BrowseTables.jl just generates a static HTML. This of course precludes editing, but allows larger tables — this was a deliberate design choice.
If you need to edit cells, and add rows/columns, I would recommend you interface with a spreadsheet, or start work there.
I think it requires Gtk to be installed as well. This is worth mentioning for Windows users.
I now I can use a spreadsheet but I was looking for something inside Julia, and if possible able to edit large datasets, most spreadsheets crash if you try to edit very large datasets. In this situation till now I’ve been forced to use databases (sql, monetdb…) but things get more complex and they don’t provide advanced functions such as Julia.
I imagine most spreadsheets can still cope with vastly more data than what one would want to edit or even just eyeball. If you only want to edit a subset, making an extract, editing that, and then remerging is possibly the best way.
TableView.jl can in principle support mutating DataFrames; it can also handle arbitrarily large datasets (row-wise, that is, but if you have a large number of columns you probably should switch to a different representation anyways).
Sweet, will this work in Juno?
Eventually yes, but I need to fix the WebIO integration again I think.
Can I make a plea that TableView not support mutating tables? The only thing I care about with a dataviewer is speed, especially for super large datasets. Any mutating feature only supports non-scripting anti-patterns and slows down the functionality as a whole.