# Trouve-KA — mapping OpenSearch bilingue
# Author: Simon-Pierre Boucher
# Contact: contact@spboucher.ai

"""Mapping et settings de l'index documents.

- Analyzers français ET anglais sur les champs texte (bilingue dès le départ).
- Filtre de synonymes bilingues (thermopompe ↔ heat pump) appliqué à la recherche
  seulement : pas de traduction à l'ingestion.
- Refresh 1 s : une page indexée est cherchable en ~1 seconde (principe cardinal).
"""

# Synonymes bilingues de départ — extensibles sans réindexation (search-time)
BILINGUAL_SYNONYMS = [
    "thermopompe, heat pump",
    "subvention, grant, subsidy",
    "emploi, job, employment",
    "entreprise, company, business",
    "université, university",
    "cégep, cegep, college",
    "garderie, daycare, cpe",
    "impôt, tax, taxes",
    "santé, health",
    "plombier, plumber",
    "électricien, electrician",
    "déneigement, snow removal",
    "assurance, insurance",
    "logement, housing, apartment",
    "ville, city, municipalité, municipality",
    "gouvernement, government",
    "permis, permit, license, licence",
]

INDEX_SETTINGS = {
    "settings": {
        "number_of_shards": 1,
        "number_of_replicas": 0,
        "refresh_interval": "1s",
        "analysis": {
            "filter": {
                "french_elision": {
                    "type": "elision",
                    "articles_case": True,
                    "articles": ["l", "m", "t", "qu", "n", "s", "j", "d", "c", "jusqu", "quoiqu", "lorsqu", "puisqu"],
                },
                "french_stop": {"type": "stop", "stopwords": "_french_"},
                "french_stemmer": {"type": "stemmer", "language": "light_french"},
                "english_stop": {"type": "stop", "stopwords": "_english_"},
                "english_stemmer": {"type": "stemmer", "language": "light_english"},
                "bilingual_synonyms": {
                    "type": "synonym_graph",
                    "lenient": True,
                    "synonyms": BILINGUAL_SYNONYMS,
                },
            },
            "analyzer": {
                "fr_text": {
                    "tokenizer": "standard",
                    "filter": ["french_elision", "lowercase", "asciifolding", "french_stop", "french_stemmer"],
                },
                "fr_search": {
                    "tokenizer": "standard",
                    "filter": ["french_elision", "lowercase", "asciifolding", "bilingual_synonyms",
                               "french_stop", "french_stemmer"],
                },
                "en_text": {
                    "tokenizer": "standard",
                    "filter": ["lowercase", "asciifolding", "english_stop", "english_stemmer"],
                },
                "en_search": {
                    "tokenizer": "standard",
                    "filter": ["lowercase", "asciifolding", "bilingual_synonyms",
                               "english_stop", "english_stemmer"],
                },
            },
        },
    },
    "mappings": {
        "properties": {
            "url": {"type": "keyword"},
            "canonical_url": {"type": "keyword"},
            "domain": {"type": "keyword"},
            "title": {
                "type": "text",
                "analyzer": "fr_text",
                "search_analyzer": "fr_search",
                "fields": {"en": {"type": "text", "analyzer": "en_text", "search_analyzer": "en_search"}},
            },
            "description": {
                "type": "text",
                "analyzer": "fr_text",
                "search_analyzer": "fr_search",
                "fields": {"en": {"type": "text", "analyzer": "en_text", "search_analyzer": "en_search"}},
            },
            "body": {
                "type": "text",
                "analyzer": "fr_text",
                "search_analyzer": "fr_search",
                "fields": {"en": {"type": "text", "analyzer": "en_text", "search_analyzer": "en_search"}},
            },
            "headings": {
                "type": "text",
                "analyzer": "fr_text",
                "search_analyzer": "fr_search",
                "fields": {"en": {"type": "text", "analyzer": "en_text", "search_analyzer": "en_search"}},
            },
            "language": {"type": "keyword"},
            "page_quebec_score": {"type": "float"},
            "domain_quebec_score": {"type": "float"},
            "locations": {"type": "keyword"},
            "organizations": {"type": "keyword"},
            "people": {"type": "keyword"},
            "categories": {"type": "keyword"},
            "published_at": {"type": "date"},
            "crawled_at": {"type": "date"},
            "authority_score": {"type": "float"},
            "freshness_score": {"type": "float"},
            "quality_score": {"type": "float"},
            "spam_score": {"type": "float"},
            # embedding (dense_vector/knn) ajouté à l'étape sémantique (§17.7),
            # le mapping est extensible sans réindexation pour un nouveau champ.
        }
    },
}