Font

NotoNastaliqUrdu-Regular.ttf

Where the font came from, and when

Site: https://github.com/notofonts/nastaliq/releases/download/NotoNastaliqUrdu-v3.007/NotoNastaliqUrdu-v3.007.zip
Date: 2022-08-04

Font Version

Version 3.007

Issue

If we change the language of the text to (e.g.) Persian, and then use the digits with a subtending mark, then the localized forms of digits do not appear:

hb-view -o out-3.007-fa.png --language=fa NotoNastaliqUrdu-Regular-3.007.ttf "؀۴۶۷ ؁۴۶۷ ؂۴۶۷ ؃۴۶۷ ؄۴۶۷"

Output from v3.007:

As far as I can tell, this used to work correctly before v3.001 (example below uses v2.001):

hb-view -o out-2.001-fa.png --language=fa NotoNastaliqUrdu-Regular-2.001.ttf "؀۴۶۷ ؁۴۶۷ ؂۴۶۷ ؃۴۶۷ ؄۴۶۷"

Output from v2.001:

Font

NotoNastaliqUrdu-Regular.ttf

Where the font came from, and when

https://github.com/googlefonts/noto-fonts/tree/master/hinted/ttf/
installed today

2022-01-28

Font Version

2.000;GOOG;NotoNastaliqUrdu-Regular

OS name and version

macOS 12.1

Application name and version

Browsers

Issue

The sound /e/ is represented in Kashmiri using:
یٚ [U+06CC ARABIC LETTER FARSI YEH + U+065A ARABIC VOWEL SIGN SMALL V ABOVE]

The small sign v above appears to be missing from the Noto Nastaliq Urdu font, and the result is that when used medially this letter fails to join to the left. This is obviously a big problem for using the Noto font for Kashmiri text.

Character data

You can find a list of lemmas containing this sequence at https://r12a.github.io/scripts/arabic/ks_vocab?q=%DB%8C%D9%9A.

Here are the top 4 from the list, using the Noto Nastaliq Urdu font.

Here's what it looks like using SIL's Awami Nastaliq font:

notofonts/noto-fonts#1646

Font

'NotoNastaliqUrdu-Regular.ttf'

Site: https://www.google.com/get/noto/#Noto%20Nastaliq%20Urdu

Issue

bad rendering for third character.

Screenshot

hb-view NotoNastaliqUrdu-Regular.ttf -u "U+0634;U+0634;U+0634;U+0634;U+0634;"

hb-view NotoNastaliqUrdu-Regular.ttf -u "U+0634;U+0634;U+0634;U+0634;U+0634;U+0634;"

hb-shape NotoNastaliqUrdu-Regular.ttf -u "U+0634;U+0634;U+0634;U+0634;U+0634;U+0634;"

[ThreeDotsUpAboveNS=5@1753,-149+0|SeenFin=5+2131|ThreeDotsUpAboveNS=4@481,18+0|SeenMed.SWinAoutT2=4@0,932+248|ThreeDotsUpAboveNS=3@1404,477+0|SeenMed.inT2outD1=3@0,997+1838|ThreeDotsUpAboveNS=2@567,652+0|SeenMed.SWinBoutT2=2@0,1514+266|Th
reeDotsUpAboveNS=1@567,720+0|SeenMed.SWinBoutT2=1@0,1582+266|ThreeDotsUpAboveNS=0@1408,1146+0|sp0=0+0|SeenIni.outD1=0@0,1650+1978]

Font

Name: Noto Nastaliq Urdu
Version: Version 2.000
Glyph count: 1138

Issue

Anchor placement in SeenMed.SWinDoutT2 leads to incorrect shaping:

$ shape NotoNastaliqUrdu-Regular.ttf -u '6be 633 633 633'
[SeenFin=3+1040|SeenMed.SWinDoutT2=2@0,455+182|SeenMed.inT2outD1=1@0,461+897|sp0=0+0|HehDoIni.outT2=0@0,713+557]

It's not entirely clear where the correct placement for that anchor is; in fact, I'm not sure the thicknesses actually match anywhere, which makes me suspect that the problem might be with the connected shaping rules.

