Merge pull request #1138 from cmu-delphi/release/0.1.4

Release 0.1.4

Author: krivard

.github/workflows/python-ci.yml

@@ -5,7 +5,7 @@ name: Python package
 
 on:
   push:
-    branches: [ main, prod, 'release/*' ]
+    branches: [ main, prod ]
   pull_request:
     types: [ opened, synchronize, reopened, ready_for_review ]
     branches: [ main, prod ]

.github/workflows/r-ci.yml

@@ -0,0 +1,63 @@
+# This workflow uses actions that are not certified by GitHub.
+# They are provided by a third-party and are governed by
+# separate terms of service, privacy policy, and support
+# documentation.
+#
+# See https://github.com/r-lib/actions/tree/master/examples#readme for
+# additional example workflows available for the R community.
+
+name: R facebook survey
+
+on:
+  push:
+    branches: [ main, prod ]
+  pull_request:
+    types: [ opened, synchronize, reopened, ready_for_review ]
+    branches: [ main, prod ]
+
+jobs:
+  build:
+    runs-on: ubuntu-20.04
+    if: github.event.pull_request.draft == false
+    strategy:
+      matrix:
+        r-version: [4.0]
+    defaults:
+      run:
+        working-directory: facebook/delphiFacebook
+
+    steps:
+      - uses: actions/checkout@v2
+      - name: Set up R ${{ matrix.r-version }}
+        uses: r-lib/actions/setup-r@v1
+        with:
+          r-version: ${{ matrix.r-version }}
+      - name: Install linux dependencies
+        run: |
+            sudo apt-get install libcurl4-openssl-dev
+      - name: Get date
+        id: get-date
+        run: |
+          echo "::set-output name=date::$(/bin/date -u "+%Y%m%d")"
+      - name: Cache R packages
+        uses: actions/cache@v2
+        with:
+          path: ${{ env.R_LIBS_USER }}
+          key: ${{ runner.os }}-r-facebook-survey-${{ steps.get-date.outputs.date }}
+          restore-keys: |
+            ${{ runner.os }}-r-facebook-survey-
+      - name: Install R dependencies
+        run: |
+          if ( packageVersion("readr") != "1.4.0" ) {
+            install.packages("devtools")
+            devtools::install_version("readr", version = "1.4.0")
+          }
+          install.packages("remotes")
+          remotes::update_packages(c("rcmdcheck", "mockr"), upgrade="always")
+          dependency_list <- remotes::dev_package_deps(dependencies=TRUE)
+          remotes::update_packages(dependency_list$package[dependency_list$package != "readr"], upgrade="always")
+        shell: Rscript {0}
+      - name: Check
+        run: |
+          rcmdcheck::rcmdcheck(args = c("--no-manual", "--test-dir=unit-tests"), error_on = "error")
+        shell: Rscript {0}

_delphi_utils_python/.bumpversion.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.1.3
+current_version = 0.1.4
 commit = False
 tag = False
 tag_name = delphi-utils/v{new_version}

_delphi_utils_python/delphi_utils/__init__.py

@@ -14,4 +14,4 @@
 from .signal import add_prefix
 from .nancodes import Nans
 
-__version__ = "0.1.3"
+__version__ = "0.1.4"

_delphi_utils_python/delphi_utils/validator/report.py

@@ -7,18 +7,22 @@
 class ValidationReport:
     """Class for reporting the results of validation."""
 
-    def __init__(self, errors_to_suppress: List[ValidationFailure]):
+    def __init__(self, errors_to_suppress: List[ValidationFailure], data_source: str = ""):
         """Initialize a ValidationReport.
 
