Search for Connections Mod

[mjaro]

2 minutes ago, fluffy82 said:

If you need Dutch input, let me know!

Of course YES! I'ii be very glad to include Dutch translations.

Michał

[mjaro]

37 minutes ago, Ken Roy said:

I am not sure about your box shadow since it is not shown on all boxes. 

Wrrrr. You are right. There is one line missing in the cfg (to be copied to genstyle)

Aftter #pathlist .pers {

should be  before its closing }

    box-shadow: 4px 4px 4px var(--cl1);

Edit: And the color variable  definition should be put earlier

    --cl1: #bba28c; /* box shadow */

Sorry  

Edited February 18, 2021 by mjaro
missing --cl1

[Michel KIRSCH]

4 hours ago, mjaro said:

Is it possible that the warning is dependent on the PHP version?

Most of the time, this is server or ISP dependent.

Michel

[tngrlkrz]

Michal,

Shadows now working fine, but no shadows on the top boxes; essentially those with connecting lines, which happens only when there is step up from a box and step down  to a box from an upper box. You likely avoided in order to keep the lines are visible.  TNG's other charts can shadow all boxes because the shadow is less, and the vertical space between boxes is not as small.  So it's understandable. 

I tested with a 3px instead of 4px shadow and it looks a bit better in the compressed diagrams.

And '*', though not so noticeable, but better than the '@', but should be easy to modify, thanks!

Ron

[tngrlkrz]

Some sort of issue with v3a.  With a close relationship, such as grandchild to grandparent, getting a 'hang' on the connection search.

 

[mjaro]

1 hour ago, tngrlkrz said:

And '*', though not so noticeable, but better than the '@',

I've found even better alternative: ● 

[mjaro]

57 minutes ago, tngrlkrz said:

Some sort of issue with v3a.  With a close relationship, such as grandchild to grandparent, getting a 'hang' on the connection search.

 

I see. I had left direct line unchanged after changing other cases. In 10 minutes I'll supply corrected file.

[mjaro]

8 minutes ago, mjaro said:

I see. I had left direct line unchanged after changing other cases. In 10 minutes I'll supply corrected file.

Here it is. 

Sorry for errors, thank you for testing!

connections.php

[tngrlkrz]

That fixed it, direct connections now fine. Thanks.

[mjaro]

2 hours ago, tngrlkrz said:

no shadows on the top boxes; essentially those with connecting lines.

Is it still bad? It is not intentional.

Ron, could you please look at the browser if such a box receives css from genstyle.php  #pathlist .paths  (the last item in it is the shadow)

For me it gets it and looks as other boxes.

[mjaro]

5 minutes ago, manofmull said:

The boxes are not hyperlinked, just the names

The correction is still on the wait list  

[tngrlkrz]

1 hour ago, mjaro said:

Is it still bad? It is not intentional.

Ron, could you please look at the browser if such a box receives css from genstyle.php  #pathlist .paths  (the last item in it is the shadow)

For me it gets it and looks as other boxes.

No not bad at all.  Yes, coming from genstyle.php.  Shadows are only absent if there is a pair,  person and spouse, side by side below, requiring connecting lines to those two boxes. Otherwise the top box is shadowed.  I am fine with that.

[tngrlkrz]

25 minutes ago, manofmull said:

TNG obsession with shadows!

But you expect mod authors  to offer an option when TNG does not?  I would want the mod to have the same look, as Michal states it.

[Kare]

1 time siden, manofmull skrev:

FWIW I'm getting shadows on some but not all boxes. I happen to think they're not needed at all.

Have you considered what the USERS of your website want?

[tngrlkrz]

Michal,

Here's an example of when a top box has shadow and when it doesn't.  If there's no step up and back down (1st diagram below), then no shadow on the top box.  This is understandable and likely appropriate because the connecting lines might be obscured by the shadow.  BTW...in the example, the shadow had been reduced from 4px, to 3px, which I think looks better.

BTW...I know, colors don't match...I intentionally using the 'out of the box' mod for this example, except for the shadow reduction.

[tngrlkrz]

55 minutes ago, manofmull said:

Why is it I'm jumped on when I make a suggestion?

Not the intent to make you feel that way. But you came with such strong words 'severely loathe' that can generate some reaction. 😑  I just know so many charts are shadowed, and you will need a lot of mods or changes to the numerous charts that have TNG shadows.  And thanks for suggesting draggability to Darrin,  and also being able to zoom them as well. Liked that so much I authored a small mod to do that my way with mouse wheel zoom.

(and yes, in private mod, can remove the containers, and also give an option to users to remove the photos for a condensed look).

