update bloom filter for the new prefix bf

[LeftHandCold]

User description

What type of PR is this?

• API-change
• BUG
• Improvement
• Documentation
• Feature
• Test and CI
• Code Refactoring

Which issue(s) this PR fixes:

issue #16628 #16629

What this PR does / why we need it:

support prefix bloom filter for object reader and writer

---

PR Type

Feature, Tests

---

Description

• Added support for prefix bloom filter in object reader and writer.
• Introduced methods to set and get bloom filter type in BlockHeader.
• Updated ReadOneBF and WriteBF methods to handle bloom filter types.
• Enhanced BlockWriter to support various bloom filter types and added corresponding tests.
• Added factory method for creating bloom filters by type.
• Defined PrefixFn struct and added ErrPrefix error.

---

Changes walkthrough 📝

	Relevant files
Enhancement	meta.go Add bloom filter type handling in BlockHeader                        pkg/objectio/meta.go Added constants for bloom filter type offset and length. Introduced methods to set and get bloom filter type in BlockHeader. +11/-1    reader.go Update ReadOneBF to handle bloom filter type                          pkg/objectio/reader.go Modified ReadOneBF to use the new bloom filter type. +2/-1      writer.go Modify WriteBF to include bloom filter type                            pkg/objectio/writer.go Updated WriteBF to include bloom filter type. +2/-1      writer.go Extend BlockWriter to support various bloom filter types  pkg/vm/engine/tae/blockio/writer.go Added pkType and prefix fields to BlockWriter. Introduced SetPrimaryKeyWithType method. Enhanced WriteBatch to handle different bloom filter types. +30/-2    filter.go Add factory method for creating bloom filters by type        pkg/vm/engine/tae/index/filter.go Introduced NewEmptyBloomFilterWithType to create bloom filters based on type. +12/-0    types.go Add prefix filter error and PrefixFn struct                            pkg/vm/engine/tae/index/types.go Added ErrPrefix error. Defined PrefixFn struct. +6/-0
Tests	writer_test.go Add tests for writing blocks with bloom filter types          pkg/vm/engine/tae/blockio/writer_test.go Added tests for writing blocks with different bloom filter types. +111/-0

---

💡 PR-Agent usage:
Comment /help on the PR to get a list of all available PR-Agent tools and their descriptions

[codiumai-pr-agent-pro]

PR-Agent was enabled for this repository. To continue using it, please link your git user with your CodiumAI identity here.

PR Review 🔍

⏱️ Estimated effort to review [1-5]	4, because the PR introduces significant changes across multiple files, including new methods for handling different types of bloom filters, updates to object reading and writing logic, and modifications to block writing strategies. The complexity of bloom filter handling and the potential interactions between different components increase the review effort.
🧪 Relevant tests	Yes
⚡ Possible issues	Error Handling: The implementation assumes that the prefix functions are always provided when required (e.g., for hybrid and prefix bloom filters). If not properly handled, this might lead to runtime errors or incorrect behavior.
	Bloom Filter Initialization: The NewEmptyBloomFilterWithType function returns nil if the type is not recognized, which could lead to nil pointer dereferences if not checked.
🔒 Security concerns	No

[codiumai-pr-agent-pro]

PR-Agent was enabled for this repository. To continue using it, please link your git user with your CodiumAI identity here.

PR Code Suggestions ✨

Category	Suggestion	Score
Best practice	Add an error check after calling NewBlockWriterNew in the test function To ensure that the err variable is properly checked, consider adding an error check after calling NewBlockWriterNew in the TestWriter_WriteBlockAndBF function. pkg/vm/engine/tae/blockio/writer_test.go [228] -writer, _ := NewBlockWriterNew(service, name, 0, nil) +writer, err := NewBlockWriterNew(service, name, 0, nil) +require.NoError(t, err) Suggestion importance[1-10]: 8 Why: Proper error handling is crucial in tests to ensure that any setup failures are caught early, preventing misleading test results and facilitating debugging.	8
Possible issue	Initialize the prefix slice with a capacity equal to the length of the prefix variadic parameter To avoid potential issues with uninitialized or nil slices, consider initializing the prefix slice with a capacity equal to the length of the prefix variadic parameter. pkg/objectio/meta.go [85-86] -w.prefix = make([]index.PrefixFn, 0) -w.prefix = append(w.prefix, prefix...) +w.prefix = make([]index.PrefixFn, len(prefix)) +copy(w.prefix, prefix) Suggestion importance[1-10]: 7 Why: Initializing the slice with a predefined capacity that matches the expected number of elements can improve memory allocation efficiency and performance.	7
	Return a default value or an error instead of nil for unexpected t values To handle unexpected t values more gracefully, consider returning an error or a default value instead of nil in the NewEmptyBloomFilterWithType function. pkg/vm/engine/tae/index/filter.go [47-49] -} else { - return nil +default: + return &bloomFilter{} // or return an error } Suggestion importance[1-10]: 7 Why: Returning a default value or an error instead of nil can prevent potential runtime errors due to nil dereference, improving the robustness of the code.	7
Maintainability	Use a switch statement instead of multiple if-else statements for checking the pkType To improve readability and maintainability, consider using a switch statement instead of multiple if-else statements when checking the pkType. pkg/vm/engine/tae/blockio/writer.go [145-160] -if w.pkType == index.BF { +switch w.pkType { +case index.BF: bf, err = index.NewBloomFilter(columnData) -} else if w.pkType == index.PBF { +case index.PBF: if len(w.prefix) < 1 { return nil, index.ErrPrefix } prefix := w.prefix[0] bf, err = index.NewPrefixBloomFilter(columnData, prefix.Id, prefix.Fn) -} else if w.pkType == index.HBF { +case index.HBF: if len(w.prefix) < 2 { return nil, index.ErrPrefix } prefixL1 := w.prefix[0] prefixL2 := w.prefix[1] bf, err = index.NewHybridBloomFilter(columnData, prefixL1.Id, prefixL1.Fn, prefixL2.Id, prefixL2.Fn) } Suggestion importance[1-10]: 6 Why: Using a switch statement enhances readability and maintainability of the code by providing a clearer structure for conditional branches.	6