Following up, I think it's important to understand the places the importer is used for my team (and probably others who use a mix of tools)
The importers are rarely used as a "one shot" tool to convert a design permanently from a different CAD tool to KiCad. Instead, we use the importers persistently. For example, we directly add non-KiCad libraries to the table (which triggers parsing them in many different places, as Seth mentioned). But, we also use KiCad as a viewer for non-KiCad board and schematic files. We open these files over and over again, and having that open process blocked by a dialog is a real interruption. I would expect dialogs to only appear when a serious error has happened that is not recoverable. Information that KiCad does not support a certain feature from an external file is not even an error, let alone a serious one. -Jon On Sat, May 4, 2024, 10:20 Jon Evans <[email protected]> wrote: > I agree that the information should be exposed somewhere. But I think > interrupting the load process with a dialog is a nag. It gets in the way of > using KiCad as an efficient way to view files designed in other programs. > > On Sat, May 4, 2024, 09:52 Jeff Young <[email protected]> wrote: > >> I don’t agree. A nag dialog would be something that told you we don’t >> have an editing feature that Altium supports, not that we can’t express the >> same end result as we found in the Altium file. >> >> There are certainly border-line cases. Complaining about not being able >> to import a datum dimension is a bit closer to an editing feature. But if >> it’s something like pad-stacks, then I’m going to have to make changes >> somewhere to get the design right. So I’d rather know about those. >> >> > On 4 May 2024, at 14:29, Jon Evans <[email protected]> wrote: >> > >> >> The only place we disagree is what to do in the meantime for >> schematics and boards. >> > >> > Using wxLog to pop up dialogs reporting that KiCad does not support >> > some feature of some other EDA tool is inappropriate in my opinion. >> > It goes against the KiCad guideline of not showing nag dialogs. I >> > don't think boards/schematics are any different from libraries in this >> > regard. >> > >> > On Sat, May 4, 2024 at 9:24 AM Jeff Young <[email protected]> wrote: >> >> >> >> We do report context info (at least line number and offset) in some of >> the other importers. >> >> >> >> I agree that the system could be improved. >> >> And I agree that we have to do something about the library importing. >> >> >> >> The only place we disagree is what to do in the meantime for >> schematics and boards. >> >> >> >> Cheers, >> >> Jeff. >> >> >> >>> On 4 May 2024, at 13:04, Jon Evans <[email protected]> wrote: >> >>> >> >>> Jeff, >> >>> >> >>> In both cases I disagree that the current warning system should >> >>> appear. The warning system we have today is really not that helpful, >> >>> and I don't want to bring it back the same way for boards or for >> >>> libraries. >> >>> The advanced_config option was something Seth and I discussed as a >> >>> simple stopgap for V8 if we are looking at something better for V9. >> >>> >> >>> A more helpful warning system: >> >>> - Would not use pop-up dialogs that block the UI to let people know >> >>> that any warnings exist (instead would show up in the info bar, the >> >>> same way our previous-format-version warning does) >> >>> - Would have a more rich system of logging, so that people always get >> >>> context about what part of their file had an issue >> >>> >> >>> I think if we want to show this info, we need a new kind of REPORTER >> >>> that can capture context rather than just a string, and a "post-import >> >>> report" dialog launched from the infobar. >> >>> >> >>> The library loader can keep track of these on a per-library basis so >> >>> that the infobar can be shown in the symbol/footprint editor after >> >>> launch. >> >>> >> >>> -Jon >> >>> >> >>> On Sat, May 4, 2024 at 4:53 AM Jeff Young <[email protected]> wrote: >> >>>> >> >>>> Presumably we can tell the difference (in the code) between a >> schematic or layout import and a library import? >> >>>> >> >>>> It would seem to me like a schematic or layout import should show >> the warnings (even without an advanced setting or the like). >> >>>> >> >>>> Personally I also think the library importer should show the >> warnings the very first time. Maybe we could do that by setting a plugin >> option (m_warningsShown or something) after reading the library? I think >> those are saved per-row in the lib table, but I’m not 100% sure. >> >>>> >> >>>> Cheers, >> >>>> Jeff. >> >>>> >> >>>> >> >>>> On 