LZ4_getIndexOnHash(U32 h, const void* tableBase, tableType_t tableType)
|
714
|
+
{
|
715
|
+
LZ4_STATIC_ASSERT(LZ4_MEMORY_USAGE > 2);
|
716
|
+
if (tableType == byU32) {
|
717
|
+
const U32* const hashTable = (const U32*) tableBase;
|
718
|
+
assert(h < (1U << (LZ4_MEMORY_USAGE-2)));
|
719
|
+
return hashTable[h];
|
720
|
+
}
|
721
|
+
if (tableType == byU16) {
|
722
|
+
const U16* const hashTable = (const U16*) tableBase;
|
723
|
+
assert(h < (1U << (LZ4_MEMORY_USAGE-1)));
|
724
|
+
return hashTable[h];
|
725
|
+
}
|
726
|
+
assert(0); return 0; /* forbidden case */
|
727
|
+
}
|
728
|
+
|
729
|
+
static const BYTE* LZ4_getPositionOnHash(U32 h, const void* tableBase, tableType_t tableType, const BYTE* srcBase)
|
730
|
+
{
|
731
|
+
if (tableType == byPtr) { const BYTE* const* hashTable = (const BYTE* const*) tableBase; return hashTable[h]; }
|
732
|
+
if (tableType == byU32) { const U32* const hashTable = (const U32*) tableBase; return hashTable[h] + srcBase; }
|
733
|
+
{ const U16* const hashTable = (const U16*) tableBase; return hashTable[h] + srcBase; } /* default, to ensure a return */
|
734
|
+
}
|
735
|
+
|
736
|
+
LZ4_FORCE_INLINE const BYTE*
|
737
|
+
LZ4_getPosition(const BYTE* p,
|
738
|
+
const void* tableBase, tableType_t tableType,
|
739
|
+
const BYTE* srcBase)
|
740
|
+
{
|
741
|
+
U32 const h = LZ4_hashPosition(p, tableType);
|
742
|
+
return LZ4_getPositionOnHash(h, tableBase, tableType, srcBase);
|
743
|
+
}
|
744
|
+
|
745
|
+
LZ4_FORCE_INLINE void
|
746
|
+
LZ4_prepareTable(LZ4_stream_t_internal* const cctx,
|
747
|
+
const int inputSize,
|
748
|
+
const tableType_t tableType) {
|
749
|
+
/* If the table hasn't been used, it's guaranteed to be zeroed out, and is
|
750
|
+
* therefore safe to use no matter what mode we're in. Otherwise, we figure
|
751
|
+
* out if it's safe to leave as is or whether it needs to be reset.
|
752
|
+
*/
|
753
|
+
if ((tableType_t)cctx->tableType != clearedTable) {
|
754
|
+
assert(inputSize >= 0);
|
755
|
+
if ((tableType_t)cctx->tableType != tableType
|
756
|
+
|| ((tableType == byU16) && cctx->currentOffset + (unsigned)inputSize >= 0xFFFFU)
|
757
|
+
|| ((tableType == byU32) && cctx->currentOffset > 1 GB)
|
758
|
+
|| tableType == byPtr
|
759
|
+
|| inputSize >= 4 KB)
|
760
|
+
{
|
761
|
+
DEBUGLOG(4, "LZ4_prepareTable: Resetting table in %p", cctx);
|
762
|
+
MEM_INIT(cctx->hashTable, 0, LZ4_HASHTABLESIZE);
|
763
|
+
cctx->currentOffset = 0;
|
764
|
+
cctx->tableType = (U32)clearedTable;
|
765
|
+
} else {
|
766
|
+
DEBUGLOG(4, "LZ4_prepareTable: Re-use hash table (no reset)");
|
767
|
+
}
|
768
|
+
}
|
769
|
+
|
770
|
+
/* Adding a gap, so all previous entries are > LZ4_DISTANCE_MAX back,
|
771
|
+
* is faster than compressing without a gap.
|
772
|
+
* However, compressing with currentOffset == 0 is faster still,
|
773
|
+
* so we preserve that case.
|
774
|
+
*/
|
775
|
+
if (cctx->currentOffset != 0 && tableType == byU32) {
|
776
|
+
DEBUGLOG(5, "LZ4_prepareTable: adding 64KB to currentOffset");
|
777
|
+
cctx->currentOffset += 64 KB;
|
778
|
+
}
|
779
|
+
|
780
|
+
/* Finally, clear history */
|
781
|
+
cctx->dictCtx = NULL;
|
782
|
+
cctx->dictionary = NULL;
|
783
|
+
cctx->dictSize = 0;
|
784
|
+
}
|
785
|
+
|
786
|
+
/** LZ4_compress_generic() :
|
787
|
+
* inlined, to ensure branches are decided at compilation time.
|
788
|
+
* Presumed already validated at this stage:
|
789
|
+
* - source != NULL
|
790
|
+
* - inputSize > 0
|
791
|
+
*/
|
792
|
+
LZ4_FORCE_INLINE int LZ4_compress_generic_validated(
|
793
|
+
LZ4_stream_t_internal* const cctx,
|
794
|
+
const char* const source,
|
795
|
+
char* const dest,
|
796
|
+
const int inputSize,
|
797
|
+
int* inputConsumed, /* only written when outputDirective == fillOutput */
|
798
|
+
const int maxOutputSize,
|
799
|
+
const limitedOutput_directive outputDirective,
|
800
|
+
const tableType_t tableType,
|
801
|
+
const dict_directive dictDirective,
|
802
|
+
const dictIssue_directive dictIssue,
|
803
|
+
const int acceleration)
|
804
|
+
{
|
805
|
+
int result;
|
806
|
+
const BYTE* ip = (const BYTE*) source;
|
807
|
+
|
808
|
+
U32 const startIndex = cctx->currentOffset;
|
809
|
+
const BYTE* base = (const BYTE*) source - startIndex;
|
810
|
+
const BYTE* lowLimit;
|
811
|
+
|
812
|
+
const LZ4_stream_t_internal* dictCtx = (const LZ4_stream_t_internal*) cctx->dictCtx;
|
813
|
+
const BYTE* const dictionary =
|
814
|
+
dictDirective == usingDictCtx ? dictCtx->dictionary : cctx->dictionary;
|
815
|
+
const U32 dictSize =
|
816
|
+
dictDirective == usingDictCtx ? dictCtx->dictSize : cctx->dictSize;
|
817
|
+
const U32 dictDelta = (dictDirective == usingDictCtx) ? startIndex - dictCtx->currentOffset : 0; /* make indexes in dictCtx comparable with index in current context */
|
818
|
+
|
819
|
+
int const maybe_extMem = (dictDirective == usingExtDict) || (dictDirective == usingDictCtx);
|
820
|
+
U32 const prefixIdxLimit = startIndex - dictSize; /* used when dictDirective == dictSmall */
|
821
|
+
const BYTE* const dictEnd = dictionary ? dictionary + dictSize : dictionary;
|
822
|
+
const BYTE* anchor = (const BYTE*) source;
|
823
|
+
const BYTE* const iend = ip + inputSize;
|
824
|
+
const BYTE* const mflimitPlusOne = iend - MFLIMIT + 1;
|
825
|
+
const BYTE* const matchlimit = iend - LASTLITERALS;
|
826
|
+
|
827
|
+
/* the dictCtx currentOffset is indexed on the start of the dictionary,
|
828
|
+
* while a dictionary in the current context precedes the currentOffset */
|
829
|
+
const BYTE* dictBase = (dictionary == NULL) ? NULL :
|
830
|
+
(dictDirective == usingDictCtx) ?
|
831
|
+
dictionary + dictSize - dictCtx->currentOffset :
|
832
|
+
dictionary + dictSize - startIndex;
|
833
|
+
|
834
|
+
BYTE* op = (BYTE*) dest;
|
835
|
+
BYTE* const olimit = op + maxOutputSize;
|
836
|
+
|
837
|
+
U32 offset = 0;
|
838
|
+
U32 forwardH;
|
839
|
+
|
840
|
+
DEBUGLOG(5, "LZ4_compress_generic_validated: srcSize=%i, tableType=%u", inputSize, tableType);
|
841
|
+
assert(ip != NULL);
|
842
|
+
/* If init conditions are not met, we don't have to mark stream
|
843
|
+
* as having dirty context, since no action was taken yet */
|
844
|
+
if (outputDirective == fillOutput && maxOutputSize < 1) { return 0; } /* Impossible to store anything */
|
845
|
+
if ((tableType == byU16) && (inputSize>=LZ4_64Klimit)) { return 0; } /* Size too large (not within 64K limit) */
|
846
|
+
if (tableType==byPtr) assert(dictDirective==noDict); /* only supported use case with byPtr */
|
847
|
+
assert(acceleration >= 1);
|
848
|
+
|
849
|
+
lowLimit = (const BYTE*)source - (dictDirective == withPrefix64k ? dictSize : 0);
|
850
|
+
|
851
|
+
/* Update context state */
|
852
|
+
if (dictDirective == usingDictCtx) {
|
853
|
+
/* Subsequent linked blocks can't use the dictionary. */
|
854
|
+
/* Instead, they use the block we just compressed. */
|
855
|
+
cctx->dictCtx = NULL;
|
856
|
+
cctx->dictSize = (U32)inputSize;
|
857
|
+
} else {
|
858
|
+
cctx->dictSize += (U32)inputSize;
|
859
|
+
}
|
860
|
+
cctx->currentOffset += (U32)inputSize;
|
861
|
+
cctx->tableType = (U32)tableType;
|
862
|
+
|
863
|
+
if (inputSize<LZ4_minLength) goto _last_literals; /* Input too small, no compression (all literals) */
|
864
|
+
|
865
|
+
/* First Byte */
|
866
|
+
LZ4_putPosition(ip, cctx->hashTable, tableType, base);
|
867
|
+
ip++; forwardH = LZ4_hashPosition(ip, tableType);
|
868
|
+
|
869
|
+
/* Main Loop */
|
870
|
+
for ( ; ; ) {
|
871
|
+
const BYTE* match;
|
872
|
+
BYTE* token;
|
873
|
+
const BYTE* filledIp;
|
874
|
+
|
875
|
+
/* Find a match */
|
876
|
+
if (tableType == byPtr) {
|
877
|
+
const BYTE* forwardIp = ip;
|
878
|
+
int step = 1;
|
879
|
+
int searchMatchNb = acceleration << LZ4_skipTrigger;
|
880
|
+
do {
|
881
|
+
U32 const h = forwardH;
|
882
|
+
ip = forwardIp;
|
883
|
+
forwardIp += step;
|
884
|
+
step = (searchMatchNb++ >> LZ4_skipTrigger);
|
885
|
+
|
886
|
+
if (unlikely(forwardIp > mflimitPlusOne)) goto _last_literals;
|
887
|
+
assert(ip < mflimitPlusOne);
|
888
|
+
|
889
|
+
match = LZ4_getPositionOnHash(h, cctx->hashTable, tableType, base);
|
890
|
+
forwardH = LZ4_hashPosition(forwardIp, tableType);
|
891
|
+
LZ4_putPositionOnHash(ip, h, cctx->hashTable, tableType, base);
|
892
|
+
|
893
|
+
} while ( (match+LZ4_DISTANCE_MAX < ip)
|
894
|
+
|| (LZ4_read32(match) != LZ4_read32(ip)) );
|
895
|
+
|
896
|
+
} else { /* byU32, byU16 */
|
897
|
+
|
898
|
+
const BYTE* forwardIp = ip;
|
899
|
+
int step = 1;
|
900
|
+
int searchMatchNb = acceleration << LZ4_skipTrigger;
|
901
|
+
do {
|
902
|
+
U32 const h = forwardH;
|
903
|
+
U32 const current = (U32)(forwardIp - base);
|
904
|
+
U32 matchIndex = LZ4_getIndexOnHash(h, cctx->hashTable, tableType);
|
905
|
+
assert(matchIndex <= current);
|
906
|
+
assert(forwardIp - base < (ptrdiff_t)(2 GB - 1));
|
907
|
+
ip = forwardIp;
|
908
|
+
forwardIp += step;
|
909
|
+
step = (searchMatchNb++ >> LZ4_skipTrigger);
|
910
|
+
|
911
|
+
if (unlikely(forwardIp > mflimitPlusOne)) goto _last_literals;
|
912
|
+
assert(ip < mflimitPlusOne);
|
913
|
+
|
914
|
+
if (dictDirective == usingDictCtx) {
|
915
|
+
if (matchIndex < startIndex) {
|
916
|
+
/* there was no match, try the dictionary */
|
917
|
+
assert(tableType == byU32);
|
918
|
+
matchIndex = LZ4_getIndexOnHash(h, dictCtx->hashTable, byU32);
|
919
|
+
match = dictBase + matchIndex;
|
920
|
+
matchIndex += dictDelta; /* make dictCtx index comparable with current context */
|
921
|
+
lowLimit = dictionary;
|
922
|
+
} else {
|
923
|
+
match = base + matchIndex;
|
924
|
+
lowLimit = (const BYTE*)source;
|
925
|
+
}
|
926
|
+
} else if (dictDirective == usingExtDict) {
|
927
|
+
if (matchIndex < startIndex) {
|
928
|
+
DEBUGLOG(7, "extDict candidate: matchIndex=%5u < startIndex=%5u", matchIndex, startIndex);
|
929
|
+
assert(startIndex - matchIndex >= MINMATCH);
|
930
|
+
assert(dictBase);
|
931
|
+
match = dictBase + matchIndex;
|
932
|
+
lowLimit = dictionary;
|
933
|
+
} else {
|
934
|
+
match = base + matchIndex;
|
935
|
+
lowLimit = (const BYTE*)source;
|
936
|
+
}
|
937
|
+
} else { /* single continuous memory segment */
|
938
|
+
match = base + matchIndex;
|
939
|
+
}
|
940
|
+
forwardH = LZ4_hashPosition(forwardIp, tableType);
|
941
|
+
LZ4_putIndexOnHash(current, h, cctx->hashTable, tableType);
|
942
|
+
|
943
|
+
DEBUGLOG(7, "candidate at pos=%u (offset=%u \n", matchIndex, current - matchIndex);
|
944
|
+
if ((dictIssue == dictSmall) && (matchIndex < prefixIdxLimit)) { continue; } /* match outside of valid area */
|
945
|
+
assert(matchIndex < current);
|
946
|
+
if ( ((tableType != byU16) || (LZ4_DISTANCE_MAX < LZ4_DISTANCE_ABSOLUTE_MAX))
