enable additional json (#54)

* implement additional personal data

* enable additional json file

* Update documents.csv

* Update .test documents.csv

* error correction

.tests/config/pep/documents.csv

@@ -1,4 +1,4 @@
-sample_name,fhir_metadata,compressed_docs
-1,data/1111111111.otto_normalverbraucher_jpgs.patient.json,data/1111111111.otto_normalverbraucher_jpgs.tar.lz4
-2,data/2222222222.maxime_mustermann_jpgs.patient.json,data/2222222222.maxime_mustermann_jpgs.tar.lz4
-3,data/v00076462.maxime_mustermann_tiff.patient.json,data/v00076462.maxime_mustermann_tiff.tar.lz4
+sample_name,fhir_metadata,compressed_docs,additional_metadata
+1,data/1111111111.otto_normalverbraucher_jpgs.patient.json,data/1111111111.otto_normalverbraucher_jpgs.tar.lz4,
+2,data/2222222222.maxime_mustermann_jpgs.patient.json,data/2222222222.maxime_mustermann_jpgs.tar.lz4,
+3,data/v00076462.maxime_mustermann_tiff.patient.json,data/v00076462.maxime_mustermann_tiff.tar.lz4,

config/pep/documents.csv

@@ -1,2 +1,2 @@
-sample_name,fhir_metadata,compressed_docs
-1,data/fhire_meta_data.json,data/path_to.lz4
+sample_name,fhir_metadata,compressed_docs,additional_metadata
+1,data/fhire_meta_data.json,data/path_to.lz4,data/path_to_additional_metadata.json

workflow/rules/common.smk

@@ -31,11 +31,18 @@ def get_compressed_docs(wildcards):
 def get_fhir_metadata(wildcards):
     return pep.sample_table.loc[wildcards.id][["fhir_metadata"]]
 
+def get_additional_metadata(wildcards):
+    if "additional_metadata" in pep.sample_table.columns:
+        if type(pep.sample_table.loc[wildcards.id]["additional_metadata"]) == str:
+            return pep.sample_table.loc[wildcards.id][["additional_metadata"]]
+        else:
+            return []
+    else:
+        return []
 
 def get_all_ids():
     return pep.sample_table["sample_name"].to_list()
 
-
 def get_image_paths_for_id(wildcards):
     with checkpoints.fix_file_ext.get(id=wildcards.id).output[0].open() as f:
         paths = pd.read_csv(f, sep="\n", header=None, squeeze=True)

workflow/rules/extract-data.smk

@@ -1,14 +1,14 @@
 rule extract_personal_data:
     input:
         get_fhir_metadata,
+        get_additional_metadata,
     output:
         temp("results/{id}/tmp/personal-data.json"),
     log:
         "logs/{id}/extract_personal_data.log",
     script:
         "../scripts/extract-personal-data.py"
 
-
 rule extract_lz4_docs:
     input:
         get_compressed_docs,

workflow/scripts/extract-personal-data.py

@@ -20,7 +20,7 @@ def parse_meta_data(json_path: str) -> defaultdict:
     with open(json_path) as json_file:
         data = json.load(json_file)
 
-    # select the patient reosurce from the bundel data export
+    # select the patient resource from the bundel data export
     for ele in data.get("entry", {}):
         # iterate of entries
         for key, value in ele.get("resource", {}).items():
@@ -141,6 +141,13 @@ def variate_personal_data(personal_data: dict, first_name_count: int) -> default
 
     return personal_data
 
+def add_additional_personal_data(add_json_path: str, personal_data: dict) -> defaultdict:
+    # if an additional data file exist, this data will be added to the personal data json file
+    with open(add_json_path) as json_file:
+        additional_data = json.load(json_file)
+    personal_data.update(additional_data)
+
+    return personal_data
 
 def save_personal_data(personal_data: dict, out_path: str):
     """Save the final dic with the personal data as json.
@@ -169,7 +176,12 @@ def save_personal_data(personal_data: dict, out_path: str):
 
     # personal_data = {key: value.lower().strip() for key, value in personal_data.items()}
     var_data = {key: value.lower().strip() for key, value in var_data.items()}
-    save_personal_data(var_data, snakemake.output[0])
+
+    if len(snakemake.input) > 1:
+        add_data = add_additional_personal_data(snakemake.input[1], var_data)
+        save_personal_data(add_data, snakemake.output[0])
+    else:
+        save_personal_data(var_data, snakemake.output[0])
 
 # def enrich_personal_data(personal_data: dict) -> dict:
 #     personal_data["names_combined"] = ",".join(