bugFree UCS Outline Fonts - Bugs: bug #34537, Serbian Localized Glyphs


bug #34537: Serbian Localized Glyphs

Submitted by:  Alessandro Ceschini <aleces>
Submitted on:  Wed 12 Oct 2011 04:49:15 PM UTC  
Category:  individual character(s) Severity:  3 - Normal
Item Group:  character substitution issue Status:  Fix posted
Privacy:  Public Assigned to:  Steve White <Stevan_White>
Open/Closed:  Open Release:  2010-09-19

Add a New Comment(Rich Markup)

You are not logged in

Please log in, so followups can be emailed to you.


( Jump to the original submission)

Tue 17 Mar 2015 03:44:45 PM UTC, comment #40:

Hi Steve,

I found Be looks a little unnatural in FreeSerif Italic when I compare with file #24191.

(file #33380)

Hsing-yen He <hammerandsickle>
Thu 03 May 2012 10:56:45 PM UTC, comment #39:

Unfortunately, the wrong de (identical to latin g), was being used in FreeSerifBoldItalic. This is now fixed, but not in the latest release. I hope to make a maintenance release soon, to fix these little errors.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Tue 22 Nov 2011 04:12:22 PM UTC, comment #38:

Hi Steve!

OK, OK, I know you're going to kill me for that :), but the lines for TE and SHA should be longer, so that they're as long as the main glyph.

And, that's most important, not a detail: I and Zoran had already warned you that the glyph for DE is a single-story (this time I know the name :) Hooray!), not a double-story g. I know the choice is up to you, but please DON'T release the font with that glyph because it's a gross fault. Also see file #24191 for a visual look.

That's it. Now you can kill me ;)

Bye Bye

Alessandro Ceschini <aleces>
Tue 22 Nov 2011 03:45:24 PM UTC, comment #37:

Hi Steve,

Yea, I saw the label "Fix posted" and I thought something had changed. Sorry

I'll take a look and let you know.

Alessandro Ceschini <aleces>
Tue 22 Nov 2011 03:34:48 PM UTC, comment #36:


Fix posted means the fix is in SVN. It has been there for a long time.
(Oh, I recently changed the label from "Fix in source" to read "Fix posted" just to make the label shorter. -- is this what you mean?)

As far as I know we're done, yes. But I usually close reports only after a release is made. (That's the reason for the "Fix posted" label.)

I thought you saw the last snapshot. Yes the lines for te and sha have been made longer, last week, I think.

As to releases: that is my decision.
In every release, there are things not quite ready.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Tue 22 Nov 2011 11:10:30 AM UTC, comment #35:

Hi Steve

Fix posted means you deem this issue as closed? I thought we weren't done with the shape of macrons, or whatever.

I know you're very busy, just take your time, but I think the development is too premature to include it in the new release.


Alessandro Ceschini <aleces>
Mon 07 Nov 2011 01:48:24 PM UTC, comment #34:

BE is correct, that's right, maybe, but that's a stylistic refinement, you could give a "swash" to Italic glyph like what it is in file #24191 : 207px-Serbian_Cyrillic_Italic.svg.png
The present Italic shape looks a little too dull. But it's otherwise correct.

As far as macrons and overlines are concerned, I'm sorry I'm an ignoramus of typography. That's you area of expertise. You tell me. :)

Alessandro Ceschini <aleces>
Mon 07 Nov 2011 01:40:14 PM UTC, comment #33:

Ok, as far as glyph shapes are concerned:

Zoran, you're right, Serbian Italic DE should look like g like it is in non-serif fonts, not a serif g. I just didn't mind it because I was concentrated on TE and SHA.
So again, to sum up glyph shapes issues:
BE upright & italic: OK
GE: macron shorter
DE: must look like non-serif g
TE: macron longer
SHA: macron under longer

As far as the implementation of SHA is concerned, salt is OK, please add hist too because it's also appropriate (glyph more used in the past), and we're done.

The argument is that character variations allow for a finer tuning, but since SHA is the only thing in the Serbian salt/hist, that's OK, it doesn't raise problematic issues.

