Summary: | Face assignement need to use both mouse and keyboard to name persons | ||
---|---|---|---|
Product: | [Applications] digikam | Reporter: | Michael Holtermann <linux> |
Component: | Faces-Workflow | Assignee: | Digikam Developers <digikam-bugs-null> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | caulier.gilles, jflemer, mick, ss2hmi |
Priority: | NOR | ||
Version: | 5.2.0 | ||
Target Milestone: | --- | ||
Platform: | Ubuntu | ||
OS: | Linux | ||
Latest Commit: | https://commits.kde.org/digikam/949e7bf27067af3ebb48685b31c7b58a3e57b736 | Version Fixed In: | 5.4.0 |
Sentry Crash Report: | |||
Attachments: | Small listbox, meaning of two entries is not clear |
I can confirm two distinct problems: 1) The listbox can be too small. It is resized according to the size hint, which is calculated by the current style. Apparently, if the text has two lines, only the first one is taken into account. Looks like a slight style/Qt bug 2) Mouse press is ignored. It is filtered out by KCompletionBox's event filter. At the moment, I dont know why it works in the icon view, but not in the preview. I need to do further debugging. Marcel, I confirm 1/ and 2/ For 1/, there is a another entry in bugzilla if i remember. For 2/, with Qt5 port, we will have a new way to code completion. Gilles I'm willing to work on this issue given minimal direction from core developers. *** Bug 372096 has been marked as a duplicate of this bug. *** *** Bug 372761 has been marked as a duplicate of this bug. *** Git commit 949e7bf27067af3ebb48685b31c7b58a3e57b736 by Maik Qualmann. Committed on 29/12/2016 at 11:42. Pushed by mqualmann into branch 'master'. fix completion mode for manual face tagging Related: bug 374185, bug 372481, bug 368124, bug 360155 FIXED-IN: 5.4.0 M +6 -1 NEWS M +40 -2 libs/tags/addtagscombobox.cpp M +8 -0 libs/tags/addtagscombobox.h M +7 -3 libs/tags/tagscompleter.cpp M +10 -0 utilities/facemanagement/assignnamewidget.cpp M +2 -0 utilities/facemanagement/assignnamewidget.h M +9 -1 utilities/facemanagement/facegroup.cpp https://commits.kde.org/digikam/949e7bf27067af3ebb48685b31c7b58a3e57b736 |
Created attachment 64806 [details] Small listbox, meaning of two entries is not clear Version: 2.1.1 (using KDE 4.7.1) OS: Linux Hello, to tag a person on a photo, I have to use bot keyboard and mouse to select the keyword from the listbox. It would be very nice if the workflow could be refined so that one could tag using the keyboard (entering a new name, hit enter) or mouse (type the start of an existing name/keyword, click the correct keyword and click on "OK"). See below for details. Kind regards, Michael Reproducible: Always Steps to Reproduce: 1. Draw a new face tag or select an automatically created face tag 2. Click on the edit box 3. Type a name, list box with matching keywords appears --> List box is very small, see attached screenshot. The meaning of both entries is not clear 4. Click on an existing keyword --> This does not complete the entered name 5. Complete name by keyboard 6. Hit enter --> Does not enable the submit button, I must click the listbox again Alternative: 1. Start typing 2. Ho down with cursor and hit enter on existing keyword --> List box disappear --> Input box is empty --> No name/keyword is stored Actual Results: - The listbox is useless if you wish to select existing keywords. - The listbox is too small. - Hitting "Enter" does not save the current entry. - Mouse select does not select the current entry Expected Results: - If the enter key is hit, the current entry in the input box should be saved, maybe dependent on the preferences (save changes in the right pane without notification, IIRC) --> Start typing, select entry with cursor keys, hit enter. Finish. - If an existing keyword is selected by mouse, the input box must contain that keyword, and the submit button must be enabled. --> Start typing, select entry with mouse click, click submit (or hit enter). Finish.