create-mathoid-ng.js
ActivePublic
Actions

Authored by Eevans on Nov 6 2017, 4:35 PM.

Tags

None

Referenced Files

	F10644544: Masterwork From Distant Lands
	Nov 6 2017, 4:35 PM

Subscribers

None

	"use strict";


	const crypto = require('crypto');
	const colors = require('colors/safe');
	const P = require('bluebird');
	const yargs = require('yargs');

	const green = colors.green;
	const red = colors.red;
	const yellow = colors.yellow;


	function hashKey(key) {
	return new crypto.Hash('sha1')
	.update(key)
	.digest()
	.toString('base64')
	// Replace [+/] from base64 with _ (illegal in Cassandra)
	.replace(/[+/]/g, '_')
	// Remove base64 padding, has no entropy
	.replace(/=+$/, '');
	}


	function getValidPrefix(key) {
	const prefixMatch = /^[a-zA-Z0-9_]+/.exec(key);
	if (prefixMatch) {
	return prefixMatch[0];
	} else {
	return '';
	}
	}


	function makeValidKey(key, length) {
	const origKey = key;
	key = key.replace(/_/g, '__')
	.replace(/\./g, '_');
	if (!/^[a-zA-Z0-9_]+$/.test(key)) {
	// Create a new 28 char prefix
	const validPrefix = getValidPrefix(key).substr(0, length * 2 / 3);
	return validPrefix + hashKey(origKey).substr(0, length - validPrefix.length);
	} else if (key.length > length) {
	return key.substr(0, length * 2 / 3) + hashKey(origKey).substr(0, length / 3);
	} else {
	return key;
	}
	}


	function keyspaceName(name, table) {
	const reversedName = name.toLowerCase().split('.').reverse().join('.');
	const prefix = makeValidKey(reversedName, Math.max(26, 48 - table.length - 3));
	// 6 chars _hash_ to prevent conflicts between domains & table names
	const res = `${prefix}_T_${makeValidKey(table, 48 - prefix.length - 3)}`;
	return res;
	}


	function cassID(name) {
	if (/^[a-zA-Z0-9_]+$/.test(name)) {
	return `"${name}"`;
	} else {
	return `"${name.replace(/"/g, '""')}"`;
	}
	}

	const kvTables = {
	'mathoid_ng.input': 'blob'
	};

	const krvTables = {
	'mathoid_ng.hash_table': 'text',
	'mathoid_ng.check': 'blob',
	'mathoid_ng.mml': 'blob',
	'mathoid_ng.svg': 'blob',
	'mathoid_ng.png': 'blob'
	};

	const storages = [
	'enwiki',
	'commons',
	'wikipedia',
	'others'
	];

	const qKs = `CREATE KEYSPACE IF NOT EXISTS <keyspace> WITH replication = {'class': 'NetworkTopologyStrategy', 'codfw': '3', 'eqiad': '3'} AND durable_writes = true;`;

	const qMeta = `CREATE TABLE IF NOT EXISTS <keyspace>.meta (
	key text PRIMARY KEY,
	value text
	) WITH bloom_filter_fp_chance = 0.1
	AND caching = {'keys': 'ALL', 'rows_per_partition': 'NONE'}
	AND comment = ''
	AND compaction = {'class': 'org.apache.cassandra.db.compaction.LeveledCompactionStrategy'}
	AND compression = {'chunk_length_in_kb': '64', 'class': 'org.apache.cassandra.io.compress.LZ4Compressor'}
	AND crc_check_chance = 1.0
	AND dclocal_read_repair_chance = 0.1
	AND default_time_to_live = 0
	AND gc_grace_seconds = 864000
	AND max_index_interval = 2048
	AND memtable_flush_period_in_ms = 0
	AND min_index_interval = 128
	AND read_repair_chance = 0.0
	AND speculative_retry = '99PERCENTILE';`;

	const qData = `CREATE TABLE IF NOT EXISTS <keyspace>.data (
	"_domain" text,
	key text,
	rev int,
	tid timeuuid,
	"content-location" text,
	"content-type" text,
	tags set<text>,
	value <type>,
	PRIMARY KEY (("_domain", key), rev, tid)
	) WITH CLUSTERING ORDER BY (rev DESC, tid DESC)
	AND bloom_filter_fp_chance = 0.01
	AND caching = {'keys': 'ALL', 'rows_per_partition': 'NONE'}
	AND comment = ''
	AND compaction = {'class': 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy', 'max_threshold': '32', 'min_threshold': '4'}
	AND compression = {'chunk_length_in_kb': '256', 'class': 'org.apache.cassandra.io.compress.DeflateCompressor'}
	AND crc_check_chance = 1.0
	AND dclocal_read_repair_chance = 0.1
	AND default_time_to_live = 0
	AND gc_grace_seconds = 86400
	AND max_index_interval = 2048
	AND memtable_flush_period_in_ms = 0
	AND min_index_interval = 128
	AND read_repair_chance = 0.0
	AND speculative_retry = '99PERCENTILE';`;