Alessandro Ceschini <aleces>
Mon 07 Nov 2011 01:32:48 PM UTC, comment #32:

Hi Zoran,

The replacement "be" for Serbian (named 'serb_be' in FreeSerifItalic and FreeSerifBoldItalic) is of the form Alessandro shows: the top tail connects to the circular part at the middle of the top, rather than on the left side as in the Russian version.

It could be the feature that does this replacement is not enabled in your software. Note that we discussed earlier, FireFox and other browsers have a bug that renders it inoperable--they have acknowledged the bug.

As to the "macrons" or "overlines": Can you refer me to a standard, or a typographic reference, that makes this stylistic imperative clear? (A reference to another font serves only as an illustration, not a standard.)


Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Mon 07 Nov 2011 01:23:46 PM UTC, comment #31:


The only 'salt' feature in FreeSerifItalc that is implemented for Serbian is this alternate SHA.

It still appears to me to be the appropriate choice, even given what you said. If someone told you something else was preferred, I would have to see the argument, but what you have written seems to be based on a misapprehension of what is in FreeFont, and perhaps a misunderstanding of the technology.

Could it be, that you aren't aware that OpenType features are contingent upon the text language?

It sounds as if you are visualizing a usage scenario that I can't imagine. I would imagine that the user would normally just turn on the 'salt' feature, to get the stylistic alternative. Very user-friendly.

I only mentioned the runs of text to illustrate that one can turn features on and off if necessary NOT that one must turn features on and off. Normally one can turn a feature on for one run of text and off for another, if desired. That is, if you wanted to sometimes show the modern style SHA, and sometimes the other style, you could do that. But I'm afraid you're mixing the issues here.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Mon 07 Nov 2011 12:50:48 PM UTC, comment #30:

Hi Zoran. Nice to find you here too.
1. It seems to me issue 1 has been solved, you likely downloaded an older build version, try here: https://docs.google.com/#folders/0Bwgv3Jc6xeHPNTA0NjhjMjUtNzFhZi00NTFhLWEwNzAtMDgwN2FhMWM0Mjg0
2. I already raised issue 2 (see #25), the lengthening also involves not just Italic TE but (optional) Italic SHA.
3. That's true, I didn't notice that, it is shorter.

To sum it up the length of macrons is as follows:
above Italic GE < above Italic I < above Italic TE & SHA


Alessandro Ceschini <aleces>
Mon 07 Nov 2011 12:39:44 PM UTC, comment #29:

I downloaded FreeSerifItalic.sfd Revision 1869, open with FontForge.

1. Cyrillic Small letter de for Serbian glyph is WRONG. Attached file #24191 by Alessandro Ceschini is example how this glyph look.
2. "macron" over cyrillic small letter te must cover whole letter te.
3. "macron" over cyrillic small letter ge must be "shorter"

4. I put some comments on same topics for Liberation font ( with pictures ) https://bugzilla.redhat.com/show_bug.cgi?id=657849

5. I put some comments on same topics for Linux Libertine font

6. Mister Alessandro Ceschini put some comments on

Ih You visit anyone of these http links ( with pictures ) , You can find how Serbian Glyphs look, Italic fonts , with locl feature , OpenType.

Adobe fonts Minion an Myriad are best example for Serbian Cyrillic glyphs, Italic OpenType fonts.

Best Regards.

Зоран В. Петровић <zoran_v_petrovic>
Sun 06 Nov 2011 08:24:52 PM UTC, comment #28:

