sibi-dst 0.3.40__py3-none-any.whl → 0.3.43__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- sibi_dst/df_helper/__init__.py +2 -0
- sibi_dst/df_helper/_artifact_updater_multi_wrapper.py +262 -0
- sibi_dst/df_helper/_df_helper.py +5 -2
- sibi_dst/df_helper/_parquet_artifact.py +8 -2
- sibi_dst/df_helper/_parquet_reader.py +5 -1
- sibi_dst/df_helper/backends/sqlalchemy/_db_connection.py +1 -0
- sibi_dst/osmnx_helper/__init__.py +2 -2
- sibi_dst/osmnx_helper/v1/basemaps/__init__.py +0 -0
- sibi_dst/osmnx_helper/{basemaps → v1/basemaps}/router_plotter.py +85 -30
- sibi_dst/osmnx_helper/v2/__init__.py +0 -0
- sibi_dst/osmnx_helper/v2/base_osm_map.py +153 -0
- sibi_dst/osmnx_helper/v2/basemaps/__init__.py +0 -0
- sibi_dst/osmnx_helper/v2/basemaps/utils.py +0 -0
- sibi_dst/utils/__init__.py +3 -0
- sibi_dst/utils/data_utils.py +66 -25
- sibi_dst/utils/data_wrapper.py +222 -285
- sibi_dst/utils/date_utils.py +118 -113
- sibi_dst/utils/df_utils.py +7 -0
- sibi_dst/utils/log_utils.py +57 -18
- sibi_dst/utils/parquet_saver.py +4 -2
- sibi_dst/utils/phone_formatter.py +127 -0
- sibi_dst/utils/storage_manager.py +14 -7
- sibi_dst-0.3.43.dist-info/METADATA +194 -0
- {sibi_dst-0.3.40.dist-info → sibi_dst-0.3.43.dist-info}/RECORD +29 -22
- sibi_dst-0.3.40.dist-info/METADATA +0 -62
- /sibi_dst/osmnx_helper/{basemaps → v1}/__init__.py +0 -0
- /sibi_dst/osmnx_helper/{base_osm_map.py → v1/base_osm_map.py} +0 -0
- /sibi_dst/osmnx_helper/{basemaps → v1/basemaps}/calendar_html.py +0 -0
- /sibi_dst/osmnx_helper/{utils.py → v1/utils.py} +0 -0
- {sibi_dst-0.3.40.dist-info → sibi_dst-0.3.43.dist-info}/WHEEL +0 -0
sibi_dst/df_helper/__init__.py
CHANGED
@@ -3,11 +3,13 @@ from __future__ import annotations
|
|
3
3
|
from ._df_helper import DfHelper
|
4
4
|
from ._parquet_artifact import ParquetArtifact
|
5
5
|
from ._parquet_reader import ParquetReader
|
6
|
+
from ._artifact_updater_multi_wrapper import ArtifactUpdaterMultiWrapper
|
6
7
|
#from .data_cleaner import DataCleaner
|
7
8
|
|
8
9
|
__all__ = [
|
9
10
|
'DfHelper',
|
10
11
|
'ParquetArtifact',
|
11
12
|
'ParquetReader',
|
13
|
+
'ArtifactUpdaterMultiWrapper',
|
12
14
|
#'DataCleaner'
|
13
15
|
]
|
@@ -0,0 +1,262 @@
|
|
1
|
+
import asyncio
|
2
|
+
import logging
|
3
|
+
import datetime
|
4
|
+
import psutil
|
5
|
+
import time
|
6
|
+
from functools import total_ordering
|
7
|
+
from collections import defaultdict
|
8
|
+
from contextlib import asynccontextmanager
|
9
|
+
import signal
|
10
|
+
from sibi_dst.utils import Logger
|
11
|
+
|
12
|
+
@total_ordering
|
13
|
+
class PrioritizedItem:
|
14
|
+
def __init__(self, priority, artifact):
|
15
|
+
self.priority = priority
|
16
|
+
self.artifact = artifact
|
17
|
+
|
18
|
+
def __lt__(self, other):
|
19
|
+
return self.priority < other.priority
|
20
|
+
|
21
|
+
def __eq__(self, other):
|
22
|
+
return self.priority == other.priority
|
23
|
+
|
24
|
+
class ArtifactUpdaterMultiWrapper:
|
25
|
+
def __init__(self, wrapped_classes=None, debug=False, **kwargs):
|
26
|
+
self.wrapped_classes = wrapped_classes or {}
|
27
|
+
self.debug = debug
|
28
|
+
self.logger = kwargs.setdefault('logger',Logger.default_logger(logger_name=self.__class__.__name__))
|
29
|
+
self.logger.set_level(logging.DEBUG if debug else logging.INFO)
|
30
|
+
|
31
|
+
today = datetime.datetime.today()
|
32
|
+
self.today_str = today.strftime('%Y-%m-%d')
|
33
|
+