Dots and dictrics may need some review. In some combinations dots and diacritics could overlap or touch each other. the word "اسْتَوَى" in :هُوَ الَّذِي خَلَقَ لَكُم مَّا فِي الأَرْضِ جَمِيعًا اسْتَوَى ثُمَّ إِلَى السَّمَاء فَسَوَّاهُنَّ سَبْعَ سَمَاوَاتٍ وَهُوَ بِكُلِّ شَيْءٍ shows this behavior. Since Arabic heavily depends on dielectrics on daily use and there extensive use in Quran I would think this is one of the issues we should look into.

Issue found in:


NotoNastaliqUrdu Version 2.000

https://github.com/googlefonts/noto-fonts/tree/master/unhinted/ttf/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.ttf

Characters data:
تتَوَ تثَوَ تثَوَتغَؤَ اسْتَوَى تتَوُ تثَوُ تثَوَتغَؤُ اسْتَوُ تتَوٌ تثَوٌ تثَوَتغَؤٌ اسْتَوٌى تتَمٌو تثَمٌو تثَمَتغَؤٌ اسْتَمٌو

Screenshot:

Font

NotoNastaliqUrdu-Regular.ttf

Where the font came from, and when

https://github.com/googlefonts/noto-fonts/tree/master/hinted/ttf/
installed today

2022-01-28

Font Version

2.000;GOOG;NotoNastaliqUrdu-Regular

OS name and version

macOS 12.1

Application name and version

Browsers

Issue

The Noto Nastaliq Urdu font produces s-shaped glyphs for combining hamzas. Kashmiri, however, generally uses the rounded shapes. It would be good to use rounded hamzas if the font detects that the language of the content is Kashmiri.

The following shows hamzas using the Noto font:

Here is the same text using SIL's Awami Nastaliq font (with appropriate font settings to produce the expected glyphs):

Note that the wavy hamza does have the rounded shape.

The images above were taken from https://r12a.github.io/scripts/arabic/kashmiri#vowel_letters

For sample lemmas containing hamzas, see https://r12a.github.io/scripts/arabic/ks_vocab?q=%D8%A5%7C%D9%95%7C%D8%A3%7C%D9%94

Font

NotoNastaliqUrdu-Regular.otf

Where the font came from, and when

Site: https://raw.githubusercontent.com/googlefonts/noto-fonts/29aa92a9a0768be2d58cf4c590adb5c18b8247c6/unhinted/otf/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.otf
Date: 2021-12-09

Font version

Version 2.001

Issue

The ijam above a lam are too high when the lam precedes another lam. When a meem separates the lams, for example, the ijam are positioned correctly.

Character data

ڷل ڷمل
U+06B7 ARABIC LETTER LAM WITH THREE DOTS ABOVE
U+0644 ARABIC LETTER LAM
U+0020 SPACE
U+06B7 ARABIC LETTER LAM WITH THREE DOTS ABOVE
U+0645 ARABIC LETTER MEEM
U+0644 ARABIC LETTER LAM
ࣇل ࣇمل
U+08C7 ARABIC LETTER LAM WITH SMALL ARABIC LETTER TAH ABOVE
U+0644 ARABIC LETTER LAM
U+0020 SPACE
U+08C7 ARABIC LETTER LAM WITH SMALL ARABIC LETTER TAH ABOVE
U+0645 ARABIC LETTER MEEM
U+0644 ARABIC LETTER LAM

Screenshot

This is regarding:

• U+0601 ARABIC SIGN SANAH
• U+0602 ARABIC FOOTNOTE MARKER
• U+0603 ARABIC SIGN SAFHA

All of these marks are used in conjunction with digits that follow the marks. When rendered correctly (as in e.g. BabelPad), the digits sit on top of the marks. In Firefox and Chrome, however, digits are displayed before the marks.

U+0602 ARABIC FOOTNOTE MARKER behaves correctly in IE and Word 2010. However, U+0601 ARABIC SIGN SANAH and U+0603 ARABIC SIGN SAFHA make IE and Word act weird: a line containing these marks is not rendered at all in IE, while Word doesn't display any characters in a line before (and including) these marks. Remove these marks, and the lines are rendered without any issues in both IE and Word.