Hi Steve
All began when I was discussing implementation of Serbian glyphs at the EB Garamond (an experimental opensource font) Forum: https://github.com/georgd/EB-Granjon/issues/1
At the beginning I was arguing for a salt/hist lookup, just like what we'd agreed back here. But then the developer convinced me that character variants were the best implementation for SHA because turning salt or hist on would have forced the user to enable other archaic or alternative glyph variants he just did not wish to.
I'm sorry I talked about old-style numerals, that was not really a good example, you'll have to excuse my ignorance of Open-Type features, I'm not really into it, I only use locl and some ligatures, that's all. Just forget about it, it's not about old-style numerals, it's about other variant glyphs being turned on.
Now, what do the salt & hist lookups contain in your font? I don't think SHA is the only stuff in there, I mean if I turn salt or hist ON, I'll be forced to swallow indiscriminately whatever the lookup provide me with just because I want that one glyph! You talk about "runs" of text, and you seem to imply by that that I should disable the lookups whenever I encounter some character that would be affected by it and then re-enable it afterwards. Have I got it right? Well, that's not really user-friendly to say the least, and moreover, that's frigging painful in very long texts :(
About the software: both I and the developer of EB Garamond happen to use XeLaTeX as a reference. Indeed, when he came out with "character variants", at first I protested, I wanted salt/hist because I didn't know how to implement cv's in XeLaTeX, but then he told me how and it's very simple.
By the way, I don't think this is a platform or software-related issue, salt & hist are OK, don't remove them by no ways, but for guys who just want ONLY that particular glyph and not the other features in the lookups, character variants are the preferred choice.
I hope I've made my point clearer.

Alessandro Ceschini <aleces>
Sun 06 Nov 2011 07:46:33 PM UTC, comment #27:

Hi Alessandro,

These are technical points that involve implementation of application software. (What software are you referring to, anyway?)

I am unconvinced of the argument.

Old-style numerals are implemented as an 'onum' lookup in FreeFont, not 'salt' or 'hist', so your example doesn't seem to make sense.

Where are you getting this information?

But anyway, regarding **NECESSARILY**
Normally one enables features within a run of text.
So turning on a feature in one run of text would not turn it on in another run. So turning on the special SHA in one run would not necessarily turn it on in another place--depending on how your applicaton software implements it, of course.

It therefore still appears we have chosen the appropriate feature.

Do I understand? Maybe not.

This is very application-dependent. You will have to explain to me how your proposals would be better in your application. For the applications I have in mind (page layout applications, future web browsers), I think the present choice makes better sense.

How would a character variant make it easier for a user to turn on old-style SHA in their text, in whatever application you use?


Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Sun 06 Nov 2011 06:24:16 PM UTC, comment #26:

OK, about SHA, I've been told the best and safest way to implement it is character variations. Because, I've been told, salt or hist would trigger, say old-style numerals and not everyone who wants the Serbian Italic SHA NECESSARILY wants these other features too.
You can keep it in salt and hist if you deem it appropriate, but a character variation would be nice for those who want JUST that, and nothing else.
I hope you understand what I'm saying.

Alessandro Ceschini <aleces>
Wed 02 Nov 2011 01:42:00 PM UTC, comment #25:

Did some testing for Free Serif Italic fonts:
Great! The optional Italic SHA works!
1) Italic TE and (optional) SHA should not have, respectively a macron above and beneath, but rather something like an overline, that covers the entire glyph. Just take a look at the DejaVu Serif implementation.
2) Optional SHA is implemented in salt but not in hist.

Otherwise as far as Serbian glyphs are concerned, UCS are at the moment among the most complete fonts available and for certain THE most complete free font. Not even DejaVu supports Serbian Italic SHA. Wow! :)

Alessandro Ceschini <aleces>
Mon 24 Oct 2011 09:29:19 PM UTC, comment #24:


The applying tags are:
locl of course and then salt and I'd also add hist, because I've been told this was more widespread in the past. The problem is, can the two tags be combined so that locl+salt/hist ONLY give the expected glyph?
locl alone is to be deprecated, as it would impose on the user what should be left to his discretion.
salt/hist alone is crappy because, say, a Russian user wanting to have some old-style Russian glyph would be forced to utilize a Serbian glyph that's completely inappropriate for Russian typesetting.

Alessandro Ceschini <aleces>
Mon 24 Oct 2011 08:55:10 PM UTC, comment #23:

Neither do I. Frankly, I don't even know where to ask. What do you suggest? Do you know some OpenType forum? I only deal with XeLaTeX, that is an implementation of OpenType, I never deal with raw stuff.