self.current_year_starts_on_str = datetime.date(today.year, 1, 1).strftime('%Y-%m-%d')
|
34
|
+
self.parquet_start_date = kwargs.get('parquet_start_date', self.current_year_starts_on_str)
|
35
|
+
self.parquet_end_date = kwargs.get('parquet_end_date', self.today_str)
|
36
|
+
|
37
|
+
# track concurrency and locks
|
38
|
+
self.locks = {}
|
39
|
+
self.worker_heartbeat = defaultdict(float)
|
40
|
+
|
41
|
+
# graceful shutdown handling
|
42
|
+
loop = asyncio.get_event_loop()
|
43
|
+
self.register_signal_handlers(loop)
|
44
|
+
|
45
|
+
# dynamic scaling config
|
46
|
+
self.min_workers = kwargs.get('min_workers', 1)
|
47
|
+
self.max_workers = kwargs.get('max_workers', 8)
|
48
|
+
self.memory_per_worker_gb = kwargs.get('memory_per_worker_gb', 1) # default 2GB per worker
|
49
|
+
self.monitor_interval = kwargs.get('monitor_interval', 10) # default monitor interval in seconds
|
50
|
+
self.retry_attempts = kwargs.get('retry_attempts', 3)
|
51
|
+
self.update_timeout_seconds = kwargs.get('update_timeout_seconds', 600)
|
52
|
+
self.lock_acquire_timeout_seconds = kwargs.get('lock_acquire_timeout_seconds', 10)
|
53
|
+
|
54
|
+
def register_signal_handlers(self, loop):
|
55
|
+
for sig in (signal.SIGINT, signal.SIGTERM):
|
56
|
+
loop.add_signal_handler(sig, lambda: asyncio.create_task(self.shutdown()))
|
57
|
+
|
58
|
+
async def shutdown(self):
|
59
|
+
self.logger.info("Shutdown signal received. Cleaning up...")
|
60
|
+
tasks = [t for t in asyncio.all_tasks() if t is not asyncio.current_task()]
|
61
|
+
[task.cancel() for task in tasks]
|
62
|
+
await asyncio.gather(*tasks, return_exceptions=True)
|
63
|
+
self.logger.info("Shutdown complete.")
|
64
|
+
|
65
|
+
def get_lock_for_artifact(self, artifact):
|
66
|
+
artifact_key = artifact.__class__.__name__
|
67
|
+
if artifact_key not in self.locks:
|
68
|
+
self.locks[artifact_key] = asyncio.Lock()
|
69
|
+
return self.locks[artifact_key]
|
70
|
+
|
71
|
+
def get_artifacts(self, data_type):
|
72
|
+
if data_type not in self.wrapped_classes:
|
73
|
+
raise ValueError(f"Unsupported data type: {data_type}")
|
74
|
+
|
75
|
+
return [
|
76
|
+
artifact_class(
|
77
|
+
parquet_start_date=self.parquet_start_date,
|
78
|
+
parquet_end_date=self.parquet_end_date,
|
79
|
+
logger=self.logger,
|
80
|
+
debug=self.debug
|
81
|
+
)
|
82
|
+
for artifact_class in self.wrapped_classes[data_type]
|
83
|
+
]
|
84
|
+
|
85
|
+
def estimate_complexity(self, artifact):
|
86
|
+
try:
|
87
|
+
if hasattr(artifact, 'get_size_estimate'):
|
88
|
+
return artifact.get_size_estimate()
|
89
|
+
except Exception as e:
|
90
|
+
self.logger.warning(f"Failed to estimate complexity for {artifact}: {e}")
|
91
|
+
return 1 # default
|
92
|
+
|
93
|
+
def prioritize_tasks(self, artifacts):
|
94
|
+
queue = asyncio.PriorityQueue()
|
95
|
+
for artifact in artifacts:
|
96
|
+
complexity = self.estimate_complexity(artifact)
|
97
|
+
# we invert the complexity to ensure higher complexity -> higher priority
|
98
|
+
# if you want high complexity first, store negative complexity in the priority queue
|
99
|
+
# or if the smaller number means earlier processing, just keep as is
|
100
|
+
queue.put_nowait(PrioritizedItem(complexity, artifact))
|
101
|
+
return queue
|
102
|
+
|
103
|
+
async def resource_monitor(self, queue, workers):
|
104
|
+
"""Monitor system resources and adjust worker count while queue is not empty."""