Screenshots are attached. Test HTML page here: http://files.saadatmand.pk/type/noto-nastaliq-marks.html

Platform: Windows 7
Software: Firefox 38.0.5, Chrome 43.0.2357.81 m, IE 9&11, Word 2010

▲ BabelPad

▲ Firefox

▲ Chrome

▲ IE

▲ Microsoft Word

Some Arabic characters in Noto Nastaliq Urdu Regular are rendered in a different typeface from others. Characters affected include those in the Arabic Extended-A Unicode block as well as newly added characters in the Arabic and Arabic Supplement Unicode blocks.
The following image shows the issue:

The dot of U+0768 is too high compared to U+0646 in Noto Nastaliq Urdu (both regular and bold). The following image shows the issue:

In Noto Nastaliq Urdu, if the Arabic sign sanah (U+0601) is encoded first before the number, the subtending feature does not work. If it is encoded after the number, the subtending feature works.
With Arabic sign sanah encoded after the digits:

With Arabic sign sanah encoded before the digits:

For comparison, in Awami Nastaliq, the Arabic sign sanah must be encoded first before the digits. For example:

Title

Noto Nastaliq support for Balti Graf U+08C8

Font

NotoNastaliqUrdu-Regular.ttf
NotoNastaliqUrdu-Bold.ttf

Issue

This is actually a request not an issue. Recently Balti Graf U 08C8 is added in Unicode 14 version.
Can you please add this alphabet in Noto Nastaliq font? It's shape is similar to Arabic Gaaf.

Fonts

NotoNastaliqUrdu-Regular.ttf
Came from: https://github.com/googlefonts/noto-fonts/blob/f01b460a7125a4025988a59e0616973ce9eb91c5/unhinted/ttf/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.ttf
Version: 3.002

NotoNastaliqUrdu-Regular.ttf
Came from: https://github.com/googlefonts/noto-fonts/blob/d5aa70fe8b425321411bb92d5cc901f87ac727b9/unhinted/ttf/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.ttf
Version: 3.003

Issue

The swash variants of U+0633 ARABIC LETTER SEEN (س) or U+0634 ARABIC LETTER SEEN (ش) (originally reported in #46) were fixed in notofonts/noto-source@88c7d85. However, this introduced a regression for the following sequences:

سس شش
سسر ششر

Screenshots

v3.002 (swash correctly appearing):

v3.003 (swash not appearing anymore):

Font

NotoNastaliq.ttc

Font Version

14.0d2e19

OS name and version

macOS Big Sur

Application name and version

web browsers

Issue

The font has language-specific features that can be displayed in a browser such as Firefox. In particular, if the language of content in an HTML page is set to 'ur' the sukun is rendered as an inverted v, and the hamza is rendered in its 's' shaped form. These alternate renderings are also needed in Kashmiri, but setting the language of the text to Kashmiri doesn't produce that glyph.

Please add ks to the list of languages recognised by the font to cause selection of these glyphs.

Character data

lang="ur"

lang="ks"

واریُْل
و U+0648: ARABIC LETTER WAW
ا U+0627: ARABIC LETTER ALEF
ر U+0631: ARABIC LETTER REH
ی U+06CC: ARABIC LETTER FARSI YEH
ُ U+064F: ARABIC DAMMA​
ْ U+0652: ARABIC SUKUN​
ل U+0644: ARABIC LETTER LAM

lang="ur"

lang="ks"

وَنْدٕ
و U+0648: ARABIC LETTER WAW
َ U+064E: ARABIC FATHA​
ن U+0646: ARABIC LETTER NOON
ْ U+0652: ARABIC SUKUN​
د U+062F: ARABIC LETTER DAL
ٕ U+0655: ARABIC HAMZA BELOW​

Fonts

NotoNastaliqUrdu-Regular.ttf
Came from: https://github.com/googlefonts/noto-fonts/blob/58e9e4f4f0abd3104437c0341ffaa53170c02d34/unhinted/ttf/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.ttf
Version: 2.001

NotoNastaliqUrdu-Regular.ttf
Came from: https://github.com/googlefonts/noto-fonts/blob/00bf5493fd1e7e0e3a481d7c003c15961f93483f/unhinted/ttf/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.ttf
Version: 3.001

Issue

In v3.001, the smaller top stroke of U+06AF ARABIC LETTER GAF (گ) is misplaced in a couple of initial gaf glyphs. This was correct in v2.001.

