freifunk/ffmap-backend
8
0
Fork 0
Code Issues Pull requests Packages Projects Releases Wiki Activity

Compare commits

base: freifunk:dev
Branches Tags
freifunk:dev
freifunk:firstseen
freifunk:detailed-rrds
freifunk:master
freifunk:feature-ffmap-d3-jq
freifunk:influxdb
freifunk:legacy
freifunk:rewrite
freifunk:alfred-data
freifunk:v2
freifunk:v1
..
compare: freifunk:rewrite
Branches Tags
freifunk:dev
freifunk:firstseen
freifunk:detailed-rrds
freifunk:master
freifunk:feature-ffmap-d3-jq
freifunk:influxdb
freifunk:legacy
freifunk:rewrite
freifunk:alfred-data
freifunk:v2
freifunk:v1

15 commits
dev ... rewrite

Author SHA1 Message Date
Jan-Philipp Litza cd1329963a Alfred input: Pass -z switch to alfred-json 2014-09-23 22:31:51 +02:00
Jan-Philipp Litza 322860be7e Add MACs from mesh_interfaces as alises 2014-09-20 12:42:53 +02:00
Jan-Philipp Litza 66112061d6 Fix adding of nodes with multiple matching alises 2014-09-20 12:42:40 +02:00
Jan-Philipp Litza f08aaaff4e Fix fuzzy MAC matching 2014-09-07 12:17:36 +02:00
Jan-Philipp Litza 6d452fc149 d3json: obscure client MACs 2014-09-07 12:17:22 +02:00
Jan-Philipp Litza 5fba69de7a d3json: make output more similar to pre-rewrite version 2014-07-31 16:31:14 +02:00
Jan-Philipp Litza 446bc98403 input alfred: unify nodeinfo and stats datatypes 2014-07-31 11:41:26 +02:00
Jan-Philipp Litza e54e7467fc Create package ffmap, add wiki input, remove old code 2014-07-08 14:20:47 +02:00
Jan-Philipp Litza f5e3705eec Began rewrite with more modular design 2014-07-07 23:27:21 +02:00
Jan-Philipp Litza 54402ce089 mkmap.sh: Add locking around script call 2014-07-06 20:09:42 +02:00
Jan-Philipp Litza ee51547664 mkmap.sh: Remove Lübeck-specific stuff 2014-07-06 20:04:24 +02:00
Jan-Philipp Litza 89e4c63700 alfred.py: Make geo attribute setting more robust 2014-07-06 20:03:10 +02:00
Jan-Philipp Litza 7075d8481c NodeRRD: add many more DS, rrd.py: generate neighbor counts 2014-07-06 20:03:10 +02:00
Jan-Philipp Litza 43e70191f1 RRD: Fix updating of DS 2014-07-06 20:03:10 +02:00
Jan-Philipp Litza 6fc1423124 Make handling of node attributes more flexible. 
This commit makes Nodes special dicts that return None-like objects for
inexistent keys, making it a dynamic attribute store.

Also, it removes the D3MapBuilder and moves its logic to the Node and
Link classes' newly introduced export() method. Only they need to be
changed to populate the final nodes.json with more attributes.
 2014-07-06 20:03:10 +02:00
36 changed files with 789 additions and 1400 deletions
Show stats Expand all files Collapse all files

7
.gitignore vendored
View file

@ -1,8 +1 @@
# script-generated
aliases*.json
nodedb/
# python bytecode / cache
*.pyc
pycache/
__pycache__/

6
.travis.yml
View file

@ -1,6 +0,0 @@
sudo: false
language: python
python:
- "3.4"
install: "pip install pep8"
script: "pep8 --ignore=E501 *.py lib/*.py"

148
README.md
View file

@ -1,117 +1,53 @@
# Data for Freifunk Map, Graph and Node List
[![Build Status](https://travis-ci.org/ffnord/ffmap-backend.svg?branch=master)](https://travis-ci.org/ffnord/ffmap-backend)
ffmap-backend gathers information on the batman network by invoking
batctl
and
batadv-vis
as root (via sudo) and has this information placed into a target directory
as the file "nodes.json" and also updates the directory "nodes" with graphical
representations of uptimes and the number of clients connecting.
ffmap-backend gathers information on the batman network by invoking :
* batctl (might require root),
* alfred-json and
* batadv-vis
The output will be written to a directory (`-d output`).
Run `backend.py --help` for a quick overview of all available options.
For the script's regular execution add the following to the crontab:
The target directory is suggested to host all information for interpreting those
node descriptions, e.g. as provided by https://github.com/ffnord/ffmap-d3.git .
When executed without root privileges, we suggest to grant sudo permissions
within wrappers of those binaries, so no further changes are required in other
scripts:
<pre>
* * * * * backend.py -d /path/to/output -a /path/to/aliases.json --vpn ae:7f:58:7d:6c:2a d2:d0:93:63:f7:da
$ cat <<EOCAT > $HOME/batctl
#!/bin/sh
exec sudo /usr/sbin/batctl $*
EOCAT
</pre>
# Dependencies
and analogously for batadv-vis. The entry for /etc/sudoers could be
whateveruser ALL=(ALL:ALL) NOPASSWD: /usr/sbin/batctl,/usr/sbin/batadv-vis,/usr/sbin/alfred-json
- Python 3
- Python 3 Package [Networkx](https://networkx.github.io/)
- [alfred-json](https://github.com/tcatm/alfred-json)
- rrdtool (if run with `--with-rrd`)
The destination directory can be made directly available through apache:
<pre>
$ cat /etc/apache2/site-enabled/000-default
...
<Directory /home/whateverusername/www/>
Options Indexes FollowSymLinks MultiViews
AllowOverride None
Order allow,deny
allow from all
</Directory>
...
# Running as unprivileged user
$ cat /etc/apache2/conf.d/freifunk
Alias /map /home/ffmap/www/
Alias /firmware /home/freifunk/autoupdates/
</pre>
Some information collected by ffmap-backend requires access to specific system resources.
To execute, run
python3 -mffmap.run --input-alfred --input-badadv --output-d3json ../www/nodes.json
The script expects above described sudo-wrappers in the $HOME directory of the user executing
the script. If those are not available, an error will occurr if not executed as root. Also,
the tool realpath optionally allows to execute the script from anywhere in the directory tree.
Make sure the user you are running this under is part of the group that owns the alfred socket, so
alfred-json can access the alfred daemon.
# ls -al /var/run/alfred.sock
srw-rw---- 1 root alfred 0 Mar 19 22:00 /var/run/alfred.sock=
# adduser map alfred
Adding user `map' to group `alfred' ...
Adding user map to group alfred
Done.
$ groups
map alfred
Running batctl requires passwordless sudo access, because it needs to access the debugfs to retrive
the gateway list.
# echo 'map ALL = NOPASSWD: /usr/sbin/batctl' | tee /etc/sudoers.d/map
map ALL = NOPASSWD: /usr/sbin/batctl
# chmod 0440 /etc/sudoers.d/map
That should be everything. The script automatically detects if it is run in unprivileged mode and
will prefix `sudo` where necessary.
# Data format
## nodes.json
{ 'nodes': {
node_id: { 'flags': { flags },
'firstseen': isoformat,
'lastseen': isoformat,
'nodeinfo': {...}, # copied from alfred type 158
'statistics': {
'uptime': double, # seconds
'memory_usage': double, # 0..1
'clients': double,
'rootfs_usage': double, # 0..1
'loadavg': double,
'gateway': mac
}
},
...
}
'timestamp': isoformat
}
### flags (bool)
- online
- gateway
## Old data format
If you want to still use the old [ffmap-d3](https://github.com/ffnord/ffmap-d3)
front end, you can use the file `ffmap-d3.jq` to convert the new output to the
old one:
```
jq -n -f ffmap-d3.jq \
--argfile nodes nodedb/nodes.json \
--argfile graph nodedb/graph.json \
> nodedb/ffmap-d3.json
```
Then point your ffmap-d3 instance to the `ffmap-d3.json` file.
# Removing owner information
If you'd like to redact information about the node owner from `nodes.json`,
you may use a filter like [jq]. In this case, specify an output directory
different from your webserver directory, e.g.:
./backend.py -d /ffmap-data
Don't write to files generated in there. ffmap-backend uses them as its
database.
After running ffmap-backend, copy `graph.json` to your webserver. Then,
filter `nodes.json` using `jq` like this:
jq '.nodes = (.nodes | with_entries(del(.value.nodeinfo.owner)))' \
< /ffmap-data/nodes.json > /var/www/data/nodes.json
This will remove owner information from nodes.json before copying the data
to your webserver.
[jq]: https://stedolan.github.io/jq/
For the script's regular execution add the following to the crontab:
<pre>
*/5 * * * * python3 -mffmap.run --input-alfred --input-badadv --output-d3json /home/ffmap/www/nodes.json
</pre>