|
105
|
+
while True:
|
106
|
+
# break if queue done
|
107
|
+
if queue.empty():
|
108
|
+
await asyncio.sleep(0.5)
|
109
|
+
if queue.empty():
|
110
|
+
break
|
111
|
+
|
112
|
+
try:
|
113
|
+
available_memory = psutil.virtual_memory().available
|
114
|
+
worker_memory_bytes = self.memory_per_worker_gb * (1024 ** 3)
|
115
|
+
max_workers_by_memory = available_memory // worker_memory_bytes
|
116
|
+
|
117
|
+
# figure out how many workers we can sustain
|
118
|
+
# note: we also cap by self.max_workers
|
119
|
+
optimal_workers = min(psutil.cpu_count(), max_workers_by_memory, self.max_workers)
|
120
|
+
|
121
|
+
# ensure at least self.min_workers is used
|
122
|
+
optimal_workers = max(self.min_workers, optimal_workers)
|
123
|
+
|
124
|
+
current_worker_count = len(workers)
|
125
|
+
|
126
|
+
if optimal_workers > current_worker_count:
|
127
|
+
# we can add more workers if queue is not empty
|
128
|
+
diff = optimal_workers - current_worker_count
|
129
|
+
for _ in range(diff):
|
130
|
+
worker_id = len(workers)
|
131
|
+
# create a new worker
|
132
|
+
w = asyncio.create_task(self.worker(queue, worker_id))
|
133
|
+
workers.append(w)
|
134
|
+
self.logger.info(f"Added worker {worker_id}. Total workers: {len(workers)}")
|
135
|
+
elif optimal_workers < current_worker_count:
|
136
|
+
# remove some workers
|
137
|
+
diff = current_worker_count - optimal_workers
|
138
|
+
for _ in range(diff):
|
139
|
+
w = workers.pop()
|
140
|
+
w.cancel()
|
141
|
+
self.logger.info(f"Removed a worker. Total workers: {len(workers)}")
|
142
|
+
|
143
|
+
await asyncio.sleep(self.monitor_interval)
|
144
|
+
|
145
|
+
except asyncio.CancelledError:
|
146
|
+
# monitor is being shut down
|
147
|
+
break
|
148
|
+
except Exception as e:
|
149
|
+
self.logger.error(f"Error in resource_monitor: {e}")
|
150
|
+
await asyncio.sleep(self.monitor_interval)
|
151
|
+
|
152
|
+
@asynccontextmanager
|
153
|
+
async def artifact_lock(self, artifact):
|
154
|
+
lock = self.get_lock_for_artifact(artifact)
|
155
|
+
try:
|
156
|
+
await asyncio.wait_for(lock.acquire(), timeout=self.lock_acquire_timeout_seconds)
|
157
|
+
yield
|
158
|
+
except asyncio.TimeoutError:
|
159
|
+
self.logger.error(f"Timeout acquiring lock for artifact: {artifact.__class__.__name__}")
|
160
|
+
yield # continue but no actual lock was acquired
|
161
|
+
finally:
|
162
|
+
if lock.locked():
|
163
|
+
lock.release()
|
164
|
+
|
165
|
+
async def async_update_artifact(self, artifact, **kwargs):
|
166
|
+
for attempt in range(self.retry_attempts):
|
167
|
+
try:
|
168
|
+
async with self.artifact_lock(artifact):
|
169
|
+
self.logger.info(
|
170
|
+
f"Updating artifact: {artifact.__class__.__name__}, Attempt: {attempt + 1} of {self.retry_attempts}" )
|
171
|
+
start_time = time.time()
|
172
|
+
await asyncio.wait_for(
|
173
|
+
asyncio.to_thread(artifact.update_parquet, **kwargs),
|
174
|
+
timeout=self.update_timeout_seconds
|
175
|
+
)
|
176
|
+
elapsed_time = time.time() - start_time
|
177
|
+
self.logger.info(
|
178
|
+
f"Successfully updated artifact: {artifact.__class__.__name__} in {elapsed_time:.2f}s." )
|
179
|
+
return
|
180
|
+
|
181
|
+
except asyncio.TimeoutError:
|
182
|
+
self.logger.error(f"Timeout updating artifact {artifact.__class__.__name__}, Attempt: {attempt + 1}")
|
183
|
+
except Exception as e:
|
184
|
+
self.logger.error(
|
185
|
+
f"Error updating artifact {artifact.__class__.__name__}, Attempt: {attempt + 1}: {e}" )
|
186
|
+
|
187
|
+
# exponential backoff
|
188
|
+
await asyncio.sleep(2 ** attempt)
|
189
|
+
|
190
|
+
self.logger.error(f"All retry attempts failed for artifact: {artifact.__class__.__name__}")
|
191
|
+
|
192
|
+
async def worker(self, queue, worker_id, **kwargs):
|
193
|
+
"""A worker that dynamically pulls tasks from the queue."""
