I cannot confirm that the cursor comes back. In the past it was possible to get the cursor back by clicking outside of the editor window. That no longer is possible. Now I have to close and reopen the editor, which is not acceptable, as I would have to do that every few seconds.
The cursor vanishes on multiple occasions, so I cannot say exactly when it happens. But it definitely happens every time if either code completion is triggered or if an error in the code is shown to the left of the text area or if I commit code via the checkmark for checking.
Edit: The topic may be wrong here, as this mainly happens in Datamapper, not Designer (but also in Designer).
I have updates on this. This only happens in Datamapper, so this thread is wrong in the forum category Designer.
As soon as I click the checkmark in the javascript editor to check the code, the cursor vanishes. This is reproducible every time. I can only get it back by closing and reopening the editor window which is very tiresome, fatiguing and frustrating. It makes purposeful work nearly impossible.
This happens each and every time an error message shows up in the editor. I even provided a video about the error some months ago to the german support, did he not relay it to the devs as he claimed?
I could again do a video, but I do not see what this would accomplish. I did it already and it would just show what I say: I click the checkmark, an error message pops up and the cursor is gone.
This has nothing to do with datatypes or anything specific. If I type a single character in an empty editor window and click the checkmark this is recognized as a syntax error and the cursor is gone for good. Since error messages pop up quite often after clicking the checkmark (because the editor in Datamapper is really, really bad and outdated and has a really, really bad UX) this happens all the time.
I have no doubt that this happens every time for you, but please understand that it doesn’t occur every time for everyone, which is why I was exploring the idea that it might somehow be related to the data type.
Now if you say this happens regardless of the data type, then we’ll have to look elsewhere for a possible cause.
Also, please note that I am not a Dev. I am simply trying to help you out of your situation.
I never implied that it happens for everyone, but since it happens to me and it is immediately reproducible this is of no relevance anyway. This is an existing, reproducible bug that only happens in Datamapper and in no other software or IDE I use on the same machine under the same circumstances.
I also know that this is an old Eclipse bug known for years that obviously is also present in the Eclipse-based Connect Datamapper despite the fact that it was fixed in Eclipse a long time ago.
I really cannot know what OL representative’s exact jobs here in this forums are.
@Xanathon, would you be able to share a Data Mapping Configuration file from which all sensitive information has been removed from the applied data sample file(s)? Because neither am I able to reproduce this issue.
I cannot say how fast I will be able to do this because it will be A LOT of work to provide a sanitized data file with all problematic data removed. As I said above: This problem persists through multiple versions of Datamapper and I already provided german service with a video showing the problem. Maybe you should ask Mr. Schröter about the video. I have reported this again and again.
Another detail: The cursor vanishes the exact moment I hover the cursor over the red error indicator and when the error tooltip shows up.
Are you able to reproduce the issue with a very simple Data Mapping Configuration to which a CSV data file has been applied which contains data like the following?
One with a simple csv file and simple extraction → error does not happen
One with a complex xml file but simple extraction → error does not happen
It looks like the error only appears if the datamapping gets more complex. So simple example files will not help in finding the bug, I will have to provide the complete datamapping with a sanitized datafile.
That looks annoying, and I would be eager to investigate further if we had a way to reproduce it.
Some suggestions:
Use Mockaroo to generate mock data instead of anonymizing your data manually, or
Isolate what is triggering the issue. Start with the complex extraction config that exhibits the issue, then selectively disable/remove parts of it until the issue no longer occurs.
Mockaroo: It’s not that easy. I have to work with quite complex invoice xmls for a company that ships thousands of parcels every day. Creating valid data for those in Mockaroo and inserting it in the xml file would take much longer than to sanitize a real invoice xml from all personal data.
I will try to find time to do that, since it is time-intensive and would have to be done in addition to my already heavy workload. In addition: If I remove parts of the extractions, the datamapping becomes invalid because fields used in Action steps are no longer valid.
I followed the steps from your video, but unfortunately I still cannot reproduce the issue. I even started a local server to serve up some dummy data to accommodate the xhr call in the “nodetest” action step.
I created a ticket anyway (internal ref: SHARED-90988) and attached all the relevant info. Hopefully someone from the data mapper team will take a look.
)