gluing lines at connection points

[Yacine]

Hi HiddenLayer,
welcome back.

You could not find the menu of the 2nd picture, because it is one I customized myself.
Till Visio 2007 the smart icons where organized logically and you knew where to look.
Those ribbbons are really bad.

I enclose this menu as attachment, so you won't need to search for too long.

[Yacine]

As for the connection types, their behaviour is logical too.

Type 0 is for incoming connections. That is when a connector (or a 2D-Shape) tries to connect to myself. With the DirXY fields I tell it to orient itself according to my values.

Type 1 is when I (shape) try to connect to another shape. That is, I orient myself according to DirXY.
When such a point is set to 1 it won't function as connection point for connectors trying to connect to myself.

Type 2 is the combo of 0 and 1. ( 0+1 = 2 )

I updated your drawing with the yellow and blue discs. They show the a.m. described behaviour.
Try pulling separately a yellow or a blue one to see the gluing.
Take a blue one by either its left or right side and try gluing it on the right or the left side of a yellow disc.

I hope it is somewhat comprehensible.

Y.

[wapperdude]

Here is formal definition of DirX,Y funcionality:
"In non-extended rows the DirX and DirY cells together define a direction vector that influences the rotation of shapes involved in connections using the connection point. If both are zero the point is directionless. Connection points are of the following types:

Inward (0), which means that shapes glue to them. This is the default.

Outward (1), which means these connection points will glue to inward connection points.

Both inward and outward (2), in which case the direction is the inward direction, which is reversed if used as an outward connection."

See this for more info:  https://docs.microsoft.com/en-us/office/client-developer/visio/connection-points-row-connection-points-section

[Yacine]

Thank you for tidying up.

[hidden layer]

Hi,
thanks so far. I didn't got the whole story but I can work with.

I read those articles at MS but there's something that I don't understand completely:
at https://www.oreilly.com/library/view/microsoft-visio-2013/9783848330416/ch15s16.html is listed that these values (DirXY) are for dynamic connectors only and size doesn't matter. But it does as I experienced).

at https://docs.microsoft.com/de-de/office/client-developer/visio/dirxa-cell-connection-points-section it is mentioned that these two values give a vector.
But there is no vector (aside from top, bottom, left or right).
the rare information there is really MS-like. 

This gluing shapes (your circles Yacine) together is currently not in my focus.

Let's go to a specific problem:
I have a vertical line.
At the top of this line have a connection point.

Regardless of the values for DirX, DirY and type the connector goes to the left or to the right - just from where the connector comes. Only if the other end of this connection line is almost precisely over this line it goes upwards.
Even if I give 0 to x and a million to y (which is the mathematically correct way imho) it goes left or right.

It seems to be impossible to force the direction of a connector that this goes upwards first and after a (specific?) distance sidewards. (I use static connectors only - from connection point to connection point).

as a workaround I replaced these lines with rectangles (with a connection point in the upper middle) , set GlueType to 8 and this works. The funny thing is if I give 1 to DirY it goes downwards, at -1 upwards. What strange logic behind?

You see I have lots of fun here.

Thanks for any hint

cheers!

[Yacine]

...
Let's go to a specific problem:
I have a vertical line.
At the top of this line have a connection point.

Regardless of the values for DirX, DirY and type the connector goes to the left or to the right - just from where the connector comes. Only if the other end of this connection line is almost precisely over this line it goes upwards.
Even if I give 0 to x and a million to y (which is the mathematically correct way imho) it goes left or right.

It seems to be impossible to force the direction of a connector that this goes upwards first and after a (specific?) distance sidewards. (I use static connectors only - from connection point to connection ...

Can you upload a drawing?


Maybe we need to transform the line into a 2D-shape.

[hidden layer]

Hi,

here it is. Thank you!

[Yacine]

Hi,
The edge points can't have a vertical and a horizontal orientation at the same time. You need to chose one. May be you could put two points close the edges and then choose one of them.

But the group shape itself did not have any connection point anyway. I don't know what you were expecting.

I added the missing points. cf red connection-shape.


----


I'm going to ask for a payment if it continues this way  . (Maybe a donation to Chris' forum)

[hidden layer]

Hi Yacine, thanks for your patience,

Ok, I'm fine with one direction only. Left to left, middle to top etc.

The group does not have connection points because the group ist just the frame of the 'terminal' the symbol is representing. It interacts in some bigger context (see here: http://visguy.com/vgforum/index.php?topic=8963.0).

If some of the lines above is 'active' all lower lines become 'active' and connected lines transport this information to further participants. This was the result of some rainy weekends in a hotel. Not perfect but it works. Now I started to make it a bit more 'nice'- y'know.

As mentioned above I replaced the lines with rectangles (in almost the same size) and this works. I was just wondering about and try to get a clue out of it For further experiences...

I read a lot of posts about this connecting thing but everyone who's asking run into the same situation that the did not get the system behind.

Does this answer your questions a little bit? Or at least make my nervous costume present?

again: Thanks!

[hidden layer]

hmmmm....

after fixing some problem I've illustrated to myself the behaviour and dependencies og th egluing angle of 2D-shapes.

For whom it may concern.

cheers & cu
hl