Test strings

گب گن گنت گنتی گد

Correct behaviour (in v2.001)

In v2.001, all of the above test strings have correct gaf strokes.

hb-view -u "06AF;0628;0020;06AF;0646;0020;06AF;0646;062A;0020;06AF;0646;062A;06CC;0020;06AF;062F" --margin=128 -o output-gaf-2.001.png NotoNastaliqUrdu-Regular-2.001.ttf

Incorrect behaviour (in v3.001)

In v3.001, the first three strings (from right to left) show gaf with correct strokes; the last two have incorrect strokes.

hb-view -u "06AF;0628;0020;06AF;0646;0020;06AF;0646;062A;0020;06AF;0646;062A;06CC;0020;06AF;062F" --margin=128 -o output-gaf-3.001.png NotoNastaliqUrdu-Regular-3.001.ttf

Using HarfBuzz or Pango create a PDF of text rendered in Noto Nastaliq Urdu 
[0], observe that it's broken and doesn't match the PNG output from the same 
tool.  Eg:

$ hb-view NotoNastaliqUrduDraft.ttf --font-size 24 --text-file sample 
--output-file sample.pdf

Where sample includes following text:

تمام انسان آزاد اور حقوق و عزت کے اعتبار سے 
برابر پیدا ہوۓ ہیں۔
انہیں ضمیر اور عقل ودیعت ہوئی ہے۔ اسلیۓ 
انہیں ایک دوسرے کے 
 ساتھ بھائی چارے کا سلوک کرنا چاہیۓ۔                                                                             


[0] https://noto.googlecode.com/git/fonts/alpha/NotoNastaliqUrduDraft.ttf

Original issue reported on code.google.com by [email protected] on 19 Nov 2014 at 1:25

NotoNastaliqUrdu seems to support other languages beyond Urudu. Should it really have “Urdu” in its name? Or should it be renamed to just “NotoNastaliq”?

In theory, we could release a NotoNastaliqPersian, NotoNastaliqKashmiri etc. with different default language systems but otherwise identical glyphs and lookups. This would be similar to what we currently do for NotoSansCJK. However, this would make things more complicated to maintain and to explain. Also, not sure if the marginal benefit would be worth the complexity: afaik localized letterforms are supported by all the modern platforms we care about; including Android, Firefox, and Chrome.

Font

NotoNastaliqUrdu-Regular.otf

Where the font came from, and when

Site: https://github.com/googlefonts/noto-fonts/blob/ca434f80edeb6a1ecb65f4d40d69691d01f65abd/unhinted/otf/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.otf
Date: 2022-02-02

Font version

Version 3.002

Issue

This font tries to ligate <lam, lam, heh> when certain diacritics are present. However, the FormDivineName lookup is broken, so nothing happens. That lookup should ignore all marks.

Character data

للّٰه
U+0644 ARABIC LETTER LAM
U+0644 ARABIC LETTER LAM
U+0651 ARABIC SHADDA
U+0670 ARABIC LETTER SUPERSCRIPT ALEF
U+0647 ARABIC LETTER HEH

Screenshot

Intended rendering:

Given the broadly Persianate of the Turkic linguistic cultures of Central Asia that have historically used—and in some countries still use—Arabic script, it is appropriate to use nastaliq style fonts for these languages, but so far I have been unable to find any nastaliq font that supports all the characters for all these languages. Noto Nastaliq Urdu supports several of the historical and current alphabets of some of these languages, but not all. While a specifically Persian style nastaliq would be ideal, extending Noto Nastaliq Urdu to support these characters would be a helpful first step.

