support for binlog compressed payload (zstd) — streaming refactor#178
Closed
alexpacio wants to merge 1 commit into
Closed
support for binlog compressed payload (zstd) — streaming refactor#178alexpacio wants to merge 1 commit into
alexpacio wants to merge 1 commit into
Conversation
6 tasks
…oadEventBuffer Transparently unpacks zstd/none-compressed TRANSACTION_PAYLOAD events, re-emitting their inner events as ordinary top-level events. Decompression is streamed/iterated with bounded memory, so transactions whose compressed or uncompressed image exceeds the 2GB byte[] limit (or spans 16MB packets) are handled. The per-payload streaming state machine now lives in a dedicated TransactionPayloadEventBuffer held by EventDeserializer, instead of inline in EventDeserializer (nextEvent simply drains the buffer before reading the stream). Custom TRANSACTION_PAYLOAD deserializers fall back to whole-payload materialization. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
alexpacio
force-pushed
the
mysql-binlog-zstd-compression-refactor
branch
from
bf8af64 to
df73a95
Compare
Author
|
@osheroff i've refactored it using claude in a dedicated single commit and in a separate branch based on your last suggestion. However I can't go touching manually the code more than that as I am definitely not able to handle the code manually, I'm not a java developer. |
Owner
|
Ok Alex, let’s get it as close as you can and I’ll take it across the line from there On Jun 15, 2026, at 12:14, Alessandro Bolletta ***@***.***> wrote:alexpacio left a comment (osheroff/mysql-binlog-connector-java#178)
@osheroff i've refactored it using claude in a dedicated single commit and in a separate branch based on your last suggestion. However I can't go touching manually the code more than that as I am definitely not able to handle the code manually, I'm not a java developer.
—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you were mentioned.Message ID: ***@***.***>
|
Author
|
Superseeded by #179 |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Refactor of #177 with a cleaner structure (supersedes it).
Transparently unpacks
TRANSACTION_PAYLOADevents (ZSTD / NONE) and re-emits theirinner events as ordinary top-level events, so consumers keep seeing
QUERY,TABLE_MAP, row and commit events. Decompression is streamed/iterated with boundedmemory, so a transaction whose compressed or uncompressed image exceeds the 2GB
byte[]limit (or spans 16MB packets) is handled.What changed vs #177
The per-payload streaming state machine no longer lives inline in
EventDeserializer.It moved into a dedicated
TransactionPayloadEventBufferheld byEventDeserializer:nextEvent()simply drains the buffer (hasPending()/next()) before reading thestream, and hands a
TRANSACTION_PAYLOADoff viaopen(...). This shrankEventDeserializer's diff vs master from +211 to +38 lines.Also: a custom
TRANSACTION_PAYLOADdeserializer now falls back to whole-payloadmaterialization (previously it threw).
Tests
Contract + unit tests pass, including the end-to-end
BinaryLogFileReaderIntegrationTestdecoding the real
mysql-bin.compressedresource (5 → 8 top-level events, noTRANSACTION_PAYLOADleaking through), no-prefetch streaming, restamped inner-eventcoordinates, >2GB payload handling, and cross-packet reads. The MySQL-backed
integration tests need a live server to run.
🤖 Generated with Claude Code