4d0cdc8f2f
An article embedded as a single vector, which finds the article but not the paragraph — so a citation could only ever point at the top of a page. Sections live in a JSON column and cannot carry a vector or a full-text index, so they are now projected into `article_section_index`: one row per section with its own embedding and weighted tsvector (migration x6d7e8f9a0b1). - Rows are keyed by section id, so editing a section updates it, removing one deletes it, and an unchanged section is not re-embedded on every save. - `article_section` joins the embeddable kinds, so the retry task, the full regeneration and the health report cover it without further changes. - `hybrid_ids(db, query, "article_section")` searches it like any other corpus. This is the groundwork for grouped search results (article, then the sections that matched) and for AI citations that deep-link to the right section. Tests: 113 backend green. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01PpfzbZ1QTLMeVYxM2kyq8m
42 lines
1.5 KiB
Python
42 lines
1.5 KiB
Python
"""Section-level index for articles, so a citation can point at the right part.
|
|
|
|
An article embeds as one vector, which is enough to find the article but not the
|
|
paragraph. Sections live in a JSON column, so they are projected into a derived
|
|
table that can carry its own vector and full-text index.
|
|
|
|
Revision ID: x6d7e8f9a0b1
|
|
Revises: w5c6d7e8f9a0
|
|
"""
|
|
from alembic import op
|
|
|
|
revision = "x6d7e8f9a0b1"
|
|
down_revision = "w5c6d7e8f9a0"
|
|
branch_labels = None
|
|
depends_on = None
|
|
|
|
|
|
def upgrade():
|
|
op.execute("""
|
|
CREATE TABLE IF NOT EXISTS article_section_index (
|
|
id SERIAL PRIMARY KEY,
|
|
article_id INTEGER NOT NULL REFERENCES articles(id) ON DELETE CASCADE,
|
|
section_id VARCHAR(64) NOT NULL,
|
|
title TEXT,
|
|
content TEXT,
|
|
embedding vector(1024),
|
|
embedding_model VARCHAR(120),
|
|
embedded_at TIMESTAMP,
|
|
search_vector tsvector GENERATED ALWAYS AS (
|
|
setweight(to_tsvector('english', coalesce(title, '')), 'A') ||
|
|
setweight(to_tsvector('english', coalesce(content, '')), 'B')
|
|
) STORED,
|
|
CONSTRAINT uq_article_section_index UNIQUE (article_id, section_id)
|
|
)
|
|
""")
|
|
op.execute("CREATE INDEX IF NOT EXISTS ix_asi_search ON article_section_index USING GIN (search_vector)")
|
|
op.execute("CREATE INDEX IF NOT EXISTS ix_asi_article ON article_section_index(article_id)")
|
|
op.execute("CREATE INDEX IF NOT EXISTS ix_asi_model ON article_section_index(embedding_model)")
|
|
|
|
|
|
def downgrade():
|
|
op.execute("DROP TABLE IF EXISTS article_section_index")
|