|
947
|
+
&& (matchIndex+LZ4_DISTANCE_MAX < current)) {
|
948
|
+
continue;
|
949
|
+
} /* too far */
|
950
|
+
assert((current - matchIndex) <= LZ4_DISTANCE_MAX); /* match now expected within distance */
|
951
|
+
|
952
|
+
if (LZ4_read32(match) == LZ4_read32(ip)) {
|
953
|
+
if (maybe_extMem) offset = current - matchIndex;
|
954
|
+
break; /* match found */
|
955
|
+
}
|
956
|
+
|
957
|
+
} while(1);
|
958
|
+
}
|
959
|
+
|
960
|
+
/* Catch up */
|
961
|
+
filledIp = ip;
|
962
|
+
while (((ip>anchor) & (match > lowLimit)) && (unlikely(ip[-1]==match[-1]))) { ip--; match--; }
|
963
|
+
|
964
|
+
/* Encode Literals */
|
965
|
+
{ unsigned const litLength = (unsigned)(ip - anchor);
|
966
|
+
token = op++;
|
967
|
+
if ((outputDirective == limitedOutput) && /* Check output buffer overflow */
|
968
|
+
(unlikely(op + litLength + (2 + 1 + LASTLITERALS) + (litLength/255) > olimit)) ) {
|
969
|
+
return 0; /* cannot compress within `dst` budget. Stored indexes in hash table are nonetheless fine */
|
970
|
+
}
|
971
|
+
if ((outputDirective == fillOutput) &&
|
972
|
+
(unlikely(op + (litLength+240)/255 /* litlen */ + litLength /* literals */ + 2 /* offset */ + 1 /* token */ + MFLIMIT - MINMATCH /* min last literals so last match is <= end - MFLIMIT */ > olimit))) {
|
973
|
+
op--;
|
974
|
+
goto _last_literals;
|
975
|
+
}
|
976
|
+
if (litLength >= RUN_MASK) {
|
977
|
+
int len = (int)(litLength - RUN_MASK);
|
978
|
+
*token = (RUN_MASK<<ML_BITS);
|
979
|
+
for(; len >= 255 ; len-=255) *op++ = 255;
|
980
|
+
*op++ = (BYTE)len;
|
981
|
+
}
|
982
|
+
else *token = (BYTE)(litLength<<ML_BITS);
|
983
|
+
|
984
|
+
/* Copy Literals */
|
985
|
+
LZ4_wildCopy8(op, anchor, op+litLength);
|
986
|
+
op+=litLength;
|
987
|
+
DEBUGLOG(6, "seq.start:%i, literals=%u, match.start:%i",
|
988
|
+
(int)(anchor-(const BYTE*)source), litLength, (int)(ip-(const BYTE*)source));
|
989
|
+
}
|
990
|
+
|
991
|
+
_next_match:
|
992
|
+
/* at this stage, the following variables must be correctly set :
|
993
|
+
* - ip : at start of LZ operation
|
994
|
+
* - match : at start of previous pattern occurrence; can be within current prefix, or within extDict
|
995
|
+
* - offset : if maybe_ext_memSegment==1 (constant)
|
996
|
+
* - lowLimit : must be == dictionary to mean "match is within extDict"; must be == source otherwise
|
997
|
+
* - token and *token : position to write 4-bits for match length; higher 4-bits for literal length supposed already written
|
998
|
+
*/
|
999
|
+
|
1000
|
+
if ((outputDirective == fillOutput) &&
|
1001
|
+
(op + 2 /* offset */ + 1 /* token */ + MFLIMIT - MINMATCH /* min last literals so last match is <= end - MFLIMIT */ > olimit)) {
|
1002
|
+
/* the match was too close to the end, rewind and go to last literals */
|
1003
|
+
op = token;
|
1004
|
+
goto _last_literals;
|
1005
|
+
}
|
1006
|
+
|
1007
|
+
/* Encode Offset */
|
1008
|
+
if (maybe_extMem) { /* static test */
|
1009
|
+
DEBUGLOG(6, " with offset=%u (ext if > %i)", offset, (int)(ip - (const BYTE*)source));
|
1010
|
+
assert(offset <= LZ4_DISTANCE_MAX && offset > 0);
|
1011
|
+
LZ4_writeLE16(op, (U16)offset); op+=2;
|
1012
|
+
} else {
|
1013
|
+
DEBUGLOG(6, " with offset=%u (same segment)", (U32)(ip - match));
|
1014
|
+
assert(ip-match <= LZ4_DISTANCE_MAX);
|
1015
|
+
LZ4_writeLE16(op, (U16)(ip - match)); op+=2;
|
1016
|
+
}
|
1017
|
+
|
1018
|
+
/* Encode MatchLength */
|
1019
|
+
{ unsigned matchCode;
|
1020
|
+
|
1021
|
+
if ( (dictDirective==usingExtDict || dictDirective==usingDictCtx)
|
1022
|
+
&& (lowLimit==dictionary) /* match within extDict */ ) {
|
1023
|
+
const BYTE* limit = ip + (dictEnd-match);
|
1024
|
+
assert(dictEnd > match);
|
1025
|
+
if (limit > matchlimit) limit = matchlimit;
|
1026
|
+
matchCode = LZ4_count(ip+MINMATCH, match+MINMATCH, limit);
|
1027
|
+
ip += (size_t)matchCode + MINMATCH;
|
1028
|
+
if (ip==limit) {
|
1029
|
+
unsigned const more = LZ4_count(limit, (const BYTE*)source, matchlimit);
|
1030
|
+
matchCode += more;
|
1031
|
+
ip += more;
|
1032
|
+
}
|
1033
|
+
DEBUGLOG(6, " with matchLength=%u starting in extDict", matchCode+MINMATCH);
|
1034
|
+
} else {
|
1035
|
+
matchCode = LZ4_count(ip+MINMATCH, match+MINMATCH, matchlimit);
|
1036
|
+
ip += (size_t)matchCode + MINMATCH;
|
1037
|
+
DEBUGLOG(6, " with matchLength=%u", matchCode+MINMATCH);
|
1038
|
+
}
|
1039
|
+
|
1040
|
+
if ((outputDirective) && /* Check output buffer overflow */
|
1041
|
+
(unlikely(op + (1 + LASTLITERALS) + (matchCode+240)/255 > olimit)) ) {
|
1042
|
+
if (outputDirective == fillOutput) {
|
1043
|
+
/* Match description too long : reduce it */
|
1044
|
+
U32 newMatchCode = 15 /* in token */ - 1 /* to avoid needing a zero byte */ + ((U32)(olimit - op) - 1 - LASTLITERALS) * 255;
|
1045
|
+
ip -= matchCode - newMatchCode;
|
1046
|
+
assert(newMatchCode < matchCode);
|
1047
|
+
matchCode = newMatchCode;
|
1048
|
+
if (unlikely(ip <= filledIp)) {
|
1049
|
+
/* We have already filled up to filledIp so if ip ends up less than filledIp
|
1050
|
+
* we have positions in the hash table beyond the current position. This is
|
1051
|
+
* a problem if we reuse the hash table. So we have to remove these positions
|
1052
|
+
* from the hash table.
|
1053
|
+
*/
|
1054
|
+
const BYTE* ptr;
|
1055
|
+
DEBUGLOG(5, "Clearing %u positions", (U32)(filledIp - ip));
|
1056
|
+
for (ptr = ip; ptr <= filledIp; ++ptr) {
|
1057
|
+
U32 const h = LZ4_hashPosition(ptr, tableType);
|
1058
|
+
LZ4_clearHash(h, cctx->hashTable, tableType);
|
1059
|
+
}
|
1060
|
+
}
|
1061
|
+
} else {
|
1062
|
+
assert(outputDirective == limitedOutput);
|
1063
|
+
return 0; /* cannot compress within `dst` budget. Stored indexes in hash table are nonetheless fine */
|
1064
|
+
}
|
1065
|
+
}
|
1066
|
+
if (matchCode >= ML_MASK) {
|
1067
|
+
*token += ML_MASK;
|
1068
|
+
matchCode -= ML_MASK;
|
1069
|
+
LZ4_write32(op, 0xFFFFFFFF);
|
1070
|
+
while (matchCode >= 4*255) {
|
1071
|
+
op+=4;
|
1072
|
+
LZ4_write32(op, 0xFFFFFFFF);
|
1073
|
+
matchCode -= 4*255;
|
1074
|
+
}
|
1075
|
+
op += matchCode / 255;
|
1076
|
+
*op++ = (BYTE)(matchCode % 255);
|
1077
|
+
} else
|
1078
|
+
*token += (BYTE)(matchCode);
|
1079
|
+
}
|
1080
|
+
/* Ensure we have enough space for the last literals. */
|
1081
|
+
assert(!(outputDirective == fillOutput && op + 1 + LASTLITERALS > olimit));
|
1082
|
+
|
1083
|
+
anchor = ip;
|
1084
|
+
|
1085
|
+
/* Test end of chunk */
|
1086
|
+
if (ip >= mflimitPlusOne) break;
|
1087
|
+
|
1088
|
+
/* Fill table */
|
1089
|
+
LZ4_putPosition(ip-2, cctx->hashTable, tableType, base);
|
1090
|
+
|
1091
|
+
/* Test next position */
|
1092
|
+
if (tableType == byPtr) {
|
1093
|
+
|
1094
|
+
match = LZ4_getPosition(ip, cctx->hashTable, tableType, base);
|
1095
|
+
LZ4_putPosition(ip, cctx->hashTable, tableType, base);
|
1096
|
+
if ( (match+LZ4_DISTANCE_MAX >= ip)
|
1097
|
+
&& (LZ4_read32(match) == LZ4_read32(ip)) )
|
1098
|
+
{ token=op++; *token=0; goto _next_match; }
|
1099
|
+
|
1100
|
+
} else { /* byU32, byU16 */
|
1101
|
+
|
1102
|
+
U32 const h = LZ4_hashPosition(ip, tableType);
|
1103
|
+
U32 const current = (U32)(ip-base);
|
1104
|
+
U32 matchIndex = LZ4_getIndexOnHash(h, cctx->hashTable, tableType);
|
1105
|
+
assert(matchIndex < current);
|
1106
|
+
if (dictDirective == usingDictCtx) {
|
1107
|
+
if (matchIndex < startIndex) {
|
1108
|
+
/* there was no match, try the dictionary */
|
1109
|
+
matchIndex = LZ4_getIndexOnHash(h, dictCtx->hashTable, byU32);
|
1110
|
+
match = dictBase + matchIndex;
|
1111
|
+
lowLimit = dictionary; /* required for match length counter */
|
1112
|
+
matchIndex += dictDelta;
|
1113
|
+
} else {
|
1114
|
+
match = base + matchIndex;
|
1115
|
+
lowLimit = (const BYTE*)source; /* required for match length counter */
|
1116
|
+
}
|
1117
|
+
} else if (dictDirective==usingExtDict) {
|
1118
|
+
if (matchIndex < startIndex) {
|
1119
|
+
assert(dictBase);
|
1120
|
+
match = dictBase + matchIndex;
|
1121
|
+
lowLimit = dictionary; /* required for match length counter */
|
1122
|
+
} else {
|
1123
|
+
match = base + matchIndex;
|
1124
|
+
lowLimit = (const BYTE*)source; /* required for match length counter */
|
1125
|
+
}
|
1126
|
+
} else { /* single memory segment */
|
1127
|
+
match = base + matchIndex;
|
1128
|
+
}
|
1129
|
+
LZ4_putIndexOnHash(current, h, cctx->hashTable, tableType);
|
1130
|
+
assert(matchIndex < current);
|
1131
|
+
if ( ((dictIssue==dictSmall) ? (matchIndex >= prefixIdxLimit) : 1)
|
1132
|
+
&& (((tableType==byU16) && (LZ4_DISTANCE_MAX == LZ4_DISTANCE_ABSOLUTE_MAX)) ? 1 : (matchIndex+LZ4_DISTANCE_MAX >= current))
|
1133
|
+
&& (LZ4_read32(match) == LZ4_read32(ip)) ) {
|
1134
|
+
token=op++;
|
1135
|
+
*token=0;
|
1136
|
+
if (maybe_extMem) offset = current - matchIndex;
|
1137
|
+
DEBUGLOG(6, "seq.start:%i, literals=%u, match.start:%i",
|
1138
|
+
(int)(anchor-(const BYTE*)source), 0, (int)(ip-(const BYTE*)source));
|
1139
|
+
goto _next_match;
|
1140
|
+
}
|
1141
|
+
}
|
1142
|
+
|
1143
|
+
/* Prepare next loop */
|
1144
|
+
forwardH = LZ4_hashPosition(++ip, tableType);
|
1145
|
+
|
1146
|
+
}
|
1147
|
+
|
1148
|
+
_last_literals:
|
1149
|
+
/* Encode Last Literals */
|
1150
|
+
{ size_t lastRun = (size_t)(iend - anchor);
|
1151
|
+
if ( (outputDirective) && /* Check output buffer overflow */
|
1152
|
+
(op + lastRun + 1 + ((lastRun+255-RUN_MASK)/255) > olimit)) {
|
1153
|
+
if (outputDirective == fillOutput) {
|
1154
|
+
/* adapt lastRun to fill 'dst' */
|
1155
|
+
assert(olimit >= op);
|
1156
|
+
lastRun = (size_t)(olimit-op) - 1/*token*/;
|
1157
|
+
lastRun -= (lastRun + 256 - RUN_MASK) / 256; /*additional length tokens*/
|
1158
|
+
} else {
|
1159
|
+
assert(outputDirective == limitedOutput);
|
1160
|
+
return 0; /* cannot compress within `dst` budget. Stored indexes in hash table are nonetheless fine */
|
1161
|
+
}
|
1162
|
+
}
|
1163
|
+
DEBUGLOG(6, "Final literal run : %i literals", (int)lastRun);
|
1164
|
+
if (lastRun >= RUN_MASK) {
|
1165
|
+
size_t accumulator = lastRun - RUN_MASK;
|
1166
|
+
*op++ = RUN_MASK << ML_BITS;
|
1167
|
+
for(; accumulator >= 255 ; accumulator-=255) *op++ = 255;
|
1168
|
+
*op++ = (BYTE) accumulator;
|
1169
|
+
} else {
|
1170
|
+
*op++ = (BYTE)(lastRun<<ML_BITS);
|
1171
|
+
}
|
1172
|
+
LZ4_memcpy(op, anchor, lastRun);
|
1173
|
+
ip = anchor + lastRun;
|
1174
|
+
op += lastRun;
|
1175
|
+
}
|
1176
|
+
|
1177
|
+
if (outputDirective == fillOutput) {
|
1178
|
+
*inputConsumed = (int) (((const char*)ip)-source);
|
1179
|
+
}
|
1180
|
+
result = (int)(((char*)op) - dest);
|
1181
|
+
assert(result > 0);
|
1182
|
+
DEBUGLOG(5, "LZ4_compress_generic: compressed %i bytes into %i bytes", inputSize, result);
