Merge pull request #822 from cal-itp/update-packages

update packages

_shared_utils/requirements.txt

@@ -1 +1,2 @@
 -e .
+gcsfs

_shared_utils/setup.py

@@ -3,7 +3,7 @@
 setup(
     name="shared_utils",
     packages=find_packages(),
-    version="2.1.0",
+    version="2.2.0",
     description="Shared utility functions for data analyses",
     author="Cal-ITP",
     license="Apache",

_shared_utils/shared_utils/rt_utils.py

@@ -15,14 +15,11 @@
 import numpy as np
 import pandas as pd
 import shapely
-from calitp_data.storage import get_fs
 from calitp_data_analysis.tables import tbls
 from numba import jit
 from shared_utils import geography_utils, gtfs_utils_v2, rt_dates, utils
 from siuba import *
 
-fs = get_fs()
-
 # set system time
 os.environ["TZ"] = "America/Los_Angeles"
 time.tzset()

_shared_utils/shared_utils/styleguide.py

@@ -25,7 +25,6 @@
 
 """
 import altair as alt
-from plotnine import *
 from shared_utils import calitp_color_palette as cp
 
 # --------------------------------------------------------------#
@@ -60,10 +59,6 @@
 </style>
 """
 
-# --------------------------------------------------------------#
-# Altair
-# --------------------------------------------------------------#
-
 
 def calitp_theme(
     font: str = font,
@@ -254,34 +249,3 @@ def preset_chart_config(chart: alt.Chart) -> alt.Chart:
     )
 
     return chart
-
-
-# --------------------------------------------------------------#
-# Plotnine
-# --------------------------------------------------------------#
-def preset_plotnine_config(chart):
-    chart = (
-        chart
-        + theme_538()
-        + theme(
-            plot_background=element_rect(fill=backgroundColor, color=backgroundColor),
-            panel_background=element_rect(fill=backgroundColor, color=backgroundColor),
-            panel_grid_major_y=element_line(color=axisColor, linetype="solid", size=1),
-            panel_grid_major_x=element_blank(),
-            figure_size=(7.0, 4.4),
-            title=element_text(weight="bold", size=font_size, family=font, color=blackTitle),
-            axis_title=element_text(family=labelFont, size=12, color=guideTitleColor),
-            axis_text=element_text(family=labelFont, size=10, color=guideLabelColor, margin={"r": 4}),
-            axis_title_x=element_text(margin={"t": 10}),
-            axis_title_y=element_text(margin={"r": 10}),
-            legend_title=element_text(font=labelFont, size=14, color=blackTitle, margin={"b": 10}),
-            legend_text=element_text(
-                font=labelFont,
-                size=11,
-                color=blackTitle,
-                margin={"t": 5, "b": 5, "r": 5, "l": 5},
-            ),
-        )
-    )
-
-    return chart

_shared_utils/shared_utils/utils.py

@@ -9,11 +9,11 @@
 
 import dask_geopandas as dg
 import fsspec
+import gcsfs
 import geopandas as gpd
 import requests
-from calitp_data.storage import get_fs
 
-fs = get_fs()
+fs = gcsfs.GCSFileSystem()
 
 
 def sanitize_file_path(file_name: str) -> str:

rt_segment_speeds/segment_speed_utils/array_utils.py

@@ -1,8 +1,8 @@
+"""
+Functions for working with numpy arrays.
+"""
 import numpy as np
 
-#from numba import jit
-
-#@jit(parallel=True)
 def get_index(array: np.ndarray, item) -> int:
     """
     Find the index for a certain value in an array.
@@ -13,12 +13,12 @@ def get_index(array: np.ndarray, item) -> int:
             return idx[0]
 
 
-#@jit(parallel=True)
 def subset_array_by_indices(
     array: np.ndarray, 
     start_end_tuple: tuple
 ):
     """
+    Subset an array using index positions.
     """
     lower_idx = start_end_tuple[0]
     upper_idx = start_end_tuple[-1] + 1

rt_segment_speeds/segment_speed_utils/gtfs_schedule_wrangling.py

@@ -70,7 +70,7 @@ def get_trips_with_geom(
     shapes = helpers.import_scheduled_shapes(
         analysis_date, 
         columns = ["shape_array_key", "geometry"],
-        get_pandas = False,
+        get_pandas = True,
     )
 
     trips = helpers.import_scheduled_trips(

rt_segment_speeds/segment_speed_utils/sched_rt_utils.py

@@ -1,74 +1,20 @@
 """
 Functions for bridging schedule and RT data.
-From RT data, gtfs_dataset_key is used.
-From schedule data, feed_key is used.
 
-These functions start with schedule data and add the RT gtfs_dataset_key.
+RT and schedule trips are joined using trip_instance_key.
+https://github.com/cal-itp/data-infra/pull/2489
+
 """
 import dask_geopandas as dg
 import dask.dataframe as dd
 import geopandas as gpd
 import pandas as pd
 
-from typing import List, Literal
-
-from shared_utils import schedule_rt_utils, rt_utils
+from shared_utils import rt_utils
 from segment_speed_utils import helpers
 from segment_speed_utils.project_vars import COMPILED_CACHED_VIEWS, PROJECT_CRS
 
 
-def crosswalk_scheduled_trip_grouping_with_rt_key(
-    analysis_date: str, 
-    keep_trip_cols: list = ["feed_key", "trip_id"],
-    feed_types: List[Literal["vehicle_positions", 
-                             "trip_updates", 
-                             "service_alerts"]] = ["vehicle_positions"],
-    **kwargs
-) -> pd.DataFrame:
-    """
-    Filter scheduled trips to a certain grouping 
-    (with route_id, direction_id or shape_array_key), 
-    and merge in gtfs_dataset_key that comes from fct_rt_feeds.
-    
-    This is our crosswalk that we can stick in the middle of vp or segments
-    and that allows us to get feed_key and gtfs_dataset_key
-    """
-    trips = helpers.import_scheduled_trips(
-        analysis_date, 
-        columns = keep_trip_cols,
-        **kwargs
-    )
-
-    # Get the schedule feed_key and RT gtfs_dataset_key and add it to crosswalk
-    fct_rt_feeds = (schedule_rt_utils.get_rt_schedule_feeds_crosswalk(
-            analysis_date, 
-            keep_cols = ["gtfs_dataset_key", "schedule_feed_key", "feed_type"], 
-            get_df = True,
-            custom_filtering = {"feed_type": feed_types}
-        ).rename(columns = {"schedule_feed_key": "feed_key"})
-        .drop(columns = "feed_type")
-    )
-
-    # Merge trips with fct_rt_feeds to get gtfs_dataset_key
-    if isinstance(trips, dd.DataFrame):
-        trips_with_rt_key = dd.merge(
-            trips,
-            fct_rt_feeds,
-            on = "feed_key",
-            how = "inner"
-        )
-
-    else:
-        trips_with_rt_key = pd.merge(
-            trips,
-            fct_rt_feeds,
-            on = "feed_key",
-            how = "inner"
-        )    
-
-    return trips_with_rt_key
-
-
 def get_trip_time_buckets(analysis_date: str) -> pd.DataFrame:
     """
     Assign trips to time-of-day.