As I just told @SchurigH in a private message, I found the cause of the issue. 
This will happen whenever a user presses Esc while just normally using the application. It toggles a boolean in our code which then prevents tabs from being duplicated. I’ll log a ticket and will fix it as soon as possible.
1 Like
Nice! Glad to hear your “baby” will soon be in it’s best form 
Maybe a bug? Or maybe I’m just blind. When I update query parameter inside the export CSV dialog, a hint asks me to rescan to apply these changes. But I can’t find the rescan button. It might be hidden because my UI looks not right at one point, see for yourself:
I assume this is caused by switching between office (laptop → docking station → 2x 1440p displays) and home office (laptop → docking station → 1x 4K display)? Maybe that causes the UI to glitch out.
Another question: On the download page I see version 2022.8.3 available. I have 2022.8.1 and “Check for Updates…” won’t find anything new. Is it supposed to not deliver specific smaller patches on purpose?
On the versions, .2 and .3 are minor updates which affect new user onboarding, so we did not release them into our autoupdater.
And the answer to your question about the Rescan button is… It’s related to the source so if you head up to
and click on the down-facing chevron, you’ll get a menu which includes the rescan option.
And yes, we do plan to improve that experience.
1 Like
Back from vacation, first day and after a few minutes of work with Studio, it crashed on me twice. 
First time while closing open tabs, after that I’ve updated to 2022.9 and again after closing one or two tabs (which were re-opened on startup as usual), it crashed a second time. I will submit both crash reports but is there anything else we can do/exchange/analyse here?
Crashing while managing/closing tabs is probably the only but common crash issue that happens since… ever(?) or at least months. Happens at least 2-3x per month, the more tabs the worse.
Edit: Sadly, the Visual Query Editor yellow popup thing is still there to annoy me, see the GIF in my thread-start post. In every version I hope that this is fixed ^^
If I copy paste a query into a new collection view, Visual Query Editor will convert that into its own format. Regex stuff is converted to [equals] [Regex] (which is fine) but is read-only within this editor. I can only edit the pasted query in plain text. Why is that so?
What is the query you are pasting?
You can just use one regex query like { "email" : /.*a\.b.*/i } and this will transform in Visual Editor into a read-only [equals] [RegEx] entry:
Version 2022.9.0
It should be possible to edit the RegEx by hitting the little “…” button after the read-only text box. This is so that we can show the options in a fancier format.
Confirmed, I can edit the text within the popup. Not as easy as if the input field were editable directly but it’s sufficient.
Two suggestions:
-
IntelliShell doesn’t seem to have a good “change variable name” feature. Search & replace sadly doesn’t have something like “exact match” or “match entire word”, which would be kind of a MVP for that kind of feature.
If I have two variables like syncSchool and syncSchoolNumber and later I decide that syncSchool should be syncSchoolId, I can’t use search & replace anymore, because every “syncSchool” finding would also find syncSchool within syncSchoolNumber and would mess this variable up. So I have to do it by hand. “Match entire word” (Visual Studio Code) / “Whole word” (Sublime) enable me to find just syncSchool, without syncSchoolNumber and only change these occurences. -
CSV Export seems to not be able to handle multi-level arrays. Lets assume I have a random number of arrays of same length in one big array:
[
[
1,
2,
3
],
[
4
5,
6
],
[
7,
8,
9
]
]
I would expect the CSV exporter to export:
1,2,3
4,5,6
7,8,9
because the big outer array defines the structre in which every inner array seems to hold one dataset, which is one line in CSV. Every inner array defines the data of that one line/dataset.
I get that this would be difficult to handle of not all inner arrays have the exact same length, but if they do, this CSV export would make more sense compared to the current result:
1,2,3,4,5,6,7,8,9 
Maybe I’m just doin’ it wrong 
With that I’m stuck to printing each inner array as comma separated text to copy that into a csv file by hand.
For 1. I think that’s a good suggestion to drop into https://3t-io.uservoice.com/ as a suggestion to go in front of the team.