I don’t have a complete and verified list of the missing characters that should be added, so am creating this issue as a place to log them. The following are ones I have identified so far:

KAZAKH
ٵ ٶ ٷ ۋ ٸ
ٴ
KYRGYZ
ۅ ﯗ ۉ ﯞ

[My immediate need regarded orthographies in the Soviet Republics immediately prior to the adoption of short-lived Latin script alphabets in the 1920s, so I have not yet looked closely at current use of Arabic script for these languages in Iran or other countries.]

Font

Noto Nastaliq Urdu
Came from: https://github.com/googlefonts/noto-fonts/blob/d5aa70fe8b425321411bb92d5cc901f87ac727b9/unhinted/ttf/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.ttf

Issue

According to this commit, the current version of Noto Nastaliq Urdu is 3.003, but the font file is reporting that the version is still 3.002.

I am a native speaker of the Crimean Tatar language, who now uses Cyrillic and Latin, but before the 1940s, we used Arabic script, including Nastaliq, and sometimes there is a need to write in Arabic letters.
And I faced the problem that in Urdu Noto Nastalik there is no letter ڭ, which is obligatory for the Crimean Tatar language.
By the way, this letter is also found in Ottoman, Uighur, Kazakh, Uzbek and other Turkic languages ​​that use nastalik.

I would like this letter to be added to the Urdu Noto Nastalik font.
I'm not a programmer, and not a calligrapher, but if you need non-qualified help, I'm ready to help.
The letter should look something like this:

Below codepoints missing/mispositioned anchors:
076B 0631 075C 076D 0633 075D 0639 0763 0643 0762.
This behaviour can be seen in NotoNaskhArabic NotoNaskhArabicUI and NotoNastaliqUrdu (All weights)

Issue found in:


NotoNaskhArabic Version 2.000
https://github.com/googlefonts/noto-fonts/tree/master/unhinted/ttf/NotoNaskhArabic
NotoNaskhArabicUI Version 2.000
https://github.com/googlefonts/noto-fonts/tree/master/unhinted/ttf/NotoNaskhArabicUI
NotoNastaliqUrdu Version 2.000

https://github.com/googlefonts/notofonts/tree/master/unhinted/ttf/NotoNastaliqUrdu

Characters used in the test:
تݫُݫُ ترُرُ ݜُ ݜُݜُݜُ ݭُ ݭُݭُݭُ سُ سُسُسُ ݝُ ݝُݝُݝُ عُ عُعُعُ ݣُ ݣُݣُݣُ كُ كُكُكُ ݢُ ݢُݢُݢُ

Screenshot:

The upper stroke of U+06AF is Noto Nastaliq Urdu Bold is sometimes misplaced.
The following is an image of an example:

In Noto Nastaliq Urdu, when text is set in Sindhi, the digits 4 and 7 are incorrectly shaped. The following is a screenshot of numerals in Noto Nastaliq Urdu:

Numerals in Sindhi must look like the following:

The Arabic letter waw (U+0648) is sometimes pointed in Noto Nastaliq Urdu Bold. The following is an image of an example:

The following is a partial list of missing characters in Noto Nastaliq Urdu (as of March 2022): U+0682, U+0690, U+0692, U+0694, U+0695, U+0697, U+069D, U+069F, U+06A3, U+06A5, U+06AA, U+06AC, U+06AD, U+06AE, U+06B2, U+06B4, U+06B5, U+06B6, U+06B8, U+06B9, U+06BF, U+06C5, U+06C8, U+06C9, U+06CB, U+06D1, U+0764, U+0765, U+0766.

Font

NotoNastaliqUrdu-Regular.otf

Where the font came from, and when

Site: https://github.com/googlefonts/noto-fonts/blob/f7556efcaf940cf6c2f8ae0c3386d651e930b5cc/unhinted/otf/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.otf
Date: 2022-02-03

Font version

Version 3.002

Issue

The glyphs of U+08AD ARABIC LETTER LOW ALEF in L2/11-209 consistently have their highest points at the baseline. The glyph in Noto Nastaliq Urdu has its lowest point at the baseline.

