Releases: Mercury13/unicodia
1.2.12
FONTS
Updated Old Turkic (10C47 is still old)
OTHER DATA
Fixed descriptions: Bamum, Bengali, Brahmi, Caucasian Albanian, Georgian, Inscriptional Parthian, Kharoshthi, Lepcha, Meetei, North Arabian, Nüshu
INTERNATIONALIZATION
Made a special engine for encoding script dates and minimizing number of “special” strings, checked all these dates
PROGRAM
Bounds check was accidentally on even in release mode
1.2.11
FONTS
Changed emoji to Google Noto, a prerequisite for comprehensive emoji reference
Side-effect: all emoji even in W7
Enlarged Old Gothic just a bit, and shrunken in description. It’s still a bit smaller than Permic/Italic, and A’s ascender is a bit higher than Latin ascenders
Implemented space in FunkySample
Updated Georgian, Lanna
OTHER DATA
In many places added NBSP after “not”
Stated that one Old Turkic rune is obsolete
Changed parachute icon to Noto-like
PROGRAM
Reduced ”Sample/OS style” windows just a bit
1.2.8 → 1.2.9
Correctly decapitalized “Harklean”
1.2.9 → 1.2.10
Fixed emoji 1F596 Vulcan salute (SVG renderer does not support clip paths)
1.2.10 → 1.2.11
Optimized three SVG emoji to prevent warnings (no misrenders)
No more wikitext in note for special 1DA5E Sutton mark
1.2.7
PROGRAM
Moved OS font to “OS style” header
All graphic + not ignorable chars now have OS style. AFAIR, virtual viramas
More precise definition of graphic char
Now we draw real Times New Roman tofu instead of Cambria “?”
Birman no longer falsely appears outside BMP
“ASCII” link no longer puts popup outside main window
DECAPITALIZATION
Linear B indexes like b123a → B123a. Sometimes other rules applied, but not always.
1.2.4 → 1.2.5
Implemented Old Gothic in FunkySample, just because W10’s is not “gothic” enough. Taken from Junicode
1.2.5 → 1.2.6
Syloti: link to virama
Phags-Pa: found blatant lie, wrote more information
1.2.6 → 1.2.7
Fixed “Arguable script” article, was modified because of Phags-Pa
1.2.3
WARNING: wartime releases are more frequent. Every single program improvement → release. A few days of development → release.
FONTS
Updated: Tibetan, SignWriting
OTHER DATA
In Tibetan made link to more stable BabelStone
Egyptian hiero format: wrote what’s expected in U15
Tai Xuan: made link to Yi Jing
PROGRAM
Now yellow popup sticks closer to main window
Stray <tr> in SignWriting
1.2.2 → 1.2.3
Smaller margins between main window and popup
Fixed popup on multi-monitor and taskbar on the left
1.2.1
(The war is coming, I’ll post quickly while still have internet, let’s call it Codename Donetsk)
NEW
Added several input methods: Alt codes, two software-specific and Ilya Birman’s typography layout
Now we check whether old character turns to emoji using VS16
“+25CC / +VS16” button, does same as hidden Ctrl+Shift+C/Ins
Sutton SignWriting fill/rotation demo
FONTS/MISRENDERS (no tofu here until U15)
Updated fonts: BabelStoneHan, Balinese, Cham, Kharoshthi, Khojki, Modi, Music, Newa, Sharada
HanaMin: hand-updated two misrendered chars (2CF33, 2D182) and a few dozen random ones
Simplified and hinted Lao font
Took Noto Math font for mathematical alnum characters in early Windows
Latin/Greek/Cyrillic: turned in advance U+A66E Multiocular O to U15 glyph with ten “eyes”
Latin/Greek/Cyrillic: dereferenced links in five Cyrillic chars
FUNKYSAMPLE → PROFESSIONAL
Dives Akuru: now Noto Serif Dives Akuru; updated icon
Sori/koron (Persian quartertone marks): now Noto Music
(My Cypro-Minoan is rather stylish, I won’t touch it)
DECAPITALIZATION
Three calendar runes
A few Ethiopic umlauts
National/nominal digit shapes
OTHER DATA
Typos in descriptions
Changed descriptions: Lao, Makasar
Wrote about upcoming U15: Alchemical
Found Brodie Lockard’s posts, he’s alive as of 2020 though rather old
PROGRAM
Less probability that combobox is falsely focused: we defocus it by timer, not by queued message, and synchronously preload fonts.
1.2 → 1.2.1
Bad rendering of Siddham U+1158B
1.1a
FONTS/TOFU
Installed professional Chorasmian font instead of homebrew
…Still supported by FunkySample: Dives Akuru, Toto (not mine and fully working), Ottoman Siyaq and some small holes, incl. emoji.