42
alfred_merge.py
View file

@ -1,42 +0,0 @@
#!/usr/bin/env python3
import subprocess
import json
from collections import MutableMapping
def rec_merge(d1, d2):
'''
Update two dicts of dicts recursively,
if either mapping has leaves that are non-dicts,
the second's leaf overwrites the first's.
'''
for k, v in d1.items(): # in Python 2, use .iteritems()!
if k in d2:
# this next check is the only difference!
if all(isinstance(e, MutableMapping) for e in (v, d2[k])):
d2[k] = rec_merge(v, d2[k])
# we could further check types and merge as appropriate here.
d3 = d1.copy()
d3.update(d2)
return d3
class alfred_merge:
def __init__(self,request_data_type_1 = 158, request_data_type_2 = 159):
self.request_data_type_1 = request_data_type_1
self.request_data_type_2 = request_data_type_2
def aliases(self):
output = subprocess.check_output(["/usr/local/bin/alfred-json","-z", "-r",str(self.request_data_type_1),"-f","json"])
alfred_data_1 = json.loads(output.decode("utf-8"))
output = subprocess.check_output(["/usr/local/bin/alfred-json","-z", "-r",str(self.request_data_type_2),"-f","json"])
alfred_data_2 = json.loads(output.decode("utf-8"))
return json.dumps(rec_merge(alfred_data_1, alfred_data_2))
if __name__ == "__main__":
ad = alfred_merge()
al = ad.aliases()
print(al)

36
aliases.json_sample
View file

@ -1,36 +0,0 @@
[
{
"node_id": "krtek",
"hostname": "krtek",
"location": {
"longitude": 10.74,
"latitude": 53.86
},
"network": {
"mesh": {
"bat0": {
"interfaces": {
"tunnel": [
"00:25:86:e6:f1:bf"
]
}
}
}
}
},
{
"node_id": "gw1",
"hostname": "burgtor",
"network": {
"mesh": {
"bat0": {
"interfaces": {
"tunnel": [
"52:54:00:f3:62:d9"
]
}
}
}
}
}
]

194
backend.py
View file

@ -1,194 +0,0 @@
#!/usr/bin/env python3
"""
backend.py - ffmap-backend runner
https://github.com/ffnord/ffmap-backend
"""
import argparse
import json
import os
import sys
from datetime import datetime
import networkx as nx
from networkx.readwrite import json_graph
from lib import graph, nodes
from lib.alfred import Alfred
from lib.batman import Batman
from lib.rrddb import RRD
from lib.nodelist import export_nodelist
from lib.validate import validate_nodeinfos
NODES_VERSION = 1
GRAPH_VERSION = 1
def main(params):
os.makedirs(params['dest_dir'], exist_ok=True)
nodes_fn = os.path.join(params['dest_dir'], 'nodes.json')
tmp_nodes_fn = os.path.join(params['dest_dir'], 'nodes.json.tmp')
graph_fn = os.path.join(params['dest_dir'], 'graph.json')
tmp_graph_fn = os.path.join(params['dest_dir'], 'graph.json.tmp')
nodelist_fn = os.path.join(params['dest_dir'], 'nodelist.json')
tmp_nodelist_fn = os.path.join(params['dest_dir'], 'nodelist.json.tmp')
now = datetime.utcnow().replace(microsecond=0)
# parse mesh param and instantiate Alfred/Batman instances
alfred_instances = []
batman_instances = []
for value in params['mesh']:
# (1) only batman-adv if, no alfred sock
if ':' not in value:
if len(params['mesh']) > 1:
raise ValueError(
'Multiple mesh interfaces require the use of '
'alfred socket paths.')
alfred_instances.append(Alfred(unix_sockpath=None))
batman_instances.append(Batman(mesh_interface=value))
else:
# (2) batman-adv if + alfred socket
try:
batif, alfredsock = value.split(':')
alfred_instances.append(Alfred(unix_sockpath=alfredsock))
batman_instances.append(Batman(mesh_interface=batif,
alfred_sockpath=alfredsock))
except ValueError:
raise ValueError(
'Unparseable value "{0}" in --mesh parameter.'.
format(value))
# read nodedb state from node.json
try:
with open(nodes_fn, 'r') as nodedb_handle:
nodedb = json.load(nodedb_handle)
except IOError:
nodedb = {'nodes': dict()}
# flush nodedb if it uses the old format
if 'links' in nodedb:
nodedb = {'nodes': dict()}
# set version we're going to output
nodedb['version'] = NODES_VERSION
# update timestamp and assume all nodes are offline
nodedb['timestamp'] = now.isoformat()
for node_id, node in nodedb['nodes'].items():
node['flags']['online'] = False
# integrate alfred nodeinfo
for alfred in alfred_instances:
nodeinfo = validate_nodeinfos(alfred.nodeinfo())
nodes.import_nodeinfo(nodedb['nodes'], nodeinfo,
now, assume_online=True)
# integrate static aliases data
for aliases in params['aliases']:
with open(aliases, 'r') as f:
# nodeinfo = validate_nodeinfos(json.load(f))
nodes.import_nodeinfo(nodedb['nodes'], json.load(f),
now, assume_online=False, statics=True)
nodes.reset_statistics(nodedb['nodes'])
for alfred in alfred_instances:
nodes.import_statistics(nodedb['nodes'], alfred.statistics())
# acquire gwl and visdata for each batman instance
mesh_info = []
for batman in batman_instances:
vd = batman.vis_data()
gwl = batman.gateway_list()
mesh_info.append((vd, gwl))
# update nodedb from batman-adv data
for vd, gwl in mesh_info:
nodes.import_mesh_ifs_vis_data(nodedb['nodes'], vd)
nodes.import_vis_clientcount(nodedb['nodes'], vd)
nodes.mark_vis_data_online(nodedb['nodes'], vd, now)
nodes.mark_gateways(nodedb['nodes'], gwl)
# clear the nodedb from nodes that have not been online in $prune days
if params['prune']:
nodes.prune_nodes(nodedb['nodes'], now, params['prune'])
# build nxnetworks graph from nodedb and visdata
batadv_graph = nx.DiGraph()
for vd, gwl in mesh_info:
graph.import_vis_data(batadv_graph, nodedb['nodes'], vd)
# force mac addresses to be vpn-link only (like gateways for example)
if params['vpn']:
graph.mark_vpn(batadv_graph, frozenset(params['vpn']))
def extract_tunnel(nodes):
macs = set()
for id, node in nodes.items():
try:
for mac in node["nodeinfo"]["network"]["mesh"]["bat0"]["interfaces"]["tunnel"]:
macs.add(mac)
for mac in node["nodeinfo"]["network"]["mesh"]["bat-ffhh"]["interfaces"]["tunnel"]:
macs.add(mac)
except KeyError:
pass
return macs
graph.mark_vpn(batadv_graph, extract_tunnel(nodedb['nodes']))
batadv_graph = graph.merge_nodes(batadv_graph)
batadv_graph = graph.to_undirected(batadv_graph)
# write processed data to dest dir
with open(tmp_nodes_fn, 'w') as f:
json.dump(nodedb, f)
graph_out = {'batadv': json_graph.node_link_data(batadv_graph),
'version': GRAPH_VERSION}
with open(tmp_graph_fn, 'w') as f:
json.dump(graph_out, f)
with open(tmp_nodelist_fn, 'w') as f:
json.dump(export_nodelist(now, nodedb), f)
os.rename(tmp_nodes_fn, nodes_fn)
os.rename(tmp_graph_fn, graph_fn)
os.rename(tmp_nodelist_fn, nodelist_fn)
# optional rrd graphs (trigger with --rrd)
if params['rrd']:
script_directory = os.path.dirname(os.path.realpath(__file__))
rrd = RRD(os.path.join(script_directory, 'nodedb'),
os.path.join(params['dest_dir'], 'nodes'))
rrd.update_database(nodedb['nodes'])
rrd.update_images()
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument('-a', '--aliases',
help='Read aliases from FILE',
nargs='+', default=[], metavar='FILE')
parser.add_argument('-m', '--mesh',
default=['bat0'], nargs='+',
help='Use given batman-adv mesh interface(s) (defaults'
'to bat0); specify alfred unix socket like '
'bat0:/run/alfred0.sock.')
parser.add_argument('-d', '--dest-dir', action='store',
help='Write output to destination directory',
required=True)
parser.add_argument('-V', '--vpn', nargs='+', metavar='MAC',
help='Assume MAC addresses are part of vpn')
parser.add_argument('-p', '--prune', metavar='DAYS', type=int,
help='forget nodes offline for at least DAYS')
parser.add_argument('--with-rrd', dest='rrd', action='store_true',
default=False,
help='enable the rendering of RRD graphs (cpu '
'intensive)')
options = vars(parser.parse_args())
main(options)

52
ffmap-d3.jq
View file

@ -1,52 +0,0 @@
{
"meta": {
"timestamp": $nodes.timestamp
},
"nodes": (
$graph.batadv.nodes
| map(
if has("node_id") and .node_id
then (
$nodes.nodes[.node_id] as $node
| {
"id": .id,
"uptime": $node.statistics.uptime,
"flags": ($node.flags + {"client": false}),
"name": $node.nodeinfo.hostname,
"clientcount": (if $node.statistics.clients >= 0 then $node.statistics.clients else 0 end),
"hardware": $node.nodeinfo.hardware.model,
"firmware": $node.nodeinfo.software.firmware.release,
"geo": (if $node.nodeinfo.location then [$node.nodeinfo.location.latitude, $node.nodeinfo.location.longitude] else null end),
#"lastseen": $node.lastseen,
"network": $node.nodeinfo.network
}
)
else
{
"flags": {},
"id": .id,
"geo": null,
"clientcount": 0
}
end
)
),
"links": (
$graph.batadv.links
| map(
$graph.batadv.nodes[.source].node_id as $source_id
| $graph.batadv.nodes[.target].node_id as $target_id
| select(
$source_id and $target_id and
($nodes.nodes | (has($source_id) and has($target_id)))
)
| {
"target": .target,
"source": .source,
"quality": "\(.tq), \(.tq)",
"id": ($source_id + "-" + $target_id),
"type": (if .vpn then "vpn" else null end)
}
)
)
}