Character data

ࢭیو
U+08AD ARABIC LETTER LOW ALEF
U+06CC ARABIC LETTER FARSI YEH
U+0648 ARABIC LETTER WAW

Screenshot

Font

NotoNastaliqUrdu-Regular.ttf and NotoNastaliqUrdu-Bold.ttf
Site: https://github.com/googlefonts/noto-fonts/tree/ad82c41a1b0277ec45081beb6abedba788598074/unhinted/ttf/NotoNastaliqUrdu Version: 3.003

Issues

The Nuqta is being rendered at the top along with the Tah in the isolated form of U+0768 ARABIC LETTER NOON WITH SMALL TAH when it should be at the same position as in U+0646 ARABIC LETTER NOON. Furthermore, the Tah should be very slightly higher than the Nuqta in all forms.

Regular:

Bold:

This how a Punjabi centric Nastaliq font, Roz Urdu Punjabi Nastaleeq, renders it:

The Tah in U+08C7 ARABIC LETTER LAM WITH SMALL ARABIC LETTER TAH ABOVE should be exactly at the top of the danda of the lam as instructed by Unicode:

But Noto Nastaliq renders it like this in Regular:

and Bold:

Both Isolated and Final forms are affected.

Fonts

NotoNastaliqUrdu-Regular.ttf
Came from: https://github.com/googlefonts/noto-fonts/blob/ad82c41a1b0277ec45081beb6abedba788598074/unhinted/ttf/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.ttf
Version: 2.000

NotoNastaliqUrdu-Regular.ttf
Came from: https://github.com/googlefonts/noto-fonts/blob/f01b460a7125a4025988a59e0616973ce9eb91c5/unhinted/ttf/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.ttf
Version: 3.002

Issue

In v2.000, if two U+0633 ARABIC LETTER SEEN (س) or U+0634 ARABIC LETTER SEEN (ش) came together (or if one is followed by the other), then the first س or ش would be replaced with a nice swash variant.

This has stopped working since v2.001. Please bring the swash س/ش back. 🙂

Strings

سسا سسب سسج سسد سسر سسس سسڈ سسع سسف سسق سسک سسک سسل سسم سسن سسو سسہ سسی سسے
ششا ششب ششج ششد ششر ششش ششڈ ششع ششف ششق ششک ششک ششل ششم ششن ششو ششہ ششی ششے

Screenshot

v2.000:

Since v2.001 (here shown in v3.002):

Note that for some sequences (like سسر), the swash does appear. The swash(es) also appear if more than two س or ش come together (as in سسس, and as detailed in #22 and #24.)

Title

Noto Nastaliq support for 5 Hindko characters U+08BE - U+08C2

Font

NotoNastaliqUrdu-Regular.ttf
NotoNastaliqUrdu-Bold.ttf

Issue

This is actually a request not an issue.

Can you please add following Hindko characters to Noto Nastaliq font?

1. U+08BE - ARABIC LETTER PEH WITH SMALL V. It's shape is similar to U+067E (Arabic Letter Peh)
2. U+08BF - ARABIC LETTER TEH WITH SMALL V. It's shape is similar to U+062A (Arabic Letter Teh)
3. U+08C0 - ARABIC LETTER TTEH WITH SMALL V. It's shape is similar to U+0679 (Arabic Letter Tteh)
4. U+08C1 - ARABIC LETTER TCHEH WITH SMALL V. It's shape is similar to U+0686 (Arabic Letter Tcheh)
5. U+08C2 - ARABIC LETTER KEHEH WITH SMALL V. It's shape is similar to U+06A9 (Arabic Letter Keheh)

Fonts

NotoNastaliqUrdu-Regular.ttf
Came from: https://github.com/googlefonts/noto-fonts/blob/58e9e4f4f0abd3104437c0341ffaa53170c02d34/unhinted/ttf/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.ttf
Version: 2.001

NotoNastaliqUrdu-Regular.ttf
Came from: https://github.com/googlefonts/noto-fonts/blob/00bf5493fd1e7e0e3a481d7c003c15961f93483f/unhinted/ttf/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.ttf
Version: 3.001

Issue

In v2.001, dot positioning for sequences that end in U+06D2 ARABIC LETTER YEH BARREE (ے) was mostly correct. In v3.001, there are a lot more incorrect cases (for ligatures of 4 or more characters).

