RmlUi/community

Michael R. P. Ragazzon

@mikke89

I see. Btw, the log messages will tell you the names of the fonts when they are loaded, which might be helpful sometimes because they need to match exactly.

VinaCC

@EngineLessCC

yes that is true

perhaps being able to set a fallback font not just for missing glyphs but missing fonts would be a great feature aswell

i would actually even suggest that, as this is typically behaviour of normal browsers to automatically use a different font/fallback incase some family is missing, reproducing this behaviour sounds reasonable plus a logmessage may tell about using fallback

Michael R. P. Ragazzon

@mikke89

Hm, I think it's better to consider non-existing font names an error. If users refer to a font that doesn't exist, that is indicative of an issue somewhere (either wrong name, or missing/error loading font file). We don't try to correct issues like that in RmlUi, unlike web browsers where compatibility is essential and correcting mistakes is the norm.

VinaCC

@EngineLessCC

errors would do aswell

Michael R. P. Ragazzon

@mikke89

Alright, I've tried to improve the error message now with the latest commit, hopefully it is much clearer now.

VinaCC

@EngineLessCC

i dont see an option to manually invalidate CompiledGeometryHandles
edit: nvm, global functions, forgot those. (took me a moment to find it in docs either)

VinaCC

@EngineLessCC

Any idea what could cause this? https://imgur.com/cT1cHxs
Ive noticed it being system dependend, mine and some testers' host systems including vms appear to have this issues, while others dont

another example: https://imgur.com/IijoHD6

Michael R. P. Ragazzon

@mikke89

Hmm, not sure. Looks like some kind of pixel alignment issues, either due to subpixel offset or scaling.

VinaCC

@EngineLessCC

it sometimes fixed itself when i changed the context dimensions, yet i dont understand the relation as some systems render these fine and others not. identical binary

Michael R. P. Ragazzon

@mikke89

Could be anything from projection matrix, viewport settings, context dimensions, and shader code.

VinaCC

@EngineLessCC

yeah im unsure what it could be atm, viewport dimensions match context, as for shader im using d3d9(Ex) funcs instead of shaders directly, projection matrix is basically taken from docs of rmlui on dx

Michael R. P. Ragazzon

@mikke89

Ah, hmm, if I remember correctly, dx9 has these half texel offset weirdness. Don't remember the details, but that should be something to investigate.

VinaCC

@EngineLessCC

yes! thank you for that tip! adding 0.5f to x and y translation fixed this

Michael R. P. Ragazzon

@mikke89

Great!

VinaCC

@EngineLessCC

are there reasons for the debuggers' element info window to be big yet empty on some elements?

Michael R. P. Ragazzon

@mikke89

It doesn't resize automatically, you can do that manually

Martin Suffer

@InactivatedPyschiopath

Hello, is there a link that I can see the supported tags, css properties and etc on rmlui?

Michael R. P. Ragazzon

@mikke89

@InactivatedPyschiopath Hi! Yes there is. See list of properties here: https://mikke89.github.io/RmlUiDoc/pages/rcss/property_index.html

And element list here: https://mikke89.github.io/RmlUiDoc/pages/rml/element_index.html

Martin Suffer

@InactivatedPyschiopath

@mikke89 thanks!

Michael R. P. Ragazzon

@mikke89

Just added a search box on the documentation site! Hope you find it useful, let me know if you find any issues :)

https://mikke89.github.io/RmlUiDoc/search.html

Martin Suffer

@InactivatedPyschiopath

thanks! it's great

Martin Suffer

@InactivatedPyschiopath

@mikke89 I'm trying to add openh264 to rmlui, do you have any plan for that?

Michael R. P. Ragazzon

@mikke89

Haven't worked with that library before, let us know if you have any issues integrating it. Most likely you won't need any specific support from the RmlUi side, the library is quite flexible so you should be able to integrate it from the client side.

Martin Suffer

@InactivatedPyschiopath

Haven't worked with that library before, let us know if you have any issues integrating it. Most likely you won't need any specific support from the RmlUi side, the library is quite flexible so you should be able to integrate it from the client side.