         Parameters
         ----------
         errors_to_suppress: List[ValidationFailure]
             List of ValidationFailures to ignore.
+        data_source: str
+            Name of data source as obtained from params
 
         Attributes
         ----------
         errors_to_suppress: List[ValidationFailure]
             See above
+        data_source: str
+            See above
         num_suppressed: int
             Number of errors suppressed
         total_checks: int
@@ -31,12 +35,12 @@ def __init__(self, errors_to_suppress: List[ValidationFailure]):
             Errors raised from validation failures not found in `self.errors_to_suppress`
         """
         self.errors_to_suppress = errors_to_suppress
+        self.data_source = data_source
         self.num_suppressed = 0
         self.total_checks = 0
         self.raised_errors = []
         self.raised_warnings = []
         self.unsuppressed_errors = []
-        self.summary = ""
 
     def add_raised_error(self, error):
         """Add an error to the report.
@@ -74,21 +78,25 @@ def add_raised_warning(self, warning):
         """
         self.raised_warnings.append(warning)
 
-    def set_summary(self):
-        """Represent summary of report as a string."""
-        out_str = f"{self.total_checks} checks run\n"
-        out_str += f"{len(self.unsuppressed_errors)} checks failed\n"
-        out_str += f"{self.num_suppressed} checks suppressed\n"
-        out_str += f"{len(self.raised_warnings)} warnings\n"
-        self.summary = out_str
-
     def log(self, logger=None):
         """Log errors and warnings."""
         if logger is None:
             logger = get_structured_logger(__name__)
 
-        self.set_summary()
-        logger.info(self.summary)
+        if self.success():
+            logger.info("Validation run successful",
+                data_source = self.data_source,
+                checks_run = self.total_checks,
+                checks_failed = len(self.unsuppressed_errors),
+                checks_suppressed = self.num_suppressed,
+                warnings = len(self.raised_warnings))
+        else:
+            logger.info("Validation run unsuccessful",
+                data_source = self.data_source,
+                checks_run = self.total_checks,
+                checks_failed = len(self.unsuppressed_errors),
+                checks_suppressed = self.num_suppressed,
+                warnings = len(self.raised_warnings))
         for error in self.unsuppressed_errors:
             logger.critical(str(error))
         for warning in self.raised_warnings:

_delphi_utils_python/delphi_utils/validator/validate.py

@@ -37,6 +37,7 @@ def __init__(self, params):
         # Date/time settings
         self.time_window = TimeWindow.from_params(validation_params["common"]["end_date"],
                                                   validation_params["common"]["span_length"])
+        self.data_source = validation_params["common"].get("data_source", "")
 
         self.static_validation = StaticValidator(validation_params)
         self.dynamic_validation = DynamicValidator(validation_params)
@@ -51,7 +52,7 @@ def validate(self):
         Returns:
             - ValidationReport collating the validation outcomes
         """
-        report = ValidationReport(self.suppressed_errors)
+        report = ValidationReport(self.suppressed_errors, self.data_source)
         frames_list = load_all_files(self.export_dir, self.time_window.start_date,
                                      self.time_window.end_date)
         self.static_validation.validate(frames_list, report)

_delphi_utils_python/setup.py

@@ -24,7 +24,7 @@
 
 setup(
     name="delphi_utils",
-    version="0.1.3",
+    version="0.1.4",
     description="Shared Utility Functions for Indicators",
     long_description=long_description,
     long_description_content_type="text/markdown",

_delphi_utils_python/tests/validator/test_report.py

@@ -40,10 +40,6 @@ def test_str(self):
         report.add_raised_warning(ImportWarning("right import"))
         report.add_raised_error(self.ERROR_1)
         report.add_raised_error(self.ERROR_2)
-        report.set_summary()
-
-        assert report.summary ==\
-            "3 checks run\n1 checks failed\n1 checks suppressed\n2 warnings\n"
 
     def test_log(self):
         """Test that the logs contain all failures and warnings."""