|
1183
|
+
return result;
|
1184
|
+
}
|
1185
|
+
|
1186
|
+
/** LZ4_compress_generic() :
|
1187
|
+
* inlined, to ensure branches are decided at compilation time;
|
1188
|
+
* takes care of src == (NULL, 0)
|
1189
|
+
* and forward the rest to LZ4_compress_generic_validated */
|
1190
|
+
LZ4_FORCE_INLINE int LZ4_compress_generic(
|
1191
|
+
LZ4_stream_t_internal* const cctx,
|
1192
|
+
const char* const src,
|
1193
|
+
char* const dst,
|
1194
|
+
const int srcSize,
|
1195
|
+
int *inputConsumed, /* only written when outputDirective == fillOutput */
|
1196
|
+
const int dstCapacity,
|
1197
|
+
const limitedOutput_directive outputDirective,
|
1198
|
+
const tableType_t tableType,
|
1199
|
+
const dict_directive dictDirective,
|
1200
|
+
const dictIssue_directive dictIssue,
|
1201
|
+
const int acceleration)
|
1202
|
+
{
|
1203
|
+
DEBUGLOG(5, "LZ4_compress_generic: srcSize=%i, dstCapacity=%i",
|
1204
|
+
srcSize, dstCapacity);
|
1205
|
+
|
1206
|
+
if ((U32)srcSize > (U32)LZ4_MAX_INPUT_SIZE) { return 0; } /* Unsupported srcSize, too large (or negative) */
|
1207
|
+
if (srcSize == 0) { /* src == NULL supported if srcSize == 0 */
|
1208
|
+
if (outputDirective != notLimited && dstCapacity <= 0) return 0; /* no output, can't write anything */
|
1209
|
+
DEBUGLOG(5, "Generating an empty block");
|
1210
|
+
assert(outputDirective == notLimited || dstCapacity >= 1);
|
1211
|
+
assert(dst != NULL);
|
1212
|
+
dst[0] = 0;
|
1213
|
+
if (outputDirective == fillOutput) {
|
1214
|
+
assert (inputConsumed != NULL);
|
1215
|
+
*inputConsumed = 0;
|
1216
|
+
}
|
1217
|
+
return 1;
|
1218
|
+
}
|
1219
|
+
assert(src != NULL);
|
1220
|
+
|
1221
|
+
return LZ4_compress_generic_validated(cctx, src, dst, srcSize,
|
1222
|
+
inputConsumed, /* only written into if outputDirective == fillOutput */
|
1223
|
+
dstCapacity, outputDirective,
|
1224
|
+
tableType, dictDirective, dictIssue, acceleration);
|
1225
|
+
}
|
1226
|
+
|
1227
|
+
|
1228
|
+
int LZ4_compress_fast_extState(void* state, const char* source, char* dest, int inputSize, int maxOutputSize, int acceleration)
|
1229
|
+
{
|
1230
|
+
LZ4_stream_t_internal* const ctx = & LZ4_initStream(state, sizeof(LZ4_stream_t)) -> internal_donotuse;
|
1231
|
+
assert(ctx != NULL);
|
1232
|
+
if (acceleration < 1) acceleration = LZ4_ACCELERATION_DEFAULT;
|
1233
|
+
if (acceleration > LZ4_ACCELERATION_MAX) acceleration = LZ4_ACCELERATION_MAX;
|
1234
|
+
if (maxOutputSize >= LZ4_compressBound(inputSize)) {
|
1235
|
+
if (inputSize < LZ4_64Klimit) {
|
1236
|
+
return LZ4_compress_generic(ctx, source, dest, inputSize, NULL, 0, notLimited, byU16, noDict, noDictIssue, acceleration);
|
1237
|
+
} else {
|
1238
|
+
const tableType_t tableType = ((sizeof(void*)==4) && ((uptrval)source > LZ4_DISTANCE_MAX)) ? byPtr : byU32;
|
1239
|
+
return LZ4_compress_generic(ctx, source, dest, inputSize, NULL, 0, notLimited, tableType, noDict, noDictIssue, acceleration);
|
1240
|
+
}
|
1241
|
+
} else {
|
1242
|
+
if (inputSize < LZ4_64Klimit) {
|
1243
|
+
return LZ4_compress_generic(ctx, source, dest, inputSize, NULL, maxOutputSize, limitedOutput, byU16, noDict, noDictIssue, acceleration);
|
1244
|
+
} else {
|
1245
|
+
const tableType_t tableType = ((sizeof(void*)==4) && ((uptrval)source > LZ4_DISTANCE_MAX)) ? byPtr : byU32;
|
1246
|
+
return LZ4_compress_generic(ctx, source, dest, inputSize, NULL, maxOutputSize, limitedOutput, tableType, noDict, noDictIssue, acceleration);
|
1247
|
+
}
|
1248
|
+
}
|
1249
|
+
}
|
1250
|
+
|
1251
|
+
/**
|
1252
|
+
* LZ4_compress_fast_extState_fastReset() :
|
1253
|
+
* A variant of LZ4_compress_fast_extState().
|
1254
|
+
*
|
1255
|
+
* Using this variant avoids an expensive initialization step. It is only safe
|
1256
|
+
* to call if the state buffer is known to be correctly initialized already
|
1257
|
+
* (see comment in lz4.h on LZ4_resetStream_fast() for a definition of
|
1258
|
+
* "correctly initialized").
|
1259
|
+
*/
|
1260
|
+
int LZ4_compress_fast_extState_fastReset(void* state, const char* src, char* dst, int srcSize, int dstCapacity, int acceleration)
|
1261
|
+
{
|
1262
|
+
LZ4_stream_t_internal* ctx = &((LZ4_stream_t*)state)->internal_donotuse;
|
1263
|
+
if (acceleration < 1) acceleration = LZ4_ACCELERATION_DEFAULT;
|
1264
|
+
if (acceleration > LZ4_ACCELERATION_MAX) acceleration = LZ4_ACCELERATION_MAX;
|
1265
|
+
|
1266
|
+
if (dstCapacity >= LZ4_compressBound(srcSize)) {
|
1267
|
+
if (srcSize < LZ4_64Klimit) {
|
1268
|
+
const tableType_t tableType = byU16;
|
1269
|
+
LZ4_prepareTable(ctx, srcSize, tableType);
|
1270
|
+
if (ctx->currentOffset) {
|
1271
|
+
return LZ4_compress_generic(ctx, src, dst, srcSize, NULL, 0, notLimited, tableType, noDict, dictSmall, acceleration);
|
1272
|
+
} else {
|
1273
|
+
return LZ4_compress_generic(ctx, src, dst, srcSize, NULL, 0, notLimited, tableType, noDict, noDictIssue, acceleration);
|
1274
|
+
}
|
1275
|
+
} else {
|
1276
|
+
const tableType_t tableType = ((sizeof(void*)==4) && ((uptrval)src > LZ4_DISTANCE_MAX)) ? byPtr : byU32;
|
1277
|
+
LZ4_prepareTable(ctx, srcSize, tableType);
|
1278
|
+
return LZ4_compress_generic(ctx, src, dst, srcSize, NULL, 0, notLimited, tableType, noDict, noDictIssue, acceleration);
|
1279
|
+
}
|
1280
|
+
} else {
|
1281
|
+
if (srcSize < LZ4_64Klimit) {
|
1282
|
+
const tableType_t tableType = byU16;
|
1283
|
+
LZ4_prepareTable(ctx, srcSize, tableType);
|
1284
|
+
if (ctx->currentOffset) {
|
1285
|
+
return LZ4_compress_generic(ctx, src, dst, srcSize, NULL, dstCapacity, limitedOutput, tableType, noDict, dictSmall, acceleration);
|
1286
|
+
} else {
|
1287
|
+
return LZ4_compress_generic(ctx, src, dst, srcSize, NULL, dstCapacity, limitedOutput, tableType, noDict, noDictIssue, acceleration);
|
1288
|
+
}
|
1289
|
+
} else {
|
1290
|
+
const tableType_t tableType = ((sizeof(void*)==4) && ((uptrval)src > LZ4_DISTANCE_MAX)) ? byPtr : byU32;
|
1291
|
+
LZ4_prepareTable(ctx, srcSize, tableType);
|
1292
|
+
return LZ4_compress_generic(ctx, src, dst, srcSize, NULL, dstCapacity, limitedOutput, tableType, noDict, noDictIssue, acceleration);
|
1293
|
+
}
|
1294
|
+
}
|
1295
|
+
}
|
1296
|
+
|
1297
|
+
|
1298
|
+
int LZ4_compress_fast(const char* source, char* dest, int inputSize, int maxOutputSize, int acceleration)
|
1299
|
+
{
|
1300
|
+
int result;
|
1301
|
+
#if (LZ4_HEAPMODE)
|
1302
|
+
LZ4_stream_t* ctxPtr = (LZ4_stream_t*)ALLOC(sizeof(LZ4_stream_t)); /* malloc-calloc always properly aligned */
|
1303
|
+
if (ctxPtr == NULL) return 0;
|
1304
|
+
#else
|
1305
|
+
LZ4_stream_t ctx;
|
1306
|
+
LZ4_stream_t* const ctxPtr = &ctx;
|
1307
|
+
#endif
|
1308
|
+
result = LZ4_compress_fast_extState(ctxPtr, source, dest, inputSize, maxOutputSize, acceleration);
|
1309
|
+
|
1310
|
+
#if (LZ4_HEAPMODE)
|
1311
|
+
FREEMEM(ctxPtr);
|
1312
|
+
#endif
|
1313
|
+
return result;
|
1314
|
+
}
|
1315
|
+
|
1316
|
+
|
1317
|
+
int LZ4_compress_default(const char* src, char* dst, int srcSize, int maxOutputSize)
|
1318
|
+
{
|
1319
|
+
return LZ4_compress_fast(src, dst, srcSize, maxOutputSize, 1);
|
1320
|
+
}
|
1321
|
+
|
1322
|
+
|
1323
|
+
/* Note!: This function leaves the stream in an unclean/broken state!
|
1324
|
+
* It is not safe to subsequently use the same state with a _fastReset() or
|
1325
|
+
* _continue() call without resetting it. */
|
1326
|
+
static int LZ4_compress_destSize_extState (LZ4_stream_t* state, const char* src, char* dst, int* srcSizePtr, int targetDstSize)
|
1327
|
+
{
|
1328
|
+
void* const s = LZ4_initStream(state, sizeof (*state));
|
1329
|
+
assert(s != NULL); (void)s;
|
1330
|
+
|
1331
|
+
if (targetDstSize >= LZ4_compressBound(*srcSizePtr)) { /* compression success is guaranteed */
|
1332
|
+
return LZ4_compress_fast_extState(state, src, dst, *srcSizePtr, targetDstSize, 1);
|
1333
|
+
} else {
|
1334
|
+
if (*srcSizePtr < LZ4_64Klimit) {
|
1335
|
+
return LZ4_compress_generic(&state->internal_donotuse, src, dst, *srcSizePtr, srcSizePtr, targetDstSize, fillOutput, byU16, noDict, noDictIssue, 1);
|
1336
|
+
} else {
|
1337
|
+
tableType_t const addrMode = ((sizeof(void*)==4) && ((uptrval)src > LZ4_DISTANCE_MAX)) ? byPtr : byU32;
|
1338
|
+
return LZ4_compress_generic(&state->internal_donotuse, src, dst, *srcSizePtr, srcSizePtr, targetDstSize, fillOutput, addrMode, noDict, noDictIssue, 1);
|
1339
|
+
} }
|
1340
|
+
}
|
1341
|
+
|
1342
|
+
|
1343
|
+
int LZ4_compress_destSize(const char* src, char* dst, int* srcSizePtr, int targetDstSize)
|
1344
|
+
{
|
1345
|
+
#if (LZ4_HEAPMODE)
|
1346
|
+
LZ4_stream_t* ctx = (LZ4_stream_t*)ALLOC(sizeof(LZ4_stream_t)); /* malloc-calloc always properly aligned */
|
1347
|
+
if (ctx == NULL) return 0;
|
1348
|
+
#else
|
1349
|
+
LZ4_stream_t ctxBody;
|
1350
|
+
LZ4_stream_t* ctx = &ctxBody;
|
1351
|
+
#endif
|
1352
|
+
|
1353
|
+
int result = LZ4_compress_destSize_extState(ctx, src, dst, srcSizePtr, targetDstSize);
|
1354
|
+
|
1355
|
+
#if (LZ4_HEAPMODE)
|
1356
|
+
FREEMEM(ctx);
|
1357
|
+
#endif
|
1358
|
+
return result;
|
1359
|
+
}
|
1360
|
+
|
1361
|
+
|
1362
|
+
|
1363
|
+
/*-******************************
|
1364
|
+
* Streaming functions
|
1365
|
+
********************************/
|
1366
|
+
|
1367
|
+
#if !defined(LZ4_STATIC_LINKING_ONLY_DISABLE_MEMORY_ALLOCATION)
|
1368
|
+
LZ4_stream_t* LZ4_createStream(void)
|
1369
|
+
{
|
1370
|
+
LZ4_stream_t* const lz4s = (LZ4_stream_t*)ALLOC(sizeof(LZ4_stream_t));
|
1371
|
+
LZ4_STATIC_ASSERT(sizeof(LZ4_stream_t) >= sizeof(LZ4_stream_t_internal));
|
1372
|
+
DEBUGLOG(4, "LZ4_createStream %p", lz4s);
|
1373
|
+
if (lz4s == NULL) return NULL;
|
1374
|
+
LZ4_initStream(lz4s, sizeof(*lz4s));
|
1375
|
+
return lz4s;
|
1376
|
+
}
|
1377
|
+
#endif
|
1378
|
+
|
1379
|
+
static size_t LZ4_stream_t_alignment(void)
|
1380
|
+
{
|
1381
|
+
#if LZ4_ALIGN_TEST
|
1382
|
+
typedef struct { char c; LZ4_stream_t t; } t_a;
|
1383
|
+
return sizeof(t_a) - sizeof(LZ4_stream_t);
|
1384
|
+
#else
|
1385
|
+
return 1; /* effectively disabled */
|
1386
|
+
#endif
|
1387
|
+
}
|
1388
|
+
|
1389
|
+
LZ4_stream_t* LZ4_initStream (void* buffer, size_t size)
|
1390
|
+
{
|
1391
|
+
DEBUGLOG(5, "LZ4_initStream");
|
1392
|
+
if (buffer == NULL) { return NULL; }
|
1393
|
+
if (size < sizeof(LZ4_stream_t)) { return NULL; }
|
1394
|
+
if (!LZ4_isAligned(buffer, LZ4_stream_t_alignment())) return NULL;
|
1395
|
+
MEM_INIT(buffer, 0, sizeof(LZ4_stream_t_internal));
|
1396
|
+
return (LZ4_stream_t*)buffer;
|
1397
|
+
}
|
1398
|
+
|
1399
|
+
/* resetStream is now deprecated,
|
1400
|
+
* prefer initStream() which is more general */
|
1401
|
+
void LZ4_resetStream (LZ4_stream_t* LZ4_stream)
|
1402
|
+
{
|
1403
|
+
