1
|
|
|
""" |
2
|
|
|
Methods for interacting with information Elodie caches about stored media. |
3
|
|
|
""" |
4
|
|
|
from builtins import map |
5
|
|
|
from builtins import object |
6
|
|
|
|
7
|
|
|
import hashlib |
8
|
|
|
import json |
9
|
|
|
import os |
10
|
|
|
import sys |
11
|
|
|
|
12
|
|
|
from math import radians, cos, sqrt |
13
|
|
|
from shutil import copyfile |
14
|
|
|
from time import strftime |
15
|
|
|
|
16
|
|
|
from elodie import constants |
17
|
|
|
|
18
|
|
|
|
19
|
|
|
class Db(object): |
20
|
|
|
|
21
|
|
|
"""A class for interacting with the JSON files created by Elodie.""" |
22
|
|
|
|
23
|
|
|
def __init__(self): |
24
|
|
|
# verify that the application directory (~/.elodie) exists, |
25
|
|
|
# else create it |
26
|
|
|
if not os.path.exists(constants.application_directory): |
27
|
|
|
os.makedirs(constants.application_directory) |
28
|
|
|
|
29
|
|
|
# If the hash db doesn't exist we create it. |
30
|
|
|
# Otherwise we only open for reading |
31
|
|
|
if not os.path.isfile(constants.hash_db): |
32
|
|
|
with open(constants.hash_db, 'a'): |
33
|
|
|
os.utime(constants.hash_db, None) |
34
|
|
|
|
35
|
|
|
self.hash_db = {} |
36
|
|
|
|
37
|
|
|
# We know from above that this file exists so we open it |
38
|
|
|
# for reading only. |
39
|
|
|
with open(constants.hash_db, 'r') as f: |
40
|
|
|
try: |
41
|
|
|
self.hash_db = json.load(f) |
42
|
|
|
except ValueError: |
43
|
|
|
pass |
44
|
|
|
|
45
|
|
|
# If the location db doesn't exist we create it. |
46
|
|
|
# Otherwise we only open for reading |
47
|
|
|
if not os.path.isfile(constants.location_db): |
48
|
|
|
with open(constants.location_db, 'a'): |
49
|
|
|
os.utime(constants.location_db, None) |
50
|
|
|
|
51
|
|
|
self.location_db = [] |
52
|
|
|
|
53
|
|
|
# We know from above that this file exists so we open it |
54
|
|
|
# for reading only. |
55
|
|
|
with open(constants.location_db, 'r') as f: |
56
|
|
|
try: |
57
|
|
|
self.location_db = json.load(f) |
58
|
|
|
except ValueError: |
59
|
|
|
pass |
60
|
|
|
|
61
|
|
|
def add_hash(self, key, value, write=False): |
62
|
|
|
"""Add a hash to the hash db. |
63
|
|
|
|
64
|
|
|
:param str key: |
65
|
|
|
:param str value: |
66
|
|
|
:param bool write: If true, write the hash db to disk. |
67
|
|
|
""" |
68
|
|
|
self.hash_db[key] = value |
69
|
|
|
if(write is True): |
70
|
|
|
self.update_hash_db() |
71
|
|
|
|
72
|
|
|
# Location database |
73
|
|
|
# Currently quite simple just a list of long/lat pairs with a name |
74
|
|
|
# If it gets many entries a lookup might take too long and a better |
75
|
|
|
# structure might be needed. Some speed up ideas: |
76
|
|
|
# - Sort it and inter-half method can be used |
77
|
|
|
# - Use integer part of long or lat as key to get a lower search list |
78
|
|
|
# - Cache a small number of lookups, photos are likely to be taken in |
79
|
|
|
# clusters around a spot during import. |
80
|
|
|
def add_location(self, latitude, longitude, place, write=False): |
81
|
|
|
"""Add a location to the database. |
82
|
|
|
|
83
|
|
|
:param float latitude: Latitude of the location. |
84
|
|
|
:param float longitude: Longitude of the location. |
85
|
|
|
:param str place: Name for the location. |
86
|
|
|
:param bool write: If true, write the location db to disk. |
87
|
|
|
""" |
88
|
|
|
data = {} |
89
|
|
|
data['lat'] = latitude |
90
|
|
|
data['long'] = longitude |
91
|
|
|
data['name'] = place |
92
|
|
|
self.location_db.append(data) |
93
|
|
|
if(write is True): |
94
|
|
|
self.update_location_db() |
95
|
|
|
|
96
|
|
|
def backup_hash_db(self): |
97
|
|
|
"""Backs up the hash db.""" |
98
|
|
|
if os.path.isfile(constants.hash_db): |
99
|
|
|
mask = strftime('%Y-%m-%d_%H-%M-%S') |
100
|
|
|
backup_file_name = '%s-%s' % (constants.hash_db, mask) |
101
|
|
|
copyfile(constants.hash_db, backup_file_name) |
102
|
|
|
return backup_file_name |
103