42
ffmap/__init__.py Normal file
View file

@ -0,0 +1,42 @@
import importlib
from ffmap.nodedb import NodeDB
def run(inputs, outputs):
"""Fill the database with given inputs and give it to given outputs.
Arguments:
inputs -- list of Input instances (with a compatible get_data(nodedb) method)
outputs -- list of Output instances (with a compatible output(nodedb) method)
"""
db = NodeDB()
for input_ in inputs:
input_.get_data(db)
for output in outputs:
output.output(db)
def run_names(inputs, outputs):
"""Fill the database with inputs and give it to outputs, each given
by names.
In contrast to run(inputs, outputs), this method expects only the
names of the modules to use, not instances thereof.
Arguments:
inputs -- list of dicts, each dict having the keys "name" with the
name of the input to use (directory name in inputs/), and
the key "options" with a dict of input-dependent options.
outputs -- list of dicts, see inputs.
"""
input_instances = []
output_instances = []
for input_ in inputs:
module = importlib.import_module(".inputs." + input_["name"], "ffmap")
input_instances.append(module.Input(**input_["options"]))
for output in outputs:
module = importlib.import_module(".outputs." + output["name"], "ffmap")
output_instances.append(module.Output(**output["options"]))
run(input_instances, output_instances)

0
nodedb/.gitkeep → ffmap/inputs/__init__.py
View file

29
ffmap/inputs/alfred.py Normal file
View file

@ -0,0 +1,29 @@
import subprocess
import json
class Input:
def __init__(self,request_data_type = 158):
self.request_data_type = request_data_type
@staticmethod
def _call_alfred(request_data_type):
return json.loads(subprocess.check_output([
"alfred-json",
"-z",
"-r", str(request_data_type),
"-f", "json",
]).decode("utf-8"))
def get_data(self, nodedb):
"""Add data from alfred to the supplied nodedb"""
nodeinfo = self._call_alfred(self.request_data_type)
statistics = self._call_alfred(self.request_data_type+1)
# merge statistics into nodeinfo to be compatible with earlier versions
for mac, node in statistics.items():
if mac in nodeinfo:
nodeinfo[mac]['statistics'] = statistics[mac]
for mac, node in nodeinfo.items():
aliases = [mac] + node.get('network', {}).get('mesh_interfaces', [])
nodedb.add_or_update(aliases, node)

100
ffmap/inputs/batadv.py Normal file
View file

@ -0,0 +1,100 @@
import subprocess
import json
class Input:
"""Fill the NodeDB with links from batadv-vis.
The links are added as lists containing the neighboring nodes, not
only their identifiers! Mind this when exporting the database, as
it probably leads to recursion.
"""
def __init__(self, mesh_interface="bat0"):
self.mesh_interface = mesh_interface
@staticmethod
def _is_similar_mac(a, b):
"""Determine if two MAC addresses are similar."""
if a == b:
return True
# Split the address into bytes
try:
mac_a = list(int(i, 16) for i in a.split(":"))
mac_b = list(int(i, 16) for i in b.split(":"))
except ValueError:
return False
# Second and third byte musn't differ
if mac_a[1] != mac_b[1] or mac_a[2] != mac_b[2]:
return False
# First byte must only differ in bits 2 and 3
if mac_a[0] | 6 != mac_b[0] | 6:
return False
# Count differing bytes after the third
c = [x for x in zip(mac_a[3:], mac_b[3:]) if x[0] != x[1]]
# No more than two additional bytes must differ
if len(c) > 2:
return False
# If no more bytes differ, they are very similar
if len(c) == 0:
return True
# If the sum of absolute differences isn't greater than 2, they
# are pretty similar
delta = sum(abs(i[0] - i[1]) for i in c)
return delta < 2
def get_data(self, nodedb):
"""Add data from batadv-vis to the supplied nodedb"""
output = subprocess.check_output([
"batadv-vis",
"-i", str(self.mesh_interface),
"-f", "jsondoc",
])
data = json.loads(output.decode("utf-8"))
# First pass
for node in data["vis"]:
# Determine possible other MAC addresses of this node by
# comparing all its client's MAC addresses to its primary
# MAC address. If they are similar, it probably is another
# address of the node itself! If it isn't, it is a real
# client.
node['aliases'] = [node["primary"]]
if 'secondary' in node:
node['aliases'].extend(node['secondary'])
real_clients = []
for mac in node["clients"]:
if self._is_similar_mac(mac, node["primary"]):
node['aliases'].append(mac)
else:
real_clients.append(mac)
node['clients'] = real_clients
# Add nodes and aliases without any information at first.
# This way, we can later link the objects themselves.
nodedb.add_or_update(node['aliases'])
# Second pass
for node in data["vis"]:
# We only need the primary address now, all aliases are
# already present in the database. Furthermore, we can be
# sure that all neighbors are in the database as well. If
# a neighbor isn't added already, we simply ignore it.
nodedb.add_or_update(
[node["primary"]],
{
"clients": node["clients"],
"neighbors": [
{
"metric": neighbor['metric'],
"neighbor": nodedb[neighbor['neighbor']],
} for neighbor in node["neighbors"]
if neighbor['neighbor'] in nodedb
]
}
)

71
ffmap/inputs/wiki.py Executable file
View file

@ -0,0 +1,71 @@
import json
import argparse
from itertools import zip_longest
from urllib.request import urlopen
from bs4 import BeautifulSoup
class Input:
def __init__(self, url="http://luebeck.freifunk.net/wiki/Knoten"):
self.url = url
def fetch_wikitable(self):
f = urlopen(self.url)
soup = BeautifulSoup(f)
table = soup.find("table")
rows = table.find_all("tr")
headers = []
data = []
def maybe_strip(x):
if isinstance(x.string, str):
return x.string.strip()
else:
return ""
for row in rows:
tds = list([maybe_strip(x) for x in row.find_all("td")])
ths = list([maybe_strip(x) for x in row.find_all("th")])
if any(tds):
data.append(tds)
if any(ths):
headers = ths
return [dict(zip(headers, d)) for d in data]
def get_data(self, nodedb):
nodes = self.fetch_wikitable()
for node in nodes:
if "MAC" not in node or not node["MAC"]:
# without MAC, we cannot merge this data with others, so
# we might as well ignore it
continue
newnode = {
"network": {
"mac": node.get("MAC").lower(),
},
"location": {
"latitude": float(node.get("GPS", " ").split(" ")[0]),
"longitude": float(node.get("GPS", " ").split(" ")[1]),
"description": node.get("Ort"),
} if " " in node.get("GPS", "") else None,
"hostname": node.get("Knotenname"),
"hardware": {
"model": node["Router"],
} if node.get("Router") else None,
"software": {
"firmware": {
"base": "LFF",
"release": node.get("LFF Version"),
},
},
"owner": {
"contact": node["Betreiber"],
} if node.get("Betreiber") else None,
}
# remove keys with None as value
newnode = {k: v for k,v in newnode.items() if v is not None}
nodedb.add_or_update([newnode["network"]["mac"]], newnode)

91
ffmap/node.py Normal file
View file

@ -0,0 +1,91 @@
from collections import defaultdict
class NoneDict:
"""Act like None but return a NoneDict for every item request.
This is similar to the behaviour of collections.defaultdict in that
even previously inexistent keys can be accessed, but nothing is
stored permanently in this class.
"""
def __repr__(self):
return 'NoneDict()'
def __bool__(self):
return False
def __getitem__(self, k):
return NoneDict()
def __json__(self):
return None
def __float__(self):
return float('NaN')
def __iter__(self):
# empty generator
return
yield
def __setitem__(self, key, value):
raise RuntimeError("NoneDict is readonly")
class Node(defaultdict):
_id = None
def __init__(self, id_=None):
self._id = id_
super().__init__(NoneDict)
def __repr__(self):
return "Node(%s)" % self.id
@property
def id(self):
return self._id
def __hash__(self):
"""Generate hash from the node's id.
WARNING: Obviously this hash doesn't cover all of the node's
data, but we need nodes to be hashable in order to eliminate
duplicates in the NodeDB.
At least the id cannot change after initialization...
"""
return hash(self.id)
def deep_update(self, other):
"""Update the dictionary like dict.update() but recursively."""
def dmerge(a, b):
for k, v in b.items():
if isinstance(v, dict) and isinstance(a.get(k), dict):
dmerge(a[k], v)
else:
a[k] = v
dmerge(self, other)
@property
def vpn_neighbors(self):
try:
vpn_neighbors = []
for neighbor in self['neighbors']:
if neighbor['neighbor']['vpn']:
vpn_neighbors.append(neighbor)
return vpn_neighbors
except TypeError:
return []
def export(self):
"""Generate a serializable dict of the node.
In particular, this replaces any references to other nodes by
their id to prevent circular references.
"""
ret = dict(self)
if "neighbors" in self:
ret["neighbors"] = []
for neighbor in self["neighbors"]:
new_neighbor = {}
for key, val in neighbor.items():
if isinstance(val, Node):
new_neighbor[key] = val.id
else:
new_neighbor[key] = val
ret["neighbors"].append(new_neighbor)
if "id" not in ret:
ret["id"] = self.id
return ret