Some test strings and examples are below. I also made a test page that lists valid Urdu ligatures ending in yeh barree (taken from CRULP/CLE) and allows switching between versions 2.001 and 3.001 of Noto Nastaliq Urdu.

Test strings for examples below

نیچے جھپٹتے بھیجتے کھینچتے

Correct behaviour (in v2.001)

In v2.001, dots for various characters are positioned correctly and without collision.

hb-view -u "0646;06CC;0686;06D2;0020;062C;06BE;067E;0679;062A;06D2;0020;0628;06BE;06CC;062C;062A;06D2;0020;06A9;06BE;06CC;0646;0686;062A;06D2" --margin=128 -o output-lig-2.001.png NotoNastaliqUrdu-Regular-2.001.ttf

Incorrect behaviour (in v3.001)

In v3.001, dots are incorrectly positioned or not even present.

hb-view -u "0646;06CC;0686;06D2;0020;062C;06BE;067E;0679;062A;06D2;0020;0628;06BE;06CC;062C;062A;06D2;0020;06A9;06BE;06CC;0646;0686;062A;06D2" --margin=128 -o output-lig-3.001.png NotoNastaliqUrdu-Regular-3.001.ttf

Please see more test cases on the test page. There are a few sequences (like پچھلے) which show incorrect dot positioning in both fonts.

Font

NotoNastaliqUrdu-Regular.otf

Where the font came from, and when

Site: https://github.com/googlefonts/noto-fonts/blob/f7556efcaf940cf6c2f8ae0c3386d651e930b5cc/unhinted/otf/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.otf
Date: 2022-02-03

Font version

Version 3.002

Issue

Some positional forms of some letters have partly or wholly invisible glyphs.

Character data

[ؠ]
U+005B LEFT SQUARE BRACKET
U+0620 ARABIC LETTER KASHMIRI YEH
U+005D RIGHT SQUARE BRACKET
[ݺل لݺل]
U+005B LEFT SQUARE BRACKET
U+077A ARABIC LETTER YEH BARREE WITH EXTENDED ARABIC-INDIC DIGIT TWO ABOVE
U+0644 ARABIC LETTER LAM
U+0020 SPACE
U+0644 ARABIC LETTER LAM
U+077A ARABIC LETTER YEH BARREE WITH EXTENDED ARABIC-INDIC DIGIT TWO ABOVE
U+0644 ARABIC LETTER LAM
U+005D RIGHT SQUARE BRACKET
[ݻل لݻل]
U+005B LEFT SQUARE BRACKET
U+077B ARABIC LETTER YEH BARREE WITH EXTENDED ARABIC-INDIC DIGIT THREE ABOVE
U+0644 ARABIC LETTER LAM
U+0020 SPACE
U+0644 ARABIC LETTER LAM
U+077B ARABIC LETTER YEH BARREE WITH EXTENDED ARABIC-INDIC DIGIT THREE ABOVE
U+0644 ARABIC LETTER LAM
U+005D RIGHT SQUARE BRACKET
[ࢬ]
U+005B LEFT SQUARE BRACKET
U+08AC ARABIC LETTER ROHINGYA YEH
U+005D RIGHT SQUARE BRACKET

Screenshot

Title

Noto Nastaliq support for U+08C7 LAM WITH SMALL TAH

Font

NotoNastaliqUrdu-Regular.ttf
NotoNastaliqUrdu-Bold.ttf

Issue

This is actually a request not an issue. Can you please add U+08C7 (ARABIC LETTER LAM WITH SMALL ARABIC
LETTER TAH ABOVE) to Noto Nastaliq font? It's shape is similar to Arabic Laam (U+0644).

see google/fonts#4272

Consider the following Persian text I encountered online:

ویدئو ﻫﺎﯼ ﺗﻮپ

If I set the font to Noto Nastaleeq, I get this weird mixture of Naksh and Nastaleeq:

In the string above, the second part uses Arabic Presentation Forms:

ﻫﺎﯼ ﺗﻮپ

I’m assuming Noto Nastaleeq does not support these, but but my system font does. Noto Nastaleeq should support these.