For 2. I’m not sure what you are looking for. Multilevel arrays may be embedded in a document…
For a document with a multi-level array like:
{
“name” : “fred”,
“data” : [
[
1.0,
2.0,
3.0
],
[
4.0,
5.0,
6.0
],
[
7.0,
8.0,
9.0
]
],
“_id” : ObjectId(“637f6b84a75fec438dffb79a”)
}
Exporting it as CSV gives:
“_id”,name,data.0.0,data.0.1,data.0.2,data.1.0,data.1.1,data.1.2,data.2.0,data.2.1,data.2.2
637f6b84a75fec438dffb79a,fred,1.0,2.0,3.0,4.0,5.0,6.0,7.0,8.0,9.0
With the structure recorded in the headers. Breaking it over multiple lines would break it being a CSV file where each row represents an entire document… CSV lacks any structural formatting, it’s hardly a format, so the smart move is to push that information into the headers,
- I’ve added the “rename variable” topic to uservoice.com.
- CSV - yeah, we might have different use cases here. In your example, you try to bring a complex JSON object into the CSV “format”. In that case, this export might fit.
In my case, there is no hierarchy, just one level of information. Like so:
I import that data as array (as text) into IntelliShell:
[
["5beaead05xxxxxxxxxx";"d8976646-98aa-4b2e-xxxx-xxxxxxxx";"b9c72efc-831e-xxxx-xxxxxxxx";"xxxxxxxxx";"5beaeacxxxxxxxx";"5bf6c0xxxxxxxx";true],
["6beaead05xxxxxxxxxx";"e8976646-98aa-4b2e-xxxx-xxxxxxxx";"c9c72efc-831e-xxxx-xxxxxxxx";"xxxxxxxxx";"5beaeacxxxxxxxx";"5bf6c0xxxxxxxx";true],
["7beaead05xxxxxxxxxx";"f8976646-98aa-4b2e-xxxx-xxxxxxxx";"d9c72efc-831e-xxxx-xxxxxxxx";"xxxxxxxxx";"5beaeacxxxxxxxx";"5bf6c0xxxxxxxx";true],
...
]
Now I add some data to each array (which is one dataset/document, so one line in CSV) and need to export it again. But with the current CSV export, all data of all documents will be exported into one line, because the containing array is seen as 1 document, not all the inside arrays.
But maybe a) not really solvable because edge case and b) maybe I do it wrong.
New suggestion: If I open IntelliShell, there is a default .find(“”) statement and the cursor places itself inside the quotes. If I then select a bookmark to load, the code gets loaded within the quotes, the default query is “around” my imported bookmark code and therefore produces error. I would suggest: If the default query was not changed and a bookmark is loaded → delete the query before pasting in the bookmark code.
Well, yes, you are putting all the data into an array, which is, quite literally all int the same document. What you might want to consider is using aggregation to unwind each line of the array into its own document and then using that aggregation as the source for the CSV export.
Would an option just to disable the default query being inserted also work or do you rely on it as a starting point?
I’m considering a couple of other scenarios of bookmark insertion, not just with the default query.
Yes, that would do. I’m never using the default query anyways.
Sorry, I only noticed these new replies just now but I thought I’d chime in anyway.
New suggestion: If I open IntelliShell, there is a default .find(“”) statement and the cursor places itself inside the quotes. If I then select a bookmark to load, the code gets loaded within the quotes, the default query is “around” my imported bookmark code and therefore produces error. I would suggest: If the default query was not changed and a bookmark is loaded → delete the query before pasting in the bookmark code.
You can open the IntelliShell on the collection’s database instead of on the collection itself. That way, no default query should be there.
In addition to that, we also changed how bookmarks behave in the IntelliShell starting with 2023.1. Bookmarks now aren’t pasted but rather replace the script content (much like files do).
1 Like
Thanks for these additions. In the newest version there is also a new IntelliShell option for that: Edit → Preferences → IntelliShell → [ ] Automatically execute the default query when opening a new IntelliShell tab
Just leave it blank and voila! 
1 Like
Suggestion: Please let us submit the “Follow reference” popup by pressing [Enter] in the “Select target” input field. Currently [Tab] and [Space] is the fastest way and not so intuitive.
Can you give Ctrl + Enter a shot? A lot of our dialogs (if not all) ignore Enter as we have many multi-line-editor dialogs.