暂存

2026-05-01 23:40:14 +08:00
commit b8f599a617
3867 changed files with 478663 additions and 0 deletions
@@ -0,0 +1,89 @@
+#!/usr/bin/env php
+<?php
+
+use support\Db;
+
+require __DIR__ . '/../vendor/autoload.php';
+require __DIR__ . '/../support/bootstrap.php';
+require __DIR__ . '/../vendor/webman/database/src/support/Db.php';
+
+$statements = [
+    <<<SQL
+CREATE TABLE IF NOT EXISTS archives (
+    id BIGSERIAL PRIMARY KEY,
+    archive_uid VARCHAR(26) NOT NULL UNIQUE,
+    title TEXT,
+    summary TEXT,
+    year INTEGER,
+    author TEXT,
+    source TEXT,
+    series TEXT,
+    tags JSONB NOT NULL DEFAULT '[]'::jsonb,
+    metadata JSONB NOT NULL DEFAULT '{}'::jsonb,
+    content TEXT,
+    raw TEXT,
+    chunks JSONB NOT NULL DEFAULT '[]'::jsonb,
+    created_time TIMESTAMPTZ NOT NULL DEFAULT CURRENT_TIMESTAMP,
+    updated_time TIMESTAMPTZ NOT NULL DEFAULT CURRENT_TIMESTAMP
+)
+SQL,
+    'ALTER TABLE archives ADD COLUMN IF NOT EXISTS summary TEXT',
+    <<<SQL
+CREATE TABLE IF NOT EXISTS chunks (
+    id BIGSERIAL PRIMARY KEY,
+    chunk_uid VARCHAR(64) NOT NULL UNIQUE,
+    archive_uid VARCHAR(26) NOT NULL,
+    chunk_index INTEGER NOT NULL,
+    page_start INTEGER,
+    page_end INTEGER,
+    text TEXT,
+    length INTEGER,
+    embedding_status INTEGER NOT NULL DEFAULT 0,
+    embedding_ref JSONB,
+    embedding_model TEXT,
+    CONSTRAINT chunks_archive_uid_foreign
+        FOREIGN KEY (archive_uid)
+        REFERENCES archives (archive_uid)
+        ON DELETE CASCADE,
+    CONSTRAINT chunks_archive_index_unique
+        UNIQUE (archive_uid, chunk_index)
+)
+SQL,
+    'CREATE INDEX IF NOT EXISTS archives_year_index ON archives (year)',
+    'CREATE INDEX IF NOT EXISTS archives_series_index ON archives (series)',
+    'CREATE INDEX IF NOT EXISTS archives_tags_gin_index ON archives USING GIN (tags)',
+    'CREATE INDEX IF NOT EXISTS archives_metadata_gin_index ON archives USING GIN (metadata)',
+    'CREATE INDEX IF NOT EXISTS chunks_archive_uid_index ON chunks (archive_uid)',
+    'CREATE INDEX IF NOT EXISTS chunks_page_range_index ON chunks (archive_uid, page_start, page_end)',
+    'CREATE INDEX IF NOT EXISTS chunks_embedding_status_index ON chunks (embedding_status)',
+    <<<SQL
+CREATE OR REPLACE FUNCTION set_updated_time()
+RETURNS TRIGGER AS $$
+BEGIN
+    NEW.updated_time = CURRENT_TIMESTAMP;
+    RETURN NEW;
+END;
+$$ LANGUAGE plpgsql
+SQL,
+    'DROP TRIGGER IF EXISTS archives_set_updated_time ON archives',
+    <<<SQL
+CREATE TRIGGER archives_set_updated_time
+BEFORE UPDATE ON archives
+FOR EACH ROW
+EXECUTE FUNCTION set_updated_time()
+SQL,
+];
+
+try {
+    Db::connection()->getPdo();
+    echo 'Database connection ok: ' . config('database.default') . PHP_EOL;
+
+    foreach ($statements as $statement) {
+        Db::statement($statement);
+    }
+
+    echo 'Tables initialized: archives, chunks' . PHP_EOL;
+} catch (Throwable $exception) {
+    fwrite(STDERR, $exception::class . ': ' . $exception->getMessage() . PHP_EOL);
+    exit(1);
+}