VEuPathDB / EdaPhaseOneNewIssuesOLD

A skeletal repo that contains new EDA issues that are not yet initially reviewed by the dev team
Apache License 2.0
0 stars 0 forks source link

Back end/Subsetting: Categorical vars with many values cause `count` to fail when all/many are checked. #41

Closed bobular closed 3 years ago

bobular commented 3 years ago

count request for PRISM2 variable Non-malaria medication is failing with a 500 when all items are selected. https://localhost:3000/eda-subsetting-service/studies/PRISM0002-1/entities/PCO_0000024/count

{
    "filters": [
        {
            "entityId": "EUPATH_0000738",
            "stringSet": [
                "Amoxicillin caps (250 mg)",
                "Amoxicillin syrup (125 mg/ 5 ml)",
                "Ampicillin ",
                "Ampiclox caps (500 mg)",
                "Ampiclox syrup (250 mg/ 5 ml)",
                "Aprinox (Bendrofluazide) ",
                "Artenum/Artemether (25mg)",
                "Artesunate ",
                "Atenolol ",
// SNIPPED //
                "Quinine injection (300 mg/ 2 ml)",
                "Ringer's lactate solution ",
                "Salbutamol tablets (4 mg)",
                "Tenofovir/Lamivudine (TDF/3TC) (300/300)",
                "Tetracycline eye ointment ",
                "Tramadol ",
                "Trimethoprim-Sulfamethoxazole (TS) ",
                "Vitamin A caps (200000 iu)",
                "Vitamin B complex ",
                "Vitamin C / Ascorbic Acid tablets (300 mg)",
                "Zinc gluconate "
            ],
            "type": "stringSet",
            "variableId": "EUPATH_0000059"
        }
    ]
}

That's a multiValued variable, but I think that's a red herring.

In SCORE S. mansoni RCT the village variable also has lots of values and selecting all on that also causes a 500 error and zero count to show up in the Entity Diagram.

image

steve-fischer-200 commented 3 years ago

this is caused by the single-quote character in one of the values. there is an issue filed for that bug, so this issue is being closed as a duplicate. thanks though!