Font issues

[drempel]

I don't know if this is an FGU or generally issue with FG, or the third party content on the store (I haven't boot up classic in a long time to dig) but I'm noticing that a fair number of third party modules or add-on's have that "missing glyph" symbol in them showing up quite a bit. Usually for some punctuation character such as a quotation or a colon. I thought maybe it was just old modules that hadn't been updated, but I just picked up the LoreSmyth Inns and Shops modules which say they were made in the last year and they have them as well. That's something I wouldn't expect. A recent module with glyph issues. Makes me think that there's something wrong with FGU or my install of it.

Anyone else seeing issues like this or is it just me?

Attaching some examples from "Remarkable Shops and Their Wares". It's not just LoreSmyth though like I said, all 0oneGames modules I have (6 or 7? I dunno, didn't count) all have this issue too. It's all usable still, just looks bad and makes it a little bit harder to read sometimes.

42618

42619

[Talyn]

It's a common issue. FGU is extremely picky with its character encoding while FGClassic was not. A lot of the custom characters that publishers use such an en dashes, em dashes, 'smart' quotes, etc. won't work on both clients so the developer has to know to essentially downgrade all the characters.

Thank you for the report, though! :) One of the forum moderators will make sure the developer learns of this to patch the product.

[Zacchaeus]

I've alerted James.

[drempel]

Ok cool guys :).

To be clear its both the remarkable inns, and the remarkable shops that have the issue. And I like I said I think every single 0one Games adventure module has it, but I know for sure "A Dead Reckoning", "Black Magic", "Entrees, Entrails, Canapes, and Corpses", "One Night at the Red Vampire", "Tabernacle of the Forsaken", and "Threads of the Orb Weaver" all do. Those are the ones I own.

What text engine did FGC use versus FGU? I take it that it's hard to improve FGU's to be on par with FGC's feature set.

[LordEntrails]

What text engine did FGC use versus FGU? I take it that it's hard to improve FGU's to be on par with FGC's feature set.
Both FGC and FGU data is supposed to be encoded per ISO-8859-1. But, FGC actually allowed characters outside that definition, FGU does not.

If you look at the header of an FG module, you will find that. Since FGU uses a commercial xml parser, it actually enforces that standard. FGC, since it had a self-developed parser, it did not (and I believe would display the wider ASCII character set).

[drempel]

Oh, ok so it sounds like if (I'm reading this correctly) FGU is basically enforcing 8bit single char encoded strings while FGC was using the more modern and flexible UTF8 (multibyte) character strings. That right?

Why would FGU want to limit itself to just those characters? Why not all of Unicode? UTF8 is backwards compatible with it is it not? Seems like an odd limitation for software these days (especially since it looks like the old version was indeed using more a lot more unicode).

Eh, doesn't matter I guess that much, would make life easier for people converting, and localization would be easier for regions that don't use those characters (although I suspect FGU's text rendering can't handle right to left for Arabic localization anyways). Sorry, software engineer in me is just scratching my head on this one. EOD though I'm loving the product and don't want to take anything away from it.

[Sulimo]

The issue is not actually UTF-8.

The issue is that FGC could use text encoding that used Windows-1252 encoding (https://en.wikipedia.org/wiki/Windows-1252) (specifically characters 127 through 159), instead of ISO-8859-1 (https://en.wikipedia.org/wiki/ISO/IEC_8859-1) which is what FGU supports (even though technically FGC only supported ISO-8859-1, it allowed Windows-1252 encoding). Windows-1252 has extra characters that are not in ISO-8859-1, so if those characters are used, you will see what you show in your screen shots, the specific characters in that screen shot that are problematic are probably 145-148 (in the chart), changing them to ISO-8859-1 22 and 27 is probably what needs to be done. However, it has to be done inside the module xml.

I would guess that going with ISO-8859-1 instead of Windows-1252 is probably necessary for cross-platform compatibility (I seem to recall that something with .NET was involved, but I cannot recall that discussion).

Previous discussion here (https://www.fantasygrounds.com/forums/showthread.php?53438-Special-characters-issue-found-in-Crypt-of-the-Sun-Lord).

[drempel]

ah, ok that makes more sense then as to the state of things. Thanks for taking the time :)

[LordEntrails]

Thanks Sulimo, I couldn't remember the details.