60
ffmap/nodedb.py Normal file
View file

@ -0,0 +1,60 @@
from .node import Node
class AmbiguityError(Exception):
"""Indicate the ambiguity of identifiers.
This exception is raised if there is more than one match for a set
of identifiers.
Attributes:
identifiers -- set of ambiguous identifiers
"""
identifiers = []
def __init__(self, identifiers):
self.identifiers = identifiers
def __str__(self):
return "Ambiguous identifiers: %s" % ", ".join(self.identifiers)
class NodeDB(dict):
def add_or_update(self, ids, other=None):
"""Add or update a node in the database.
Searches for an already existing node and updates it, or adds a new
one if no existing one is found. Raises an AmbiguityException if
more than one different nodes are found matching the criteria.
Arguments:
ids -- list of possible identifiers (probably MAC addresses) of the
node
other -- dict of values to update in an existing node or add to
the new one. Defaults to None, in which case no values
are added or updated, only the aliases of the
(possibly freshly created) node are updated.
"""
# Find existing node, if any
node = None
node_id = None
for id_ in ids:
if id_ == node_id:
continue
if id_ in self:
if node is not None and node is not self[id_]:
raise AmbiguityError([node_id, id_])
node = self[id_]
node_id = id_
# If no node was found, create a new one
if node is None:
node = Node(ids[0])
# Update the node with the given properties using its own update method.
if other is not None:
node.deep_update(other)
# Add new aliases if any
for id_ in ids:
self[id_] = node

1
ffmap/outputs/__init__.py Normal file
View file

@ -0,0 +1 @@

91
ffmap/outputs/d3json.py Normal file
View file

@ -0,0 +1,91 @@
import json
from datetime import datetime
__all__ = ["Exporter"]
class CustomJSONEncoder(json.JSONEncoder):
"""
JSON encoder that uses an object's __json__() method to convert it
to something JSON-compatible.
"""
def default(self, obj):
try:
return obj.__json__()
except AttributeError:
pass
return super().default(obj)
class Output:
def __init__(self, filepath="nodes.json"):
self.filepath = filepath
@staticmethod
def generate(nodedb):
indexes = {}
nodes = []
count = 0
for node in set(nodedb.values()):
node_export = node.export()
node_export["flags"] = {
"gateway": "vpn" in node and node["vpn"],
"client": False,
"online": True
}
nodes.append(node_export)
indexes[node.id] = count
count += 1
links = {}
for node in set(nodedb.values()):
for neighbor in node.get("neighbors", []):
key = (neighbor["neighbor"].id, node.id)
rkey = tuple(reversed(key))
if rkey in links:
links[rkey]["quality"] += ","+neighbor["metric"]
else:
links[key] = {
"source": indexes[node.id],
"target": indexes[neighbor["neighbor"].id],
"quality": neighbor["metric"],
"type": "vpn" if neighbor["neighbor"]["vpn"] or node["vpn"] else None,
"id": "-".join((node.id, neighbor["neighbor"].id)),
}
clientcount = 0
for client in node.get("clients", []):
nodes.append({
"id": "%s-%s" % (node.id, clientcount),
"flags": {
"client": True,
"online": True,
"gateway": False
}
})
indexes[client] = count
links[(node.id, client)] = {
"source": indexes[node.id],
"target": indexes[client],
"quality": "TT",
"type": "client",
"id": "%s-%i" % (node.id, clientcount),
}
count += 1
clientcount += 1
return {
"nodes": nodes,
"links": list(links.values()),
"meta": {
"timestamp": datetime.utcnow()
.replace(microsecond=0)
.isoformat()
}
}
def output(self, nodedb):
with open(self.filepath, "w") as nodes_json:
json.dump(
self.generate(nodedb),
nodes_json,
cls=CustomJSONEncoder
)

30
ffmap/outputs/rrd.py Normal file
View file

@ -0,0 +1,30 @@
import os
from ffmap.rrd.rrds import NodeRRD, GlobalRRD
class Output:
def __init__(self, directory="nodedb"):
self.directory = directory
try:
os.mkdir(self.directory)
except OSError:
pass
def output(self, nodedb):
nodes = set(nodedb.values())
clients = 0
nodecount = 0
for node in nodes:
clients += len(node.get("clients", []))
nodecount += 1
NodeRRD(
os.path.join(
self.directory,
str(node.id).replace(':', '') + '.rrd'
),
node
).update()
GlobalRRD(os.path.join(self.directory, "nodes.rrd")).update(
nodecount,
clients
)

86
lib/RRD.py → ffmap/rrd/__init__.py
View file

@ -1,20 +1,19 @@
import subprocess
import re
import io
import os
from tempfile import TemporaryFile
from operator import xor, eq
from functools import reduce
from itertools import starmap
import math
class RRDIncompatibleException(Exception):
"""
Is raised when an RRD doesn't have the desired definition and cannot be
upgraded to it.
"""
pass
class RRDOutdatedException(Exception):
"""
Is raised when an RRD doesn't have the desired definition, but can be
@ -26,8 +25,7 @@ if not hasattr(__builtins__, "FileNotFoundError"):
class FileNotFoundError(Exception):
pass
class RRD(object):
class RRD:
"""
An RRD is a Round Robin Database, a database which forgets old data and
aggregates multiple records into new ones.
@ -51,7 +49,7 @@ class RRD(object):
def _exec_rrdtool(self, cmd, *args, **kwargs):
pargs = ["rrdtool", cmd, self.filename]
for k, v in kwargs.items():
for k,v in kwargs.items():
pargs.extend(["--" + k, str(v)])
pargs.extend(args)
subprocess.check_output(pargs)
@ -59,7 +57,7 @@ class RRD(object):
def __init__(self, filename):
self.filename = filename
def ensure_sanity(self, ds_list, rra_list, **kwargs):
def ensureSanity(self, ds_list, rra_list, **kwargs):
"""
Create or upgrade the RRD file if necessary to contain all DS in
ds_list. If it needs to be created, the RRAs in rra_list and any kwargs
@ -67,13 +65,13 @@ class RRD(object):
database are NOT modified!
"""
try:
self.check_sanity(ds_list)
self.checkSanity(ds_list)
except FileNotFoundError:
self.create(ds_list, rra_list, **kwargs)
except RRDOutdatedException:
self.upgrade(ds_list)
def check_sanity(self, ds_list=()):
def checkSanity(self, ds_list=()):
"""
Check if the RRD file exists and contains (at least) the DS listed in
ds_list.
@ -83,11 +81,8 @@ class RRD(object):
info = self.info()
if set(ds_list) - set(info['ds'].values()) != set():
for ds in ds_list:
if ds.name in info['ds'] and\
ds.type != info['ds'][ds.name].type:
raise RRDIncompatibleException(
"{} is {} but should be {}".format(
ds.name, ds.type, info['ds'][ds.name].type))
if ds.name in info['ds'] and ds.type != info['ds'][ds.name].type:
raise RRDIncompatibleException("%s is %s but should be %s" % (ds.name, ds.type, info['ds'][ds.name].type))
else:
raise RRDOutdatedException()
@ -110,10 +105,8 @@ class RRD(object):
if ds.name in info['ds']:
old_ds = info['ds'][ds.name]
if info['ds'][ds.name].type != ds.type:
raise RuntimeError(
"Cannot convert existing DS '{}'"
"from type '{}' to '{}'".format(
ds.name, old_ds.type, ds.type))
raise RuntimeError('Cannot convert existing DS "%s" from type "%s" to "%s"' %
(ds.name, old_ds.type, ds.type))
ds.index = old_ds.index
new_ds[ds.index] = ds
else:
@ -123,11 +116,12 @@ class RRD(object):
dump = subprocess.Popen(
["rrdtool", "dump", self.filename],
stdout=subprocess.PIPE)
stdout=subprocess.PIPE
)
restore = subprocess.Popen(
["rrdtool", "restore", "-", self.filename + ".new"],
stdin=subprocess.PIPE)
stdin=subprocess.PIPE
)
echo = True
ds_definitions = True
for line in dump.stdout:
@ -149,17 +143,19 @@ class RRD(object):
<value>%s</value>
<unknown_sec> %i </unknown_sec>
</ds>
""" % (ds.name,
ds.type,
ds.args[0],
ds.args[1],
ds.args[2],
ds.last_ds,
ds.value,
ds.unknown_sec), "utf-8"))
""" % (
ds.name,
ds.type,
ds.args[0],
ds.args[1],
ds.args[2],
ds.last_ds,
ds.value,
ds.unknown_sec)
, "utf-8"))
if b'</cdp_prep>' in line:
restore.stdin.write(added_ds_num * b"""
restore.stdin.write(added_ds_num*b"""
<ds>
<primary_value> NaN </primary_value>
<secondary_value> NaN </secondary_value>
@ -173,7 +169,7 @@ class RRD(object):
restore.stdin.write(
line.replace(
b'</row>',
(added_ds_num * b'<v>NaN</v>') + b'</row>'
(added_ds_num*b'<v>NaN</v>')+b'</row>'
)
)
@ -241,8 +237,7 @@ class RRD(object):
for line in out.splitlines():
base = info
for match in self._info_regex.finditer(line):
section, key, name, value = match.group(
"section", "key", "name", "value")
section, key, name, value = match.group("section", "key", "name", "value")
if section and key:
try:
key = int(key)
@ -263,8 +258,7 @@ class RRD(object):
base[name] = value
dss = {}
for name, ds in info['ds'].items():
ds_obj = DS(name, ds['type'], ds['minimal_heartbeat'],
ds['min'], ds['max'])
ds_obj = DS(name, ds['type'], ds['minimal_heartbeat'], ds['min'], ds['max'])
ds_obj.index = ds['index']
ds_obj.last_ds = ds['last_ds']
ds_obj.value = ds['value']
@ -273,14 +267,12 @@ class RRD(object):
info['ds'] = dss
rras = []
for rra in info['rra'].values():
rras.append(RRA(rra['cf'], rra['xff'],
rra['pdp_per_row'], rra['rows']))
rras.append(RRA(rra['cf'], rra['xff'], rra['pdp_per_row'], rra['rows']))
info['rra'] = rras
self._cached_info = info
return info
class DS(object):
class DS:
"""
DS stands for Data Source and represents one line of data points in a Round
Robin Database (RRD).
@ -292,7 +284,6 @@ class DS(object):
last_ds = 'U'
value = 0
unknown_sec = 0
def __init__(self, name, dst, *args):
self.name = name
self.type = dst
@ -302,7 +293,7 @@ class DS(object):
return "DS:%s:%s:%s" % (
self.name,
self.type,
":".join(map(str, self._nan_to_u_args()))
":".join(map(str, self._nan_to_U_args()))
)
def __repr__(self):
@ -314,23 +305,22 @@ class DS(object):
)
def __eq__(self, other):
return all(starmap(eq, zip(self.compare_keys(), other.compare_keys())))
return all(starmap(eq, zip(self._compare_keys(), other._compare_keys())))
def __hash__(self):
return reduce(xor, map(hash, self.compare_keys()))
return reduce(xor, map(hash, self._compare_keys()))
def _nan_to_u_args(self):
def _nan_to_U_args(self):
return tuple(
'U' if type(arg) is float and math.isnan(arg)
else arg
for arg in self.args
)
def compare_keys(self):
return self.name, self.type, self._nan_to_u_args()
def _compare_keys(self):
return (self.name, self.type, self._nan_to_U_args())
class RRA(object):
class RRA:
def __init__(self, cf, *args):
self.cf = cf
self.args = args

