aboutsummaryrefslogtreecommitdiffstats
path: root/sync.py
blob: 15d19837ff5fd0e9faf78f728c123c009c0dfcf9 (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
#!/usr/bin/env python3

import psycopg2
from psycopg2 import sql
psycopg2.extensions.register_type(psycopg2.extensions.UNICODE)

import requests
from requests.adapters import HTTPAdapter
from requests.packages.urllib3.util.retry import Retry

import json

import time

import logging
from config import get_config
schema_name = get_config()['db_schema']

#logging.basicConfig(level=logging.DEBUG)

def open_db():
    global conn
    db_conn_str = get_config()['db_conn_str']
    conn = psycopg2.connect(db_conn_str)

column_cache = {}
def get_columns(table_name):
    global column_cache
    if table_name in column_cache:
        return column_cache[table_name]
    sql_str = """
        SELECT column_name FROM information_schema.columns
        WHERE table_schema = %s AND table_name = %s
        """
    cur = conn.cursor()
    cur.execute(sql_str, (schema_name, table_name))
    rows = cur.fetchall()
    ret = [row[0] for row in rows]
    column_cache[table_name] = ret
    return ret

def test_table(table_name):
    cols = get_columns(table_name)
    try:
        assert('id' in cols)
        assert(cols[0] == 'id')
        assert('updated' in cols)
        assert('data' in cols)
    except AssertionError:
        print(f'Invalid or missing table schema for {table_name:s}')
        raise

def gen_sql(table_name, cols):
    updates = []
    for col in cols[1:]:
        if col == 'updated':
            stmt = sql.SQL('updated = CASE WHEN EXCLUDED.deleted IS NOT NULL THEN t.updated ELSE EXCLUDED.updated END')
        else:
            stmt = sql.SQL('{} = EXCLUDED.{}').format(sql.Identifier(col), sql.Identifier(col))
        updates.append(stmt)

    comp = sql.SQL("""
            INSERT INTO {}.{} AS t
                ({})
                VALUES ({})
                ON CONFLICT ({})
                DO UPDATE SET {};
            """).format(sql.Identifier(schema_name),
                    sql.Identifier(table_name),
                    sql.SQL(', ').join(map(sql.Identifier, cols)),
                    sql.SQL(', ').join(map(sql.Placeholder, cols)),
                    sql.Identifier(cols[0]),
                    sql.SQL(', ').join(updates))
    return comp

def update_object(kind, obj):
    cols = get_columns(kind)
    comp = gen_sql(kind, cols)
    t = {}
    for key in cols:
        if key in obj:
            t[key] = obj[key]
        else:
            t[key] = None
    t['data'] = json.dumps(obj)
    if t['status'] == 'deleted':
        t['deleted'] = t['updated']
    cur = conn.cursor()
    try:
        cur.execute(comp, t)
    except:
        print(kind, t['id'], t['data'])
        raise

def last_updated(kind):
    comp = sql.SQL('SELECT EXTRACT(EPOCH FROM MAX(updated)) - 1 FROM {}.{};'
                ).format(sql.Identifier(schema_name), sql.Identifier(kind))
    cur = conn.cursor()
    cur.execute(comp)
    last = cur.fetchone()[0]
    if last is None: return None
    return int(last)

def fetch_objects(s, kind, extra_params):
    endpoint = f'https://peeringdb.com/api/{kind:s}'
    params = {
            'depth': 0,
            'status__in': 'ok,pending,deleted',
            'since': 1,
            **extra_params }

    r = s.get(endpoint, params=params )
    objs = json.loads(r.text)['data']
    for obj in objs:
        update_object(kind, obj)
    return len(objs)

def initial_sync(s, kind):
    initial_step = 100
    step = initial_step
    lows = 0
    low = 0
    while lows < 5:
        high = low + step + 1
        n = fetch_objects(s, kind, { 'id__gt': low, 'id__lt': high } )
        low += step
        if n > 0:
            step = initial_step
            lows = 0
        else:
            step *= 2
            lows += 1
    fetch_objects(s, kind, { 'id__gt': low } )

def sync_table(s, kind):
    test_table(kind)
    endpoint = f'https://peeringdb.com/api/{kind:s}'
    last = last_updated(kind)
    if last is None:
        last = int(time.time()) - 3600
        initial_sync(s, kind)
    fetch_objects(s, kind, { 'since': last } )

def main():
    open_db()

    s = requests.Session()
    retries = Retry(total=5, backoff_factor=1, status_forcelist=[ 502, 503, 504 ])
    s.mount('https://', HTTPAdapter(max_retries=retries))
    s.auth = get_config()['auth']

    req_agent = s.headers.get('User-Agent')
    s.headers.update({'User-Agent': f'peeringdb-simplesync/0.1 {req_agent:s}'})
    r = s.get('https://peeringdb.com/apidocs/')
    s.headers.update({'Accept': 'application/json'})

    ignored = [ 'as_set' ]

    apidoc = json.loads(r.text)
    for key in apidoc['api']:
        if key[0] == '_' or key in ignored: continue
        try:
            sync_table(s, key)
        except AssertionError:
            print(f'skipping {key:s}...')

    conn.commit()

main()