From 953720ce5473f4953881c4c7e4b68f0bc192944d Mon Sep 17 00:00:00 2001
From: Marty Oehme <marty.oehme@gmail.com>
Date: Thu, 15 Feb 2024 16:32:46 +0100
Subject: [PATCH] refactor(code): Move finding table validities into model
 module

All findings tables can use the validities functionality to add strength of
evidence (internal/external) to themselves. Generalized the function to
work for any main findings csv (to dataframe) table not just institional
findings.
---
 scoping_review.qmd                | 26 +++-------------
 src/model/strength_of_findings.py | 51 +++++++++++++++++++++++++++++++
 2 files changed, 55 insertions(+), 22 deletions(-)
 create mode 100644 src/model/strength_of_findings.py

diff --git a/scoping_review.qmd b/scoping_review.qmd
index 7b41996..dfb3c54 100644
--- a/scoping_review.qmd
+++ b/scoping_review.qmd
@@ -631,30 +631,12 @@ g = sns.PairGrid(validities[["internal_validity", "external_validity", "identifi
 ```{python}
 #| label: tbl-findings-institutional
 #| tbl-cap: Main findings summary institutional policies
+from src.model import strength_of_findings as findings
+
 findings_institutional = pd.read_csv("02-data/supplementary/findings-institutional.csv")
-from src import prep_data
-import math
+fd_df = findings.add_validities(findings_institutional, by_intervention)
 
-EVIDENCE_STRENGH=["\-","\-","\-","\+","\+","\+","++","++","++","++","++","++","++","++","++","++"]
-
-valid_subset = prep_data.calculate_validities(by_intervention)[["internal_validity", "external_validity", "citation"]].fillna(1.0).drop_duplicates(subset=["citation"]).sort_values("internal_validity")
-def combined_validities(df_in, column: str = "internal_validity"):
-    if not isinstance(df_in, str):
-        return
-    combined = 0.0
-    for study in df_in.split(";"):
-        new = valid_subset.loc[valid_subset["citation"] == study, column]
-        if len(new) > 0 and not math.isnan(new.iat[0]):
-            combined += new.iat[0]
-    if combined:
-        return EVIDENCE_STRENGH[int(combined)] + f" ({str(combined)})"
-    return "\-"
-def combined_external(df_in, column: str = "external_validity"):
-    return combined_validities(df_in, column)
-
-findings_institutional["internal_validity"] = findings_institutional["studies"].apply(combined_validities)
-findings_institutional["external_validity"] = findings_institutional["studies"].apply(combined_external)
-md(tabulate(findings_institutional[["area of policy",  "internal_validity", "external_validity", "findings", "channels"]].fillna(""), showindex=False, headers="keys", tablefmt="grid"))
+md(tabulate(fd_df[["area of policy",  "internal_validity", "external_validity", "findings", "channels"]].fillna(""), showindex=False, headers="keys", tablefmt="grid"))
 ```
 
 {{< landscape >}}
diff --git a/src/model/strength_of_findings.py b/src/model/strength_of_findings.py
new file mode 100644
index 0000000..f078cd6
--- /dev/null
+++ b/src/model/strength_of_findings.py
@@ -0,0 +1,51 @@
+import math
+
+from pandas import DataFrame
+
+from src import prep_data
+
+
+def _binned_strength(strength: float) -> str:
+    if strength < 3.0:
+        return r"\-"
+    if strength < 6.0:
+        return r"\+"
+    return r"\++"
+
+
+def _combined_validities(
+    apply_to: DataFrame, by_intervention: DataFrame, column: str = "internal_validity"
+):
+    if not isinstance(apply_to, str):
+        return
+    combined = 0.0
+    for study in apply_to.split(";"):
+        new = by_intervention.loc[by_intervention["citation"] == study, column]
+        if len(new) > 0 and not math.isnan(new.iat[0]):
+            combined += new.iat[0]
+    if combined:
+        return _binned_strength(combined)
+    return r"\-"
+
+
+def add_validities(
+    findings_df: DataFrame, studies_by_intervention: DataFrame
+) -> DataFrame:
+    valid_subset = (
+        prep_data.calculate_validities(studies_by_intervention)[
+            ["internal_validity", "external_validity", "citation"]
+        ]
+        .fillna(1.0)
+        .drop_duplicates(subset=["citation"])
+        .sort_values("internal_validity")
+    )
+
+    def apply_internal(df):
+        return _combined_validities(df, valid_subset, "internal_validity")
+
+    def apply_external(df):
+        return _combined_validities(df, valid_subset, "external_validity")
+
+    findings_df["internal_validity"] = findings_df["studies"].apply(apply_internal)
+    findings_df["external_validity"] = findings_df["studies"].apply(apply_external)
+    return findings_df