jsvine / visidata-plugins Goto Github PK

Deduplicating rows doesn't work in the pre-release. In this case the action didn't do anything, but in other cases it multiplies the rows (I can't figure out any pattern here)

Ctrl+E gets "no error".

Context: saulpw/visidata#1134

Visidata now has a clean_names options which is similar to the normcol plugin. In the thread, Saul points out a way to revert the column renames after doing some work on a sheet. This would be a useful feature for this plugin.

Hi, I'm trying to install the plugin, I've tried this in visidata 2.4 as well as after upgrading to 2.6.1.

Apparently the 2.x version strips whitespace before hashing, and encodes to utf-8 ; too bad the docs weren't properly updated. There's now a utility in visidata/dev/vdhash.py to create a similar hash. I ran it and it generated this hash:

94cc1b0afcf7dbcbe93ca421104a2252c411448d59ba75df00ae455d3577934a

I tried setting in the open-plugins settings, and it fixed the issue. If you're happy, I can submit a PR to saulpw/visidata to correct the plugin hash there, as well as maybe some docs (if I can find them)

This is a feature request. The vdnormcol plugin is very useful since it fixes column names to be referenced in any Visidata command that takes a Python express. This seems like it should be built-in to Visidata.

However, it is possible for a column name to conflict with an existing global. For example a column with the name "time" would be left unchanged by vdnormcol, but would hide the time module.

Request: vdnormcol should also check columns against existing global names.