Updated fonts: BabelStone Han, Bassa, Canadian, Dogra, Latin/Greek/Cyrillic, Old Uyghur
Implemented ALL remaining tofu, no tofu in W10
FunkySample’s metrics are more in harmony with Noto
Fixed most blatant bugs of FunkySample
Moved all new characters and Cyrillic enclosing marks from Noto Latin/Greek/Cyrillic to FunkySample
More chars of General Punctuation are drawn with Cambria
DECAPITALIZATION
Fixed “Hebrew accent Foo”, “Hebrew point Foo”, “Syriac letter Gibberish Foo”, “Modifier letter Georgian Foo”, "Tag Number Foo"
Cuneiform keywords “Tenu” (slanted), “Gunu” (4 extra wedges), “Sheshig” (additional SHE sign), “Nutillu” (unfinisned) are written with initial capital to show that they are suffixes of sign name (used to be all caps)
Manually reviewed words “Ligature”, “Allah”
ICONS
Cuneiform: better contrast
Dogra: obsolete glyph, changed to Ka, despite looks like Gurmukhi
Egyptian hiero: low-contrast, has some features of Segoe font missing in Noto
Tags: now look more in harmony with Unicode
Variation selectors: better contrast
OTHER DATA
More on Chorasmian, Khmer
Stated when Egyptian hieros disappeared
Added “Default-ignorable” property
Changed “pk:” (popup/block) pseudo-protocol: Cyrillic “pk:8” → “pk:400”. Added these links
BUGS/IMPROVEMENTS
Nicer search of non-existing codepoints
Egyptian sticks 133FA…40C are now displayed correctly
Drew tag symbol on E0000 Tags block
More DPI-agnostic method of drawing control characters
More precise way of checking whether we have character A in font B
DIFFERENCES FROM INITIAL 1.1
U+1DCA etc were badly displayed, well-known flaw of Noto
1.0
FONTS/TOFU
Replaced HanaMin fonts with other version, smaller and neater → −10M of archive
Sorry, removed in W7 Lydian, Lycian, Carian, Old Gothic, South Arabian (were in old HanaMin)
Took nice Orkhon runes from Google Noto instead (for some reason dislike Segoe Historic)
Closed U14 tofu: CJK (in W7 still remaining some), Music, Kaithi, Brahmi, Ahom, Old Uyghur
Removed unused and non-functional Nandinagari from FunkySample
Fixed misrenders: virtual virama in both Gondi, Miao tones, music G-clef octave up/down, Duployan overlap
Updated fonts: Canadian aboriginal, Tirhuta
Rehinted Oriya
Made Tagalog U14 chars more functional
Switched custom Egyptian hieroglyphs TTF → standard OTF
Same for Anatolian hieroglyphs, though TTF was standard, not custom
Remaining tofu: Latin F and G, Cypro-Minoan (187 total)
OTHER DATA
Descriptions for everything, incl. three tofu blocks
Reviewed texts, lots of changes. Blatant lies found: Devanagari, Grantha, Gunjala, Malayalam, Modi, Sharada
Removed KUVT and Vyaz terms, as a preparation to globalization
Indent is always three NBSPs now
Alchemical symbols international → Europe
Chinese divination international → CJK (icon only, do not collapse and/or count as CJK tofu)
Put RLM to Kharoshthi and Hanifi descriptions → now text direction works properly
BUGS/IMPROVEMENTS
Decimal search
Relaxed dependence on font loading order: Noto Symbol2 is loaded first whenever we load other font
Tofu table now has fixed rows
Made base for custom antialiasing (no antialiasing itself, and not enabled right now)
Simplified engine a bit
Better search if you are looking for non-existing codepoint
You can copy any codepoint, not just existing
0.4
Did not start on some W7 computers — Qt 6.2 had strange dependency to non-standard M$ library. Reverted to 6.1.3
Replaced Balinese font with high-contrast one (aka serif, though Balinese has no serifs)
Updated Scheherazade font to 3.3 → fixed Arabic tofu except presentation forms
Fixed misrender of 20E7 annuity symbol
Fixed small U14 tofu: Latin D, Balinese, umlauts, punctuation, Glagolitic, Arabic presentation forms
More detailed tofu statistics, now accessible from UI, not just by hidden hotkey
More blocks in tofu stats are considered CJK
No more tofu in Base Plane, even in W7
2230 tofu CJK, 229 rest
0.3
New scripts: Ottoman Siyaq numbers, Tangsa, U14 emoji
Replaced funky Nandinagari with fully functional (Google Noto)
Important milestone: no more U12 tofu, in U13 only CJK remain
Six remaining U14 tofu blocks (Arabic B, Latin F/G, Kana B, Cypro-Minoan, Uyghur) are marked with special icon
Fixed tofu counting, Unicodia falsely reported 51 chars of Tamil Supplement as tofu
Reduced probability of “falsely-focused” combobox
When copy-paste and Alt-Tab to other software, “Copied” window no more hangs
2217 CJK tofu, 341 rest