Merge /python/python into /python

Author: uw0s

.github/workflows/workflow.yml

@@ -8,7 +8,6 @@ jobs:
         uses: actions/checkout@v4
       - name: Check code and build
         run: |
-          cd python/python && make clean check all && cd ..
           cd python/ && make clean check all && cd ..
   pip:
     needs: [make]

README

@@ -10,7 +10,6 @@ git clone https://github.com/valab-certh/meddisc $HOME/github.com/valab-certh/me
 cd $HOME/github.com/valab-certh/meddisc
 
 # Development
-cd python/python/ && make clean check all && cd ../..
 cd python/ && make clean check all && cd ..
 
 # Staging

python/main.py

@@ -1260,7 +1260,7 @@ def dicom_deidentifier(
     else:
         custom_config_df = None
     action_groups_df = pd.read_csv(
-        filepath_or_buffer="./python/tmp/action-groups-dcm.csv",
+        filepath_or_buffer="./tmp/action-groups-dcm.csv",
         index_col=0,
     )
     with Path("./tmp/session-data/user-options.json").open() as file:
@@ -1362,6 +1362,128 @@ async def handle_submit_button_click(user_options: UserOptionsClass) -> list[Any
     return dicom_pair_fps
 
 
+def generate_action_groups() -> None:
+    threshold = 8
+    acceptable_actions = {"", "K", "C", "Z", "X"}
+    with request.urlopen(
+        "https://dicom.nema.org/medical/dicom/current/output/chtml/part15/chapter_e.html",
+    ) as response:
+        html = response.read()
+    df_table = pd.read_html(html)[3].fillna("")
+    dicom_tags = df_table["Tag"].str.replace("[(,)]", "", regex=True).to_list()
+    basic_profile = ["X" for i in df_table["Basic Prof."].to_list()]
+    retain_safe_private_option = df_table["Rtn. Safe Priv. Opt."].to_list()
+    retain_uids_option = df_table["Rtn. UIDs Opt."].to_list()
+    retain_device_identity_option = df_table["Rtn. Dev. Id. Opt."].to_list()
+    retain_patient_characteristics_option = [
+        "K" if i == "C" else i for i in df_table["Rtn. Pat. Chars. Opt."].to_list()
+    ]
+    retain_long_modified_dates_option = df_table["Rtn. Long. Modif. Dates Opt."]
+    retain_description_option = [
+        "K" if i == "C" else i for i in df_table["Clean Desc. Opt."].to_list()
+    ]
+    output = [
+        basic_profile,
+        retain_safe_private_option,
+        retain_uids_option,
+        retain_device_identity_option,
+        retain_patient_characteristics_option,
+        retain_long_modified_dates_option,
+        retain_description_option,
+    ]
+    dicom_tag_to_nema_action = dict(zip(dicom_tags, list(map(list, zip(*output)))))
+    dicom_tag_to_nema_action_df = pd.DataFrame(dicom_tag_to_nema_action).transpose()
+    dicom_tag_to_nema_action_df.columns = [
+        "Default",
+        "Rtn. Safe Priv. Opt.",
+        "Rtn. UIDs Opt.",
+        "Rtn. Dev. Id. Opt.",
+        "Rtn. Pat. Chars. Opt.",
+        "Rtn. Long. Modif. Dates Opt.",
+        "Clean Desc. Opt.",
+    ]
+    dicom_tag_to_nema_action_df.insert(
+        0,
+        "Name",
+        ["" for i in range(len(dicom_tag_to_nema_action_df))],
+    )
+    dicom_tag_to_nema_action_df.loc["00100010", "Default"] = "Z"
+    dicom_tag_to_nema_action_df.loc["00100020", "Default"] = "Z"
+    dicom_tag_to_nema_action_df = dicom_tag_to_nema_action_df.rename(
+        columns={"Clean Desc. Opt.": "Rtn. Desc. Opt."},
+    )
+    dicom_tag_to_nema_action_df.insert(
+        loc=6,
