search

VariantEffect / mavedb-ui

MaveDB UI
GNU Affero General Public License v3.0
1 stars 1 forks source link

Drop empty hgvs columns from table preview #95

Open afrubin opened 9 months ago

afrubin commented 9 months ago

The data table preview currently shows all three possible hgvs columns (hgvs_nt, hgvs_splice, hgvs_pro) even when only one is used by the dataset.

It makes sense to provide all of these columns in the csv output for consistency, but when showing the data table preview it seems reasonable to omit the unpopulated columns. hgvs_splice in particular is seldom used, and we have many datasets with only hgvs_pro variants.

ashsny commented 9 months ago

I suspect the reason these columns exist in the UI is because the UI is using the same API endpoint as CSV download does

On Thu, Dec 21, 2023 at 4:15 PM Alan Rubin @.***> wrote:

The data table preview currently shows all three possible hgvs columns (hgvs_nt, hgvs_splice, hgvs_pro) even when only one is used by the dataset.

It makes sense to provide all of these columns in the csv output for consistency, but when showing the data table preview it seems reasonable to omit the unpopulated columns. hgvs_splice in particular is seldom used, and we have many datasets with only hgvs_pro variants.

— Reply to this email directly, view it on GitHub https://github.com/VariantEffect/mavedb-ui/issues/95, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAKTI3TGILGORVB2JVELYHTYKTGJRAVCNFSM6AAAAABA7DWO66VHI2DSMVQWIX3LMV43ASLTON2WKOZSGA2TGMJZHEZTGNA . You are receiving this because you are subscribed to this thread.Message ID: @.***>

EstelleDa commented 8 months ago

I suspect the reason these columns exist in the UI is because the UI is using the same API endpoint as CSV download does …

Yes, you're correct. In fact, we have the function of showing data in table preview first. I think downloading CSV file can use it directly so I just used the function of showing data in table preview for downloading CSV file.

afrubin commented 8 months ago

Originally we decided to keep all of the hgvs columns in csv output to make it easier for users who are downloading a lot of datasets, but we could drop the null columns and make the users identify which columns are present.

It seems like making that change to the csv logic would solve this issue as well, right?