Alessandro Ceschini <aleces>
Mon 24 Oct 2011 08:27:56 PM UTC, comment #22:


In SVN: implemented a 'salt' "stylistic alternative" for 'sha', localized for Serbian.

I have never seen these things in action; perhaps you can tell me how you use them in your environment.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Mon 24 Oct 2011 06:43:15 PM UTC, comment #21:

Hi Steve

1) Yeah, Serbian optional Italic SHA should be like Serbian Italic TE but the underline should go under, instead of above (as in Italic TE). Or in other words, we should add an underline under Russian SHA to make it look more Serbian.
I know how to switch on "alternative forms", what I don't know is whether it's possible to create/switch on "alternative forms" as a subset of "localized forms". For that variant is valid only for Serbian, not for other Cyrillic-based languages.

2) Yes, removing the substitution making it the only glyph for GJE. that's it. I'm sorry I didn't manage to make my point clearer before, but now that you agree with me I think it would be OK for me to ask the DejaVu team to do the same in their font.

Alessandro Ceschini <aleces>
Mon 24 Oct 2011 03:12:14 PM UTC, comment #20:

Hi again Alessandro,

First: Regarding an "alternative form" for 'sha': I found a remark about such a thing, used to emphasize the distinction from the italic 'te' -- consisting of the normal 'sha' with an underline. Is this what you're talking about? I can do that...


The "alternative forms" features are used by specialized software (especially professional page-layout applications) to present the user with a list of alternative glyphs for the same character.

Second: It seems we were totally misunderstanding one another with the gje. Maybe I get it now.
You're complaining of the substitution lookup for gje in FreeSerifItalic, replacing it with a Serbian ge with macron and accent. I see--this is just wrong, as I understand it. I'm not sure why I put it there in the first place.
I will remove that substitution from FreeSerifItalic, leaving only the Macedonian italic form as the main character.
Let me know if that is what you expected.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Mon 24 Oct 2011 02:26:43 PM UTC, comment #19:

1) OK, just take your time, no need to go overboard, you've done a pretty good job here, normally (Ubuntu bugs, for instance), they'd have taken a year to just confirm the bug, let alone solve it.
2) Try TeXWorks with XeLaTeX, that's what I use to typeset Serbian texts.
3) I understand what you're saying: the standard glyph is just indicative, localizations affects GLYPHS, not CHARACTERS. That's true, but I was saying, what's the purpose of having a localized glyph for a character that's in use in JUST ONE language? And of having it linked to a standard glyph that is WRONG and therefore to be DEPRECATED? The shape of DejaVu Macedonian Localized Italic U+0453 CYRILLIC SMALL LETTER GJE is correct, so let's just replace with it the "standard" glyph that is, as a matter of fact, an aberration, a Russian glyph for a Macedonian-only character. No need to add a localized glyph, let's just do away with the WRONG "standard" glyph. In DejaVu as well as in UCS Fonts. So, NO flaw in the glyph, I'm just discussing the implementation. My humble opinion is that you just adopt the localized Macedonian glyph for GJE as the standard glyph.
4) Yeah, DejaVu is not the cutest of all free fonts, but it's beyond any doubt a well-established reference with plenty of glyphs. I repeat it, I don't find any fault with the glyph, only with the implementation of it.
5) No reference, I confess I don't even know how to access those "localized alternative forms".

Thanks for your efforts!

Alessandro Ceschini <aleces>
Mon 24 Oct 2011 01:59:32 PM UTC, comment #18:

Hi Alessandro,

1) The FireFox developers have acknowledged my bug report as being real. This is very good. I could do more, but I have too much to do now to become a FireFox developer as well.

2) Unfortunately, none of the several other apps I tried is correctly implementing the features necessary for Serbian letters. This is disappointing, but not surprising. All we can do is report the issues to the developers. Improvements do happen.

3) I have already implemented the Macedonian letter. I still do not understand your objection. In FreeFont there are several cases of special letter forms for individual languages, and this is precisely the purpose of localized forms. I suspect that you are misuderstanding the technology (we can discuss that elsewhere if you like). Unless you have some further objection, the letter variant stays.

