Skip to content

Spelling {#architecture_spelling}

Spelling is the name-authority layer above Database. It loads Scid spelling files, stores canonical names and aliases for each database name category, and answers correction and metadata queries without mutating the database itself.

@ref architecture_spelling_name_normalization "Name Normalization" handles the general prefix, infix and suffix rewrite rules. @ref architecture_spelling_spell_checker "Spell Checker" is the loaded index over canonical names and aliases. @ref architecture_spelling_player_metadata "Player Metadata" is the player-only side channel for title, country, dates, peak rating, biography and rating history.

@startuml spelling-overview skinparam backgroundColor #FFFFFF skinparam shadowing false skinparam linetype ortho skinparam defaultFontName Cantarel skinparam defaultFontSize 10 skinparam roundcorner 8 skinparam Nodesep 68 skinparam Ranksep 54 skinparam ArrowColor #4B5563 skinparam RectangleBorderColor #6B7280 skinparam RectangleBackgroundColor #EEF2FF

left to right direction

rectangle "Spelling file" as File #EEF2FF

rectangle "Spell checker" as Checker #FFF7ED

rectangle "Name normalization" as Normalization #EEF2FF

rectangle "Canonical name lookup" as Lookup #EEF2FF

rectangle "Player metadata" as Metadata #EEF2FF

rectangle "Database names" as DatabaseNames #EEF2FF

File -[hidden]right- Checker Checker -[hidden]right- Lookup Normalization -[hidden]right- Metadata Metadata -[hidden]right- DatabaseNames Checker -[hidden]down- Metadata

File --> Checker : loads Checker --> Normalization : rules Checker --> Lookup : aliases Checker --> Metadata : player records DatabaseNames --> Checker : queries Checker --> DatabaseNames : canonical names @enduml

\htmlonly

\endhtmlonly - @subpage architecture_spelling_name_normalization "Name Normalization" - @subpage architecture_spelling_spell_checker "Spell Checker" - @subpage architecture_spelling_player_metadata "Player Metadata" \htmlonly

\endhtmlonly

@ref architecture_spelling_name_normalization "Name normalization" is a broad cleanup pass. A caller can apply spelling-file Prefix, Infix and Suffix rules for a database name category without performing canonical alias lookup.

@ref architecture_spelling_spell_checker "Spell checker" owns the loaded strings. It builds sorted alias indexes per database name category, removes section-specific comparison characters from lookup keys, and returns borrowed canonical-name pointers.

@ref architecture_spelling_player_metadata "Player metadata" is available only for player names. It interprets the comment plus Bio and Elo lines attached to canonical player entries, again returning borrowed views owned by the loaded spell checker.