Add an indicator to the Footprint Properties dialog that the 3D model file doesn't exist (or there's some other problem)

Description

Having a file path for 3D model in every footprint in the official library even when there's no corresponding 3D model file may be a technically good idea, but for usability, especially for newcomers, it's confusing. Many times someone has wondered in the user forum why they can't see the 3d model even though it should be there.

This would be helped if there was some visible indicator in the Footprint Properties dialog's 3D Settings when the file doesn't exist or there's some other problem.

There are many possibilities. My own proposition would be to have an extra column in the "3D Model(s)" table. It would have an icon showing the status of the file: OK, warning, error (as they often are, a green tick mark, a yellow triangle with exclamation mark, a red circle with exclamation mark). Hovering over or clicking that mark would show the problem explanations. Warning could be shown if the model exists and can be read and rendered but has some problem (for example there's a STEP model but the model is scaled). Error would be shown if the model doesn't exist or can't be read or interpreted or rendered.

KiCad Version

Application: Pcbnew
Version: (5.99.0-791-g06c979dfa), release build
Libraries:
    wxWidgets 3.0.4
    libcurl/7.66.0 OpenSSL/1.1.1d (Schannel) zlib/1.2.11 brotli/1.0.7 libidn2/2.2.0 libpsl/0.21.0 (+libidn2/2.1.1) nghttp2/1.39.2
Platform: Windows 8 (build 9200), 64-bit edition, 64 bit, Little endian, wxMSW
Build Info:
    Build date: Jan 24 2020 06:58:39
    wxWidgets: 3.0.4 (wchar_t,wx containers,compatible with 2.8)
    Boost: 1.71.0
    OpenCASCADE Community Edition: 6.9.1
    Curl: 7.66.0
    Compiler: GCC 9.2.0 with C++ ABI 1013

Build settings:
    KICAD_SCRIPTING=ON
    KICAD_SCRIPTING_MODULES=ON
    KICAD_SCRIPTING_PYTHON3=OFF
    KICAD_SCRIPTING_WXPYTHON=ON
    KICAD_SCRIPTING_WXPYTHON_PHOENIX=OFF
    KICAD_SCRIPTING_ACTION_MENU=ON
    BUILD_GITHUB_PLUGIN=ON
    KICAD_USE_OCE=ON
    KICAD_USE_OCC=OFF
    KICAD_SPICE=ON

added priorityundecided statusnew labels

added 3d-viewer pcbnew prioritywishlist statusconfirmed labels and removed priorityundecided statusnew labels

I don't know about the warnings, since I don't know if there is a good way of doing the STEP file comparison (or if that is something we should do), but definitely indicating if the file exists/is readable is good. Perhaps that could also just be by turning the text/row color red to make it stand out.

I thought about row color, too, but alone it doesn't necessarily tell much to the user. The three symbols I described are familiar from many other programs. But the point is to tell the user as clearly as possible that there is some problem and therefore the 3D view doesn't work as expected.

KiCad can use .wrl and .step files and I believe can detect some problems because the files are necessarily read and parsed somehow. It's done with a 3rd party library, I think, but doesn't it give some error messages or codes when it can't parse files or finds problems in them? I don't understand what you mean by "STEP file comparison".

Isn't the fact that there is no 3D model visible in the 3D model viewer indication that the 3D model wasn't found? I certainly want to avoid nagware in this case.

I don't think it's nagware. Throwing messages at a user is nagware. Adding a visible mark telling that there's something wrong isn't nagware. Actually this would be replacing nagware. As far as I know previously messages were shown in some situations which were nagware. In my proposition the user has to move the mouse onto the icon on purpose to see the message. I carefully crafted the proposition that way to avoid nagware.

The problem is real, based on following the user forum and seeing several questions about non-visible 3D models. Really, people don't expect to have a file name in the official library footprint and yet not to have the file. (EDIT: To put this in other words: they expect to have the file if there's a file name. Therefore they presume there's some bug or something else wrong when they don't see the 3D model in the view.)

