| c66b5ec6 | 05-Apr-2026 |
Andreas Gohr <andi@splitbrain.org> |
SearchIndex: rewrite Lock as static registry with reference counting
Replace the instance-based Lock class with a static registry that
tracks held locks per-process with reference counting. This sol
SearchIndex: rewrite Lock as static registry with reference counting
Replace the instance-based Lock class with a static registry that
tracks held locks per-process with reference counting. This solves
three problems:
- Split indexes (w3, w4, ...) share a single lock name and now
coordinate naturally via the registry
- Multiple callers can acquire the same lock without conflict
- Indexes enforce their own writability through lock()/unlock()
methods on AbstractIndex
The Lock registry manages both the filesystem lock (mkdir) and the
in-process tracking. The first acquire creates the directory, subsequent
acquires increment the refcount. Release decrements, and only removes
the directory when the count reaches zero.
Note: I am not sure if implementing this as a static object is a great
idea or if we should pass an instance through the collection to the
indexes...
show more ...
|
| d92c078c | 05-Apr-2026 |
Andreas Gohr <andi@splitbrain.org> |
SearchIndex: add DirectCollection for 1:1 entity-token mappings
Introduce DirectCollection as a third collection type alongside
FrequencyCollection and LookupCollection. Direct collections store
exa
SearchIndex: add DirectCollection for 1:1 entity-token mappings
Introduce DirectCollection as a third collection type alongside
FrequencyCollection and LookupCollection. Direct collections store
exactly one token per entity at the entity's position in the token
index (entity.RID === token.RID), with no frequency or reverse indexes.
AbstractCollection now accepts optional frequency/reverse index names
(default to '') and skips locking empty index names.
Adds PageTitleCollection as the first concrete direct collection
for the page -> title mapping.
show more ...
|
| f2bbffb5 | 05-Apr-2026 |
Andreas Gohr <andi@splitbrain.org> |
SearchIndex: extract Collection base class hierarchy
Introduce AbstractCollection as the shared base for all index
collections, with FrequencyCollection and LookupCollection as
the two abstract subc
SearchIndex: extract Collection base class hierarchy
Introduce AbstractCollection as the shared base for all index
collections, with FrequencyCollection and LookupCollection as
the two abstract subclasses differing only in how tokens are
counted (frequency vs dedup).
Key design decisions:
- splitByLength is a constructor parameter on AbstractCollection
controlling whether token/frequency indexes use length-based
file splitting. This is independent of the collection type.
- The reverse index format is self-describing: entries with *
have a group prefix (split), entries without don't (non-split).
No branching needed in parse/format methods.
- addEntity, resolveTokens, updateIndexes, and reverse index
handling all live in AbstractCollection. Subclasses only
implement countTokens().
Concrete collections: PageFulltextCollection (frequency, split),
MediaCollection and ReferencesCollection (lookup, non-split).
Renames FulltextCollection -> PageFulltextCollection and
FulltextCollectionSearch -> FrequencyCollectionSearch.
show more ...
|
| 2f70db90 | 04-Dec-2025 |
WillForan <willforan@gmail.com> |
fix: 32bit IP tests w/string of decimal representation, overflows
Math in PHP is hard!
sprintf("%.0f",0x7FFFFFFFFFFFFFFF) == sprintf("%.0f",0x7FFFFFFFFFFFFF00)
Changes
* 32bit gets own version o
fix: 32bit IP tests w/string of decimal representation, overflows
Math in PHP is hard!
sprintf("%.0f",0x7FFFFFFFFFFFFFFF) == sprintf("%.0f",0x7FFFFFFFFFFFFF00)
Changes
* 32bit gets own version of tests where expected values are strings
* decimalToBinary32 to replace `sprintf("%032b%032b"...)`, avoids overflow
* overflow check in ipv4 too
* refactor
* partsTo64 for 32bit parts into dec value as str (bcmath)
* Ip32::$b32 as class constant
* condition always PHP_INT_SIZE == 4 for 32bit (instead of == 8 for 64)
show more ...
|
| bc997a9d | 30-Oct-2025 |
Andreas Gohr <gohr@cosmocode.de> |
SearchIndex: TupleOps now work with frequencies of 1
We have indexes where we simply track that a relation between entity and
token exists, but there is no frequency. The frequency is always 1.
For
SearchIndex: TupleOps now work with frequencies of 1
We have indexes where we simply track that a relation between entity and
token exists, but there is no frequency. The frequency is always 1.
For those indexes we do not store *1 as frequency but omit it
completely.
The TupleOps class now can work with such indexes and will also store
frequencies of 1 like this.
show more ...
|