115
ffmap/rrd/rrds.py Normal file
View file

@ -0,0 +1,115 @@
import os
import subprocess
from ffmap.node import Node
from . import RRD, DS, RRA
class NodeRRD(RRD):
ds_list = [
DS('upstate', 'GAUGE', 120, 0, 1),
DS('clients', 'GAUGE', 120, 0, float('NaN')),
DS('neighbors', 'GAUGE', 120, 0, float('NaN')),
DS('vpn_neighbors', 'GAUGE', 120, 0, float('NaN')),
DS('loadavg', 'GAUGE', 120, 0, float('NaN')),
DS('rx_bytes', 'DERIVE', 120, 0, float('NaN')),
DS('rx_packets', 'DERIVE', 120, 0, float('NaN')),
DS('tx_bytes', 'DERIVE', 120, 0, float('NaN')),
DS('tx_packets', 'DERIVE', 120, 0, float('NaN')),
DS('mgmt_rx_bytes', 'DERIVE', 120, 0, float('NaN')),
DS('mgmt_rx_packets', 'DERIVE', 120, 0, float('NaN')),
DS('mgmt_tx_bytes', 'DERIVE', 120, 0, float('NaN')),
DS('mgmt_tx_packets', 'DERIVE', 120, 0, float('NaN')),
DS('forward_bytes', 'DERIVE', 120, 0, float('NaN')),
DS('forward_packets', 'DERIVE', 120, 0, float('NaN')),
]
rra_list = [
RRA('AVERAGE', 0.5, 1, 120), # 2 hours of 1 minute samples
RRA('AVERAGE', 0.5, 5, 1440), # 5 days of 5 minute samples
RRA('AVERAGE', 0.5, 60, 720), # 30 days of 1 hour samples
RRA('AVERAGE', 0.5, 720, 730), # 1 year of 12 hour samples
]
def __init__(self, filename, node = None):
"""
Create a new RRD for a given node.
If the RRD isn't supposed to be updated, the node can be omitted.
"""
self.node = node
super().__init__(filename)
self.ensureSanity(self.ds_list, self.rra_list, step=60)
@property
def imagename(self):
return os.path.basename(self.filename).rsplit('.', 2)[0] + ".png"
def update(self):
values = {
'upstate': 1,
'clients': float(len(self.node.get('clients', []))),
'neighbors': float(len(self.node.get('neighbors', []))),
'vpn_neighbors': float(len(self.node.vpn_neighbors)),
'loadavg': float(self.node['statistics']['loadavg']),
}
for item in ('rx', 'tx', 'mgmt_rx', 'mgmt_tx', 'forward'):
try:
values[item + '_bytes'] = int(self.node['statistics']['traffic'][item]['bytes'])
except TypeError:
pass
try:
values[item + '_packets'] = int(self.node['statistics']['traffic'][item]['packets'])
except TypeError:
pass
super().update(values)
def graph(self, directory, timeframe):
"""
Create a graph in the given directory. The file will be named
basename.png if the RRD file is named basename.rrd
"""
args = ['rrdtool','graph', os.path.join(directory, self.imagename),
'-s', '-' + timeframe ,
'-w', '800',
'-h', '400',
'-l', '0',
'-y', '1:1',
'DEF:clients=' + self.filename + ':clients:AVERAGE',
'VDEF:maxc=clients,MAXIMUM',
'CDEF:c=0,clients,ADDNAN',
'CDEF:d=clients,UN,maxc,UN,1,maxc,IF,*',
'AREA:c#0F0:up\\l',
'AREA:d#F00:down\\l',
'LINE1:c#00F:clients connected\\l',
]
subprocess.check_output(args)
class GlobalRRD(RRD):
ds_list = [
# Number of nodes available
DS('nodes', 'GAUGE', 120, 0, float('NaN')),
# Number of client available
DS('clients', 'GAUGE', 120, 0, float('NaN')),
]
rra_list = [
RRA('AVERAGE', 0.5, 1, 120), # 2 hours of 1 minute samples
RRA('AVERAGE', 0.5, 60, 744), # 31 days of 1 hour samples
RRA('AVERAGE', 0.5, 1440, 1780),# ~5 years of 1 day samples
]
def __init__(self, filepath):
super().__init__(filepath)
self.ensureSanity(self.ds_list, self.rra_list, step=60)
def update(self, nodeCount, clientCount):
super().update({'nodes': nodeCount, 'clients': clientCount})
def graph(self, filename, timeframe):
args = ["rrdtool", 'graph', filename,
'-s', '-' + timeframe,
'-w', '800',
'-h' '400',
'DEF:nodes=' + self.filename + ':nodes:AVERAGE',
'LINE1:nodes#F00:nodes\\l',
'DEF:clients=' + self.filename + ':clients:AVERAGE',
'LINE2:clients#00F:clients',
]
subprocess.check_output(args)

69
ffmap/run.py Normal file
View file

@ -0,0 +1,69 @@
#!/usr/bin/env python3
import argparse
import sys
from ffmap import run_names
class MyAction(argparse.Action):
def __call__(self, parser, namespace, values, option_string=None):
if self.dest.startswith(("input_", "output_")):
collection_name = self.dest.split("_")[0] + "s"
name = self.dest.split("_", 1)[1]
if not hasattr(namespace, collection_name):
setattr(namespace, collection_name, [])
collection = getattr(namespace, collection_name)
collection.append({
"name": name,
"options": {self.metavar.lower(): values}
if values is not None else {}
})
else:
raise Exception("Unexpected dest=" + self.dest)
def parser_add_myarg(parser, name, metavar="OPT", help=None):
parser.add_argument("--" + name,
metavar=metavar,
type=str,
nargs='?',
const=None,
action=MyAction,
help=help)
parser = argparse.ArgumentParser(
description="""Merge node data from multiple sources and generate
various output formats from this data""",
)
input_group = parser.add_argument_group("Inputs", description="""
Inputs are used in the order given on the command line, where later
inputs can overwrite attributes of earlier inputs if named equally,
but the first input encountering a node sets its id, which is
immutable afterwards.
The same input can be given multiple times, probably with different
options.
""")
output_group = parser.add_argument_group("Outputs")
parser_add_myarg(input_group, 'input-alfred', metavar="REQUEST_DATA_TYPE",
help="read node details from A.L.F.R.E.D.")
parser_add_myarg(input_group, 'input-wiki', metavar="URL",
help="read node details from a Wiki page")
parser_add_myarg(input_group, 'input-batadv', metavar="MESH_INTERFACE",
help="add node's neighbors and clients from batadv-vis")
parser_add_myarg(output_group, 'output-d3json', metavar="FILEPATH",
help="generate JSON file compatible with ffmap-d3")
parser_add_myarg(output_group, 'output-rrd', metavar="DIRECTORY",
help="update RRDs with statistics, one global and one per node")
args = parser.parse_args()
if "inputs" not in args or not args.inputs:
parser.print_help(sys.stderr)
sys.stderr.write("\nERROR: No input has been defined!\n")
sys.exit(1)
if "outputs" not in args or not args.outputs:
parser.print_help(sys.stderr)
sys.stderr.write("\nERROR: No output has been defined!\n")
sys.exit(1)
run_names(inputs=args.inputs, outputs=args.outputs)

29
gateway.json
View file

@ -1,29 +0,0 @@
[
{
"node_id": "deadbfff0101",
"hostname": "gw01"
},
{
"node_id": "deadbeef0505",
"hostname": "gw02.hamburg.freifunk.net",
"network": {
"mac": "de:ad:be:ef:05:05",
"mesh": {
"bat0": {
"interfaces": {
"tunnel": [
"de:ad:be:ff:05:05",
"de:ad:be:fc:05:05",
"de:ad:bf:ff:05:05"
]
}
}
}
}
},
{
"node_id": "00163efb9d8d",
"hostname": "gw03"
}
]