+        column="Offset Long. Modif. Dates Opt.",
+        value=dicom_tag_to_nema_action_df["Rtn. Long. Modif. Dates Opt."],
+    )
+    dicom_tag_to_nema_action_df.insert(
+        loc=7,
+        column="Remove Long. Modif. Dates Opt.",
+        value=dicom_tag_to_nema_action_df["Rtn. Long. Modif. Dates Opt."],
+    )
+    dicom_tag_to_nema_action_df = dicom_tag_to_nema_action_df.replace(
+        to_replace={"Rtn. Long. Modif. Dates Opt.": "C"},
+        value="K",
+    )
+    dicom_tag_to_nema_action_df = dicom_tag_to_nema_action_df.replace(
+        to_replace={"Remove Long. Modif. Dates Opt.": "C"},
+        value="X",
+    )
+    dicom_tag_to_nema_action_df = dicom_tag_to_nema_action_df.replace(
+        to_replace={
+            "Rtn. Safe Priv. Opt.": "C",
+            "Rtn. UIDs Opt.": "C",
+            "Rtn. Dev. Id. Opt.": "C",
+            "Rtn. Pat. Chars. Opt.": "C",
+            "Rtn. Desc. Opt.": "C",
+        },
+        value="K",
+    )
+    for action in ("D", "X/Z", "X/Z/D", "X/D", "Z/D"):
+        dicom_tag_to_nema_action_df = dicom_tag_to_nema_action_df.replace(
+            to_replace=action,
+            value="X",
+        )
+    for tag_idx in dicom_tag_to_nema_action_df.index:
+        tag = "(" + tag_idx[0:4] + "," + tag_idx[4:9] + ")"
+        if len(tag_idx) > threshold:
+            if tag_idx != "ggggeeee where gggg is odd":
+                msg = "Tag index error"
+                raise ValueError(msg)
+            tag = "(" + tag_idx[0:4] + "," + tag_idx[4:8] + ")" + tag_idx[8:]
+        dicom_tag_to_nema_action_df.loc[tag_idx, "Name"] = df_table.loc[
+            df_table["Tag"] == tag,
+            "Attribute Name",
+        ].item()
+        if tag_idx in ["00100010", "00100020"]:
+            dicom_tag_to_nema_action_df.loc[tag_idx, "Default"] = "Z"
+        elif (
+            "K" not in dicom_tag_to_nema_action_df.loc[tag_idx].to_numpy().tolist()[1:]
+        ):
+            dicom_tag_to_nema_action_df.loc[tag_idx, "Default"] = "X"
+    dicom_tag_to_nema_action_df = dicom_tag_to_nema_action_df.sort_values(by="Name")
+
+    patient_id_actions = dicom_tag_to_nema_action_df.loc["00100010"].to_numpy()[2:]
+    patient_name_actions = dicom_tag_to_nema_action_df.loc["00100020"].to_numpy()[2:]
+    if not (
+        all(action == "" for action in patient_id_actions)
+        and all(action == "" for action in patient_name_actions)
+    ):
+        msg = "Patient ID and name actions must remain empty except the default one"
+        raise ValueError(
+            msg,
+        )
+
+    dicom_tag_to_nema_action_arr = dicom_tag_to_nema_action_df.iloc[:, 1:].to_numpy()
+    action_set = set(dicom_tag_to_nema_action_arr.flatten())
+    if not action_set <= acceptable_actions:
+        msg = "Unacceptable values found in action table"
+        raise ValueError(msg)
+
+    dicom_tag_to_nema_action_df.to_csv("./tmp/action-groups-dcm.csv")
+
+
 def ndarray_size(arr: NDArray[Any]) -> int:
     return arr.itemsize * arr.size
 
@@ -1374,6 +1496,9 @@ def meddisc() -> None:
     ]
     for directory in tmp_directories:
         directory.mkdir(parents=True, exist_ok=True)
+    csv_path = Path("tmp/action-groups-dcm.csv")
+    if not csv_path.exists():
+        generate_action_groups()
     if os.getenv("STAGING"):
         if not Path("tmp/fullchain.pem").exists():
             subprocess.run(