|
194
|
+
while True:
|
195
|
+
try:
|
196
|
+
prioritized_item = await queue.get()
|
197
|
+
if prioritized_item is None:
|
198
|
+
break
|
199
|
+
artifact = prioritized_item.artifact
|
200
|
+
# heartbeat
|
201
|
+
self.worker_heartbeat[worker_id] = time.time()
|
202
|
+
|
203
|
+
await self.async_update_artifact(artifact, **kwargs)
|
204
|
+
|
205
|
+
except asyncio.CancelledError:
|
206
|
+
self.logger.info(f"Worker {worker_id} shutting down gracefully.")
|
207
|
+
break
|
208
|
+
except Exception as e:
|
209
|
+
self.logger.error(f"Error in worker {worker_id}: {e}")
|
210
|
+
finally:
|
211
|
+
queue.task_done()
|
212
|
+
|
213
|
+
async def process_tasks(self, queue, initial_workers, **kwargs):
|
214
|
+
"""Start a set of workers and a resource monitor to dynamically adjust them."""
|
215
|
+
# create initial workers
|
216
|
+
workers = []
|
217
|
+
for worker_id in range(initial_workers):
|
218
|
+
w = asyncio.create_task(self.worker(queue, worker_id, **kwargs))
|
219
|
+
workers.append(w)
|
220
|
+
|
221
|
+
# start resource monitor
|
222
|
+
monitor_task = asyncio.create_task(self.resource_monitor(queue, workers))
|
223
|
+
|
224
|
+
# wait until queue is done
|
225
|
+
try:
|
226
|
+
await queue.join()
|
227
|
+
finally:
|
228
|
+
# cancel resource monitor
|
229
|
+
monitor_task.cancel()
|
230
|
+
# all workers done
|
231
|
+
for w in workers:
|
232
|
+
w.cancel()
|
233
|
+
await asyncio.gather(*workers, return_exceptions=True)
|
234
|
+
|
235
|
+
async def update_data(self, data_type, **kwargs):
|
236
|
+
self.logger.info(f"Processing wrapper group: {data_type} with {kwargs}")
|
237
|
+
artifacts = self.get_artifacts(data_type)
|
238
|
+
queue = self.prioritize_tasks(artifacts)
|
239
|
+
|
240
|
+
# compute initial worker count (this can be low if memory is low initially)
|
241
|
+
initial_workers = self.calculate_initial_workers(len(artifacts))
|
242
|
+
self.logger.info(f"Initial worker count: {initial_workers} for {len(artifacts)} artifacts")
|
243
|
+
|
244
|
+
total_start_time = time.time()
|
245
|
+
await self.process_tasks(queue, initial_workers, **kwargs)
|
246
|
+
total_time = time.time() - total_start_time
|
247
|
+
self.logger.info(f"Total processing time: {total_time:.2f} seconds.")
|
248
|
+
|
249
|
+
def calculate_initial_workers(self, artifact_count: int) -> int:
|
250
|
+
"""Compute the initial number of workers before resource_monitor can adjust."""
|
251
|
+
self.logger.info("Calculating initial worker count...")