DEBUGLOG(5, "LZ4_resetStream (ctx:%p)", LZ4_stream);
|
1404
|
+
MEM_INIT(LZ4_stream, 0, sizeof(LZ4_stream_t_internal));
|
1405
|
+
}
|
1406
|
+
|
1407
|
+
void LZ4_resetStream_fast(LZ4_stream_t* ctx) {
|
1408
|
+
LZ4_prepareTable(&(ctx->internal_donotuse), 0, byU32);
|
1409
|
+
}
|
1410
|
+
|
1411
|
+
#if !defined(LZ4_STATIC_LINKING_ONLY_DISABLE_MEMORY_ALLOCATION)
|
1412
|
+
int LZ4_freeStream (LZ4_stream_t* LZ4_stream)
|
1413
|
+
{
|
1414
|
+
if (!LZ4_stream) return 0; /* support free on NULL */
|
1415
|
+
DEBUGLOG(5, "LZ4_freeStream %p", LZ4_stream);
|
1416
|
+
FREEMEM(LZ4_stream);
|
1417
|
+
return (0);
|
1418
|
+
}
|
1419
|
+
#endif
|
1420
|
+
|
1421
|
+
|
1422
|
+
#define HASH_UNIT sizeof(reg_t)
|
1423
|
+
int LZ4_loadDict (LZ4_stream_t* LZ4_dict, const char* dictionary, int dictSize)
|
1424
|
+
{
|
1425
|
+
LZ4_stream_t_internal* dict = &LZ4_dict->internal_donotuse;
|
1426
|
+
const tableType_t tableType = byU32;
|
1427
|
+
const BYTE* p = (const BYTE*)dictionary;
|
1428
|
+
const BYTE* const dictEnd = p + dictSize;
|
1429
|
+
const BYTE* base;
|
1430
|
+
|
1431
|
+
DEBUGLOG(4, "LZ4_loadDict (%i bytes from %p into %p)", dictSize, dictionary, LZ4_dict);
|
1432
|
+
|
1433
|
+
/* It's necessary to reset the context,
|
1434
|
+
* and not just continue it with prepareTable()
|
1435
|
+
* to avoid any risk of generating overflowing matchIndex
|
1436
|
+
* when compressing using this dictionary */
|
1437
|
+
LZ4_resetStream(LZ4_dict);
|
1438
|
+
|
1439
|
+
/* We always increment the offset by 64 KB, since, if the dict is longer,
|
1440
|
+
* we truncate it to the last 64k, and if it's shorter, we still want to
|
1441
|
+
* advance by a whole window length so we can provide the guarantee that
|
1442
|
+
* there are only valid offsets in the window, which allows an optimization
|
1443
|
+
* in LZ4_compress_fast_continue() where it uses noDictIssue even when the
|
1444
|
+
* dictionary isn't a full 64k. */
|
1445
|
+
dict->currentOffset += 64 KB;
|
1446
|
+
|
1447
|
+
if (dictSize < (int)HASH_UNIT) {
|
1448
|
+
return 0;
|
1449
|
+
}
|
1450
|
+
|
1451
|
+
if ((dictEnd - p) > 64 KB) p = dictEnd - 64 KB;
|
1452
|
+
base = dictEnd - dict->currentOffset;
|
1453
|
+
dict->dictionary = p;
|
1454
|
+
dict->dictSize = (U32)(dictEnd - p);
|
1455
|
+
dict->tableType = (U32)tableType;
|
1456
|
+
|
1457
|
+
while (p <= dictEnd-HASH_UNIT) {
|
1458
|
+
LZ4_putPosition(p, dict->hashTable, tableType, base);
|
1459
|
+
p+=3;
|
1460
|
+
}
|
1461
|
+
|
1462
|
+
return (int)dict->dictSize;
|
1463
|
+
}
|
1464
|
+
|
1465
|
+
void LZ4_attach_dictionary(LZ4_stream_t* workingStream, const LZ4_stream_t* dictionaryStream)
|
1466
|
+
{
|
1467
|
+
const LZ4_stream_t_internal* dictCtx = (dictionaryStream == NULL) ? NULL :
|
1468
|
+
&(dictionaryStream->internal_donotuse);
|
1469
|
+
|
1470
|
+
DEBUGLOG(4, "LZ4_attach_dictionary (%p, %p, size %u)",
|
1471
|
+
workingStream, dictionaryStream,
|
1472
|
+
dictCtx != NULL ? dictCtx->dictSize : 0);
|
1473
|
+
|
1474
|
+
if (dictCtx != NULL) {
|
1475
|
+
/* If the current offset is zero, we will never look in the
|
1476
|
+
* external dictionary context, since there is no value a table
|
1477
|
+
* entry can take that indicate a miss. In that case, we need
|
1478
|
+
* to bump the offset to something non-zero.
|
1479
|
+
*/
|
1480
|
+
if (workingStream->internal_donotuse.currentOffset == 0) {
|
1481
|
+
workingStream->internal_donotuse.currentOffset = 64 KB;
|
1482
|
+
}
|
1483
|
+
|
1484
|
+
/* Don't actually attach an empty dictionary.
|
1485
|
+
*/
|
1486
|
+
if (dictCtx->dictSize == 0) {
|
1487
|
+
dictCtx = NULL;
|
1488
|
+
}
|
1489
|
+
}
|
1490
|
+
workingStream->internal_donotuse.dictCtx = dictCtx;
|
1491
|
+
}
|
1492
|
+
|
1493
|
+
|
1494
|
+
static void LZ4_renormDictT(LZ4_stream_t_internal* LZ4_dict, int nextSize)
|
1495
|
+
{
|
1496
|
+
assert(nextSize >= 0);
|
1497
|
+
if (LZ4_dict->currentOffset + (unsigned)nextSize > 0x80000000) { /* potential ptrdiff_t overflow (32-bits mode) */
|
1498
|
+
/* rescale hash table */
|
1499
|
+
U32 const delta = LZ4_dict->currentOffset - 64 KB;
|
1500
|
+
const BYTE* dictEnd = LZ4_dict->dictionary + LZ4_dict->dictSize;
|
1501
|
+
int i;
|
1502
|
+
DEBUGLOG(4, "LZ4_renormDictT");
|
1503
|
+
for (i=0; i<LZ4_HASH_SIZE_U32; i++) {
|
1504
|
+
if (LZ4_dict->hashTable[i] < delta) LZ4_dict->hashTable[i]=0;
|
1505
|
+
else LZ4_dict->hashTable[i] -= delta;
|
1506
|
+
}
|
1507
|
+
LZ4_dict->currentOffset = 64 KB;
|
1508
|
+
if (LZ4_dict->dictSize > 64 KB) LZ4_dict->dictSize = 64 KB;
|
1509
|
+
LZ4_dict->dictionary = dictEnd - LZ4_dict->dictSize;
|
1510
|
+
}
|
1511
|
+
}
|
1512
|
+
|
1513
|
+
|
1514
|
+
int LZ4_compress_fast_continue (LZ4_stream_t* LZ4_stream,
|
1515
|
+
const char* source, char* dest,
|
1516
|
+
int inputSize, int maxOutputSize,
|
1517
|
+
int acceleration)
|
1518
|
+
{
|
1519
|
+
const tableType_t tableType = byU32;
|
1520
|
+
LZ4_stream_t_internal* const streamPtr = &LZ4_stream->internal_donotuse;
|
1521
|
+
const char* dictEnd = streamPtr->dictSize ? (const char*)streamPtr->dictionary + streamPtr->dictSize : NULL;
|
1522
|
+
|
1523
|
+
DEBUGLOG(5, "LZ4_compress_fast_continue (inputSize=%i, dictSize=%u)", inputSize, streamPtr->dictSize);
|
1524
|
+
|
1525
|
+
LZ4_renormDictT(streamPtr, inputSize); /* fix index overflow */
|
1526
|
+
if (acceleration < 1) acceleration = LZ4_ACCELERATION_DEFAULT;
|
1527
|
+
if (acceleration > LZ4_ACCELERATION_MAX) acceleration = LZ4_ACCELERATION_MAX;
|
1528
|
+
|
1529
|
+
/* invalidate tiny dictionaries */
|
1530
|
+
if ( (streamPtr->dictSize < 4) /* tiny dictionary : not enough for a hash */
|
1531
|
+
&& (dictEnd != source) /* prefix mode */
|
1532
|
+
&& (inputSize > 0) /* tolerance : don't lose history, in case next invocation would use prefix mode */
|
1533
|
+
&& (streamPtr->dictCtx == NULL) /* usingDictCtx */
|
1534
|
+
) {
|
1535
|
+
DEBUGLOG(5, "LZ4_compress_fast_continue: dictSize(%u) at addr:%p is too small", streamPtr->dictSize, streamPtr->dictionary);
|
1536
|
+
/* remove dictionary existence from history, to employ faster prefix mode */
|
1537
|
+
streamPtr->dictSize = 0;
|
1538
|
+
streamPtr->dictionary = (const BYTE*)source;
|
1539
|
+
dictEnd = source;
|
1540
|
+
}
|
1541
|
+
|
1542
|
+
/* Check overlapping input/dictionary space */
|
1543
|
+
{ const char* const sourceEnd = source + inputSize;
|
1544
|
+
if ((sourceEnd > (const char*)streamPtr->dictionary) && (sourceEnd < dictEnd)) {
|
1545
|
+
streamPtr->dictSize = (U32)(dictEnd - sourceEnd);
|
1546
|
+
if (streamPtr->dictSize > 64 KB) streamPtr->dictSize = 64 KB;
|
1547
|
+
if (streamPtr->dictSize < 4) streamPtr->dictSize = 0;
|
1548
|
+
streamPtr->dictionary = (const BYTE*)dictEnd - streamPtr->dictSize;
|
1549
|
+
}
|
1550
|
+
}
|
1551
|
+
|
1552
|
+
/* prefix mode : source data follows dictionary */
|
1553
|
+
if (dictEnd == source) {
|
1554
|
+
if ((streamPtr->dictSize < 64 KB) && (streamPtr->dictSize < streamPtr->currentOffset))
|
1555
|
+
return LZ4_compress_generic(streamPtr, source, dest, inputSize, NULL, maxOutputSize, limitedOutput, tableType, withPrefix64k, dictSmall, acceleration);
|
1556
|
+
else
|
1557
|
+
return LZ4_compress_generic(streamPtr, source, dest, inputSize, NULL, maxOutputSize, limitedOutput, tableType, withPrefix64k, noDictIssue, acceleration);
|
1558
|
+
}
|
1559
|
+
|
1560
|
+
/* external dictionary mode */
|
1561
|
+
{ int result;
|
1562
|
+
if (streamPtr->dictCtx) {
|
1563
|
+
/* We depend here on the fact that dictCtx'es (produced by
|
1564
|
+
* LZ4_loadDict) guarantee that their tables contain no references
|
1565
|
+
* to offsets between dictCtx->currentOffset - 64 KB and
|
1566
|
+
* dictCtx->currentOffset - dictCtx->dictSize. This makes it safe
|
1567
|
+
* to use noDictIssue even when the dict isn't a full 64 KB.
|
1568
|
+
*/
|
1569
|
+
if (inputSize > 4 KB) {
|
1570
|
+
/* For compressing large blobs, it is faster to pay the setup
|
1571
|
+
* cost to copy the dictionary's tables into the active context,
|
1572
|
+
* so that the compression loop is only looking into one table.
|
1573
|
+
*/
|
1574
|
+
LZ4_memcpy(streamPtr, streamPtr->dictCtx, sizeof(*streamPtr));
|
1575
|
+
result = LZ4_compress_generic(streamPtr, source, dest, inputSize, NULL, maxOutputSize, limitedOutput, tableType, usingExtDict, noDictIssue, acceleration);
|
1576
|
+
} else {
|
1577
|
+
result = LZ4_compress_generic(streamPtr, source, dest, inputSize, NULL, maxOutputSize, limitedOutput, tableType, usingDictCtx, noDictIssue, acceleration);
|
1578
|
+
}
|
1579
|
+
} else { /* small data <= 4 KB */
|
1580
|
+
if ((streamPtr->dictSize < 64 KB) && (streamPtr->dictSize < streamPtr->currentOffset)) {
|
1581
|
+
result = LZ4_compress_generic(streamPtr, source, dest, inputSize, NULL, maxOutputSize, limitedOutput, tableType, usingExtDict, dictSmall, acceleration);
|
1582
|
+
} else {
|
1583
|
+
result = LZ4_compress_generic(streamPtr, source, dest, inputSize, NULL, maxOutputSize, limitedOutput, tableType, usingExtDict, noDictIssue, acceleration);
|
1584
|
+
}
|
1585
|
+
}
|
1586
|
+
streamPtr->dictionary = (const BYTE*)source;
|
1587
|
+
streamPtr->dictSize = (U32)inputSize;
|
1588
|
+
return result;
|
1589
|
+
}
|
1590
|
+
}
|
1591
|
+
|
1592
|
+
|
1593
|
+
/* Hidden debug function, to force-test external dictionary mode */
|
1594
|
+
int LZ4_compress_forceExtDict (LZ4_stream_t* LZ4_dict, const char* source, char* dest, int srcSize)
|
1595
|
+
{
|
1596
|
+
LZ4_stream_t_internal* streamPtr = &LZ4_dict->internal_donotuse;
|
1597
|
+
int result;
|
1598
|
+
|
1599
|
+
LZ4_renormDictT(streamPtr, srcSize);
|
1600
|
+
|
1601
|
+
if ((streamPtr->dictSize < 64 KB) && (streamPtr->dictSize < streamPtr->currentOffset)) {
|
1602
|
+
result = LZ4_compress_generic(streamPtr, source, dest, srcSize, NULL, 0, notLimited, byU32, usingExtDict, dictSmall, 1);
|
1603
|
+
} else {
|
1604
|
+
result = LZ4_compress_generic(streamPtr, source, dest, srcSize, NULL, 0, notLimited, byU32, usingExtDict, noDictIssue, 1);
|
1605
|
+
}
|
1606
|
+
|
1607
|
+
streamPtr->dictionary = (const BYTE*)source;
|
1608
|
+
streamPtr->dictSize = (U32)srcSize;
|
1609
|
+
|
1610
|
+
return result;
|
1611
|
+
}
|
1612
|
+
|
1613
|
+
|
1614
|
+
/*! LZ4_saveDict() :
|
1615
|
+
* If previously compressed data block is not guaranteed to remain available at its memory location,
|
1616
|
+
* save it into a safer place (char* safeBuffer).
|
1617
|
+
* Note : no need to call LZ4_loadDict() afterwards, dictionary is immediately usable,
|
1618
|
+
* one can therefore call LZ4_compress_fast_continue() right after.
|
1619
|
+
* @return : saved dictionary size in bytes (necessarily <= dictSize), or 0 if error.
