# Mixing up the windows

**URL:** <https://gramps.discourse.group/t/mixing-up-the-windows/2004>\
**Category:** Help\
**Created:** [December 2, 2021, 8:33pm UTC](https://gramps.discourse.group/t/mixing-up-the-windows/2004 "2021-12-02T20:33:13Z")\
**Posts on this page:** 6\
**Page:** 1

<div class="post-metadata">

**Author:** ![Davesellers](https://yyz2.discourse-cdn.com/free1/user_avatar/gramps.discourse.group/davesellers/32/1457_2.png) [@Davesellers](https://gramps.discourse.group/u/Davesellers)\
**Post date:** [December 2, 2021, 8:33pm UTC](https://gramps.discourse.group/t/mixing-up-the-windows/2004/1 "2021-12-02T20:33:13Z")

</div>

This is starting to bug me…  
When I’m creating an event, and adding a source, maybe 3 windows deep, I go and hit the OK, but stupid me, the OK from the wrong window. This closes all the sub-windows and I loose the data that I have entered.  
Is there a way to prevent this? A way to force you to close the windows in the correct order?  
I could position each new window over the OK button, however I reposition the windows a lot when I want to check other entries,

---

<div class="post-metadata">

**Author:** ![emyoulation](https://yyz2.discourse-cdn.com/free1/user_avatar/gramps.discourse.group/emyoulation/32/67_2.png) [@emyoulation](https://gramps.discourse.group/u/emyoulation)\
**Post date:** [December 2, 2021, 9:17pm UTC](https://gramps.discourse.group/t/mixing-up-the-windows/2004/2 "2021-12-02T21:17:00Z")

</div>

I did some digging into this and found a surprise. In most cases, only the references/attachment portion is lost when you close spawned windows in the wrong order. The actual new records are still created. (Which is still painful but a bit more recoverable.)

In most cases, I can stitch the pieces together. Just hit the Share icons (to select from existing records), sort so the newest records are on top and select that missing piece. Then the ‘OK’ from the right level commits the attachment.

If there are a LOT of records that weren’t attached, I’ll go into the View for the Object type, sort on newest. If the References column is showing, I can work the ones with no references.

Like I said, painful but workable.

This bug has been reported & acknowledged: [12008](https://gramps-project.org/bugs/view.php?id=12008): Parents quietly abandon spawned newborn Objects

---

<div class="post-metadata">

**Author:** ![DaveSch](https://yyz2.discourse-cdn.com/free1/user_avatar/gramps.discourse.group/davesch/32/40_2.png) [@DaveSch](https://gramps.discourse.group/u/DaveSch)\
**Post date:** [December 2, 2021, 10:43pm UTC](https://gramps.discourse.group/t/mixing-up-the-windows/2004/3 "2021-12-02T22:43:36Z")

</div>

One of the things I try to do is as the windows open on top of another, I make sure the OK/Cancel buttons of the lower windows are not visible. Only the top window’s OK/Cancel is visible. As Gramps remembers the position of windows this works until I have to slide windows around on the monitor

This does not always prevent the selection of the wrong open window but more often than not it helps.

---

<div class="post-metadata">

**Author:** ![dlynch](https://avatars.discourse-cdn.com/v4/letter/d/848f3c/32.png) [@dlynch](https://gramps.discourse.group/u/dlynch)\
**Post date:** [December 3, 2021, 12:05pm UTC](https://gramps.discourse.group/t/mixing-up-the-windows/2004/4 "2021-12-03T12:05:23Z")

</div>

I used to get this too. My solution (on Windows 10) is to use AutoHotkey, in Gramps, to map enter to Cntl+O. Now I just hit enter enough times and the windows close in the correct order.

David Lynch

---

<div class="post-metadata">

**Author:** ![emyoulation](https://yyz2.discourse-cdn.com/free1/user_avatar/gramps.discourse.group/emyoulation/32/67_2.png) [@emyoulation](https://gramps.discourse.group/u/emyoulation)\
**Post date:** [December 3, 2021, 1:03pm UTC](https://gramps.discourse.group/t/mixing-up-the-windows/2004/5 "2021-12-03T13:03:50Z")

</div>

Although there’s another [bug related to Ctrl+O & data entry](https://gramps-project.org/bugs/view.php?id=12489).

It is a Schrodinger’s Cat bug.

If you have typed into a text field but have not ‘commited’ the field (by tabbing to or clicking on another field) then Gramps will be badly confused if it uses that field as a basis for any calculation. At that point, the field is both populated and unpopulated.

As a concrete example while you are roughing-in a family. You just added a father or child (the surname would’ve been pre-populated and the given name field is active by default) then you type the given name ‘John’ but hit Ctrl+O. Since Gramps tries to match the first word of the given name against the statistical genders for that name, Gramps becomes confused.

Gramps knows there is probably something in the field but cannot look inside to discover the name. But it HAS to check that name because it exists. But it doesn’t exist because it hasn’t been created.

Error… error… Norman, coordinate!

---

<div class="post-metadata">

**Author:** ![system](https://global.discourse-cdn.com/free1/uploads/gramps/original/1X/2ac1e712b7adf612e31ca08a55419f4cf37c0158.png) [@system](https://gramps.discourse.group/u/system)\
**Post date:** [January 2, 2022, 1:04pm UTC](https://gramps.discourse.group/t/mixing-up-the-windows/2004/6 "2022-01-02T13:04:30Z")

</div>

Closed on January 2, 2022, 1:04pm UTC

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.