110
generate_aliases.py
View file

@ -1,110 +0,0 @@
#!/usr/bin/env python2
from __future__ import print_function
import json
import os
import sys
if len(sys.argv) != 2:
print('usage: ' + sys.argv[0] + ' /path/to/peers')
sys.exit(1)
peersDir = sys.argv[1]
def normalizeMac(mac):
mac = mac.lower()
normalized = ''
n = 0
for c in mac:
if c != ':':
if n > 0 and n % 2 == 0:
normalized = normalized + ':'
normalized = normalized + c
n += 1
return normalized
def toAlias(peer):
alias = {}
if not (peer.has_key('name') and peer.has_key('mac')):
return None
name = peer['name']
mac = peer['mac']
alias['node_id'] = mac.replace(':', '')
alias['hostname'] = name
if peer.has_key('geo'):
geo = peer['geo']
location = {}
if geo.has_key('lon'): location['longitude'] = geo['lon']
if geo.has_key('lat'): location['latitude'] = geo['lat']
alias['location'] = location
#alias['network'] = {}
#alias['network']['mesh_interfaces'] = [mac]
return alias
aliases = []
for filename in os.listdir(peersDir):
if len(filename) == 0 or filename[0] == '.':
continue
isGateway = False
absFilename = peersDir + '/' + filename
if os.path.isfile(absFilename):
peerFile = open(absFilename, 'r')
try:
peerLines = peerFile.readlines()
peer = {}
for line in peerLines:
parts = line.split()
if len(parts) > 2:
if parts[1] == 'Knotenname:':
peer['name'] = parts[2]
elif parts[0] == 'remote':
isGateway = True
elif parts[1] == 'MAC:':
peer['mac'] = normalizeMac(parts[2])
elif parts[1] == 'Koordinaten:' and len(parts) > 3:
try:
peer['geo'] = {'lat': float(parts[2]), 'lon': float(parts[3])}
except ValueError:
print('Error in %s: Invalid coordinates: %s' % (absFilename, parts[2:4]), file = sys.stderr)
elif len(parts) == 2 and parts[0] == 'key':
keyParts = parts[1].split('"')
if len(keyParts) > 1:
peer['vpn'] = keyParts[1].lower()
if isGateway:
continue
alias = toAlias(peer)
if alias:
aliases.append(alias)
except Exception as e:
print('Error in %s, ignoring peer: %s' % (absFilename, e), file = sys.stderr)
finally:
peerFile.close()
print(json.dumps(aliases))

112
generate_aliases_v2.py
View file

@ -1,112 +0,0 @@
#!/usr/bin/env python2
from __future__ import print_function
import json
import os
import sys
if len(sys.argv) != 2:
print('usage: ' + sys.argv[0] + ' /path/to/peers')
sys.exit(1)
peersDir = sys.argv[1]
def normalizeMac(mac):
mac = mac.lower()
normalized = ''
n = 0
for c in mac:
if c != ':':
if n > 0 and n % 2 == 0:
normalized = normalized + ':'
normalized = normalized + c
n += 1
return normalized
def toAlias(peer):
alias = {}
if not (peer.has_key('name') and peer.has_key('mac')):
return None
name = peer['name']
mac = peer['mac']
alias['node_id'] = mac.replace(':', '')
alias['hostname'] = name
if peer.has_key('geo'):
geo = peer['geo']
location = {}
if geo.has_key('lon'): location['longitude'] = geo['lon']
if geo.has_key('lat'): location['latitude'] = geo['lat']
alias['location'] = location
#alias['network'] = {}
#alias['network']['mesh_interfaces'] = [mac]
return {'nodeinfo':alias}
aliases = {}
for filename in os.listdir(peersDir):
if len(filename) == 0 or filename[0] == '.':
continue
isGateway = False
absFilename = peersDir + '/' + filename
if os.path.isfile(absFilename):
peerFile = open(absFilename, 'r')
try:
peerLines = peerFile.readlines()
peer = {}
for line in peerLines:
parts = line.split()
if len(parts) > 2:
if parts[1] == 'Knotenname:':
peer['name'] = parts[2]
elif parts[0] == 'remote':
isGateway = True
elif parts[1] == 'MAC:':
peer['mac'] = normalizeMac(parts[2])
elif parts[1] == 'Koordinaten:' and len(parts) > 3:
try:
peer['geo'] = {'lat': float(parts[2]), 'lon': float(parts[3])}
except ValueError:
print('Error in %s: Invalid coordinates: %s' % (absFilename, parts[2:4]), file = sys.stderr)
elif len(parts) == 2 and parts[0] == 'key':
keyParts = parts[1].split('"')
if len(keyParts) > 1:
peer['vpn'] = keyParts[1].lower()
if isGateway:
continue
alias = toAlias(peer)
if alias:
tmpid = alias['nodeinfo']['node_id']
# alias['nodeinfo'].pop('node_id')
aliases[tmpid] = alias
except Exception as e:
print('Error in %s, ignoring peer: %s' % (absFilename, e), file = sys.stderr)
finally:
peerFile.close()
print(json.dumps(aliases))

40
lib/GlobalRRD.py
View file

@ -1,40 +0,0 @@
import os
import subprocess
from lib.RRD import DS, RRA, RRD
class GlobalRRD(RRD):
ds_list = [
# Number of nodes available
DS('nodes', 'GAUGE', 120, 0, float('NaN')),
# Number of client available
DS('clients', 'GAUGE', 120, 0, float('NaN')),
]
rra_list = [
# 2 hours of 1 minute samples
RRA('AVERAGE', 0.5, 1, 120),
# 31 days of 1 hour samples
RRA('AVERAGE', 0.5, 60, 744),
# ~5 years of 1 day samples
RRA('AVERAGE', 0.5, 1440, 1780),
]
def __init__(self, directory):
super().__init__(os.path.join(directory, "nodes.rrd"))
self.ensure_sanity(self.ds_list, self.rra_list, step=60)
# TODO: fix this, python does not support function overloading
def update(self, node_count, client_count):
super().update({'nodes': node_count, 'clients': client_count})
def graph(self, filename, timeframe):
args = ["rrdtool", 'graph', filename,
'-s', '-' + timeframe,
'-w', '800',
'-h' '400',
'DEF:nodes=' + self.filename + ':nodes:AVERAGE',
'LINE1:nodes#F00:nodes\\l',
'DEF:clients=' + self.filename + ':clients:AVERAGE',
'LINE2:clients#00F:clients']
subprocess.check_output(args)

61
lib/NodeRRD.py
View file

@ -1,61 +0,0 @@
import os
import subprocess
from lib.RRD import DS, RRA, RRD
class NodeRRD(RRD):
ds_list = [
DS('upstate', 'GAUGE', 120, 0, 1),
DS('clients', 'GAUGE', 120, 0, float('NaN')),
]
rra_list = [
# 2 hours of 1 minute samples
RRA('AVERAGE', 0.5, 1, 120),
# 5 days of 5 minute samples
RRA('AVERAGE', 0.5, 5, 1440),
# 30 days of 1 hour samples
RRA('AVERAGE', 0.5, 60, 720),
# 1 year of 12 hour samples
RRA('AVERAGE', 0.5, 720, 730),
]
def __init__(self, filename, node=None):
"""
Create a new RRD for a given node.
If the RRD isn't supposed to be updated, the node can be omitted.
"""
self.node = node
super().__init__(filename)
self.ensure_sanity(self.ds_list, self.rra_list, step=60)
@property
def imagename(self):
return "{basename}.png".format(
basename=os.path.basename(self.filename).rsplit('.', 2)[0])
# TODO: fix this, python does not support function overloading
def update(self):
super().update({'upstate': int(self.node['flags']['online']),
'clients': self.node['statistics']['clients']})
def graph(self, directory, timeframe):
"""
Create a graph in the given directory. The file will be named
basename.png if the RRD file is named basename.rrd
"""
args = ['rrdtool', 'graph', os.path.join(directory, self.imagename),
'-s', '-' + timeframe,
'-w', '800',
'-h', '400',
'-l', '0',
'-y', '1:1',
'DEF:clients=' + self.filename + ':clients:AVERAGE',
'VDEF:maxc=clients,MAXIMUM',
'CDEF:c=0,clients,ADDNAN',
'CDEF:d=clients,UN,maxc,UN,1,maxc,IF,*',
'AREA:c#0F0:up\\l',
'AREA:d#F00:down\\l',
'LINE1:c#00F:clients connected\\l']
subprocess.check_output(args)

1
lib/__init__.py
View file

@ -1 +0,0 @@
__author__ = 'hexa'

33
lib/alfred.py
View file

@ -1,33 +0,0 @@
import subprocess
import json
import os
class Alfred(object):
"""
Bindings for the alfred-json utility
"""
def __init__(self, unix_sockpath=None):
self.unix_sock = unix_sockpath
if unix_sockpath is not None and not os.path.exists(unix_sockpath):
raise RuntimeError('alfred: invalid unix socket path given')
def _fetch(self, data_type):
cmd = ['/usr/local/bin/alfred-json',
'-z',
'-f', 'json',
'-r', str(data_type)]
if self.unix_sock:
cmd.extend(['-s', self.unix_sock])
output = subprocess.check_output(cmd)
return json.loads(output.decode("utf-8")).values()
def nodeinfo(self):
return self._fetch(158)
def statistics(self):
return self._fetch(159)
def vis(self):
return self._fetch(160)