|
1620
|
+
*/
|
1621
|
+
int LZ4_saveDict (LZ4_stream_t* LZ4_dict, char* safeBuffer, int dictSize)
|
1622
|
+
{
|
1623
|
+
LZ4_stream_t_internal* const dict = &LZ4_dict->internal_donotuse;
|
1624
|
+
|
1625
|
+
DEBUGLOG(5, "LZ4_saveDict : dictSize=%i, safeBuffer=%p", dictSize, safeBuffer);
|
1626
|
+
|
1627
|
+
if ((U32)dictSize > 64 KB) { dictSize = 64 KB; } /* useless to define a dictionary > 64 KB */
|
1628
|
+
if ((U32)dictSize > dict->dictSize) { dictSize = (int)dict->dictSize; }
|
1629
|
+
|
1630
|
+
if (safeBuffer == NULL) assert(dictSize == 0);
|
1631
|
+
if (dictSize > 0) {
|
1632
|
+
const BYTE* const previousDictEnd = dict->dictionary + dict->dictSize;
|
1633
|
+
assert(dict->dictionary);
|
1634
|
+
LZ4_memmove(safeBuffer, previousDictEnd - dictSize, (size_t)dictSize);
|
1635
|
+
}
|
1636
|
+
|
1637
|
+
dict->dictionary = (const BYTE*)safeBuffer;
|
1638
|
+
dict->dictSize = (U32)dictSize;
|
1639
|
+
|
1640
|
+
return dictSize;
|
1641
|
+
}
|
1642
|
+
|
1643
|
+
|
1644
|
+
|
1645
|
+
/*-*******************************
|
1646
|
+
* Decompression functions
|
1647
|
+
********************************/
|
1648
|
+
|
1649
|
+
typedef enum { decode_full_block = 0, partial_decode = 1 } earlyEnd_directive;
|
1650
|
+
|
1651
|
+
#undef MIN
|
1652
|
+
#define MIN(a,b) ( (a) < (b) ? (a) : (b) )
|
1653
|
+
|
1654
|
+
|
1655
|
+
/* variant for decompress_unsafe()
|
1656
|
+
* does not know end of input
|
1657
|
+
* presumes input is well formed
|
1658
|
+
* note : will consume at least one byte */
|
1659
|
+
size_t read_long_length_no_check(const BYTE** pp)
|
1660
|
+
{
|
1661
|
+
size_t b, l = 0;
|
1662
|
+
do { b = **pp; (*pp)++; l += b; } while (b==255);
|
1663
|
+
DEBUGLOG(6, "read_long_length_no_check: +length=%zu using %zu input bytes", l, l/255 + 1)
|
1664
|
+
return l;
|
1665
|
+
}
|
1666
|
+
|
1667
|
+
/* core decoder variant for LZ4_decompress_fast*()
|
1668
|
+
* for legacy support only : these entry points are deprecated.
|
1669
|
+
* - Presumes input is correctly formed (no defense vs malformed inputs)
|
1670
|
+
* - Does not know input size (presume input buffer is "large enough")
|
1671
|
+
* - Decompress a full block (only)
|
1672
|
+
* @return : nb of bytes read from input.
|
1673
|
+
* Note : this variant is not optimized for speed, just for maintenance.
|
1674
|
+
* the goal is to remove support of decompress_fast*() variants by v2.0
|
1675
|
+
**/
|
1676
|
+
LZ4_FORCE_INLINE int
|
1677
|
+
LZ4_decompress_unsafe_generic(
|
1678
|
+
const BYTE* const istart,
|
1679
|
+
BYTE* const ostart,
|
1680
|
+
int decompressedSize,
|
1681
|
+
|
1682
|
+
size_t prefixSize,
|
1683
|
+
const BYTE* const dictStart, /* only if dict==usingExtDict */
|
1684
|
+
const size_t dictSize /* note: =0 if dictStart==NULL */
|
1685
|
+
)
|
1686
|
+
{
|
1687
|
+
const BYTE* ip = istart;
|
1688
|
+
BYTE* op = (BYTE*)ostart;
|
1689
|
+
BYTE* const oend = ostart + decompressedSize;
|
1690
|
+
const BYTE* const prefixStart = ostart - prefixSize;
|
1691
|
+
|
1692
|
+
DEBUGLOG(5, "LZ4_decompress_unsafe_generic");
|
1693
|
+
if (dictStart == NULL) assert(dictSize == 0);
|
1694
|
+
|
1695
|
+
while (1) {
|
1696
|
+
/* start new sequence */
|
1697
|
+
unsigned token = *ip++;
|
1698
|
+
|
1699
|
+
/* literals */
|
1700
|
+
{ size_t ll = token >> ML_BITS;
|
1701
|
+
if (ll==15) {
|
1702
|
+
/* long literal length */
|
1703
|
+
ll += read_long_length_no_check(&ip);
|
1704
|
+
}
|
1705
|
+
if ((size_t)(oend-op) < ll) return -1; /* output buffer overflow */
|
1706
|
+
LZ4_memmove(op, ip, ll); /* support in-place decompression */
|
1707
|
+
op += ll;
|
1708
|
+
ip += ll;
|
1709
|
+
if ((size_t)(oend-op) < MFLIMIT) {
|
1710
|
+
if (op==oend) break; /* end of block */
|
1711
|
+
DEBUGLOG(5, "invalid: literals end at distance %zi from end of block", oend-op);
|
1712
|
+
/* incorrect end of block :
|
1713
|
+
* last match must start at least MFLIMIT==12 bytes before end of output block */
|
1714
|
+
return -1;
|
1715
|
+
} }
|
1716
|
+
|
1717
|
+
/* match */
|
1718
|
+
{ size_t ml = token & 15;
|
1719
|
+
size_t const offset = LZ4_readLE16(ip);
|
1720
|
+
ip+=2;
|
1721
|
+
|
1722
|
+
if (ml==15) {
|
1723
|
+
/* long literal length */
|
1724
|
+
ml += read_long_length_no_check(&ip);
|
1725
|
+
}
|
1726
|
+
ml += MINMATCH;
|
1727
|
+
|
1728
|
+
if ((size_t)(oend-op) < ml) return -1; /* output buffer overflow */
|
1729
|
+
|
1730
|
+
{ const BYTE* match = op - offset;
|
1731
|
+
|
1732
|
+
/* out of range */
|
1733
|
+
if (offset > (size_t)(op - prefixStart) + dictSize) {
|
1734
|
+
DEBUGLOG(6, "offset out of range");
|
1735
|
+
return -1;
|
1736
|
+
}
|
1737
|
+
|
1738
|
+
/* check special case : extDict */
|
1739
|
+
if (offset > (size_t)(op - prefixStart)) {
|
1740
|
+
/* extDict scenario */
|
1741
|
+
const BYTE* const dictEnd = dictStart + dictSize;
|
1742
|
+
const BYTE* extMatch = dictEnd - (offset - (size_t)(op-prefixStart));
|
1743
|
+
size_t const extml = (size_t)(dictEnd - extMatch);
|
1744
|
+
if (extml > ml) {
|
1745
|
+
/* match entirely within extDict */
|
1746
|
+
LZ4_memmove(op, extMatch, ml);
|
1747
|
+
op += ml;
|
1748
|
+
ml = 0;
|
1749
|
+
} else {
|
1750
|
+
/* match split between extDict & prefix */
|
1751
|
+
LZ4_memmove(op, extMatch, extml);
|
1752
|
+
op += extml;
|
1753
|
+
ml -= extml;
|
1754
|
+
}
|
1755
|
+
match = prefixStart;
|
1756
|
+
}
|
1757
|
+
|
1758
|
+
/* match copy - slow variant, supporting overlap copy */
|
1759
|
+
{ size_t u;
|
1760
|
+
for (u=0; u<ml; u++) {
|
1761
|
+
op[u] = match[u];
|
1762
|
+
} } }
|
1763
|
+
op += ml;
|
1764
|
+
if ((size_t)(oend-op) < LASTLITERALS) {
|
1765
|
+
DEBUGLOG(5, "invalid: match ends at distance %zi from end of block", oend-op);
|
1766
|
+
/* incorrect end of block :
|
1767
|
+
* last match must stop at least LASTLITERALS==5 bytes before end of output block */
|
1768
|
+
return -1;
|
1769
|
+
}
|
1770
|
+
} /* match */
|
1771
|
+
} /* main loop */
|
1772
|
+
return (int)(ip - istart);
|
1773
|
+
}
|
1774
|
+
|
1775
|
+
|
1776
|
+
/* Read the variable-length literal or match length.
|
1777
|
+
*
|
1778
|
+
* @ip : input pointer
|
1779
|
+
* @ilimit : position after which if length is not decoded, the input is necessarily corrupted.
|
1780
|
+
* @initial_check - check ip >= ipmax before start of loop. Returns initial_error if so.
|
1781
|
+
* @error (output) - error code. Must be set to 0 before call.
|
1782
|
+
**/
|
1783
|
+
typedef size_t Rvl_t;
|
1784
|
+
static const Rvl_t rvl_error = (Rvl_t)(-1);
|
1785
|
+
LZ4_FORCE_INLINE Rvl_t
|
1786
|
+
read_variable_length(const BYTE** ip, const BYTE* ilimit,
|
1787
|
+
int initial_check)
|
1788
|
+
{
|
1789
|
+
Rvl_t s, length = 0;
|
1790
|
+
assert(ip != NULL);
|
1791
|
+
assert(*ip != NULL);
|
1792
|
+
assert(ilimit != NULL);
|
1793
|
+
if (initial_check && unlikely((*ip) >= ilimit)) { /* read limit reached */
|
1794
|
+
return rvl_error;
|
1795
|
+
}
|
1796
|
+
do {
|
1797
|
+
s = **ip;
|
1798
|
+
(*ip)++;
|
1799
|
+
length += s;
|
1800
|
+
if (unlikely((*ip) > ilimit)) { /* read limit reached */
|
1801
|
+
return rvl_error;
|
1802
|
+
}
|
1803
|
+
/* accumulator overflow detection (32-bit mode only) */
|
1804
|
+
if ((sizeof(length)<8) && unlikely(length > ((Rvl_t)(-1)/2)) ) {
|
1805
|
+
return rvl_error;
|
1806
|
+
}
|
1807
|
+
} while (s==255);
|
1808
|
+
|
1809
|
+
return length;
|
1810
|
+
}
|
1811
|
+
|
1812
|
+
/*! LZ4_decompress_generic() :
|
1813
|
+
* This generic decompression function covers all use cases.
|
1814
|
+
* It shall be instantiated several times, using different sets of directives.
|
1815
|
+
* Note that it is important for performance that this function really get inlined,
|
1816
|
+
* in order to remove useless branches during compilation optimization.
|
1817
|
+
*/
|
1818
|
+
LZ4_FORCE_INLINE int
|
1819
|
+
LZ4_decompress_generic(
|
1820
|
+
const char* const src,
|
1821
|
+
char* const dst,
|
1822
|
+
int srcSize,
|
1823
|
+
int outputSize, /* If endOnInput==endOnInputSize, this value is `dstCapacity` */
|
1824
|
+
|
1825
|
+
earlyEnd_directive partialDecoding, /* full, partial */
|
1826
|
+
dict_directive dict, /* noDict, withPrefix64k, usingExtDict */
|
1827
|
+
const BYTE* const lowPrefix, /* always <= dst, == dst when no prefix */
|
1828
|
+
const BYTE* const dictStart, /* only if dict==usingExtDict */
|
1829
|
+
const size_t dictSize /* note : = 0 if noDict */
|
1830
|
+
)
|
1831
|
+
{
|
1832
|
+
if ((src == NULL) || (outputSize < 0)) { return -1; }
|
1833
|
+
|
1834
|
+
{ const BYTE* ip = (const BYTE*) src;
|
1835
|
+
const BYTE* const iend = ip + srcSize;
|
1836
|
+
|
1837
|
+
BYTE* op = (BYTE*) dst;
|
1838
|
+
BYTE* const oend = op + outputSize;
|
1839
|
+
BYTE* cpy;
|
1840
|
+
|
1841
|
+
const BYTE* const dictEnd = (dictStart == NULL) ? NULL : dictStart + dictSize;
|
1842
|
+
|
1843
|
+
const int checkOffset = (dictSize < (int)(64 KB));
|
1844
|
+
|
1845
|
+
|
1846
|
+
/* Set up the "end" pointers for the shortcut. */
|
1847
|
+
const BYTE* const shortiend = iend - 14 /*maxLL*/ - 2 /*offset*/;
|
1848
|
+
const BYTE* const shortoend = oend - 14 /*maxLL*/ - 18 /*maxML*/;
|
1849
|
+
|
1850
|
+
const BYTE* match;
|
1851
|
+
size_t offset;
|
1852
|
+
unsigned token;
|
1853
|
+
size_t length;
|
1854
|
+
|
1855
|
+
|
1856
|
+
DEBUGLOG(5, "LZ4_decompress_generic (srcSize:%i, dstSize:%i)", srcSize, outputSize);
|
1857
|
+
|
1858
|
+
/* Special cases */
|
1859
|
+
assert(lowPrefix <= op);
|
1860
|
+
if (unlikely(outputSize==0)) {
|
1861
|
+
/* Empty output buffer */
|
1862
|
+
if (partialDecoding) return 0;
|
1863
|
+
return ((srcSize==1) && (*ip==0)) ? 0 : -1;
|
1864
|
+
}
|
1865
|
+
if (unlikely(srcSize==0)) { return -1; }
|
1866
|
+
|
1867
|
+
/* LZ4_FAST_DEC_LOOP:
|
1868
|
+
* designed for modern OoO performance cpus,
|
1869
|
+
* where copying reliably 32-bytes is preferable to an unpredictable branch.
