It's pushed on master and works fine: https://build.kde.org/job/Applications/job/kajongg/
Should I cherry-pick it to stable branch?
I can't close the revision too (maybe missing an approval?)

Cool :), j'ai juste fait quelques remarques de typo, je ne suis pas à même d'installer et tester :). Bon courage !

Replace spaces by underscores
Use relative filename instead of just filename to avoid duplicates target name rules

This one should also fix https://bugs.kde.org/show_bug.cgi?id=376303
No need to have a full path one the rule name, it's not used elsewhere.

pushed in https://commits.kde.org/gcompris/160543e4075a7765609c1f819c3a25c60f58745a
Thank you

In D22548#508823, @echarruau wrote:

Hi,
Not sure I agree with Johnny and Timothée.
I think there is as much interest in selecting the next word in alphabetical order that there is in selecting the next number. If you start with a word like "action" then present "zebra" then "boat" then "coat" then "three" then "dount", if your pupil is able to skip "zebra" and "three" then it is a nice activity.

@echarruau @timotheegiet
For me, there is no pedagogic usefulness to finding words in the good order in alphabetical_order: the child does not see all the words at start so it's just going on all the words expecting it's the first one to find (especially when showNext is false).
I think we should think on how to improve the pedagogic side of these activities instead of adding dataset (but not on the scope of this gsoc)...

last MR to be pushed: https://gitlab.com/fdroid/fdroiddata/merge_requests/5129 but GCompris is already in the store now

Thank you, it's merged in commits https://cgit.kde.org/gcompris.git/commit/?h=multiple_dataset&id=199814b94aa2f637dbe74172d480f8c406184e6c and https://cgit.kde.org/gcompris.git/commit/?h=multiple_dataset&id=d48f35b2b149ffb4a69ab3c301173d4fca0a7187

@AkshayCHD I was trying to merge it on the multiple_dataset branch but I see wordsgame activity does not have the configuration, can you update it? It is a derivate of gletters

According to https://forum.qt.io/topic/66783/textfield-and-keyboard-on-android, it seems the android keyboard pops up if we call forceActiveFocus.
I don't have a physical keyboard for android but can they test if this patch:

@echarruau can you please check the difficulty of the datasets and set the corresponding levels in "stars"?
For example, for money_back, at last level, Tux has 100€ and needs to buy 4 elements (21€, 59€, 50€ and 68€) and I'm not sure that it is not too complicated (at least for a 3 stars dataset).
Same for money_cents and money_cents_back where there are also cents (so 49.06€, 164.09€, 25.02€ and 545.08€ for example) which seems quite difficult without calculator

The diff is not complete (missing datasets for example). Just one point to fix, on last dataset of minus-hat, it's up to 10000 not 1000.
@echarruau , for higher levels, would it make sense to have an hint on the numbers ? When the coefficients are displayed, and the stars are displayed, it maybe difficult for children to count exactly?

@echarruau can you test this when you have some time to see if it is good on the pedagogic level? Instead of applying the diff, checkout the branch gsoc_akshaychd_multiple_datasets where the work is done

Good job for the activities.
There is an issue when running gletters or wordsgame:
qrc:/gcompris/src/activities/gletters/Gletters.qml:89: TypeError: Type error

https://commits.kde.org/websites/gcompris-net/5f862f4d1f2e986d2d6784d7ee9136fe0b581b08 removed the dependency for the language list

For 6, it's a regression I added with multiple dataset, I'll handle it.

I let the other mentors do a complete review, I only point out one improvement that would be great to have