4) I consider FreeFont to be well-established. All fonts have their quirks. We all do our best, but I will not follow the failings of other fonts.

5) There is also a way to implement localized, "alternative forms". Perhaps this is the way to go with the 'sha'. I'll look into it. (Do you have a reference, or an example?)


Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Mon 24 Oct 2011 01:32:18 PM UTC, comment #17:

I've been told that there should also be an OPTIONAL, that is NOT MANDATORY Serbian localized glyph for Italic U+0448 CYRILLIC SMALL LETTER SHA.

(file #24191)

Alessandro Ceschini <aleces>
Fri 14 Oct 2011 06:00:25 PM UTC, comment #16:

As far as the HTML is concerned, me too, I can't see localized glyphs from your sample and even this test page doesn't work:


It seems there's really a bug in Firefox, I fear there's nothing we can do about it, just waiting for a patch.

Alessandro Ceschini <aleces>
Fri 14 Oct 2011 05:41:50 PM UTC, comment #15:

Yeah, that's use only in Macedonian. But that's the reason why I think we'd better not touch it. Why bothering to create a localized variant for a character that only occurs in one language?
And as far as the reliability of DejaVu is concerned, I myself already complained once about the accuracy of their glyph construction https://bugs.freedesktop.org/show_bug.cgi?id=31979
So, my humble opinion (in no way binding) is that you should stick to what well-established fonts do. And they don't have that glyph, I can assure you. Check it out in FontForge to be sure about that.

Alessandro Ceschini <aleces>
Fri 14 Oct 2011 05:00:51 PM UTC, comment #14:

Hi Alessandro,

I gather from this
that gje is used only in Macedonian. That is, there is no "russian" letter gje.

It seems logical to treat it the same as the 'ge' in italics, and there is precedence for that.
So unless you have some other objection, I'll leave the special form in the italics style lookups.

Let me know if there's anything further I can do!

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Fri 14 Oct 2011 04:18:45 PM UTC, comment #13:

As far as GJE is concerned, this font does NOT occur in Serbian, actually it occurs ONLY in Macedonian, not in every other Cyrillic-based language. The point I tried to make in my last post, it's that since it's used ONLY in Macedonian, the standard glyph should already be the appropriate one for Macedonian.
I didn't know DejaVu Serif made us of it, but all I can tell you is that well-established fonts, by commercial firms, like Constantia and MinionPro, do NOT make use of it.
As far as the HTML file is concerned, I'm sorry I can't program HTML, I only know LaTeX. But I'll try to gather some piece of information about language tags in Firefox that maybe could come in handy.

Alessandro Ceschini <aleces>
Fri 14 Oct 2011 12:47:15 PM UTC, comment #12:

Regarding the browser support,

I have already submitted a bug report to Mozilla on this topic.

This is Firefox 7.0.1, and I am no way seeing Serbian letters.

Attached is an HTML file, set up to use the DejaVuSerif Italic fonts, which I presume have been tested. It doesn't show Serbian glyphs when viewed in Firefox.

If you could fix the HTML file so the character do show, that would be really helpful.

(file #24145)

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Fri 14 Oct 2011 12:39:52 PM UTC, comment #11:

Hi again,

The 'be' is now implemented in all four styles in Serif and Sans.

As to the gje, maybe I don't grasp what you're saying.

Is this character used at all in Serbian?

From the en.wikipedia.com pages, it appears that it is not used in Serbian, but is used in Macedonian.

I got the idea to apply the lookup from DejaVuSerif Italic, which does this exact thing.

It would be possible to make a separate table just for Macedonian, if you're quite sure this is wrong for Serbian.


Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Thu 13 Oct 2011 10:48:36 AM UTC, comment #10:

As far as the Serbian BE is concerned, that's perfect!!! Wow! You did a really good job!
You also draw an italic shape for U+0453 CYRILLIC SMALL LETTER GJE which is just the Serbian shape plus an accent. Problem is, that this character is, as far as I've been told (I'm a scholar of Serbian, not of Macedonian) ONLY used by the Macedonian language. So, font designers would have directly included the appropriate Macedonian glyph, provided that there was no need to draw the standard/Russian glyph. I don't know if you can grasp what I'm trying to point out here.
Also, I looked up in FontForge and no font I'm aware of does contain such a glyph for U+0453 in italic shape.
I'll investigate more and I'll let you know, but in the meantime I suggest you should drop that glyph since there is no font I know that does make use of it.