|
1870
|
+
* note : fast loop may show a regression for some client arm chips. */
|
1871
|
+
#if LZ4_FAST_DEC_LOOP
|
1872
|
+
if ((oend - op) < FASTLOOP_SAFE_DISTANCE) {
|
1873
|
+
DEBUGLOG(6, "skip fast decode loop");
|
1874
|
+
goto safe_decode;
|
1875
|
+
}
|
1876
|
+
|
1877
|
+
/* Fast loop : decode sequences as long as output < oend-FASTLOOP_SAFE_DISTANCE */
|
1878
|
+
while (1) {
|
1879
|
+
/* Main fastloop assertion: We can always wildcopy FASTLOOP_SAFE_DISTANCE */
|
1880
|
+
assert(oend - op >= FASTLOOP_SAFE_DISTANCE);
|
1881
|
+
assert(ip < iend);
|
1882
|
+
token = *ip++;
|
1883
|
+
length = token >> ML_BITS; /* literal length */
|
1884
|
+
|
1885
|
+
/* decode literal length */
|
1886
|
+
if (length == RUN_MASK) {
|
1887
|
+
size_t const addl = read_variable_length(&ip, iend-RUN_MASK, 1);
|
1888
|
+
if (addl == rvl_error) { goto _output_error; }
|
1889
|
+
length += addl;
|
1890
|
+
if (unlikely((uptrval)(op)+length<(uptrval)(op))) { goto _output_error; } /* overflow detection */
|
1891
|
+
if (unlikely((uptrval)(ip)+length<(uptrval)(ip))) { goto _output_error; } /* overflow detection */
|
1892
|
+
|
1893
|
+
/* copy literals */
|
1894
|
+
cpy = op+length;
|
1895
|
+
LZ4_STATIC_ASSERT(MFLIMIT >= WILDCOPYLENGTH);
|
1896
|
+
if ((cpy>oend-32) || (ip+length>iend-32)) { goto safe_literal_copy; }
|
1897
|
+
LZ4_wildCopy32(op, ip, cpy);
|
1898
|
+
ip += length; op = cpy;
|
1899
|
+
} else {
|
1900
|
+
cpy = op+length;
|
1901
|
+
DEBUGLOG(7, "copy %u bytes in a 16-bytes stripe", (unsigned)length);
|
1902
|
+
/* We don't need to check oend, since we check it once for each loop below */
|
1903
|
+
if (ip > iend-(16 + 1/*max lit + offset + nextToken*/)) { goto safe_literal_copy; }
|
1904
|
+
/* Literals can only be <= 14, but hope compilers optimize better when copy by a register size */
|
1905
|
+
LZ4_memcpy(op, ip, 16);
|
1906
|
+
ip += length; op = cpy;
|
1907
|
+
}
|
1908
|
+
|
1909
|
+
/* get offset */
|
1910
|
+
offset = LZ4_readLE16(ip); ip+=2;
|
1911
|
+
match = op - offset;
|
1912
|
+
assert(match <= op); /* overflow check */
|
1913
|
+
|
1914
|
+
/* get matchlength */
|
1915
|
+
length = token & ML_MASK;
|
1916
|
+
|
1917
|
+
if (length == ML_MASK) {
|
1918
|
+
size_t const addl = read_variable_length(&ip, iend - LASTLITERALS + 1, 0);
|
1919
|
+
if (addl == rvl_error) { goto _output_error; }
|
1920
|
+
length += addl;
|
1921
|
+
length += MINMATCH;
|
1922
|
+
if (unlikely((uptrval)(op)+length<(uptrval)op)) { goto _output_error; } /* overflow detection */
|
1923
|
+
if ((checkOffset) && (unlikely(match + dictSize < lowPrefix))) { goto _output_error; } /* Error : offset outside buffers */
|
1924
|
+
if (op + length >= oend - FASTLOOP_SAFE_DISTANCE) {
|
1925
|
+
goto safe_match_copy;
|
1926
|
+
}
|
1927
|
+
} else {
|
1928
|
+
length += MINMATCH;
|
1929
|
+
if (op + length >= oend - FASTLOOP_SAFE_DISTANCE) {
|
1930
|
+
goto safe_match_copy;
|
1931
|
+
}
|
1932
|
+
|
1933
|
+
/* Fastpath check: skip LZ4_wildCopy32 when true */
|
1934
|
+
if ((dict == withPrefix64k) || (match >= lowPrefix)) {
|
1935
|
+
if (offset >= 8) {
|
1936
|
+
assert(match >= lowPrefix);
|
1937
|
+
assert(match <= op);
|
1938
|
+
assert(op + 18 <= oend);
|
1939
|
+
|
1940
|
+
LZ4_memcpy(op, match, 8);
|
1941
|
+
LZ4_memcpy(op+8, match+8, 8);
|
1942
|
+
LZ4_memcpy(op+16, match+16, 2);
|
1943
|
+
op += length;
|
1944
|
+
continue;
|
1945
|
+
} } }
|
1946
|
+
|
1947
|
+
if (checkOffset && (unlikely(match + dictSize < lowPrefix))) { goto _output_error; } /* Error : offset outside buffers */
|
1948
|
+
/* match starting within external dictionary */
|
1949
|
+
if ((dict==usingExtDict) && (match < lowPrefix)) {
|
1950
|
+
assert(dictEnd != NULL);
|
1951
|
+
if (unlikely(op+length > oend-LASTLITERALS)) {
|
1952
|
+
if (partialDecoding) {
|
1953
|
+
DEBUGLOG(7, "partialDecoding: dictionary match, close to dstEnd");
|
1954
|
+
length = MIN(length, (size_t)(oend-op));
|
1955
|
+
} else {
|
1956
|
+
goto _output_error; /* end-of-block condition violated */
|
1957
|
+
} }
|
1958
|
+
|
1959
|
+
if (length <= (size_t)(lowPrefix-match)) {
|
1960
|
+
/* match fits entirely within external dictionary : just copy */
|
1961
|
+
LZ4_memmove(op, dictEnd - (lowPrefix-match), length);
|
1962
|
+
op += length;
|
1963
|
+
} else {
|
1964
|
+
/* match stretches into both external dictionary and current block */
|
1965
|
+
size_t const copySize = (size_t)(lowPrefix - match);
|
1966
|
+
size_t const restSize = length - copySize;
|
1967
|
+
LZ4_memcpy(op, dictEnd - copySize, copySize);
|
1968
|
+
op += copySize;
|
1969
|
+
if (restSize > (size_t)(op - lowPrefix)) { /* overlap copy */
|
1970
|
+
BYTE* const endOfMatch = op + restSize;
|
1971
|
+
const BYTE* copyFrom = lowPrefix;
|
1972
|
+
while (op < endOfMatch) { *op++ = *copyFrom++; }
|
1973
|
+
} else {
|
1974
|
+
LZ4_memcpy(op, lowPrefix, restSize);
|
1975
|
+
op += restSize;
|
1976
|
+
} }
|
1977
|
+
continue;
|
1978
|
+
}
|
1979
|
+
|
1980
|
+
/* copy match within block */
|
1981
|
+
cpy = op + length;
|
1982
|
+
|
1983
|
+
assert((op <= oend) && (oend-op >= 32));
|
1984
|
+
if (unlikely(offset<16)) {
|
1985
|
+
LZ4_memcpy_using_offset(op, match, cpy, offset);
|
1986
|
+
} else {
|
1987
|
+
LZ4_wildCopy32(op, match, cpy);
|
1988
|
+
}
|
1989
|
+
|
1990
|
+
op = cpy; /* wildcopy correction */
|
1991
|
+
}
|
1992
|
+
safe_decode:
|
1993
|
+
#endif
|
1994
|
+
|
1995
|
+
/* Main Loop : decode remaining sequences where output < FASTLOOP_SAFE_DISTANCE */
|
1996
|
+
while (1) {
|
1997
|
+
assert(ip < iend);
|
1998
|
+
token = *ip++;
|
1999
|
+
length = token >> ML_BITS; /* literal length */
|
2000
|
+
|
2001
|
+
/* A two-stage shortcut for the most common case:
|
2002
|
+
* 1) If the literal length is 0..14, and there is enough space,
|
2003
|
+
* enter the shortcut and copy 16 bytes on behalf of the literals
|
2004
|
+
* (in the fast mode, only 8 bytes can be safely copied this way).
|
2005
|
+
* 2) Further if the match length is 4..18, copy 18 bytes in a similar
|
2006
|
+
* manner; but we ensure that there's enough space in the output for
|
2007
|
+
* those 18 bytes earlier, upon entering the shortcut (in other words,
|
2008
|
+
* there is a combined check for both stages).
|
2009
|
+
*/
|
2010
|
+
if ( (length != RUN_MASK)
|
2011
|
+
/* strictly "less than" on input, to re-enter the loop with at least one byte */
|
2012
|
+
&& likely((ip < shortiend) & (op <= shortoend)) ) {
|
2013
|
+
/* Copy the literals */
|
2014
|
+
LZ4_memcpy(op, ip, 16);
|
2015
|
+
op += length; ip += length;
|
2016
|
+
|
2017
|
+
/* The second stage: prepare for match copying, decode full info.
|
2018
|
+
* If it doesn't work out, the info won't be wasted. */
|
2019
|
+
length = token & ML_MASK; /* match length */
|
2020
|
+
offset = LZ4_readLE16(ip); ip += 2;
|
2021
|
+
match = op - offset;
|
2022
|
+
assert(match <= op); /* check overflow */
|
2023
|
+
|
2024
|
+
/* Do not deal with overlapping matches. */
|
2025
|
+
if ( (length != ML_MASK)
|
2026
|
+
&& (offset >= 8)
|
2027
|
+
&& (dict==withPrefix64k || match >= lowPrefix) ) {
|
2028
|
+
/* Copy the match. */
|
2029
|
+
LZ4_memcpy(op + 0, match + 0, 8);
|
2030
|
+
LZ4_memcpy(op + 8, match + 8, 8);
|
2031
|
+
LZ4_memcpy(op +16, match +16, 2);
|
2032
|
+
op += length + MINMATCH;
|
2033
|
+
/* Both stages worked, load the next token. */
|
2034
|
+
continue;
|
2035
|
+
}
|
2036
|
+
|
2037
|
+
/* The second stage didn't work out, but the info is ready.
|
2038
|
+
* Propel it right to the point of match copying. */
|
2039
|
+
goto _copy_match;
|
2040
|
+
}
|
2041
|
+
|
2042
|
+
/* decode literal length */
|
2043
|
+
if (length == RUN_MASK) {
|
2044
|
+
size_t const addl = read_variable_length(&ip, iend-RUN_MASK, 1);
|
2045
|
+
if (addl == rvl_error) { goto _output_error; }
|
2046
|
+
length += addl;
|
2047
|
+
if (unlikely((uptrval)(op)+length<(uptrval)(op))) { goto _output_error; } /* overflow detection */
|
2048
|
+
if (unlikely((uptrval)(ip)+length<(uptrval)(ip))) { goto _output_error; } /* overflow detection */
|
2049
|
+
}
|
2050
|
+
|
2051
|
+
/* copy literals */
|
2052
|
+
cpy = op+length;
|
2053
|
+
#if LZ4_FAST_DEC_LOOP
|
2054
|
+
safe_literal_copy:
|
2055
|
+
#endif
|
2056
|
+
LZ4_STATIC_ASSERT(MFLIMIT >= WILDCOPYLENGTH);
|
2057
|
+
if ((cpy>oend-MFLIMIT) || (ip+length>iend-(2+1+LASTLITERALS))) {
|
2058
|
+
/* We've either hit the input parsing restriction or the output parsing restriction.
|
2059
|
+
* In the normal scenario, decoding a full block, it must be the last sequence,
|
2060
|
+
* otherwise it's an error (invalid input or dimensions).
|
2061
|
+
* In partialDecoding scenario, it's necessary to ensure there is no buffer overflow.
|
2062
|
+
*/
|
2063
|
+
if (partialDecoding) {
|
2064
|
+
/* Since we are partial decoding we may be in this block because of the output parsing
|
2065
|
+
* restriction, which is not valid since the output buffer is allowed to be undersized.
|
2066
|
+
*/
|
2067
|
+
DEBUGLOG(7, "partialDecoding: copying literals, close to input or output end")
|
2068
|
+
DEBUGLOG(7, "partialDecoding: literal length = %u", (unsigned)length);
|
2069
|
+
DEBUGLOG(7, "partialDecoding: remaining space in dstBuffer : %i", (int)(oend - op));
|
2070
|
+
DEBUGLOG(7, "partialDecoding: remaining space in srcBuffer : %i", (int)(iend - ip));
|
2071
|
+
/* Finishing in the middle of a literals segment,
|
2072
|
+
* due to lack of input.
|
2073
|
+
*/
|
2074
|
+
if (ip+length > iend) {
|
2075
|
+
length = (size_t)(iend-ip);
|
2076
|
+
cpy = op + length;
|
2077
|
+
}
|
2078
|
+
/* Finishing in the middle of a literals segment,
|
2079
|
+
* due to lack of output space.
|
2080
|
+
*/
|
2081
|
+
if (cpy > oend) {
|
2082
|
+
cpy = oend;
|
2083
|
+
assert(op<=oend);
|
2084
|
+
length = (size_t)(oend-op);
|
2085
|
+
}
|
2086
|
+
} else {
|
2087
|
+
/* We must be on the last sequence (or invalid) because of the parsing limitations
|
2088
|
+
* so check that we exactly consume the input and don't overrun the output buffer.
|
2089
|
+
*/
|
2090
|
+
if ((ip+length != iend) || (cpy > oend)) {
|
2091
|
+
DEBUGLOG(6, "should have been last run of literals")
|
2092
|
+
DEBUGLOG(6, "ip(%p) + length(%i) = %p != iend (%p)", ip, (int)length, ip+length, iend);
|
2093
|
+
DEBUGLOG(6, "or cpy(%p) > oend(%p)", cpy, oend);
|
2094
|
+
goto _output_error;
|
2095
|
+
}
|
2096
|
+
}
|
2097
|
+
LZ4_memmove(op, ip, length); /* supports overlapping memory regions, for in-place decompression scenarios */
|
2098
|
+
ip += length;
|
2099
|
+
op += length;
|
2100
|
+
/* Necessarily EOF when !partialDecoding.
|
2101
|
+
* When partialDecoding, it is EOF if we've either
|
2102
|
+
* filled the output buffer or
|
2103
|
+
* can't proceed with reading an offset for following match.