It's also possible to go wrong with file path text (it has happend to me, I just don't remember anymore how, but it involved a path variable or alias). There's no indication of what went wrong.

Also it's possible to have a step model which is internally offset so far that it's not visible at all. I would appreciate some information in that case, too. At least that would be possible with the same mechanism even if that exact situation wouldn't be taken care of in the implementation.

There just are so many things which can go wrong and cause confusion because KiCad doesn't give any information. My proposition was meant to show a way to give that information - whatever it is - without being nagware.

The problem is real, based on following the user forum and seeing several questions about non-visible 3D models. Really, people don't expect to have a file name in the official library footprint and yet not to have the file. (EDIT: To put this in other words: they expect to have the file if there's a file name. Therefore they presume there's some bug or something else wrong when they don't see the 3D model in the view.)

Simply telling them that the file is missing isn't going to stop the user questions, it will just change them. Instead of asking why they can't see the model, they will ask why it is telling them that a model doesn't exist for that footprint even though the footprint was provided by KiCad (and they will start questioning if they are missing files from the install, etc.). In that case, this information would simply be pushing the user into a different question/complaint.

It's also possible to go wrong with file path text (it has happend to me, I just don't remember anymore how, but it involved a path variable or alias). There's no indication of what went wrong.

I agree, in this case it would be nice to have some more diagnostic information, such as having a tooltip available that could say something to the effect of Unknown alias/Model not found at {full path}/etc. depending on the reason for the error.

Also it's possible to have a step model which is internally offset so far that it's not visible at all. I would appreciate some information in that case, too. At least that would be possible with the same mechanism even if that exact situation wouldn't be taken care of in the implementation.

This is bordering on the territory I think we should avoid entering: making internal assumptions about how a model should be made/designed and checking for those. While the KLC might specify models to have no offset/scaling, we should not be hard coding the KLC into the main program since users/companies can have their own requirements. When this was brought up in #2285, it was essentially agreed it should be plugin based. For things like this modifying the options available on the 3D preview might work, for instance adding a "zoom to model" button that would automatically find the model and zoom to it (then we could also display coordinates to the user, so they can see it is offset).

mentioned in issue #1781 (closed)

Accidentally, once again today... https://forum.kicad.info/t/dead-links-on-kicad-3d-models/20889

EDIT: and a bit later: https://forum.kicad.info/t/3d-model-of-3-5mm-screw-terminals/20972/5

EDIT: again: https://forum.kicad.info/t/possible-bug-default-library-incorrect-footprints-associated-with-symbols/23362/8

From my reading of that thread, the user is confused by the fact that a 3D model is specified, but the library doesn't contain one in its current release (which is the question/error that I think we would be funneling people towards if we aren't careful).

I went hunting, and found the forum thread that seems to have spun-off this issue: https://forum.kicad.info/t/replacing-default-kicad-3d-model-with-an-external-model/20860 (please correct if I am wrong). I agree with some of the points in that thread (on both sides), such as having better indications that a file is missing (this has happened to me, where I have actually typed the path wrong so it didn't appear or I have used an incorrect alias).

The main issue I have with this is the suggestion that we include something like The official KiCad library doesn’t have 3D model for every footprint but still has a 3D model file path defined for every footprint. as an error message. How do we define what an official library is to the code?

• We can't just go by the library names, since it is entirely possible for users to make a Connector footprint library if they don't use the official one (I came close, but added on the plural to get Connectors because I liked it better), and in those cases we can't use that error because it is nonsensical. This would also mean we have to update the code if new libraries are added/a reorganization occurs.
• We can't just go by the path they are located in, since they can be downloaded into other places or installed differently depending on user tastes.