98
lib/batman.py
View file

@ -1,98 +0,0 @@
import subprocess
import json
import os
import re
class Batman(object):
"""
Bindings for B.A.T.M.A.N. Advanced
commandline interface "batctl"
"""
def __init__(self, mesh_interface='bat0', alfred_sockpath=None):
self.mesh_interface = mesh_interface
self.alfred_sock = alfred_sockpath
# ensure /usr/sbin and /usr/local/sbin are in PATH
env = os.environ
path = set(env['PATH'].split(':'))
path.add('/usr/sbin/')
path.add('/usr/local/sbin')
env['PATH'] = ':'.join(path)
self.environ = env
# compile regular expressions only once on startup
self.mac_addr_pattern = re.compile(r'(([a-z0-9]{2}:){5}[a-z0-9]{2})')
def vis_data(self):
return self.vis_data_batadv_vis()
@staticmethod
def vis_data_helper(lines):
vd_tmp = []
for line in lines:
try:
utf8_line = line.decode('utf-8')
vd_tmp.append(json.loads(utf8_line))
except UnicodeDecodeError:
pass
return vd_tmp
def vis_data_batadv_vis(self):
"""
Parse "batadv-vis -i <mesh_interface> -f json"
into an array of dictionaries.
"""
cmd = ['batadv-vis', '-i', self.mesh_interface, '-f', 'json']
if self.alfred_sock:
cmd.extend(['-u', self.alfred_sock])
output = subprocess.check_output(cmd, env=self.environ)
lines = output.splitlines()
return self.vis_data_helper(lines)
def gateway_list(self):
"""
Parse "batctl -m <mesh_interface> gwl -n"
into an array of dictionaries.
"""
cmd = ['batctl', '-m', self.mesh_interface, 'gwl', '-n']
if os.geteuid() > 0:
cmd.insert(0, 'sudo')
output = subprocess.check_output(cmd, env=self.environ)
output_utf8 = output.decode('utf-8')
rows = output_utf8.splitlines()
gateways = []
# local gateway
header = rows.pop(0)
mode, bandwidth = self.gateway_mode()
if mode == 'server':
local_gw_mac = self.mac_addr_pattern.search(header).group(0)
gateways.append(local_gw_mac)
# remote gateway(s)
for row in rows:
match = self.mac_addr_pattern.search(row)
if match:
gateways.append(match.group(1))
return gateways
def gateway_mode(self):
"""
Parse "batctl -m <mesh_interface> gw"
return: tuple mode, bandwidth, if mode != server then bandwidth is None
"""
cmd = ['batctl', '-m', self.mesh_interface, 'gw']
if os.geteuid() > 0:
cmd.insert(0, 'sudo')
output = subprocess.check_output(cmd, env=self.environ)
chunks = output.decode("utf-8").split()
return chunks[0], chunks[3] if 3 in chunks else None
if __name__ == "__main__":
bc = Batman()
vd = bc.vis_data()
gw = bc.gateway_list()

84
lib/graph.py
View file

@ -1,84 +0,0 @@
from functools import reduce
from itertools import chain
import networkx as nx
from lib.nodes import build_mac_table
def import_vis_data(graph, nodes, vis_data):
macs = build_mac_table(nodes)
nodes_a = map(lambda d: 2 * [d['primary']],
filter(lambda d: 'primary' in d, vis_data))
nodes_b = map(lambda d: [d['secondary'], d['of']],
filter(lambda d: 'secondary' in d, vis_data))
graph.add_nodes_from(map(lambda a, b:
(a, dict(primary=b, node_id=macs.get(b))),
*zip(*chain(nodes_a, nodes_b))))
edges = filter(lambda d: 'neighbor' in d, vis_data)
graph.add_edges_from(map(lambda d: (d['router'], d['neighbor'],
dict(tq=float(d['label']))), edges))
def mark_vpn(graph, vpn_macs):
components = map(frozenset, nx.weakly_connected_components(graph))
components = filter(vpn_macs.intersection, components)
nodes = reduce(lambda a, b: a | b, components, set())
for node in nodes:
for k, v in graph[node].items():
v['vpn'] = True
def to_multigraph(graph):
def f(a):
node = graph.node[a]
return node['primary'] if node else a
def map_node(node, data):
return (data['primary'],
dict(node_id=data['node_id'])) if data else (node, dict())
digraph = nx.MultiDiGraph()
digraph.add_nodes_from(map(map_node, *zip(*graph.nodes_iter(data=True))))
digraph.add_edges_from(map(lambda a, b, data: (f(a), f(b), data),
*zip(*graph.edges_iter(data=True))))
return digraph
def merge_nodes(graph):
def merge_edges(data):
tq = min(map(lambda d: d['tq'], data))
vpn = all(map(lambda d: d.get('vpn', False), data))
return dict(tq=tq, vpn=vpn)
multigraph = to_multigraph(graph)
digraph = nx.DiGraph()
digraph.add_nodes_from(multigraph.nodes_iter(data=True))
edges = chain.from_iterable([[(e, d, merge_edges(
multigraph[e][d].values()))
for d in multigraph[e]] for e in multigraph])
digraph.add_edges_from(edges)
return digraph
def to_undirected(graph):
multigraph = nx.MultiGraph()
multigraph.add_nodes_from(graph.nodes_iter(data=True))
multigraph.add_edges_from(graph.edges_iter(data=True))
def merge_edges(data):
tq = max(map(lambda d: d['tq'], data))
vpn = all(map(lambda d: d.get('vpn', False), data))
return dict(tq=tq, vpn=vpn, bidirect=len(data) == 2)
graph = nx.Graph()
graph.add_nodes_from(multigraph.nodes_iter(data=True))
edges = chain.from_iterable([[(e, d, merge_edges(
multigraph[e][d].values()))
for d in multigraph[e]] for e in multigraph])
graph.add_edges_from(edges)
return graph

27
lib/nodelist.py
View file

@ -1,27 +0,0 @@
def export_nodelist(now, nodedb):
nodelist = list()
for node_id, node in nodedb["nodes"].items():
node_out = dict()
node_out["id"] = node_id
node_out["name"] = node["nodeinfo"]["hostname"]
if "location" in node["nodeinfo"]:
node_out["position"] = {"lat": node["nodeinfo"]["location"]["latitude"],
"long": node["nodeinfo"]["location"]["longitude"]}
node_out["status"] = dict()
node_out["status"]["online"] = node["flags"]["online"]
if "firstseen" in node:
node_out["status"]["firstcontact"] = node["firstseen"]
if "lastseen" in node:
node_out["status"]["lastcontact"] = node["lastseen"]
if "clients" in node["statistics"]:
node_out["status"]["clients"] = node["statistics"]["clients"]
nodelist.append(node_out)
return {"version": "1.0.1", "nodes": nodelist, "updated_at": now.isoformat()}