|
2104
|
+
*/
|
2105
|
+
if (!partialDecoding || (cpy == oend) || (ip >= (iend-2))) {
|
2106
|
+
break;
|
2107
|
+
}
|
2108
|
+
} else {
|
2109
|
+
LZ4_wildCopy8(op, ip, cpy); /* can overwrite up to 8 bytes beyond cpy */
|
2110
|
+
ip += length; op = cpy;
|
2111
|
+
}
|
2112
|
+
|
2113
|
+
/* get offset */
|
2114
|
+
offset = LZ4_readLE16(ip); ip+=2;
|
2115
|
+
match = op - offset;
|
2116
|
+
|
2117
|
+
/* get matchlength */
|
2118
|
+
length = token & ML_MASK;
|
2119
|
+
|
2120
|
+
_copy_match:
|
2121
|
+
if (length == ML_MASK) {
|
2122
|
+
size_t const addl = read_variable_length(&ip, iend - LASTLITERALS + 1, 0);
|
2123
|
+
if (addl == rvl_error) { goto _output_error; }
|
2124
|
+
length += addl;
|
2125
|
+
if (unlikely((uptrval)(op)+length<(uptrval)op)) goto _output_error; /* overflow detection */
|
2126
|
+
}
|
2127
|
+
length += MINMATCH;
|
2128
|
+
|
2129
|
+
#if LZ4_FAST_DEC_LOOP
|
2130
|
+
safe_match_copy:
|
2131
|
+
#endif
|
2132
|
+
if ((checkOffset) && (unlikely(match + dictSize < lowPrefix))) goto _output_error; /* Error : offset outside buffers */
|
2133
|
+
/* match starting within external dictionary */
|
2134
|
+
if ((dict==usingExtDict) && (match < lowPrefix)) {
|
2135
|
+
assert(dictEnd != NULL);
|
2136
|
+
if (unlikely(op+length > oend-LASTLITERALS)) {
|
2137
|
+
if (partialDecoding) length = MIN(length, (size_t)(oend-op));
|
2138
|
+
else goto _output_error; /* doesn't respect parsing restriction */
|
2139
|
+
}
|
2140
|
+
|
2141
|
+
if (length <= (size_t)(lowPrefix-match)) {
|
2142
|
+
/* match fits entirely within external dictionary : just copy */
|
2143
|
+
LZ4_memmove(op, dictEnd - (lowPrefix-match), length);
|
2144
|
+
op += length;
|
2145
|
+
} else {
|
2146
|
+
/* match stretches into both external dictionary and current block */
|
2147
|
+
size_t const copySize = (size_t)(lowPrefix - match);
|
2148
|
+
size_t const restSize = length - copySize;
|
2149
|
+
LZ4_memcpy(op, dictEnd - copySize, copySize);
|
2150
|
+
op += copySize;
|
2151
|
+
if (restSize > (size_t)(op - lowPrefix)) { /* overlap copy */
|
2152
|
+
BYTE* const endOfMatch = op + restSize;
|
2153
|
+
const BYTE* copyFrom = lowPrefix;
|
2154
|
+
while (op < endOfMatch) *op++ = *copyFrom++;
|
2155
|
+
} else {
|
2156
|
+
LZ4_memcpy(op, lowPrefix, restSize);
|
2157
|
+
op += restSize;
|
2158
|
+
} }
|
2159
|
+
continue;
|
2160
|
+
}
|
2161
|
+
assert(match >= lowPrefix);
|
2162
|
+
|
2163
|
+
/* copy match within block */
|
2164
|
+
cpy = op + length;
|
2165
|
+
|
2166
|
+
/* partialDecoding : may end anywhere within the block */
|
2167
|
+
assert(op<=oend);
|
2168
|
+
if (partialDecoding && (cpy > oend-MATCH_SAFEGUARD_DISTANCE)) {
|
2169
|
+
size_t const mlen = MIN(length, (size_t)(oend-op));
|
2170
|
+
const BYTE* const matchEnd = match + mlen;
|
2171
|
+
BYTE* const copyEnd = op + mlen;
|
2172
|
+
if (matchEnd > op) { /* overlap copy */
|
2173
|
+
while (op < copyEnd) { *op++ = *match++; }
|
2174
|
+
} else {
|
2175
|
+
LZ4_memcpy(op, match, mlen);
|
2176
|
+
}
|
2177
|
+
op = copyEnd;
|
2178
|
+
if (op == oend) { break; }
|
2179
|
+
continue;
|
2180
|
+
}
|
2181
|
+
|
2182
|
+
if (unlikely(offset<8)) {
|
2183
|
+
LZ4_write32(op, 0); /* silence msan warning when offset==0 */
|
2184
|
+
op[0] = match[0];
|
2185
|
+
op[1] = match[1];
|
2186
|
+
op[2] = match[2];
|
2187
|
+
op[3] = match[3];
|
2188
|
+
match += inc32table[offset];
|
2189
|
+
LZ4_memcpy(op+4, match, 4);
|
2190
|
+
match -= dec64table[offset];
|
2191
|
+
} else {
|
2192
|
+
LZ4_memcpy(op, match, 8);
|
2193
|
+
match += 8;
|
2194
|
+
}
|
2195
|
+
op += 8;
|
2196
|
+
|
2197
|
+
if (unlikely(cpy > oend-MATCH_SAFEGUARD_DISTANCE)) {
|
2198
|
+
BYTE* const oCopyLimit = oend - (WILDCOPYLENGTH-1);
|
2199
|
+
if (cpy > oend-LASTLITERALS) { goto _output_error; } /* Error : last LASTLITERALS bytes must be literals (uncompressed) */
|
2200
|
+
if (op < oCopyLimit) {
|
2201
|
+
LZ4_wildCopy8(op, match, oCopyLimit);
|
2202
|
+
match += oCopyLimit - op;
|
2203
|
+
op = oCopyLimit;
|
2204
|
+
}
|
2205
|
+
while (op < cpy) { *op++ = *match++; }
|
2206
|
+
} else {
|
2207
|
+
LZ4_memcpy(op, match, 8);
|
2208
|
+
if (length > 16) { LZ4_wildCopy8(op+8, match+8, cpy); }
|
2209
|
+
}
|
2210
|
+
op = cpy; /* wildcopy correction */
|
2211
|
+
}
|
2212
|
+
|
2213
|
+
/* end of decoding */
|
2214
|
+
DEBUGLOG(5, "decoded %i bytes", (int) (((char*)op)-dst));
|
2215
|
+
return (int) (((char*)op)-dst); /* Nb of output bytes decoded */
|
2216
|
+
|
2217
|
+
/* Overflow error detected */
|
2218
|
+
_output_error:
|
2219
|
+
return (int) (-(((const char*)ip)-src))-1;
|
2220
|
+
}
|
2221
|
+
}
|
2222
|
+
|
2223
|
+
|
2224
|
+
/*===== Instantiate the API decoding functions. =====*/
|
2225
|
+
|
2226
|
+
LZ4_FORCE_O2
|
2227
|
+
int LZ4_decompress_safe(const char* source, char* dest, int compressedSize, int maxDecompressedSize)
|
2228
|
+
{
|
2229
|
+
return LZ4_decompress_generic(source, dest, compressedSize, maxDecompressedSize,
|
2230
|
+
decode_full_block, noDict,
|
2231
|
+
(BYTE*)dest, NULL, 0);
|
2232
|
+
}
|
2233
|
+
|
2234
|
+
LZ4_FORCE_O2
|
2235
|
+
int LZ4_decompress_safe_partial(const char* src, char* dst, int compressedSize, int targetOutputSize, int dstCapacity)
|
2236
|
+
{
|
2237
|
+
dstCapacity = MIN(targetOutputSize, dstCapacity);
|
2238
|
+
return LZ4_decompress_generic(src, dst, compressedSize, dstCapacity,
|
2239
|
+
partial_decode,
|
2240
|
+
noDict, (BYTE*)dst, NULL, 0);
|
2241
|
+
}
|
2242
|
+
|
2243
|
+
LZ4_FORCE_O2
|
2244
|
+
int LZ4_decompress_fast(const char* source, char* dest, int originalSize)
|
2245
|
+
{
|
2246
|
+
DEBUGLOG(5, "LZ4_decompress_fast");
|
2247
|
+
return LZ4_decompress_unsafe_generic(
|
2248
|
+
(const BYTE*)source, (BYTE*)dest, originalSize,
|
2249
|
+
0, NULL, 0);
|
2250
|
+
}
|
2251
|
+
|
2252
|
+
/*===== Instantiate a few more decoding cases, used more than once. =====*/
|
2253
|
+
|
2254
|
+
LZ4_FORCE_O2 /* Exported, an obsolete API function. */
|
2255
|
+
int LZ4_decompress_safe_withPrefix64k(const char* source, char* dest, int compressedSize, int maxOutputSize)
|
2256
|
+
{
|
2257
|
+
return LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize,
|
2258
|
+
decode_full_block, withPrefix64k,
|
2259
|
+
(BYTE*)dest - 64 KB, NULL, 0);
|
2260
|
+
}
|
2261
|
+
|
2262
|
+
LZ4_FORCE_O2
|
2263
|
+
static int LZ4_decompress_safe_partial_withPrefix64k(const char* source, char* dest, int compressedSize, int targetOutputSize, int dstCapacity)
|
2264
|
+
{
|
2265
|
+
dstCapacity = MIN(targetOutputSize, dstCapacity);
|
2266
|
+
return LZ4_decompress_generic(source, dest, compressedSize, dstCapacity,
|
2267
|
+
partial_decode, withPrefix64k,
|
2268
|
+
(BYTE*)dest - 64 KB, NULL, 0);
|
2269
|
+
}
|
2270
|
+
|
2271
|
+
/* Another obsolete API function, paired with the previous one. */
|
2272
|
+
int LZ4_decompress_fast_withPrefix64k(const char* source, char* dest, int originalSize)
|
2273
|
+
{
|
2274
|
+
return LZ4_decompress_unsafe_generic(
|
2275
|
+
(const BYTE*)source, (BYTE*)dest, originalSize,
|
2276
|
+
64 KB, NULL, 0);
|
2277
|
+
}
|
2278
|
+
|
2279
|
+
LZ4_FORCE_O2
|
2280
|
+
static int LZ4_decompress_safe_withSmallPrefix(const char* source, char* dest, int compressedSize, int maxOutputSize,
|
2281
|
+
size_t prefixSize)
|
2282
|
+
{
|
2283
|
+
return LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize,
|
2284
|
+
decode_full_block, noDict,
|
2285
|
+
(BYTE*)dest-prefixSize, NULL, 0);
|
2286
|
+
}
|
2287
|
+
|
2288
|
+
LZ4_FORCE_O2
|
2289
|
+
static int LZ4_decompress_safe_partial_withSmallPrefix(const char* source, char* dest, int compressedSize, int targetOutputSize, int dstCapacity,
|
2290
|
+
size_t prefixSize)
|
2291
|
+
{
|
2292
|
+
dstCapacity = MIN(targetOutputSize, dstCapacity);
|
2293
|
+
return LZ4_decompress_generic(source, dest, compressedSize, dstCapacity,
|
2294
|
+
partial_decode, noDict,
|
2295
|
+
(BYTE*)dest-prefixSize, NULL, 0);
|
2296
|
+
}
|
2297
|
+
|
2298
|
+
LZ4_FORCE_O2
|
2299
|
+
int LZ4_decompress_safe_forceExtDict(const char* source, char* dest,
|
2300
|
+
int compressedSize, int maxOutputSize,
|
2301
|
+
const void* dictStart, size_t dictSize)
|
2302
|
+
{
|
2303
|
+
return LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize,
|
2304
|
+
decode_full_block, usingExtDict,
|
2305
|
+
(BYTE*)dest, (const BYTE*)dictStart, dictSize);
|
2306
|
+
}
|
2307
|
+
|
2308
|
+
LZ4_FORCE_O2
|
2309
|
+
int LZ4_decompress_safe_partial_forceExtDict(const char* source, char* dest,
|
2310
|
+
int compressedSize, int targetOutputSize, int dstCapacity,
|
2311
|
+
const void* dictStart, size_t dictSize)
|
2312
|
+
{
|
2313
|
+
dstCapacity = MIN(targetOutputSize, dstCapacity);
|
2314
|
+
return LZ4_decompress_generic(source, dest, compressedSize, dstCapacity,
|
2315
|
+
partial_decode, usingExtDict,
|
2316
|
+
(BYTE*)dest, (const BYTE*)dictStart, dictSize);
|
2317
|
+
}
|
2318
|
+
|
2319
|
+
LZ4_FORCE_O2
|
2320
|
+
static int LZ4_decompress_fast_extDict(const char* source, char* dest, int originalSize,
|
2321
|
+
const void* dictStart, size_t dictSize)
|
2322
|
+
{
|
2323
|
+
return LZ4_decompress_unsafe_generic(
|
2324
|
+
(const BYTE*)source, (BYTE*)dest, originalSize,
|
2325
|
+
0, (const BYTE*)dictStart, dictSize);
|
2326
|
+
}
|
2327
|
+
|
2328
|
+
/* The "double dictionary" mode, for use with e.g. ring buffers: the first part
|
2329
|
+
* of the dictionary is passed as prefix, and the second via dictStart + dictSize.
|
2330
|
+
* These routines are used only once, in LZ4_decompress_*_continue().
|
2331
|
+
*/
|
2332
|
+
LZ4_FORCE_INLINE
|
2333
|
+
int LZ4_decompress_safe_doubleDict(const char* source, char* dest, int compressedSize, int maxOutputSize,
|
2334
|
+
size_t prefixSize, const void* dictStart, size_t dictSize)
|
2335
|
+
{
|
2336
|
+
return LZ4_decompress_generic(source, dest, compressedSize, maxOutputSize,
|
2337
|
+
decode_full_block, usingExtDict,
|
2338
|
+
(BYTE*)dest-prefixSize, (const BYTE*)dictStart, dictSize);
|
2339
|
+
}
|
2340
|
+
|
2341
|
+
/*===== streaming decompression functions =====*/
|
2342
|
+
|
2343
|
+
#if !defined(LZ4_STATIC_LINKING_ONLY_DISABLE_MEMORY_ALLOCATION)
|
2344
|
+
LZ4_streamDecode_t* LZ4_createStreamDecode(void)
|
2345
|
+
{
|
2346
|
+
LZ4_STATIC_ASSERT(sizeof(LZ4_streamDecode_t) >= sizeof(LZ4_streamDecode_t_internal));
|
2347
|
+
return (LZ4_streamDecode_t*) ALLOC_AND_ZERO(sizeof(LZ4_streamDecode_t));
|
2348
|
+
}
|
2349
|
+
|
2350
|
+
int LZ4_freeStreamDecode (LZ4_streamDecode_t* LZ4_stream)
|
2351
|
+
{
|
2352
|
+
if (LZ4_stream == NULL) { return 0; } /* support free on NULL */
|
2353
|
+
FREEMEM(LZ4_stream);
|
2354
|
+
return 0;
|
2355
|
+
}
|
2356
|
+
#endif
|
2357
|
+
|
2358
|
+
/*! LZ4_setStreamDecode() :
|
2359
|
+
* Use this function to instruct where to find the dictionary.
|
2360
|
+
* This function is not necessary if previous data is still available where it was decoded.
|
2361
|
+
* Loading a size of 0 is allowed (same effect as no dictionary).
|
2362
|
+
* @return : 1 if OK, 0 if error
|
2363
|
+
*/
|
2364
|
+
int LZ4_setStreamDecode (LZ4_streamDecode_t* LZ4_streamDecode, const char* dictionary, int dictSize)
|
2365
|
+
{
|
2366
|
+
LZ4_streamDecode_t_internal* lz4sd = &LZ4_streamDecode->internal_donotuse;
|
2367
|
+
lz4sd->prefixSize = (size_t)dictSize;
|
2368
|
+
if (dictSize) {
|
2369
|
+
assert(dictionary != NULL);
|
2370
|
+
lz4sd->prefixEnd = (const BYTE*) dictionary + dictSize;
|
2371
|
+
} else {
|
2372
|
+
lz4sd->prefixEnd = (const BYTE*) dictionary;
|
2373
|
+
}
|
2374
|
+
lz4sd->externalDict = NULL;
|
2375
|
+
lz4sd->extDictSize = 0;
|
2376
|
+
return 1;
|
2377
|
+
}
|
2378
|
+
|
2379
|
+
/*! LZ4_decoderRingBufferSize() :
|
2380
|
+
* when setting a ring buffer for streaming decompression (optional scenario),
|
2381
|
+
* provides the minimum size of this ring buffer
|
2382
|
+
* to be compatible with any source respecting maxBlockSize condition.
