Introduce patchsets - #551
Draft
mason-sharp wants to merge 2 commits into
Draft
Conversation
Introduce a runtime handshake between Spock and the PostgreSQL core patches it relies on, so Spock can tell which core patch-set "generation" a server provides. - New pgNN-000-spock-patchset.diff for PG 15-19 defines an exported global `int SpockCorePatchsetVersion` (the patch-set generation; starts at 1, 0 reserved to mean a patch set predating the marker / baseline). - Spock detects it at load time via dlsym(RTLD_DEFAULT, ...) and records it into `spock_detected_patchset`. dlsym (not a hard PGDLLIMPORT extern or a weak symbol) keeps behavior identical on Linux/ELF and macOS/Mach-O and lets spock.so load regardless of the symbol's presence. - _PG_init() logs the detected generation and continues. Detection is informational only: no startup enforcement, so existing patched servers that predate the marker (reporting generation 0) keep working. The SPOCK_CORE_PATCHSET_TARGET macro records the generation this build targets for future use. - spock.core_patchset() SQL function exposes the recorded value, added to both the fresh-install (spock--6.0.0.sql) and the 5.0.10->6.0.0 upgrade script so upgraded databases get it too. - TAP test 106_core_patchset.pl covers the positive path. Enforcement and feature-gating are deferred to a future release.
📝 WalkthroughWalkthroughChangesThe patch adds core patch-set version tracking for PostgreSQL 15–19, resolves patched symbols dynamically during module initialization, routes apply operations through those symbols, exposes Core patch-set integration
Poem
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Up to standards ✅🟢 Issues
|
| Metric | Results |
|---|---|
| Duplication | 0 |
NEW Get contextual insights on your PRs based on Codacy's metrics, along with PR and Jira context, without leaving GitHub. Enable AI reviewer
TIP This summary will be updated as you push new changes.
mason-sharp
marked this pull request as draft
July 28, 2026 19:51
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/spock.c`:
- Around line 1055-1069: Require both resolved symbols in src/spock.c lines
1055-1069 by setting spock_hlc_available only when spock_remote_ts_ptr and
spock_subxact_setts are non-NULL, and report the specific missing symbol during
startup failure. In src/spock_apply_heap.c lines 776-784, remove the silent
fallback and invoke the SubTransactionIdSetCommitTsData function under this
initialization invariant, optionally asserting its pointer is non-NULL.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro Plus
Run ID: 5ce6f731-c337-45d5-b84d-641437b4949c
📒 Files selected for processing (13)
include/spock.hpatches/15/pg15-000-spock-patchset.diffpatches/16/pg16-000-spock-patchset.diffpatches/17/pg17-000-spock-patchset.diffpatches/18/pg18-000-spock-patchset.diffpatches/19/pg19-000-spock-patchset.diffsql/spock--5.0.11--6.0.0.sqlsql/spock--6.0.0.sqlsrc/spock.csrc/spock_apply.csrc/spock_apply_heap.csrc/spock_functions.ctests/tap/t/106_core_patchset.pl
Spock hard-referenced remoteTransactionStopTimestamp (HLC) and SubTransactionIdSetCommitTsData (delta apply), so spock.so failed to dlopen on unpatched PostgreSQL with a cryptic linker error before any spock code ran. Resolve both with dlsym(RTLD_DEFAULT, ...) in _PG_init() and access them through NULL-guarded pointers. With no unresolved patch symbols, the same binary loads on patched and unpatched servers alike. On unpatched PostgreSQL the HLC symbol is absent, so _PG_init() now raises a clear FATAL instead of the linker error. Degraded operation on unpatched PostgreSQL is future work. No behavior change on patched PostgreSQL.
mason-sharp
force-pushed
the
feature/SPOC-636/patchset
branch
from
July 28, 2026 23:04
dca0a79 to
5c214aa
Compare
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@src/spock_apply.c`:
- Around line 985-994: Update the commit flow around CommitTransactionCommand
and append_feedback_position to make spock_remote_ts_ptr cleanup exception-safe:
establish a cleanup block that resets *spock_remote_ts_ptr before any error can
bypass the reset, while keeping feedback processing outside the guarded region
and avoiding normal returns from it. Preserve the existing conditional timestamp
assignment and feedback behavior.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro Plus
Run ID: 8e2ab7fc-db9e-4ec7-a3dc-dfb7b4d44c1f
📒 Files selected for processing (4)
include/spock.hsrc/spock.csrc/spock_apply.csrc/spock_apply_heap.c
🚧 Files skipped from review as they are similar to previous changes (3)
- src/spock_apply_heap.c
- include/spock.h
- src/spock.c
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
1 participant
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.
Introduce patchset version numbers. Over time we can start enforcing this in the extension as appropriate. For now avoid forcing users to update Postgres.
In addition, we lay the groundwork for spock lite mode.
By checking for the existence of symbols, Spock can determine what the situation with the Postgres it is using is:
It contains a patchset version number.
It does not contain a patchset version number, but from checking symbols we can see it is pgedge-patched Postgres
It is unpatched Postgres. Instead of failing at library load time, fail at startup with a more descriptive message. In the future, we can instead output a message and operate in a mode with degraded functionality.
Original idea came from this PR: #421