No description provided.
Added the Latin Extended A code block.
The new glyphs have not been kerned.
I've added all of the missing glyphs from the Latin Extended-A code block.
I saw on your blog that someone was asking for them:
You can see a preview of the new glyphs here:
I've also added a Markdown file with extensive notes.
None of the new glyphs have been properly kerned, although I did update their side bearings appropriately.
Also, I've only updated the .ufo file in my fork. I leave it to you to update the other files.
I've also cleaned up the contours of a number of glyphs in the font using rf-contour-cleanup, a robofont extension I wrote for this purpose.
It is available here: https://github.com/charlesmchen/robofont-extensions-and-scripts
It should have no visible effect on the glyphs.
Diacritics are quite a delicate matter and for all the attention given here it seems the approach is still too simplistic. Few of the diacritics were already not so well balanced, particularly the acute or grave, and the same unfortunately continues. Now different accents around the letter ‘r’ should be weighted to its stem as well as the ogonek drawn as a natural continuation of the letter's grapheme. In fact, and I don’t know how important historical accuracy is in this case, but the diacritics themselves could be refined in the first place. This could include more seamless ogonek, proper cedilla, more distinctive vertical caron or heavier ring.
I also think that for compatibility reasons we shouldn’t mess with curves of existing set unless some major release is on its way, especially with nonessential commits like points shifting around in contours as it is seen in the diffs.
Thanks for the comments, ichosis.
If you redraw the diacritics, I'll apply them to the relevant glyphs.
This can easily be fixed.
Can you give an example of a glyph where this has been done improperly and an example of how it should be done?
These changes are not a significant part of the changeset. All of them just convert straight beziers to straight lines. I can revert them if the League agrees with you.
The thing with the ogonek is that it’s not actually being attached to a letter, but rather drawn as a continuation of the letter’s stroke, thus inheriting its movement and style. This also implies slight changes to the ogonek depending on the base letter. I could not give a better explanation than the one by Adam Twardoch at http://twardoch.com/download/polishhowto/ogonek.html with inline examples of what the ogonek should look like in all those circumstances. I also recommend checking out a great resource on understanding different diacritics at http://diacritics.typo.cz/. That being said, I hope to just fork all this and put my two cents this way.
Very interesting, thanks @ichosis.
I'm curious to see what @theleagueof thinks. If they're even considering this pull request, there would be a few options: a) omit any ogonek (and otherwise problematic) glyphs for now. b) design a new ogonek (you, me, or someone else) and incorporate that (or, as suggest, design custom ogoneks for each glyph that uses them).
In fact, I think the optimal, somewhat more disciplined and collaborative scenario for the work would be restaged as follows:
• Strip off the existing, precomposed diacritics from the Western set
• Refine outlines of the original standalone ones
• Set anchors wherever possible and populate much of Western and Extended set this way
• Hand-tweak joints of the problematic components like the ogoneks
• Compose all the remaining glyphs
In all this, what's important is all the changes could be committed successively and in small, swallowable chunks so that The League and others could easily review the steps taken, discuss and pull them as needed.
That sounds very reasonable, @ichosis. I'd be willing to do the gruntwork on most of the steps if you would be willing to do the "Refine outlines of the original standalone ones" step.
But perhaps we should wait to see what @theleagueof thinks. It doesn't look like they've ever accepted a pull request. In fact, only two of their 16 repos have any activity after the initial commit. I wonder if these projects are active and accepting unsolicited contributions.
Let me add if that if you and @theleagueof can agree on how the diacriticals should be modified, I'd also be willing to take a stab at the "Refine outlines of the original standalone ones" step if that would help.
Oh wow, there's a lot to catch up on here! I certainly didn't mean to ignore this, I think it's incredible you've done all this work, though I definitely agree with the approach @ichosis is suggesting – smaller digestible chunks will make it far easier.
Pull request are a little new to me, though I understand why this is great. So I'm definitely interested, but we just need to take care with how we do it, because downloads by non-technical people are directly affected as soon as we change anything in the repos.
So let me look over this a little more carefully & respond in a little bit!
Just to throw another wrench in here, part of the complication is that I've been working on integrating Dannci's contributions as well – https://github.com/micahbrich/league-gothic – which has a lot more glyphs. I know @ichosis has expressed that he thinks those are not top-quality drawings, so maybe we can work on improving those and seeing what we can merge between both instead of doing it all over.
Hi Micah, thanks for your comments. Let me know if I can help with whatever course you settle upon.