|
2383
|
+
* Note : in a ring buffer scenario,
|
2384
|
+
* blocks are presumed decompressed next to each other.
|
2385
|
+
* When not enough space remains for next block (remainingSize < maxBlockSize),
|
2386
|
+
* decoding resumes from beginning of ring buffer.
|
2387
|
+
* @return : minimum ring buffer size,
|
2388
|
+
* or 0 if there is an error (invalid maxBlockSize).
|
2389
|
+
*/
|
2390
|
+
int LZ4_decoderRingBufferSize(int maxBlockSize)
|
2391
|
+
{
|
2392
|
+
if (maxBlockSize < 0) return 0;
|
2393
|
+
if (maxBlockSize > LZ4_MAX_INPUT_SIZE) return 0;
|
2394
|
+
if (maxBlockSize < 16) maxBlockSize = 16;
|
2395
|
+
return LZ4_DECODER_RING_BUFFER_SIZE(maxBlockSize);
|
2396
|
+
}
|
2397
|
+
|
2398
|
+
/*
|
2399
|
+
*_continue() :
|
2400
|
+
These decoding functions allow decompression of multiple blocks in "streaming" mode.
|
2401
|
+
Previously decoded blocks must still be available at the memory position where they were decoded.
|
2402
|
+
If it's not possible, save the relevant part of decoded data into a safe buffer,
|
2403
|
+
and indicate where it stands using LZ4_setStreamDecode()
|
2404
|
+
*/
|
2405
|
+
LZ4_FORCE_O2
|
2406
|
+
int LZ4_decompress_safe_continue (LZ4_streamDecode_t* LZ4_streamDecode, const char* source, char* dest, int compressedSize, int maxOutputSize)
|
2407
|
+
{
|
2408
|
+
LZ4_streamDecode_t_internal* lz4sd = &LZ4_streamDecode->internal_donotuse;
|
2409
|
+
int result;
|
2410
|
+
|
2411
|
+
if (lz4sd->prefixSize == 0) {
|
2412
|
+
/* The first call, no dictionary yet. */
|
2413
|
+
assert(lz4sd->extDictSize == 0);
|
2414
|
+
result = LZ4_decompress_safe(source, dest, compressedSize, maxOutputSize);
|
2415
|
+
if (result <= 0) return result;
|
2416
|
+
lz4sd->prefixSize = (size_t)result;
|
2417
|
+
lz4sd->prefixEnd = (BYTE*)dest + result;
|
2418
|
+
} else if (lz4sd->prefixEnd == (BYTE*)dest) {
|
2419
|
+
/* They're rolling the current segment. */
|
2420
|
+
if (lz4sd->prefixSize >= 64 KB - 1)
|
2421
|
+
result = LZ4_decompress_safe_withPrefix64k(source, dest, compressedSize, maxOutputSize);
|
2422
|
+
else if (lz4sd->extDictSize == 0)
|
2423
|
+
result = LZ4_decompress_safe_withSmallPrefix(source, dest, compressedSize, maxOutputSize,
|
2424
|
+
lz4sd->prefixSize);
|
2425
|
+
else
|
2426
|
+
result = LZ4_decompress_safe_doubleDict(source, dest, compressedSize, maxOutputSize,
|
2427
|
+
lz4sd->prefixSize, lz4sd->externalDict, lz4sd->extDictSize);
|
2428
|
+
if (result <= 0) return result;
|
2429
|
+
lz4sd->prefixSize += (size_t)result;
|
2430
|
+
lz4sd->prefixEnd += result;
|
2431
|
+
} else {
|
2432
|
+
/* The buffer wraps around, or they're switching to another buffer. */
|
2433
|
+
lz4sd->extDictSize = lz4sd->prefixSize;
|
2434
|
+
lz4sd->externalDict = lz4sd->prefixEnd - lz4sd->extDictSize;
|
2435
|
+
result = LZ4_decompress_safe_forceExtDict(source, dest, compressedSize, maxOutputSize,
|
2436
|
+
lz4sd->externalDict, lz4sd->extDictSize);
|
2437
|
+
if (result <= 0) return result;
|
2438
|
+
lz4sd->prefixSize = (size_t)result;
|
2439
|
+
lz4sd->prefixEnd = (BYTE*)dest + result;
|
2440
|
+
}
|
2441
|
+
|
2442
|
+
return result;
|
2443
|
+
}
|
2444
|
+
|
2445
|
+
LZ4_FORCE_O2 int
|
2446
|
+
LZ4_decompress_fast_continue (LZ4_streamDecode_t* LZ4_streamDecode,
|
2447
|
+
const char* source, char* dest, int originalSize)
|
2448
|
+
{
|
2449
|
+
LZ4_streamDecode_t_internal* const lz4sd =
|
2450
|
+
(assert(LZ4_streamDecode!=NULL), &LZ4_streamDecode->internal_donotuse);
|
2451
|
+
int result;
|
2452
|
+
|
2453
|
+
DEBUGLOG(5, "LZ4_decompress_fast_continue (toDecodeSize=%i)", originalSize);
|
2454
|
+
assert(originalSize >= 0);
|
2455
|
+
|
2456
|
+
if (lz4sd->prefixSize == 0) {
|
2457
|
+
DEBUGLOG(5, "first invocation : no prefix nor extDict");
|
2458
|
+
assert(lz4sd->extDictSize == 0);
|
2459
|
+
result = LZ4_decompress_fast(source, dest, originalSize);
|
2460
|
+
if (result <= 0) return result;
|
2461
|
+
lz4sd->prefixSize = (size_t)originalSize;
|
2462
|
+
lz4sd->prefixEnd = (BYTE*)dest + originalSize;
|
2463
|
+
} else if (lz4sd->prefixEnd == (BYTE*)dest) {
|
2464
|
+
DEBUGLOG(5, "continue using existing prefix");
|
2465
|
+
result = LZ4_decompress_unsafe_generic(
|
2466
|
+
(const BYTE*)source, (BYTE*)dest, originalSize,
|
2467
|
+
lz4sd->prefixSize,
|
2468
|
+
lz4sd->externalDict, lz4sd->extDictSize);
|
2469
|
+
if (result <= 0) return result;
|
2470
|
+
lz4sd->prefixSize += (size_t)originalSize;
|
2471
|
+
lz4sd->prefixEnd += originalSize;
|
2472
|
+
} else {
|
2473
|
+
DEBUGLOG(5, "prefix becomes extDict");
|
2474
|
+
lz4sd->extDictSize = lz4sd->prefixSize;
|
2475
|
+
lz4sd->externalDict = lz4sd->prefixEnd - lz4sd->extDictSize;
|
2476
|
+
result = LZ4_decompress_fast_extDict(source, dest, originalSize,
|
2477
|
+
lz4sd->externalDict, lz4sd->extDictSize);
|
2478
|
+
if (result <= 0) return result;
|
2479
|
+
lz4sd->prefixSize = (size_t)originalSize;
|
2480
|
+
lz4sd->prefixEnd = (BYTE*)dest + originalSize;
|
2481
|
+
}
|
2482
|
+
|
2483
|
+
return result;
|
2484
|
+
}
|
2485
|
+
|
2486
|
+
|
2487
|
+
/*
|
2488
|
+
Advanced decoding functions :
|
2489
|
+
*_usingDict() :
|
2490
|
+
These decoding functions work the same as "_continue" ones,
|
2491
|
+
the dictionary must be explicitly provided within parameters
|
2492
|
+
*/
|
2493
|
+
|
2494
|
+
int LZ4_decompress_safe_usingDict(const char* source, char* dest, int compressedSize, int maxOutputSize, const char* dictStart, int dictSize)
|
2495
|
+
{
|
2496
|
+
if (dictSize==0)
|
2497
|
+
return LZ4_decompress_safe(source, dest, compressedSize, maxOutputSize);
|
2498
|
+
if (dictStart+dictSize == dest) {
|
2499
|
+
if (dictSize >= 64 KB - 1) {
|
2500
|
+
return LZ4_decompress_safe_withPrefix64k(source, dest, compressedSize, maxOutputSize);
|
2501
|
+
}
|
2502
|
+
assert(dictSize >= 0);
|
2503
|
+
return LZ4_decompress_safe_withSmallPrefix(source, dest, compressedSize, maxOutputSize, (size_t)dictSize);
|
2504
|
+
}
|
2505
|
+
assert(dictSize >= 0);
|
2506
|
+
return LZ4_decompress_safe_forceExtDict(source, dest, compressedSize, maxOutputSize, dictStart, (size_t)dictSize);
|
2507
|
+
}
|
2508
|
+
|
2509
|
+
int LZ4_decompress_safe_partial_usingDict(const char* source, char* dest, int compressedSize, int targetOutputSize, int dstCapacity, const char* dictStart, int dictSize)
|
2510
|
+
{
|
2511
|
+
if (dictSize==0)
|
2512
|
+
return LZ4_decompress_safe_partial(source, dest, compressedSize, targetOutputSize, dstCapacity);
|
2513
|
+
if (dictStart+dictSize == dest) {
|
2514
|
+
if (dictSize >= 64 KB - 1) {
|
2515
|
+
return LZ4_decompress_safe_partial_withPrefix64k(source, dest, compressedSize, targetOutputSize, dstCapacity);
|
2516
|
+
}
|
2517
|
+
assert(dictSize >= 0);
|
2518
|
+
return LZ4_decompress_safe_partial_withSmallPrefix(source, dest, compressedSize, targetOutputSize, dstCapacity, (size_t)dictSize);
|
2519
|
+
}
|
2520
|
+
assert(dictSize >= 0);
|
2521
|
+
return LZ4_decompress_safe_partial_forceExtDict(source, dest, compressedSize, targetOutputSize, dstCapacity, dictStart, (size_t)dictSize);
|
2522
|
+
}
|
2523
|
+
|
2524
|
+
int LZ4_decompress_fast_usingDict(const char* source, char* dest, int originalSize, const char* dictStart, int dictSize)
|
2525
|
+
{
|
2526
|
+
if (dictSize==0 || dictStart+dictSize == dest)
|
2527
|
+
return LZ4_decompress_unsafe_generic(
|
2528
|
+
(const BYTE*)source, (BYTE*)dest, originalSize,
|
2529
|
+
(size_t)dictSize, NULL, 0);
|
2530
|
+
assert(dictSize >= 0);
|
2531
|
+
return LZ4_decompress_fast_extDict(source, dest, originalSize, dictStart, (size_t)dictSize);
|
2532
|
+
}
|
2533
|
+
|
2534
|
+
|
2535
|
+
/*=*************************************************
|
2536
|
+
* Obsolete Functions
|
2537
|
+
***************************************************/
|
2538
|
+
/* obsolete compression functions */
|
2539
|
+
int LZ4_compress_limitedOutput(const char* source, char* dest, int inputSize, int maxOutputSize)
|
2540
|
+
{
|
2541
|
+
return LZ4_compress_default(source, dest, inputSize, maxOutputSize);
|
2542
|
+
}
|
2543
|
+
int LZ4_compress(const char* src, char* dest, int srcSize)
|
2544
|
+
{
|
2545
|
+
return LZ4_compress_default(src, dest, srcSize, LZ4_compressBound(srcSize));
|
2546
|
+
}
|
2547
|
+
int LZ4_compress_limitedOutput_withState (void* state, const char* src, char* dst, int srcSize, int dstSize)
|
2548
|
+
{
|
2549
|
+
return LZ4_compress_fast_extState(state, src, dst, srcSize, dstSize, 1);
|
2550
|
+
}
|
2551
|
+
int LZ4_compress_withState (void* state, const char* src, char* dst, int srcSize)
|
2552
|
+
{
|
2553
|
+
return LZ4_compress_fast_extState(state, src, dst, srcSize, LZ4_compressBound(srcSize), 1);
|
2554
|
+
}
|
2555
|
+
int LZ4_compress_limitedOutput_continue (LZ4_stream_t* LZ4_stream, const char* src, char* dst, int srcSize, int dstCapacity)
|
2556
|
+
{
|
2557
|
+
return LZ4_compress_fast_continue(LZ4_stream, src, dst, srcSize, dstCapacity, 1);
|
2558
|
+
}
|
2559
|
+
int LZ4_compress_continue (LZ4_stream_t* LZ4_stream, const char* source, char* dest, int inputSize)
|
2560
|
+
{
|
2561
|
+
return LZ4_compress_fast_continue(LZ4_stream, source, dest, inputSize, LZ4_compressBound(inputSize), 1);
|
2562
|
+
}
|
2563
|
+
|
2564
|
+
/*
|
2565
|
+
These decompression functions are deprecated and should no longer be used.
|
2566
|
+
They are only provided here for compatibility with older user programs.
|
2567
|
+
- LZ4_uncompress is totally equivalent to LZ4_decompress_fast
|
2568
|
+
- LZ4_uncompress_unknownOutputSize is totally equivalent to LZ4_decompress_safe
|
2569
|
+
*/
|
2570
|
+
int LZ4_uncompress (const char* source, char* dest, int outputSize)
|
2571
|
+
{
|
2572
|
+
return LZ4_decompress_fast(source, dest, outputSize);
|
2573
|
+
}
|
2574
|
+
int LZ4_uncompress_unknownOutputSize (const char* source, char* dest, int isize, int maxOutputSize)
|
2575
|
+
{
|
2576
|
+
return LZ4_decompress_safe(source, dest, isize, maxOutputSize);
|
2577
|
+
}
|
2578
|
+
|
2579
|
+
/* Obsolete Streaming functions */
|
2580
|
+
|
2581
|
+
int LZ4_sizeofStreamState(void) { return sizeof(LZ4_stream_t); }
|
2582
|
+
|
2583
|
+
int LZ4_resetStreamState(void* state, char* inputBuffer)
|
2584
|
+
{
|
2585
|
+
(void)inputBuffer;
|
2586
|
+
LZ4_resetStream((LZ4_stream_t*)state);
|
2587
|
+
return 0;
|
2588
|
+
}
|
2589
|
+
|
2590
|
+
#if !defined(LZ4_STATIC_LINKING_ONLY_DISABLE_MEMORY_ALLOCATION)
|
2591
|
+
void* LZ4_create (char* inputBuffer)
|
2592
|
+
{
|
2593
|
+
(void)inputBuffer;
|
2594
|
+
return LZ4_createStream();
|
2595
|
+
}
|
2596
|
+
#endif
|
2597
|
+
|
2598
|
+
char* LZ4_slideInputBuffer (void* state)
|
2599
|
+
{
|
2600
|
+
/* avoid const char * -> char * conversion warning */
|
2601
|
+
return (char *)(uptrval)((LZ4_stream_t*)state)->internal_donotuse.dictionary;
|
2602
|
+
}
|
2603
|
+
|
2604
|
+
#endif /* LZ4_COMMONDEFS_ONLY */
|
2605
|
+
}
|