|
252
|
+
available_memory = psutil.virtual_memory().available
|
253
|
+
self.logger.info(f"Available memory: {available_memory / (1024 ** 3):.2f} GB")
|
254
|
+
worker_memory_bytes = self.memory_per_worker_gb * (1024 ** 3)
|
255
|
+
self.logger.info(f"Memory per worker: {worker_memory_bytes / (1024 ** 3):.2f} GB")
|
256
|
+
max_workers_by_memory = available_memory // worker_memory_bytes
|
257
|
+
self.logger.info(f"Max workers by memory: {max_workers_by_memory}")
|
258
|
+
# also consider CPU count and artifact_count
|
259
|
+
initial = min(psutil.cpu_count(), max_workers_by_memory, artifact_count, self.max_workers)
|
260
|
+
self.logger.info(f"Optimal workers: {initial} CPU: {psutil.cpu_count()} Max Workers: {self.max_workers}")
|
261
|
+
return max(self.min_workers, initial)
|
262
|
+
|
sibi_dst/df_helper/_df_helper.py
CHANGED
@@ -112,6 +112,7 @@ class DfHelper:
|
|
112
112
|
:return: None
|
113
113
|
"""
|
114
114
|
self.logger.debug(f"backend used: {self.backend}")
|
115
|
+
self.logger.debug(f"kwargs passed to backend plugins: {kwargs}")
|
115
116
|
self._backend_query = self.__get_config(QueryConfig, kwargs)
|
116
117
|
self._backend_params = self.__get_config(ParamsConfig, kwargs)
|
117
118
|
if self.backend == 'django_db':
|
@@ -124,8 +125,8 @@ class DfHelper:
|
|
124
125
|
elif self.backend == 'sqlalchemy':
|
125
126
|
self.backend_sqlalchemy = self.__get_config(SqlAlchemyConnectionConfig, kwargs)
|
126
127
|
|
127
|
-
|
128
|
-
def __get_config(model: [T], kwargs: Dict[str, Any]) -> Union[T]:
|
128
|
+
|
129
|
+
def __get_config(self, model: [T], kwargs: Dict[str, Any]) -> Union[T]:
|
129
130
|
"""
|
130
131
|
Initializes a Pydantic model with the keys it recognizes from the kwargs,
|
131
132
|
and removes those keys from the kwargs dictionary.
|
@@ -135,7 +136,9 @@ class DfHelper:
|
|
135
136
|
"""
|
136
137
|
# Extract keys that the model can accept
|
137
138
|
recognized_keys = set(model.__annotations__.keys())
|
139
|
+
self.logger.debug(f"recognized keys: {recognized_keys}")
|
138
140
|
model_kwargs = {k: kwargs.pop(k) for k in list(kwargs.keys()) if k in recognized_keys}
|
141
|
+
self.logger.debug(f"model_kwargs: {model_kwargs}")
|
139
142
|
return model(**model_kwargs)
|
140
143
|
|
141
144
|
def load_parallel(self, **options):
|
@@ -1,11 +1,11 @@
|
|
1
|
+
import logging
|
1
2
|
from typing import Optional, Any, Dict
|
2
3
|
|
3
4
|
import dask.dataframe as dd
|
4
5
|
import fsspec
|
5
6
|
|
6
7
|
from sibi_dst.df_helper import DfHelper
|
7
|
-
from sibi_dst.utils import DataWrapper
|
8
|
-
from sibi_dst.utils import DateUtils
|
8
|
+
from sibi_dst.utils import DataWrapper, DateUtils, Logger
|
9
9
|
|
10
10
|
|
11
11
|
class ParquetArtifact(DfHelper):
|
@@ -82,7 +82,12 @@ class ParquetArtifact(DfHelper):
|
|
82
82
|
**kwargs,
|
83
83
|
}
|
84
84
|
self.df: Optional[dd.DataFrame] = None
|
85
|
+
self.debug = self.config.setdefault('debug', False)
|
86
|
+
self.logger = self.config.setdefault('logger',Logger.default_logger(logger_name=f'parquet_artifact_{__class__.__name__}'))
|
87
|
+
self.logger.set_level(logging.DEBUG if self.debug else logging.INFO)
|
85
88
|
self.data_wrapper_class = data_wrapper_class
|
89
|
+
self.class_params = self.config.setdefault('class_params', None)
|
90
|
+
self.load_params = self.config.setdefault('load_params', None)
|
86
91
|
self.date_field = self.config.setdefault('date_field', None)
|
87
92
|
if self.date_field is None:
|
88
93
|
raise ValueError('date_field must be set')
|
@@ -150,6 +155,7 @@ class ParquetArtifact(DfHelper):
|
|
150
155
|
|
151
156
|
def _prepare_params(self, kwargs: Dict[str, Any]) -> Dict[str, Any]:
|
152
157
|
"""Prepare the parameters for generating the Parquet file."""