|
|
|
|
104
|
|
|
def check_hash(self, key): |
105
|
|
|
"""Check whether a hash is present for the given key. |
106
|
|
|
|
107
|
|
|
:param str key: |
108
|
|
|
:returns: bool |
109
|
|
|
""" |
110
|
|
|
return key in self.hash_db |
111
|
|
|
|
112
|
|
View Code Duplication |
def checksum(self, file_path, blocksize=65536): |
|
|
|
|
113
|
|
|
"""Create a hash value for the given file. |
114
|
|
|
|
115
|
|
|
See http://stackoverflow.com/a/3431835/1318758. |
116
|
|
|
|
117
|
|
|
:param str file_path: Path to the file to create a hash for. |
118
|
|
|
:param int blocksize: Read blocks of this size from the file when |
119
|
|
|
creating the hash. |
120
|
|
|
:returns: str or None |
121
|
|
|
""" |
122
|
|
|
hasher = hashlib.sha256() |
123
|
|
|
with open(file_path, 'rb') as f: |
124
|
|
|
buf = f.read(blocksize) |
125
|
|
|
|
126
|
|
|
while len(buf) > 0: |
127
|
|
|
hasher.update(buf) |
128
|
|
|
buf = f.read(blocksize) |
129
|
|
|
return hasher.hexdigest() |
130
|
|
|
return None |
131
|
|
|
|
132
|
|
|
def get_hash(self, key): |
133
|
|
|
"""Get the hash value for a given key. |
134
|
|
|
|
135
|
|
|
:param str key: |
136
|
|
|
:returns: str or None |
137
|
|
|
""" |
138
|
|
|
if(self.check_hash(key) is True): |
139
|
|
|
return self.hash_db[key] |
140
|
|
|
return None |
141
|
|
|
|
142
|
|
|
def get_location_name(self, latitude, longitude, threshold_m): |
143
|
|
|
"""Find a name for a location in the database. |
144
|
|
|
|
145
|
|
|
:param float latitude: Latitude of the location. |
146
|
|
|
:param float longitude: Longitude of the location. |
147
|
|
|
:param int threshold_m: Location in the database must be this close to |
148
|
|
|
the given latitude and longitude. |
149
|
|
|
:returns: str, or None if a matching location couldn't be found. |
150
|
|
|
""" |
151
|
|
|
last_d = sys.maxsize |
152
|
|
|
name = None |
153
|
|
|
for data in self.location_db: |
154
|
|
|
# As threshold is quite small use simple math |
155
|
|
|
# From http://stackoverflow.com/questions/15736995/how-can-i-quickly-estimate-the-distance-between-two-latitude-longitude-points # noqa |
156
|
|
|
# convert decimal degrees to radians |
157
|
|
|
|
158
|
|
|
lon1, lat1, lon2, lat2 = list(map( |
159
|
|
|
radians, |
160
|
|
|
[longitude, latitude, data['long'], data['lat']] |
161
|
|
|
)) |
162
|
|
|
|
163
|
|
|
r = 6371000 # radius of the earth in m |
164
|
|
|
x = (lon2 - lon1) * cos(0.5 * (lat2 + lat1)) |
165
|
|
|
y = lat2 - lat1 |
166
|
|
|
d = r * sqrt(x * x + y * y) |
167
|
|
|
# Use if closer then threshold_km reuse lookup |
168
|
|
|
if(d <= threshold_m and d < last_d): |
169
|
|
|
name = data['name'] |
170
|
|
|
last_d = d |
171
|
|
|
|
172
|
|
|
return name |
173
|
|
|
|
174
|
|
|
def get_location_coordinates(self, name): |
175
|
|
|
"""Get the latitude and longitude for a location. |
176
|
|
|
|
177
|
|
|
:param str name: Name of the location. |
178
|
|
|
:returns: tuple(float), or None if the location wasn't in the database. |
179
|
|
|
""" |
180
|
|
|
for data in self.location_db: |
181
|
|
|
if data['name'] == name: |
182
|
|
|
return (data['lat'], data['long']) |
183
|
|
|
|
184
|
|
|
return None |
185
|
|
|
|
186
|
|
|
def all(self): |
187
|
|
|
"""Generator to get all entries from self.hash_db |
188
|
|
|
|
189
|
|
|
:returns tuple(string) |
190
|
|
|
""" |
191
|
|
|
for checksum, path in self.hash_db.items(): |
192
|
|
|
yield (checksum, path) |
193
|
|
|
|
194
|
|
|
def reset_hash_db(self): |
195
|
|
|
self.hash_db = {} |
196
|
|
|
|
197
|
|
|
def update_hash_db(self): |
198
|
|
|
"""Write the hash db to disk.""" |
199
|
|
|
with open(constants.hash_db, 'w') as f: |
200
|
|
|
json.dump(self.hash_db, f) |
201
|
|
|
|
202
|
|
|
def update_location_db(self): |
203
|
|
|
"""Write the location db to disk.""" |
204
|
|
|
with open(constants.location_db, 'w') as f: |
205
|
|
|
json.dump(self.location_db, f) |
206
|
|
|
|