@mikke89 yes I could decode and get the frame I guess I will face some issues to draw them in rmlui thanks for the help I will try to make it stable and send a commit to rmlui, lightweight mp4 files are extremely helpful for creating cool backgrounds.

Jeff425

@Jeff425

Hey @mikke89 I am getting a formatting error with tables Element has a display type 'table-row-group', but is not located in a table. It will not be formatted. In element tbody < table.commandBox < body < #root#main using the style sheet from https://mikke89.github.io/RmlUiDoc/pages/rcss/tables.html and my table is formatted as such

<table class='commandBox' data-model='commandModel'><tbody>
<tr class='commandLog'><td>{{commandLog}}</td></tr>
<tr class='commandInput'><td><input type='text' data-value='commandInput' /></td></tr>
</tbody></table>

I feel like I am missing something obvious (or if I need to change the table RCSS)

Jeff425

@Jeff425

Also love that you are working on flexbox support. Such an easy to use responsive layout

Jeff425

@Jeff425

Ah it seems using position absolute was causing issues with the table

Michael R. P. Ragazzon

@mikke89

@Jeff425 That's not the best error message for this situation certainly! Which element had absolute position here, the table or tbody?

Thanks, just getting started on flexbox, we'll see how it goes :)

Marco Bortolin

@barotto

I had the same problem with a table with absolute position. I solved putting the table inside an absolutely positioned div.

Michael R. P. Ragazzon

@mikke89

@barotto I don't think I ever tested this actually, thanks for letting me know, I'll see if we can do something about it. At least a better error message.

Marco Bortolin

@barotto

Just did a cursory look at the source code and I think this happens because when a table has absolute position it's added as such in the block_context_box obj but the LayoutTable::FormatTable won't be called in the CloseAbsoluteElements, and table elements have Static position. I also think putting a table inside an absolute div would be an acceptable workaround for this case, assuming a more meaningful error message is also provided.

Michael R. P. Ragazzon

@mikke89

Ah, right, that's a good start. I'll see if there is an easy fix, otherwise I'll just resort to an error message with your work-around.

Jeff425

@Jeff425

@mikke89 sorry for late reply, yeah it was on the table element itself. I ended up doing the same thing barotto did to fix it

Jeff425

@Jeff425

Is it possible to drag an element from one document to another?

Jeff425

@Jeff425

Another question: Is it possible to remove focus when ProcessMouseButtonDown(0, ...) is called while hovering over no elements?

Jeff425

@Jeff425

And no documents

Michael R. P. Ragazzon

@mikke89

@Jeff425 Yes, see the drag sample to do exactly this. As for removing focus, you could always do that manually if nothing captures the mouse button.

Jeff425

@Jeff425

Thanks for the help so far! Using context->GetRootElement()->Focus(); to remove focus but that doesn't seem to remove it from the input element. UnfocusDocument doesn't work as I don't have any other document on the screen. Is there an easy way to remove focus that I am missing?

Michael R. P. Ragazzon

@mikke89

@Jeff425 No problem! :) I think you have the right idea with context->GetRootElement()->Focus();, this is where I would start. Not sure why it doesn't work, maybe we want to do something about that? You could also try Focus on the document itself.

Jeff425

@Jeff425

I created a blank document to focus for this situation. That seems to work. Would be nice to have a context->Unfocus(); function as an easier solution

Marco Bortolin

@barotto

Hi, I need to use full paths with <img src="..."> on linux, is there any particular reason why the leading '/' character is stripped off from the texture source?

Michael R. P. Ragazzon

@mikke89

@barotto Yeah, I don't know why that is done. The whole path handling could need some improvements, see this issue here: mikke89/RmlUi#161

I'm also a bit afraid of changing it, because it will for sure break things for someone. And there are plenty of workarounds.

Marco Bortolin

@barotto

Thanks for the pointer. I can see why you're reluctant to touch the whole path handling system, that would certainly be a can of worms... For now I'll fix this issue adding an extra '/'.

Michael R. P. Ragazzon

@mikke89

Appreciate your understanding :)

Where communities thrive

People

Repo info

Activity