|
158
|
+
kwargs = {**self.config, **kwargs}
|
153
159
|
return {
|
154
160
|
'class_params': kwargs.pop('class_params', None),
|
155
161
|
'date_field': kwargs.pop('date_field', self.date_field),
|
@@ -1,10 +1,11 @@
|
|
1
|
+
import logging
|
1
2
|
from typing import Optional
|
2
3
|
|
3
4
|
import dask.dataframe as dd
|
4
5
|
import fsspec
|
5
6
|
|
6
7
|
from sibi_dst.df_helper import DfHelper
|
7
|
-
|
8
|
+
from sibi_dst.utils import Logger
|
8
9
|
|
9
10
|
class ParquetReader(DfHelper):
|
10
11
|
"""
|
@@ -53,6 +54,9 @@ class ParquetReader(DfHelper):
|
|
53
54
|
**kwargs,
|
54
55
|
}
|
55
56
|
self.df: Optional[dd.DataFrame] = None
|
57
|
+
self.debug = self.config.setdefault('debug', False)
|
58
|
+
self.logger = self.config.setdefault('logger', Logger.default_logger(logger_name=self.__class__.__name__))
|
59
|
+
self.logger.set_level(logging.DEBUG if self.debug else logging.INFO)
|
56
60
|
self.parquet_storage_path = self.config.setdefault('parquet_storage_path', None)
|
57
61
|
if self.parquet_storage_path is None:
|
58
62
|
raise ValueError('parquet_storage_path must be set')
|
File without changes
|
@@ -1,20 +1,21 @@
|
|
1
1
|
from __future__ import annotations
|
2
|
-
from sibi_dst.osmnx_helper.utils import get_distance_between_points, add_arrows
|
2
|
+
from sibi_dst.osmnx_helper.v1.utils import get_distance_between_points, add_arrows
|
3
3
|
from collections import defaultdict
|
4
4
|
import folium
|
5
5
|
from folium.plugins import AntPath
|
6
6
|
import networkx as nx
|
7
7
|
|
8
8
|
from sibi_dst.osmnx_helper import BaseOsmMap
|
9
|
-
from sibi_dst.osmnx_helper.basemaps.calendar_html import calendar_html
|
9
|
+
from sibi_dst.osmnx_helper.v1.basemaps.calendar_html import calendar_html
|
10
10
|
|
11
11
|
class RoutePlotter(BaseOsmMap):
|
12
12
|
def __init__(self, osmnx_graph, df, **kwargs):
|
13
13
|
self.action_field = kwargs.pop('action_field', '')
|
14
14
|
self.action_groups = kwargs.pop('action_groups', {})
|
15
15
|
self.action_styles = kwargs.pop('action_styles', {})
|
16
|
-
self.use_ant_path = kwargs.pop('use_ant_path',
|
17
|
-
self.show_calendar = kwargs.pop('show_calendar',
|
16
|
+
self.use_ant_path = kwargs.pop('use_ant_path', False)
|
17
|
+
self.show_calendar = kwargs.pop('show_calendar', False)
|
18
|
+
self.show_order_markers = kwargs.pop('show_order_markers', False)
|
18
19
|
self.show_map_title = kwargs.pop('show_map_title', True)
|
19
20
|
self.sort_keys = kwargs.pop('sort_keys', None)
|
20
21
|
self.main_route_layer = folium.FeatureGroup(name="Main Route")
|
@@ -24,6 +25,8 @@ class RoutePlotter(BaseOsmMap):
|
|
24
25
|
self.actions = []
|
25
26
|
self.action_group_counts = {action_group: 0 for action_group in self.action_groups}
|
26
27
|
self.marker_count = 1
|
28
|
+
# Add a snapping threshold (in meters) to avoid drawing nodes/markers that are too close.
|
29
|
+
self.snap_distance = kwargs.pop('snap_distance', 30)
|
27
30
|
kwargs.update({'calc_nearest_nodes': True})
|
28
31
|
kwargs['dt_field'] = 'date_time'
|
29
32
|
super().__init__(osmnx_graph, df, **kwargs)
|
@@ -36,6 +39,8 @@ class RoutePlotter(BaseOsmMap):
|
|
36
39
|
self._calculate_routes()
|
37
40
|
self._plot_routes()
|
38
41
|
self._add_markers()
|
42
|
+
if self.show_order_markers:
|
43
|
+
self._add_order_markers()
|
39
44
|
self.main_route_layer.add_to(self.osm_map)
|
40
45
|
if self.show_calendar:
|
41
46
|
self._add_calendar()
|
@@ -57,8 +62,8 @@ class RoutePlotter(BaseOsmMap):
|
|
57
62
|
self.route_polylines.append((polyline, color))
|
58
63
|
for action_group, action_markers in markers.items():
|
59
64
|
self.markers[action_group].extend(action_markers)
|
60
|
-
self.action_group_counts[action_group] +=
|
61
|
-
self.marker_count +=
|
65
|
+
self.action_group_counts[action_group] += len(action_markers)
|
66
|
+
self.marker_count += len(action_markers)
|
62
67
|
if self.verbose:
|
63
68
|
print("Route and marker calculation complete.")