202
lib/nodes.py
View file

@ -1,202 +0,0 @@
from collections import Counter, defaultdict
from datetime import datetime
from functools import reduce
def build_mac_table(nodes):
macs = dict()
for node_id, node in nodes.items():
try:
macs[node['network']['mac']] = node_id
except KeyError:
pass
try:
for mac in node['nodeinfo']['network']['mesh_interfaces']:
macs[mac] = node_id
except KeyError:
pass
try:
for mac in node['nodeinfo']['network']['mesh']['bat0']['interfaces']['wireless']:
macs[mac] = node_id
except KeyError:
pass
try:
for mac in node['nodeinfo']['network']['mesh']['bat0']['interfaces']['tunnel']:
macs[mac] = node_id
except KeyError:
pass
try:
for mac in node['nodeinfo']['network']['mesh']['bat-ffhh']['interfaces']['tunnel']:
macs[mac] = node_id
except KeyError:
pass
try:
for mac in node['nodeinfo']['network']['mesh']['bat0']['interfaces']['other']:
macs[mac] = node_id
except KeyError:
pass
return macs
def prune_nodes(nodes, now, days):
prune = []
for node_id, node in nodes.items():
if 'lastseen' not in node:
prune.append(node_id)
continue
lastseen = datetime.strptime(node['lastseen'], '%Y-%m-%dT%H:%M:%S')
delta = (now - lastseen).days
if delta >= days:
prune.append(node_id)
for node_id in prune:
del nodes[node_id]
def mark_online(node, now):
node['lastseen'] = now.isoformat()
node.setdefault('firstseen', now.isoformat())
node['flags']['online'] = True
def overrideFields(dest, src, fields):
for field in fields:
if field in src:
dest[field] = src[field]
else:
dest.pop(field, None)
def import_nodeinfo(nodes, nodeinfos, now, assume_online=False, statics=False):
for nodeinfo in filter(lambda d: 'node_id' in d, nodeinfos):
node = nodes.setdefault(nodeinfo['node_id'], {'flags': {'online': False, 'gateway': False}})
if statics:
node['nodeinfo'] = node.setdefault('nodeinfo', {})
overrideFields(node['nodeinfo'], nodeinfo, ['hostname', 'location', 'node_id'])
else:
node['nodeinfo'] = nodeinfo
if assume_online:
mark_online(node, now)
def reset_statistics(nodes):
for node in nodes.values():
node['statistics'] = {'clients': 0}
def import_statistics(nodes, stats):
def add(node, statistics, target, source, f=lambda d: d):
try:
node['statistics'][target] = f(reduce(dict.__getitem__,
source,
statistics))
except (KeyError, TypeError, ZeroDivisionError):
pass
macs = build_mac_table(nodes)
stats = filter(lambda d: 'node_id' in d, stats)
stats = filter(lambda d: d['node_id'] in nodes, stats)
for node, stats in map(lambda d: (nodes[d['node_id']], d), stats):
add(node, stats, 'clients', ['clients', 'total'])
add(node, stats, 'gateway', ['gateway'], lambda d: macs.get(d, d))
add(node, stats, 'uptime', ['uptime'])
add(node, stats, 'loadavg', ['loadavg'])
add(node, stats, 'memory_usage', ['memory'],
lambda d: 1 - d['free'] / d['total'])
add(node, stats, 'rootfs_usage', ['rootfs_usage'])
add(node, stats, 'traffic', ['traffic'])
def import_mesh_ifs_vis_data(nodes, vis_data):
macs = build_mac_table(nodes)
mesh_ifs = defaultdict(lambda: set())
for line in filter(lambda d: 'secondary' in d, vis_data):
primary = line['of']
mesh_ifs[primary].add(primary)
mesh_ifs[primary].add(line['secondary'])
def if_to_node(ifs):
a = filter(lambda d: d in macs, ifs)
a = map(lambda d: nodes[macs[d]], a)
try:
return next(a), ifs
except StopIteration:
return None
mesh_nodes = filter(lambda d: d, map(if_to_node, mesh_ifs.values()))
for v in mesh_nodes:
node = v[0]
ifs = set()
try:
ifs = ifs.union(set(node['nodeinfo']['network']['mesh_interfaces']))
except KeyError:
pass
try:
ifs = ifs.union(set(node['nodeinfo']['network']['mesh']['bat0']['interfaces']['wireless']))
except KeyError:
pass
try:
ifs = ifs.union(set(node['nodeinfo']['network']['mesh']['bat0']['interfaces']['tunnel']))
except KeyError:
pass
try:
ifs = ifs.union(set(node['nodeinfo']['network']['mesh']['bat-ffhh']['interfaces']['tunnel']))
except KeyError:
pass
try:
ifs = ifs.union(set(node['nodeinfo']['network']['mesh']['bat0']['interfaces']['other']))
except KeyError:
pass
node['nodeinfo']['network']['mesh_interfaces'] = list(ifs | v[1])
def import_vis_clientcount(nodes, vis_data):
macs = build_mac_table(nodes)
data = filter(lambda d: d.get('label', None) == 'TT', vis_data)
data = filter(lambda d: d['router'] in macs, data)
data = map(lambda d: macs[d['router']], data)
for node_id, clientcount in Counter(data).items():
nodes[node_id]['statistics'].setdefault('clients', clientcount)
def mark_gateways(nodes, gateways):
macs = build_mac_table(nodes)
gateways = filter(lambda d: d in macs, gateways)
for node in map(lambda d: nodes[macs[d]], gateways):
node['flags']['gateway'] = True
def mark_vis_data_online(nodes, vis_data, now):
macs = build_mac_table(nodes)
online = set()
for line in vis_data:
if 'primary' in line:
online.add(line['primary'])
elif 'secondary' in line:
online.add(line['secondary'])
elif 'gateway' in line:
# This matches clients' MACs.
# On pre-Gluon nodes the primary MAC will be one of it.
online.add(line['gateway'])
for mac in filter(lambda d: d in macs, online):
mark_online(nodes[macs[mac]], now)

54
lib/rrddb.py
View file

@ -1,54 +0,0 @@
#!/usr/bin/env python3
import time
import os
from lib.GlobalRRD import GlobalRRD
from lib.NodeRRD import NodeRRD
class RRD(object):
def __init__(self,
database_directory,
image_path,
display_time_global="7d",
display_time_node="1d"):
self.dbPath = database_directory
self.globalDb = GlobalRRD(self.dbPath)
self.imagePath = image_path
self.displayTimeGlobal = display_time_global
self.displayTimeNode = display_time_node
self.currentTimeInt = (int(time.time()) / 60) * 60
self.currentTime = str(self.currentTimeInt)
try:
os.stat(self.imagePath)
except OSError:
os.mkdir(self.imagePath)
def update_database(self, nodes):
online_nodes = dict(filter(
lambda d: d[1]['flags']['online'], nodes.items()))
client_count = sum(map(
lambda d: d['statistics']['clients'], online_nodes.values()))
self.globalDb.update(len(online_nodes), client_count)
for node_id, node in online_nodes.items():
rrd = NodeRRD(os.path.join(self.dbPath, node_id + '.rrd'), node)
rrd.update()
def update_images(self):
self.globalDb.graph(os.path.join(self.imagePath, "globalGraph.png"),
self.displayTimeGlobal)
nodedb_files = os.listdir(self.dbPath)
for file_name in nodedb_files:
if not os.path.isfile(os.path.join(self.dbPath, file_name)):
continue
node_name = os.path.basename(file_name).split('.')
if node_name[1] == 'rrd' and not node_name[0] == "nodes":
rrd = NodeRRD(os.path.join(self.dbPath, file_name))
rrd.graph(self.imagePath, self.displayTimeNode)

19
lib/validate.py
View file

@ -1,19 +0,0 @@
import json
def validate_nodeinfos(nodeinfos):
result = []
for nodeinfo in nodeinfos:
if validate_nodeinfo(nodeinfo):
result.append(nodeinfo)
return result
def validate_nodeinfo(nodeinfo):
if 'location' in nodeinfo:
if 'latitude' not in nodeinfo['location'] or 'longitude' not in nodeinfo['location']:
return False
return True

7
mkmap.sh
View file

@ -1,7 +0,0 @@
#!/bin/bash
FFMAPPATH='/opt/ffmap-backend/'
PEERS="/etc/fastd/ffhh-mesh-vpn/peers"
python2 $FFMAPPATH/generate_aliases.py $PEERS > $FFMAPPATH/aliases.json
#python3 $FFMAPPATH/backend.py -d /var/www/meshviewer/ --aliases $FFMAPPATH/aliases.json $FFMAPPATH/gateway.json -m bat0:/var/run/alfred.sock -p 30 --vpn de:ad:be:ff:01:01 --vpn de:ad:be:ff:05:05 --vpn de:ad:be:ff:05:06 --vpn de:ad:be:ff:03:03 --vpn de:ad:be:ff:22:22 --vpn de:ad:be:ff:22:23 --vpn de:ad:be:ff:88:88 --vpn de:ad:be:ff:88:89 --vpn de:ad:bf:ff:88:88 --vpn de:ad:bf:ff:22:22 --vpn de:ad:bf:ff:03:03 --vpn de:ad:bf:ff:05:05 --vpn de:ad:bf:ff:01:01 --vpn de:ad:be:fc:03:03 --vpn 00:16:3e:53:75:0d --vpn de:ad:be:fc:05:05 --vpn de:ad:be:fc:01:01 --vpn de:ad:be:ef:03:03 --vpn de:ad:be:ef:01:01 --vpn de:ad:be:ef:05:05 --vpn 00:16:3e:fb:9d:8d --vpn 00:16:3e:fb:9d:9d
python3 $FFMAPPATH/backend.py -d /var/www/meshviewer/ --aliases $FFMAPPATH/aliases.json $FFMAPPATH/gateway.json -m bat0:/var/run/alfred.sock -p 30 --vpn de:ad:be:ff:01:01 de:ad:be:ff:05:05 de:ad:be:ff:05:06 de:ad:be:ff:03:03 de:ad:be:ff:22:22 de:ad:be:ff:22:23 de:ad:be:ff:88:88 de:ad:be:ff:88:89 de:ad:bf:ff:88:88 de:ad:bf:ff:22:22 de:ad:bf:ff:03:03 de:ad:bf:ff:05:05 de:ad:bf:ff:01:01 de:ad:be:fc:03:03 00:16:3e:53:75:0d de:ad:be:fc:05:05 de:ad:be:fc:01:01 de:ad:be:ef:03:03 de:ad:be:ef:01:01 de:ad:be:ef:05:05 00:16:3e:fb:9d:8d 00:16:3e:fb:9d:9d

32
node_number.py
View file

@ -1,32 +0,0 @@
#!/usr/bin/env python
#Bibliotheken importieren
import time
import datetime
import json
import urllib2
#Datei oeffnen
Datei = urllib2.urlopen('https://map.hamburg.freifunk.net/nodes.json')
Datei_Sued = urllib2.urlopen('https://map.hamburg.freifunk.net/hhsued/mv1/nodes.json')
Text = Datei.read()
Knotenzahl = Text.count('"online": true')
Text = Datei_Sued.read()
Knotenzahl = Knotenzahl + Text.count('"online":true')
#Zeit holen
thetime = datetime.datetime.now().isoformat()
ffhh = None
#Freifunk API-Datei einladen und JSON lesen
with open('/var/www/meta/ffhh.json', 'r') as fp:
ffhh = json.load(fp)
#Attribute Zeitstempel und Knotenanzahl setzen
ffhh['state']['lastchange'] = thetime
ffhh['state']['nodes'] = Knotenzahl
#Freifunk API-Datein mit geaenderten werten schreiben
with open('/var/www/meta/ffhh.json', 'w') as fp:
json.dump(ffhh, fp, indent=2, separators=(',', ': '))

10
setup.py Normal file
View file

@ -0,0 +1,10 @@
#!/usr/bin/env python3
from distutils.core import setup
setup(name='FFmap',
version='0.1',
description='Freifunk map backend',
url='https://github.com/ffnord/ffmap-backend',
packages=['ffmap', 'ffmap.inputs', 'ffmap.outputs', 'ffmap.rrd'],
)