From 371590b578d6748001948cda21bb735087cd307c Mon Sep 17 00:00:00 2001
From: Noah Levitt <nlevitt@archive.org>
Date: Thu, 30 Jun 2016 15:16:04 -0500
Subject: [PATCH] command line utility brozzler-ensure-tables, creates
 rethinkdb tables if they don't already exist... brozzler normally creates
 them on demand at startup, but if multiple instances are starting up at the
 same time, you can end up with duplicate broken tables, so it's a good idea
 to use this utility when spinning up a cluster

---
 brozzler/cli.py | 33 +++++++++++++++++++++++++++++++++
 setup.py        |  3 ++-
 2 files changed, 35 insertions(+), 1 deletion(-)

diff --git a/brozzler/cli.py b/brozzler/cli.py
index 9d4cca4..972d521 100644
--- a/brozzler/cli.py
+++ b/brozzler/cli.py
@@ -38,12 +38,18 @@ import warnings
 import yaml
 
 def _add_common_options(arg_parser):
+    arg_parser.add_argument(
+            '-q', '--quiet', dest='log_level',
+            action='store_const', default=logging.INFO, const=logging.WARN)
     arg_parser.add_argument(
             '-v', '--verbose', dest='log_level',
             action='store_const', default=logging.INFO, const=logging.DEBUG)
     arg_parser.add_argument(
             '--trace', dest='log_level',
             action='store_const', default=logging.INFO, const=brozzler.TRACE)
+    # arg_parser.add_argument(
+    #         '-s', '--silent', dest='log_level',
+    #         action='store_const', default=logging.INFO, const=logging.CRITICAL)
     arg_parser.add_argument(
             '--version', action='version',
             version='brozzler %s - %s' % (
@@ -266,3 +272,30 @@ def brozzler_worker():
                 th.join()
 
     logging.info("brozzler-worker is all done, exiting")
+
+def brozzler_ensure_tables():
+    '''
+    Creates rethinkdb tables if they don't already exist. Brozzler
+    (brozzler-worker, brozzler-new-job, etc) normally creates the tables it
+    needs on demand at startup, but if multiple instances are starting up at
+    the same time, you can end up with duplicate broken tables. So it's a good
+    idea to use this utility at an early step when spinning up a cluster.
+    '''
+    arg_parser = argparse.ArgumentParser(
+            prog=os.path.basename(sys.argv[0]),
+            formatter_class=argparse.ArgumentDefaultsHelpFormatter)
+    _add_rethinkdb_options(arg_parser)
+    _add_common_options(arg_parser)
+
+    args = arg_parser.parse_args(args=sys.argv[1:])
+    _configure_logging(args)
+
+    r = rethinkstuff.Rethinker(
+            args.rethinkdb_servers.split(','), args.rethinkdb_db)
+
+    # services table
+    rethinkstuff.ServiceRegistry(r)
+
+    # sites, pages, jobs tables
+    brozzler.frontier.RethinkDbFrontier(r)
+
diff --git a/setup.py b/setup.py
index 93d3cfb..0607dcb 100644
--- a/setup.py
+++ b/setup.py
@@ -21,7 +21,7 @@ import setuptools
 
 setuptools.setup(
         name='brozzler',
-        version='1.1.dev38',
+        version='1.1.dev39',
         description='Distributed web crawling with browsers',
         url='https://github.com/internetarchive/brozzler',
         author='Noah Levitt',
@@ -36,6 +36,7 @@ setuptools.setup(
                 'brozzler-new-job=brozzler.cli:brozzler_new_job',
                 'brozzler-new-site=brozzler.cli:brozzler_new_site',
                 'brozzler-worker=brozzler.cli:brozzler_worker',
+                'brozzler-ensure-tables=brozzler.cli:brozzler_ensure_tables',
                 'brozzler-webconsole=brozzler.webconsole:run',
             ],
         },