A transform feels too heavy sometimes, and clutters the canvas too much. For example: the simple renaming or reordering of columns. It breaks its own effect when you decide to shuffle transforms around or insert a transform somewhere, thus requiring more housekeeping which distracts from the real work.
And please on Macos use some other hotkey for the magnify cursor instead of the default Macos combo for minimizing a window. I minimize windows all the time using keypresses, and the way Easydatatransform steals this hotkey is very distracting.
If we include renaming and reordering of columns in every transform, it would result in a few less transforms. But it would mean a lot of duplication between transforms (including more clutter in the UI) and it would make it harder to see what is going on.
Easy Data Transform should be more robust when disconnecting and reconnecting transforms since v1.47.1.
We try to keep the keyboard shortcuts the same on Windows and Mac, for people who work across both. We missed that Cmd-M is minimize on Mac. Perhaps Alt+M would be a better choice?
No doubt it would. There are some more UX-optimizations I could think of but since this is a feature request thread and I am only a very beginner, I wonât add noise to the thread.
New Compare transform needs an output report of non-matching data rows; Characters TAB shows numbers of events (symbols, whitespaces, Non-ASCII, etc.) and it be helpful to have a built-in selectable report showing the rows with the events;
Disable is used with output files, but a pause or stop switch for the middle of a large transform file with New Column transform would provide for updating a Value without running the second half of the transform to make the change; and Create an expansive REGEX library within the EDT documentation.
I reuse transform files and work on dirty data with 8+ million rowsâŚ
You can also use the new Compare Data feature for a quick visual check.
Possibly we could list the first 5 rows that contain that symbol when you hover over the cell in the Characters tab. But showing every row would require a potentially massive amount of memory. Maybe much more than the original dataset.
Canât you just use a Filter to find rows with a certain symbol?
Canât you just turn off Run>Auto run? Then you can choose which transforms to run.
Regex is a whole massive thing on itâs own. Also I suspect different customers use very different regexes. Any regex that is commonly used we will probably include in a transform (e.g. the verify phone number rule in Verify). But would like to add a way that you can create your own library of regex to share between .transforms.
I did vote, but there is one feature I would like to see which wasnât really touched on, and thatâs support for documentation.
Years ago I designed a system based around the OMGâs MOF architecture, and I realised that it would be possible to create a good documentation base from the metadata - the metameta level being key.
The information to feed Crystal Reports was created by XSLT to transform the metadata XML into something usable in Crystal. The Transform files have all the basis of a pretty good documentation set ⌠crazy idea ???
PS adding the node ID text to the transform box would help the doc as well.
Do you mean auto generating documentation from a particular transform? That was on the list as âExportable audit logâ. The idea would be to generate some human readable documentation from a .transform. Probably in HTML or PDF format.
XSLT is also somewhere on the wishlist.
The unique id for a node is shown in the âDetailsâ tab and in the tooltip when you hover over the node.