fix: clean up metatable on row filter change and add edge-case test coverage#38
Merged
fix: clean up metatable on row filter change and add edge-case test coverage#38
Conversation
test: add row filter edge-case coverage
test/unit.c — 6 new test functions:
1. do_test_row_filter_clear — Tests cloudsync_clear_filter (previously
zero coverage): set filter, insert data, clear filter, verify
unfiltered operations are tracked, sync to peer
2. do_test_row_filter_complex_expressions — Tests AND + comparison,
string literals with escaped quotes, IS NULL, IN (...), and sync
roundtrip with compound filters
3. do_test_row_filter_row_transition — Tests rows moving out of filter
scope (UPDATE makes row non-matching → no metadata generated) and
into filter scope (UPDATE makes row matching → metadata created),
plus bidirectional sync with transitions
4. do_test_row_filter_change — Tests changing filter expression after
data exists: old metadata persists, new filter applies to new
operations, updates under old filter conditions are not tracked
5. do_test_row_filter_composite_pk_multi_table — Tests composite
primary keys with filters, multiple tables with different filters
(including string literal in filter), sync roundtrip, and
update/delete on composite PK rows
6. do_test_row_filter_prefill — Tests pre-existing data inserted
before cloudsync_init and cloudsync_set_filter: verifies initial
metatable fill includes all rows (no filter during refill), new
operations after set_filter respect the filter, and sync roundtrip
transfers all pre-existing rows plus only matching new rows
test/postgresql/47_row_filter_advanced.sql — 10 assertions covering
clear_filter, complex expressions, IS NULL, row transitions, and filter
change
test/postgresql/48_row_filter_multi_table.sql — 8 assertions covering
composite PKs, multi-column filters with string literals, multi-table
sync roundtrip, and update/delete on composite PKs
test/postgresql/49_row_filter_prefill.sql — 8 assertions covering
pre-existing data before init+set_filter, matching/non-matching insert
behavior after filter, sync roundtrip with refill metadata, and
composite PK prefill with filter
cloudsync_set_filter and cloudsync_clear_filter now DELETE all rows from the metatable and call cloudsync_refill_metatable after updating the filter setting. This ensures the metatable only contains rows matching the active filter, preventing stale non-matching rows from syncing. Added SQL_DELETE_ALL_FROM_CLOUDSYNC_TABLE with platform-specific format specifiers (%w for SQLite, %s for PostgreSQL) to avoid undefined behavior from using SQLite's %w in PostgreSQL's vsnprintf.
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.
Previously, cloudsync_set_filter and cloudsync_clear_filter only updated
the filter setting and recreated triggers — they did not touch the
metatable. This left stale metadata for rows that no longer matched the
active filter, causing non-matching rows to sync to other nodes.
Now both functions DELETE all rows from the metatable and call
cloudsync_refill_metatable after changing the filter. This ensures the
metatable only contains entries for rows matching the current filter
(or all rows when the filter is cleared).
Core changes:
refills from scratch using the current filter setting
format specifiers (%w for SQLite, %s for PostgreSQL) to match how each
platform formats the meta_ref identifier
the SQLite and PostgreSQL implementations
Test coverage:
expressions (AND, IS NULL), row entering/exiting filter via UPDATE,
filter change after data, composite PK with multi-column filters,
multi-table roundtrip sync, and pre-existing data prefill
latest filter after set_filter and clear_filter calls