 

[tngrlkrz]

1 minute ago, manofmull said:

mini photos

Don't misunderstand.  I like having images as well, but on many of my pages with images of any sort, for a more compressed look, I present the option  to view the page either way.

[mjaro]

10 hours ago, tngrlkrz said:

Michal,

Here's an example of when a top box has shadow and when it doesn't.  If there's no step up and back down (1st diagram below), then no shadow on the top box.  This is understandable and likely appropriate because the connecting lines might be obscured by the shadow.  BTW...in the example, the shadow had been reduced from 4px, to 3px, which I think looks better.

BTW...I know, colors don't match...I intentionally using the 'out of the box' mod for this example, except for the shadow reduction.

Ron, in the connections_v13.0.2.3a.cfg there is lack of definition of --cl1 color (to be copied to genstyle).

In the genstyle.css I see you've added it "by hand" for .pers (am I right?), and the same is needed for .spantxt. (finally, it should be defined just before --cl2).

Edit: I see it wasn't explained clearly in my older post starting with "Wrrrr". Mybe it's time to release an update ? (or next "beta"?)

Michał

[tngrlkrz]

21 hours ago, mjaro said:

Aftter #pathlist .pers {

should be  before its closing }

    box-shadow: 4px 4px 4px var(--cl1);

Edit: And the color variable  definition should be put earlier

    --cl1: #bba28c; /* box shadow */

Hi Michal,  ..I followed your instructions and the shadows appeared  as I showed in my last snapshot.  Not sure what you meant 'by hand' since I put them into the cfg to be copied to genstyle.css.  However, I did reduce the shadow by 1px.

#pathlist .pers {
    min-width: 110px;
    max-width: 250px;
    border: 1px solid black;
    border-radius: 5px;
    text-align: center;
    background-color: var(--bg2);
    --cl1: #bba28c; /* box shadow */    
    box-shadow: 3px 3px 3px var(--cl1);

#pathlist .spantxt {
    display:inline-block;
    min-width: 200px; /* prawie 2*pers */
    padding: 1px 5px 1px 5px;
    border: 1px solid black; /* here is visible border for parent */
    border-radius: 5px;
    background-color: var(--bg2);
    box-shadow: 3px 3px 3px var(--cl1);

Is there more needed? Things looked good to me, except for the unshadowed box.

 

[fluffy82]

Not sure if it is an "error" in the mod, or if I just misunderstand the use of it.

When I search for example between my (maternal) 2nd GG-father and my (paternal) 2nd cousin 4x removed, with "max. steps" and "max. results" set to 99, the mod gives only one result (path): going down from person 1 to my mother, then from my father upwards to my 5x GG-father, and back down to person 2. That is indeed the shortest path between them.

But... They are also related through my 26x GG father, who is a direct ancestor of both. They are only 44 steps apart, so less than the max. of 99. Why doesn't this show as an existing connection? And even though I put "99" steps as maximum, the result mentions that it stopped searching after 16 steps (the one it found).

Are direct relationships intentionally ignored?

[mjaro]

53 minutes ago, tngrlkrz said:

Is there more needed?

Still .spantxt lacks of the --cl1 definition. Set it either locally as you did in .pers, or better -what was intended- in the #pathlist (to be visible to all nested elements including .pers and .spantxt).

Sorry for that my stupid omission this is causing so much troubles  

[tngrlkrz]

2 minutes ago, mjaro said:

Still .spantxt lacks of the --cl1 definition. Set it either locally as you did in .pers, or better -what was intended- in the #pathlist (to be visible to all nested elements including .pers and .spantxt).

Sorry for that my stupid omission this is causing so much troubles  

No worries.  I am puttering around with TNG 13.0.3 mod issues (your mod is OK) right now.

[mjaro]

58 minutes ago, fluffy82 said:

Are direct relationships intentionally ignored?

No, they are not ignored. Probably 99 search runs is just not enough. Please check in the search summary (below the last diagram) how long the longest checked path was (I guess <<26). Maybe 999 runs will be enough?

[tngrlkrz]

OK Michal,

Finally got all boxes with shadows! Sorry it took me so long to get it right.  Looks good.  Only mild suggestion would to put a little more vertical space between boxes, but not necessary.  Even a 3px shadow nearly reaches the box below.

 

[mjaro]

I've just released mod version 13.0.2.3a. 

Most of the suggested changes has been introduced, incl.Dutch translation.

Big thanks for all helpers here end "my beta-testers" ! 

Michał

P.S. I still have a long ToDo list, and wait for next improvement proposals!