|
64
69
|
|
@@ -70,7 +75,7 @@ class RoutePlotter(BaseOsmMap):
|
|
70
75
|
|
71
76
|
def _calculate_route(self, i):
|
72
77
|
if self.verbose:
|
73
|
-
print(f"Calculating for item:{i}")
|
78
|
+
print(f"Calculating for item: {i}")
|
74
79
|
orig = self.nearest_nodes[i]
|
75
80
|
dest = self.nearest_nodes[i + 1]
|
76
81
|
try:
|
@@ -81,17 +86,31 @@ class RoutePlotter(BaseOsmMap):
|
|
81
86
|
lats, lons = zip(*[(self.G.nodes[node]['y'] + offset, self.G.nodes[node]['x']) for node in route])
|
82
87
|
color = 'blue' if i < self.max_distance_index else 'red'
|
83
88
|
polyline = list(zip(lats, lons))
|
89
|
+
# Apply node snapping to the polyline to remove points that are too close.
|
90
|
+
polyline = self._snap_polyline(polyline)
|
84
91
|
markers = self._calculate_markers(i)
|
85
92
|
return polyline, color, markers
|
86
93
|
except nx.NetworkXNoPath:
|
87
94
|
if self.verbose:
|
88
|
-
print(f"Item:{i}-No path found for {orig} to {dest}")
|
95
|
+
print(f"Item: {i} - No path found for {orig} to {dest}")
|
89
96
|
return None, None, {}
|
90
97
|
except nx.NodeNotFound:
|
91
98
|
if self.verbose:
|
92
|
-
print(f"Item:{i}-No path found for {orig} to {dest}")
|
99
|
+
print(f"Item: {i} - No path found for {orig} to {dest}")
|
93
100
|
return None, None, {}
|
94
101
|
|
102
|
+
def _snap_polyline(self, polyline: list[tuple[float, float]]) -> list[tuple[float, float]]:
|
103
|
+
"""
|
104
|
+
Returns a filtered polyline where consecutive points closer than snap_distance are removed.
|
105
|
+
"""
|
106
|
+
if not polyline:
|
107
|
+
return polyline
|
108
|
+
snapped_polyline = [polyline[0]]
|
109
|
+
for point in polyline[1:]:
|
110
|
+
if get_distance_between_points(snapped_polyline[-1], point, 'm') >= self.snap_distance:
|
111
|
+
snapped_polyline.append(point)
|
112
|
+
return snapped_polyline
|
113
|
+
|
95
114
|
def _calculate_markers(self, i):
|
96
115
|
# Calculate markers for action groups
|
97
116
|
markers = defaultdict(list)
|
@@ -110,24 +129,26 @@ class RoutePlotter(BaseOsmMap):
|
|
110
129
|
def _plot_routes(self):
|
111
130
|
if self.verbose:
|
112
131
|
print("Plotting routes and markers...")
|
113
|
-
# self.action_group_counts = {action_group: 0 for action_group in self.feature_groups.keys()}
|
114
132
|
for polyline, color in self.route_polylines:
|
115
133
|
if self.use_ant_path:
|
116
134
|
AntPath(
|
117
135
|
locations=polyline,
|
118
136
|
color=color,
|
119
|
-
weight=3,
|
120
|
-
opacity=10,
|
121
|
-
#
|
122
|
-
delay=1000, # Slower animation to reduce flickering
|
123
|
-
# dash_array=[20, 30] # Adjust dash pattern if needed
|
137
|
+
weight=3, # Increased line thickness
|
138
|
+
opacity=10, # Increased opacity
|
139
|
+
delay=1000, # Slower animation to reduce flickering
|
124
140
|
).add_to(self.main_route_layer)
|
125
141
|
else:
|
126
142
|
folium.PolyLine(locations=polyline, color=color).add_to(self.main_route_layer)
|
127
143
|
self.osm_map = add_arrows(self.osm_map, polyline, color, n_arrows=3)
|
128
|
-
# Plot markers for action groups
|
144
|
+
# Plot markers for action groups with snapping to avoid drawing too many nearby markers.
|
129
145
|
for action_group, action_markers in self.markers.items():
|
146
|
+
seen_positions = []
|
130
147
|
for location, tooltip, popup_data, action_style in action_markers:
