Warn when someone uploads a csv file that has columns not in the db

* wqflask/wqflask/metadata_edits.py: Import "extract_invalid_csv_headers"
and "get_allowable_sampledata_headers".
(display_phenotype_metadata): Pass the allowable headers to the
template.
(update_phenotype): If a user uploads data with a column header that's
not in the db, don't upload the file, and send a warning message.
* wqflask/wqflask/templates/edit_phenotype.html: List the allowable
headers in the template.

2 files changed, 22 insertions, 2 deletions

diff --git a/wqflask/wqflask/metadata_edits.py b/wqflask/wqflask/metadata_edits.py
index 49205b76..30acf4d4 100644
--- a/wqflask/wqflask/metadata_edits.py
+++ b/wqflask/wqflask/metadata_edits.py
@@ -28,6 +28,8 @@ from gn3.authentication import AdminRole
 from gn3.authentication import get_highest_user_access_role
 from gn3.csvcmp import create_dirs_if_not_exists
 from gn3.csvcmp import csv_diff
+from gn3.csvcmp import extract_invalid_csv_headers
+from gn3.csvcmp import get_allowable_sampledata_headers
 from gn3.csvcmp import remove_insignificant_edits
 from gn3.db import diff_from_dict
 from gn3.db import fetchall
@@ -40,10 +42,10 @@ from gn3.db.phenotypes import Probeset
 from gn3.db.phenotypes import Publication
 from gn3.db.phenotypes import PublishXRef
 from gn3.db.phenotypes import probeset_mapping
-from gn3.db.sample_data import get_trait_csv_sample_data
-from gn3.db.sample_data import update_sample_data
 from gn3.db.sample_data import delete_sample_data
+from gn3.db.sample_data import get_trait_csv_sample_data
 from gn3.db.sample_data import insert_sample_data
+from gn3.db.sample_data import update_sample_data
 
 
 metadata_edit = Blueprint('metadata_edit', __name__)
@@ -187,6 +189,7 @@ def display_phenotype_metadata(dataset_id: str, name: str):
             publication=_d.get("publication"),
             dataset_id=dataset_id,
             resource_id=request.args.get("resource-id"),
+            headers=get_allowable_sampledata_headers(conn),
             version=os.environ.get("GN_VERSION"),
         )
 
@@ -239,6 +242,18 @@ def update_phenotype(dataset_id: str, name: str):
                     delta_csv=(delta_csv := file_.read().decode()),
                     tmp_dir=TMPDIR),
                 epsilon=0.001)
+            headers = get_allowable_sampledata_headers(conn)
+            invalid_headers = extract_invalid_csv_headers(
+                    allowed_headers=headers,
+                    csv_text=delta_csv)
+            if invalid_headers:
+                flash("You have invalid headers: "
+                      f"""{', '.join(invalid_headers)}.  Valid headers """
+                      f"""are: {', '.join(headers)}""",
+                  "warning")
+                return redirect(
+                    f"/datasets/{dataset_id}/traits/{name}"
+                    f"?resource-id={request.args.get('resource-id')}")
         # Edge case where the csv file has not been edited!
         if not any(diff_data.values()):
             flash("You have not modified the csv file you downloaded!",
diff --git a/wqflask/wqflask/templates/edit_phenotype.html b/wqflask/wqflask/templates/edit_phenotype.html
index 76d15043..5458247e 100644
--- a/wqflask/wqflask/templates/edit_phenotype.html
+++ b/wqflask/wqflask/templates/edit_phenotype.html
@@ -224,6 +224,11 @@
     <div class="form-group">
         <input type = "file" class="col-sm-4 control-label" name = "file" />
     </div>
+    <div class="col-xs-6">
+        <p>
+            Note: Current allowable sample headers are: {{ ', '.join(headers) }}
+        </p>
+    </div>
     <div class="controls center-block" style="width: max-content;">
         <input name="inbred-set-id" class="changed" type="hidden" value="{{ publish_xref.inbred_set_id }}"/>
         <input name="phenotype-id" class="changed" type="hidden" value="{{ publish_xref.phenotype_id }}"/>