ansible/templates/sir_complainsalot-params-prod.json.j2

@@ -46,6 +46,18 @@
       "max_age":6,
       "maintainers": ["U01AP8GSWG3","U01069KCRS7"],
       "retired-signals": ["raw_pct_negative","smoothed_pct_negative","raw_tests_per_device","smoothed_tests_per_device"]
+    },
+    "nchs-mortality": {
+      "max_age":16,
+      "maintainers": []
+    },
+    "covid-act-now": {
+      "max_age":5,
+      "maintainers": []
+    },
+    "hhs": {
+      "max_age":8,
+      "maintainers": []
     }
   }
 }

combo_cases_and_deaths/delphi_combo_cases_and_deaths/run.py

@@ -322,6 +322,7 @@ def run_module(params):
     variants = [tuple((metric, geo_res)+sensor_signal(metric, sensor, smoother))
                 for (metric, geo_res, sensor, smoother) in
                 product(METRICS, GEO_RESOLUTIONS, SENSORS, SMOOTH_TYPES)]
+    variants = [i for i in variants if not ("7dav" in i[2] and "cumulative" in i[2])]
     params = configure(variants, params)
     logger = get_structured_logger(
         __name__, filename=params["common"].get("log_filename"),

combo_cases_and_deaths/tests/receiving/.gitkeep

@@ -1,13 +1,16 @@
 """Tests for running combo cases and deaths indicator."""
 from datetime import date
 from itertools import product
+import os
 import unittest
 from unittest.mock import patch, call
 import pandas as pd
 import numpy as np
 
 from delphi_combo_cases_and_deaths.run import (
-    extend_raw_date_range, get_updated_dates,
+    run_module,
+    extend_raw_date_range,
+    get_updated_dates,
     sensor_signal,
     combine_usafacts_and_jhu,
     compute_special_geo_dfs,
@@ -244,6 +247,50 @@ def test_no_nation_jhu(mock_covidcast_signal):
                       "sample_size": [None]},)
     )
 
+@patch("delphi_combo_cases_and_deaths.run.combine_usafacts_and_jhu")
+def test_output_files(mock_combine):
+    params = {
+        "common": {
+            "export_dir": "./receiving"
+        },
+        "indicator": {
+            "export_start_date": [2020, 4, 1],
+            "source":"indicator-combination",
+            "wip_signal": ""
+        }
+    }
+    mock_combine.return_value = pd.DataFrame(
+            {
+                "geo_id": ["01000"],
+                "val": [10],
+                "timestamp": [pd.to_datetime("2021-01-04")],
+                "issue": [pd.to_datetime("2021-01-04")],
+                "se": 0,
+                "sample_size": 0
+            },
+            index=[1]
+        )
+    run_module(params)
+    csv_files = [f for f in os.listdir("receiving") if f.endswith(".csv")]
+    dates = ["20210104"]
+    geos = ["county", "hrr", "msa", "state", "hhs", "nation"]
+
+    # enumerate metric names.
+    metrics = []
+    for event, span, stat in product(["deaths", "confirmed"],
+                                     ["cumulative", "incidence"],
+                                     ["num", "prop"]):
+        metrics.append("_".join([event, span, stat]))
+        metrics.append("_".join([event, "7dav", span, stat]))
+
+    expected_files = []
+    for date in dates:
+        for geo in geos:
+            for metric in metrics:
+                if "7dav" in metric and "cumulative" in metric:
+                    continue
+                expected_files += [date + "_" + geo + "_" + metric + ".csv"]
+    assert set(csv_files) == set(expected_files)
 
 if __name__ == '__main__':
     unittest.main()

combo_cases_and_deaths/tests/test_run.py

@@ -189,7 +189,7 @@ write_rollup <- function(newly_seen_files, seen_file, output_df, output_file) {
 
 args <- commandArgs(TRUE)
 
-if (length(args) < 2) {
+if (length(args) != 2) {
   stop("Usage: Rscript contingency-combine.R path/to/individual/files/ path/to/rollup/files/")
 }
 

facebook/contingency-combine.R

@@ -18,6 +18,7 @@ export(get_filenames_in_range)
 export(get_range_prev_full_month)
 export(get_range_prev_full_period)
 export(get_range_prev_full_week)
+export(get_sparse_filenames)
 export(jeffreys_se)
 export(join_weights)
 export(load_archive)