Alessandro Ceschini <aleces>
Thu 13 Oct 2011 09:00:43 AM UTC, comment #9:

I used XeLaTeX, and I can tell you that Firefox 7.0.1 DOES support Serbian glyphs, but only if you set the language tag to "SRB": on the Serbian Wikipedia, for instance, with the font Ubuntu, on Ubuntu 11.04, I do see the Serbian BE.
But I heard there were some problems if the font was not complete in its support: i.e. if you inserted locl glyphs only for regular/italic and not bold/bold italic, for instance. Even if you use, say only regular in your test, Firefox still demands from you a complete set of localized glyphs. Maybe this could help.
OpenOffice, unfortunately doesn't support OpenType yet, so setting the language tag to Serbian won't have any effect on the glyphs.
As far as the shape of Serbian BE is concerned, yeah, I think the construction you described is OK.

Alessandro Ceschini <aleces>
Thu 13 Oct 2011 08:47:34 AM UTC, comment #8:

Hi again Alessandro,

First, I have a question: In what software do you see this feature actually enabled? I have tried it in Firefox and LibreOffice, but setting the language to Serbian had no effect there.

You are right, in both sans and mono, the 'be' differs from delta.
All I can say is, I was rather tired when I looked before...

Allright, so we need a special character for the Serbian 'be', I would say, like the delta in form with a curved hook over an o shape, but the hook should terminate horizontally, with no flourish at the end. I'll work on this.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Thu 13 Oct 2011 06:45:10 AM UTC, comment #7:

Sorry, I botched the last comment I meant:
Well, actually, after having taken a look at how other Serified fonts behave in this case, (look in the pdf) Serbian BE should NOT have a descendant on the upper stroke (I hope you understand what I say since I don't know specialized typographical expressions).

Alessandro Ceschini <aleces>
Thu 13 Oct 2011 06:43:52 AM UTC, comment #6:

Well, actually, after having taken a look at how other Serified fonts behave in this case (take the identification of the glyphs i wrong, Serbian BE should NOT have a descendant on the upper stroke (I hope you understand what I say since I don't know specialized typographical expressions).

Alessandro Ceschini <aleces>
Thu 13 Oct 2011 06:35:09 AM UTC, comment #5:

As far as FreeSerif is concerned, the glyphs are OK.
I looked up in FontForge, and actually neither in FreeMono nor in FreeSans is U+0431 identical with U+03B4. And even if it was the case, that would be inappropriate for Russian and most other languages written in Cyrillic script. That works only for Serbian and, yes, Macedonian, I didn't know that, but it turns out it uses the same glyphs as Serbian.
As far as the shape of U+03B4 in FreeSans is concerned, I think it's acceptable for U+0431; Ubuntu (name of font) uses a rather similar glyph for both.
In case you needed guidance for construction rules, I posted a comparison of U+0431 (localized Serbian & Macedonian) and U+034B in all the fonts I know of that support the locl feature.

(file #24140)

Alessandro Ceschini <aleces>
Wed 12 Oct 2011 10:35:27 PM UTC, comment #4:

Here's the result for the latest FreeSerif, using the direct Pango rendering command:
pango-view --font "FreeSerif 32" --markup Serbian.html
on this file
<span lang="ru"> б г д п т </span>
<span lang="ru"> <i>б г д п т</i> </span>
<span lang="sr"> б г д п т </span>
<span lang="sr"> <i>б г д п т</i> </span>

(also works with lang="mk")

(file #24138)

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Wed 12 Oct 2011 09:30:29 PM UTC, comment #3:

Hi Alessandro,

Neither Sans nor Mono has an "italic".
They both have a simple "oblique".

The be in mono is already identical with the greek delta.