	const qDataRevision = `CREATE TABLE IF NOT EXISTS <keyspace>.data (
	"_domain" text,
	key text,
	ts timestamp,
	rev int,
	PRIMARY KEY (("_domain", key), ts)
	) WITH CLUSTERING ORDER BY (ts DESC)
	AND bloom_filter_fp_chance = 0.1
	AND caching = {'keys': 'ALL', 'rows_per_partition': 'NONE'}
	AND comment = ''
	AND compaction = {'class': 'org.apache.cassandra.db.compaction.LeveledCompactionStrategy'}
	AND compression = {'chunk_length_in_kb': '64', 'class': 'org.apache.cassandra.io.compress.LZ4Compressor'}
	AND crc_check_chance = 1.0
	AND dclocal_read_repair_chance = 0.1
	AND default_time_to_live = 864000
	AND gc_grace_seconds = 864000
	AND max_index_interval = 2048
	AND memtable_flush_period_in_ms = 0
	AND min_index_interval = 128
	AND read_repair_chance = 0.0
	AND speculative_retry = '99PERCENTILE';`;

	const qDataRender = `CREATE TABLE IF NOT EXISTS <keyspace>.data (
	"_domain" text,
	key text,
	rev int,
	ts timestamp,
	tid timeuuid,
	PRIMARY KEY (("_domain", key), rev, ts)
	) WITH CLUSTERING ORDER BY (rev DESC, ts DESC)
	AND bloom_filter_fp_chance = 0.1
	AND caching = {'keys': 'ALL', 'rows_per_partition': 'NONE'}
	AND comment = ''
	AND compaction = {'class': 'org.apache.cassandra.db.compaction.LeveledCompactionStrategy'}
	AND compression = {'chunk_length_in_kb': '64', 'class': 'org.apache.cassandra.io.compress.LZ4Compressor'}
	AND crc_check_chance = 1.0
	AND dclocal_read_repair_chance = 0.1
	AND default_time_to_live = 864000
	AND gc_grace_seconds = 864000
	AND max_index_interval = 2048
	AND memtable_flush_period_in_ms = 0
	AND min_index_interval = 128
	AND read_repair_chance = 0.0
	AND speculative_retry = '99PERCENTILE';`;

	const qKvData = `CREATE TABLE IF NOT EXISTS <keyspace>.data (
	"_domain" text,
	key text,
	tid timeuuid,
	headers blob,
	value <type>,
	PRIMARY KEY (("_domain", key))
	) WITH CLUSTERING ORDER BY (rev DESC, tid DESC)
	AND bloom_filter_fp_chance = 0.01
	AND caching = {'keys': 'ALL', 'rows_per_partition': 'NONE'}
	AND comment = ''
	AND compaction = {'class': 'org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy', 'max_threshold': '32', 'min_threshold': '4'}
	AND compression = {'chunk_length_in_kb': '64', 'class': 'org.apache.cassandra.io.compress.LZ4Compressor'}
	AND crc_check_chance = 1.0
	AND dclocal_read_repair_chance = 0.1
	AND default_time_to_live = 0
	AND gc_grace_seconds = 86400
	AND max_index_interval = 2048
	AND memtable_flush_period_in_ms = 0
	AND min_index_interval = 128
	AND read_repair_chance = 0.0
	AND speculative_retry = '99PERCENTILE';`;


	const argv = yargs.usage('Usage: $0 [options]')
	.options('h', {alias: 'help'})
	.argv;

	if (argv.help) {
	yargs.showHelp();
	process.exit(0);
	}

	return P.each(storages, (storage) => {
	return P.each(Object.keys(krvTables), (table) => {
	const keyspace = cassID(keyspaceName(storage, table));
	console.log(`\n-- Keyspace: ${keyspace}`);
	console.log(qKs.replace('<keyspace>', keyspace));
	console.log(`\n-- Meta: ${keyspace}.meta`);
	console.log(qMeta.replace('<keyspace>', keyspace));
	console.log(`\n-- Data: ${keyspace}.data`);
	let dataCQL = qData;
	if(/render/.test(table)) {
	dataCQL = qDataRender;
	} else if(/revision/.test(table)) {
	dataCQL = qDataRevision;
	}
	console.log(dataCQL.replace('<keyspace>', keyspace).replace('<type>', krvTables[table]));
	console.log('\n-- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~');
	});
	});

Event Timeline

Eevans edited the content of this paste. (Show Details)Nov 6 2017, 4:35 PM

Eevans changed the title of this paste from untitled to Masterwork From Distant Lands.

Eevans updated the paste's language from autodetect to autodetect.

Eevans changed the title of this paste from Masterwork From Distant Lands to create-mathoid-ng.js.Nov 6 2017, 4:35 PM

Eevans updated the paste's language from autodetect to js.

create-mathoid-ng.jsActivePublicActions

Event Timeline

create-mathoid-ng.js
ActivePublic
Actions