We could introduce library metadata of sorts, that would contain information on the library location, creator, official status, etc; but I don't know if that would actually allow us to fix the problem in a better way.

added fp-editor label

mentioned in issue #4567 (closed)

Again: https://forum.kicad.info/t/3d-model-not-showing/24172/8.

Let me explain this from another viewpoint.

Ian said:

Simply telling them that the file is missing isn't going to stop the user questions, it will just change them. Instead of asking why they can't see the model, they will ask why it is telling them that a model doesn't exist for that footprint even though the footprint was provided by KiCad (and they will start questioning if they are missing files from the install, etc.). In that case, this information would simply be pushing the user into a different question/complaint.

Which question is easier to answer:

• "Why I don't see the 3D model for this footprint?" or
• "Why doesn't KiCad have the file even though there is the model name in the footprint?"

With the first question we can't know what is going on before we ask more questions about the situation. With the second question we can immediately know what is happening and what is the answer. I would rather answer the second question. It's very frustrating to go back and forth with questions and counterquestions every time somebody is perplexed about the lacking model. For some reason people don't even seem to answer the simple question "did you check if the file is actually in your file system".

We could write a FAQ article about non-visible 3D models, but we should then explain everything which can go wrong. It would be so much easier to just explain shortly that in the official library there are 3D model file names in the footprints to enable easier adding of the models without need to modify the footprint files -- if the UI just would show the user an indicator about potential problems with a specific model so that they could ask the right question.

I was one of the guys that was initially confused about this and agree that having some information about the 3D models would be good.

I'd be happy to work on this after finishing my CADSTAR importer (I still need at least 5 more weeks on that) Hopefully I have time get this in just before v6 feature freeze - however I can't see any agreement in this issue on how to proceed?

I propose "error" tooltip wording below:

3D Model does not exist.

Please note KiCad Official libraries include a link to a 3D model file that might not yet exist. Refer to KiCad Library Convention F9.3 paragraph 1

Help improve our libraries by contributing!

Also, I agree with Ian that it would be a bit too complicated to show the "warning" error if the 3D model is scaled. That could be a future improvement (separate issue / wish).

EDIT: the text about KiCad Official libraries could only be shown when the path start with "${KISYS3DMOD}"

I think the simplest thing is to just warn when a path is not found. No extra logic to try to detect the special case of the official libraries. The notice that if you don't download or install the 3D models for the official libraries that you will see these error messages should go on the forums, documentation, and maybe KiCad website -- I don't think we need to include it in the software itself.

Isn't the fact that there is no 3D model visible in the 3D model viewer indication that the 3D model wasn't found?

Not specific enough for efficient troubleshooting, as Eeli points out. Using an infobar or something to show the warning (and the path we are looking in) I think would not be a nuisance.

mentioned in issue #5360 (closed)

added to epic kicad&6

Isn't the fact that no 3D model is shown in the preview panel a pretty good indicator that the 3D model cannot be found? I suppose we could add some other indicators but this dialog panel is already pretty busy.

I think being able to tell apart a missing file vs. a file that shows up as empty due to a problem is useful

I was going to see about adding a column to the front of the rows that will display a simple error symbol when the file can't be found (I was going to do this along with updating the library grids to do the same thing). Then we can also have a tooltip when hovering over that icon to give an actual message. That column will be hidden normally if there are no issues.

changed milestone to %6.0.0-rc1

assigned to @imcinerney

Here's a fresh topic in the forum: https://forum.kicad.info/t/3d-models-not-rendering/24651

ATM there's only the initial post there and I have no clue what the problem might be. It's these kinds of situations where some indicator would be good.

Yet another user who thought that because there's a model name defined there would exist the model file, too: https://forum.kicad.info/t/missing-kisys3dmod-package-csp-3dshapes/24974/3?u=eelik

changed title from Add an indicator to the footprint Preferences dialog that the 3D model file doesn't exist (or there's some other problem) to Add an indicator to the Footprint Properties dialog that the 3D model file doesn't exist (or there's some other problem)