Ah. The delta in Sans is... rather strange in form. I doubt you really want to use it for a be. Have a look.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Wed 12 Oct 2011 07:03:01 PM UTC, comment #2:

Be>greek delta in ALL styles, serif mono and sans, italic and upright.
As far as mono is concerned, the inclusion of italic glyphs depends on whether its italic chooses serif-like glyphs (like Arial) or it just behaves like a sans font.

Alessandro Ceschini <aleces>
Wed 12 Oct 2011 06:54:00 PM UTC, comment #1:

As I understood it (and implemented it) this applies only to the serif face -- the sans and mono faces don't make such fine distinctions.

In all styles,

be -> greek delta

In italic

de -> latin g
ge -> latin i-macron
pe -> latin u-macron
te -> latin turned m-macron

The only letter not already available in Unicode was a properly turned m (well there is on, but the serifs are in the wrong places, as it it really just turned). So I constructed that glyph.

Fix in SVN.

Steve White <Stevan_White>
Project AdministratorIn charge of this item.
Wed 12 Oct 2011 04:49:15 PM UTC, original submission:

Could you please add the localized Serbian glyphs in the Cyrillic set? You'll find in the pdf a comparison between the standard a.k.a. Russian glyphs and the Serbian localized ones. Let me know if you need more information.

Alessandro Ceschini <aleces>


(Note: upload size limit is set to 16384 kB, after insertion of the required escape characters.)

Attach Files:

Attached Files
file #24321:  20111107-CVS-FreeSerfItalic.sfd.png added by zoran_v_petrovic (100KiB - image/png - Tested FontForge sfd file for FreeSerifItalic font)
file #24191:  207px-Serbian_Cyrillic_Italic.svg.png added by aleces (16KiB - image/png - The last letter is SHA)
file #24145:  Serbian.html added by Stevan_White (626B - text/html - test of HTML lang tag for Serbian)
file #24141:  Serbian1.png added by Stevan_White (20KiB - image/png - Latest in SVN)
file #24140:  Serbian BE versus Greek DELTA.pdf added by aleces (69KiB - application/pdf)
file #24138:  Serbian.png added by Stevan_White (16KiB - image/png)
file #24137:  Russian vs. Serbian Glyphs.pdf added by aleces (25KiB - application/pdf - Russian versus Serbian glyphs)


Depends on the following items: None found

Items that depend on this one: None found


Carbon-Copy List
  • -email is unavailable- added by hammerandsickle (Updated the item)
  • -email is unavailable- added by zoran_v_petrovic (Posted a comment)
  • -email is unavailable- added by Stevan_White (Posted a comment)

    Do you think this task is very important?
    If so, you can add your encouragement to it.
    This task has 0 encouragements so far.

    Only logged-in users can vote.




    Follow 12 latest changes.

    Date Changed by Updated Field Previous Value => Replaced by
    2015-03-17 hammerandsickle Attached File- => Added A problem of Serbian Be.png, #33380
    2011-11-25 aleces Carbon-CopyRemoved 85679 => -
    2011-11-07 zoran_v_petrovic Attached File- => Added 20110201-peca-SerbianGlyphs-Microsoft-OpenType.gif, #24322
    2011-11-07 zoran_v_petrovic Attached File- => Added 20111107-CVS-FreeSerfItalic.sfd.png, #24321
    2011-10-24 aleces Attached File- => Added 207px-Serbian_Cyrillic_Italic.svg.png, #24191
    2011-10-14 Stevan_White Attached File- => Added Serbian.html, #24145
    2011-10-13 Stevan_White Attached File- => Added Serbian1.png, #24141
    2011-10-13 aleces Attached File- => Added Serbian BE versus Greek DELTA.pdf, #24140
    2011-10-12 Stevan_White Attached File- => Added Serbian.png, #24138
    2011-10-12 Stevan_White StatusNone => Fix posted
        Assigned toNone => Stevan_White
    2011-10-12 aleces Attached File- => Added Russian vs. Serbian Glyphs.pdf, #24137

    Back to the top

    Powered by Savane 3.3