4 May 2024, at 09:47, 'Stefan Weber' via KiCad Developers < >> [email protected]> wrote: >> >>>> >> >>>> Hi, >> >>>> >> >>>> I recently started a discussion with @Jon Evans in my MR >> https://gitlab.com/kicad/code/kicad/-/merge_requests/1935 regarding the >> visibility of Altium importer warning & errors for not yet supported >> schematic and layout features. >> >>>> >> >>>> Since commit >> https://gitlab.com/kicad/code/kicad/-/commit/83be5ba63d82e237b8944ee6ee7c639215b14027 >> the Kicad Altium Importer does no longer shows warning & errors for not yet >> supported schematic and layout features in release builds. >> >>>> >> >>>> As Jon mentioned this was implemented to get rid of those annoying >> warnings which pop-up every time you use an Altium library which contains a >> not yet supported schematic and layout features. This I absolutely >> understand. >> >>>> >> >>>> I also do not expect that importing data from other EDA tools will >> be 100% accurate. But as the importer already knows it's limits I >> personally would like to have the possibility to get those warnings and >> errors back when I am importing an Altium schematic or layout. >> >>>> >> >>>> Jon mentioned: >> >>>> >> >>>> "I would be fine with adding some new way to show these optionally, >> in a less obtrusive way." >> >>>> >> >>>> "We try to avoid settings wherever possible. It is best to come up >> with a solution that can be enabled for everyone, meaning that people who >> want the information can get it in a way that does not slow down people who >> don't want the information. >> >>>> One option would be to create a "post-import report" that can >> optionally be launched from an infobar notification, for example." >> >>>> >> >>>> "One pretty simple option would be an advanced_config flag that >> changes the behavior of the WXLOG_REPORTER to use logging calls that are >> present in release builds" >> >>>> >> >>>> I am curious what's the opinion of others. >> >>>> >> >>>> Best regards, >> >>>> Stefan >> >>>> >> >>>> -- >> >>>> You received this message because you are subscribed to the Google >> Groups "KiCad Developers" group. >> >>>> To unsubscribe from this group and stop receiving emails from it, >> send an email to [email protected]. >> >>>> To view this discussion on the web visit >> https://groups.google.com/a/kicad.org/d/msgid/devlist/f49a2c0b-5a6f-4ed9-a40d-20f2a385d0een%40kicad.org >> . >> >>>> >> >>>> >> >>>> -- >> >>>> You received this message because you are subscribed to the Google >> Groups "KiCad Developers" group. >> >>>> To unsubscribe from this group and stop receiving emails from it, >> send an email to [email protected]. >> >>>> To view this discussion on the web visit >> https://groups.google.com/a/kicad.org/d/msgid/devlist/2533D70F-B95B-4548-97C7-335F4F9944CF%40rokeby.ie >> . >> >>> >> >>> -- >> >>> You received this message because you are subscribed to the Google >> Groups "KiCad Developers" group. >> >>> To unsubscribe from this group and stop receiving emails from it, >> send an email to [email protected]. >> >>> To view this discussion on the web visit >> https://groups.google.com/a/kicad.org/d/msgid/devlist/CA%2BqGbCDQhM7pm0-c14hFm8qf-YqpmEq5L8NynBoxCv3V1XLkRg%40mail.gmail.com >> . >> >> >> >> -- >> >> You received this message because you are subscribed to the Google >> Groups "KiCad Developers" group. >> >> To unsubscribe from this group and stop receiving emails from it, send >> an email to [email protected]. >> >> To view this discussion on the web visit >> https://groups.google.com/a/kicad.org/d/msgid/devlist/8302318D-0FB7-4D29-877C-E802C9644FC1%40rokeby.ie >> . >> > >> > -- >> > You received this message because you are subscribed to the Google >> Groups "KiCad Developers" group. >> > To unsubscribe from this group and stop receiving emails from it, send >> an email to [email protected]. >> > To view this discussion on the web visit >> https://groups.google.com/a/kicad.org/d/msgid/devlist/CA%2BqGbCAA%2B-ufmPWw0t0mzgq1368LcHKLrm53q-6STjz%3D_TZJPA%40mail.gmail.com >> . >> >> -- >> You received this message because you are subscribed to the Google Groups >> "KiCad Developers" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> To view this discussion on the web visit >> https://groups.google.com/a/kicad.org/d/msgid/devlist/92A29E29-8AAC-4EFF-8A2E-57ECD5C7C553%40rokeby.ie >> . >> > -- You received this message because you are subscribed to the Google Groups "KiCad Developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/a/kicad.org/d/msgid/devlist/CA%2BqGbCCCYWp9SEzrvAsrQXvBD9gx1GwStoa3N7cfzSQpT8AR_w%40mail.gmail.com.