|
148
|
+
# Skip marker if a nearby marker (within snap_distance) has already been added.
|
149
|
+
if any(get_distance_between_points(location, pos, 'm') < self.snap_distance for pos in seen_positions):
|
150
|
+
continue
|
151
|
+
seen_positions.append(location)
|
131
152
|
folium.Marker(
|
132
153
|
location=location,
|
133
154
|
popup=folium.Popup(popup_data, max_width=600),
|
@@ -145,11 +166,14 @@ class RoutePlotter(BaseOsmMap):
|
|
145
166
|
def _add_markers(self):
|
146
167
|
if self.verbose:
|
147
168
|
print("Adding markers...")
|
148
|
-
# Add start marker
|
169
|
+
# Add a start marker
|
149
170
|
start_popup = folium.Popup(f"Start of route at {self.dt[0]}", max_width=300)
|
150
|
-
folium.Marker(
|
151
|
-
|
152
|
-
|
171
|
+
folium.Marker(
|
172
|
+
location=self.gps_points[0],
|
173
|
+
popup=start_popup,
|
174
|
+
icon=folium.Icon(icon='flag-checkered', prefix='fa')
|
175
|
+
).add_to(self.osm_map)
|
176
|
+
# Add an end marker with total distance info
|
153
177
|
folium.Marker(
|
154
178
|
self.gps_points[-1],
|
155
179
|
popup=f"End of Route at {self.dt[self.max_time_index]}. Total Distance Travelled: {self.total_distance / 1000:.2f} km",
|
@@ -165,15 +189,15 @@ class RoutePlotter(BaseOsmMap):
|
|
165
189
|
def _add_map_title(self):
|
166
190
|
if self.map_html_title and self.show_map_title:
|
167
191
|
title_html = f'''
|
168
|
-
<div style="position: fixed;
|
169
|
-
top: 10px;
|
170
|
-
left: 50%;
|
192
|
+
<div style="position: fixed;
|
193
|
+
top: 10px;
|
194
|
+
left: 50%;
|
171
195
|
transform: translate(-50%, 0%);
|
172
|
-
z-index: 9999;
|
173
|
-
font-size: 24px;
|
174
|
-
font-weight: bold;
|
175
|
-
background-color: white;
|
176
|
-
padding: 10px;
|
196
|
+
z-index: 9999;
|
197
|
+
font-size: 24px;
|
198
|
+
font-weight: bold;
|
199
|
+
background-color: white;
|
200
|
+
padding: 10px;
|
177
201
|
border: 2px solid black;
|
178
202
|
border-radius: 5px;">
|
179
203
|
{self.map_html_title}
|
@@ -181,6 +205,37 @@ class RoutePlotter(BaseOsmMap):
|
|
181
205
|
'''
|
182
206
|
self.osm_map.get_root().html.add_child(folium.Element(title_html))
|
183
207
|
|
208
|
+
def _add_order_markers(self):
|
209
|
+
"""Adds numbered markers to indicate the visit order."""
|
210
|
+
order_feature_group = folium.FeatureGroup(name="Visit Order")
|
211
|
+
for idx, location in enumerate(self.gps_points):
|
212
|
+
# Create a DivIcon with the number (starting at 1)
|
213
|
+
icon = folium.DivIcon(
|
214
|
+
icon_size=(24, 24),
|
215
|
+
icon_anchor=(12, 12),
|
216
|
+
html=f'''
|
217
|
+
<div style="
|
218
|
+
font-size: 12pt;
|
219
|
+
color: black;
|
220
|
+
background-color: white;
|
221
|
+
border: 1px solid black;
|
222
|
+
border-radius: 50%;
|
223
|
+
width: 24px;
|
224
|
+
height: 24px;
|
225
|
+
text-align: center;
|
226
|
+
line-height: 24px;">
|
227
|
+
{idx + 1}
|
228
|
+
</div>
|
229
|
+
'''
|
230
|
+
)
|
231
|
+
folium.Marker(
|
232
|
+
location=location,
|
233
|
+
icon=icon,
|
234
|
+
tooltip=f"GPS Set No. {idx + 1}: {self.dt[idx]}"
|
235
|
+
).add_to(order_feature_group)
|
236
|
+
|
237
|
+
order_feature_group.add_to(self.osm_map)
|
238
|
+
|
184
239
|
def _get_data(self, index):
|
185
|
-
#
|
186
|
-
...
|
240
|
+
# Implement in subclass to populate popups
|
241
|
+
...
|
File without changes
|