Font

NotoNastaliqUrdu-Regular.otf

Where the font came from, and when

Site: https://github.com/googlefonts/noto-fonts/blob/f7556efcaf940cf6c2f8ae0c3386d651e930b5cc/unhinted/otf/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.otf
Date: 2022-02-03

Font version

Version 3.002

Issue

The curly fatha and kasra are practically indistinguishable from their straight counterparts.

Character data

دࣤدࣦدࣧدࣩ
U+062F ARABIC LETTER DAL
U+08E4 ARABIC CURLY FATHA
U+062F ARABIC LETTER DAL
U+08E6 ARABIC CURLY KASRA
U+062F ARABIC LETTER DAL
U+08E7 ARABIC CURLY FATHATAN
U+062F ARABIC LETTER DAL
U+08E9 ARABIC CURLY KASRATAN

دَدِدًدٍ
U+062F ARABIC LETTER DAL
U+064E ARABIC FATHA
U+062F ARABIC LETTER DAL
U+0650 ARABIC KASRA
U+062F ARABIC LETTER DAL
U+064B ARABIC FATHATAN
U+062F ARABIC LETTER DAL
U+064D ARABIC KASRATAN

Screenshot

Font

NotoNastaliqUrdu-Regular.otf

Where the font came from, and when

Site: https://github.com/googlefonts/noto-fonts/blob/97e5bf0bbaf85c14498696b408d60b517b0e9e70/unhinted/otf/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.otf
Date: 2022-02-08

Font version

Version 3.002

Issue

The initial and medial glyphs for U+06BD should actually have three dots below. The final and isolated glyphs should have the rasm of noon.

Character data

ڽڽڽ ڽ
U+06BD ARABIC LETTER NOON WITH THREE DOTS ABOVE
U+06BD ARABIC LETTER NOON WITH THREE DOTS ABOVE
U+06BD ARABIC LETTER NOON WITH THREE DOTS ABOVE
U+0020 SPACE
U+06BD ARABIC LETTER NOON WITH THREE DOTS ABOVE

Screenshot

Font

Name: Noto Nastaliq Urdu
Version: Version 2.000
Glyph count: 1138
Source: https://github.com/googlefonts/noto-fonts/blob/master/unhinted/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.ttf

Issue

Repeated beh glyphs in a yeh-barree sequence clash:

shape ../NotoNastaliqUrdu-Regular.ttf 'ببے بببے'
[YBc1=7@764,-282+0|YBc2=7@1098,-159+0|YehBarreeFin_4=7+355|OneDotBelowNS=6@20,-120+0|BehxMed.inT2outD2YB=6@0,349+182|OneDotBelowAltNS=5@231,69+0|BehxMed.inT1outT2=5@0,406+184|OneDotBelowYB=4@73,306+0|sp5=4+0|BehxIni=4@0,471+541|space=3+132|YBc1=2@764,-282+0|YehBarreeFin_3=2+355|OneDotBelowNS=1@20,-120+0|BehxMed.inT2outD2YB=1@0,349+182|OneDotBelowYB=0@127,255+0|sp10=0+0|BehxIni.outT2=0@0,406+766]

There are two bugs here. The longer sequence clashes because the repeated-beh-handling lookups (161 and 162) are replacing the first OneDotBelowNS with a OneDotBelowAltNS even in a yeh barree context.

On further investigation, the shorter sequence clashes because all beh ... yeh-barree sequences clash when the beh is not immediately next to the YB:

hb-view  -O png ../NotoNastaliqUrdu-Regular.ttf 'بلے سبلے بسے بنلے سبفے بکے'

Title

Should be brief but contain essential information.

Font

NotoNastaliqUrdu-Regular.ttf

Where the font came from, and when

https://fonts.google.com/noto/specimen/Noto+Nastaliq+Urdu

Font Version

The latest

OS name and version

Windows 11 22H2

Application name and version

Notepad（UWP Version）

Issue

Wrong display of ۋ.
（While in modern days,Uyghur is almost always written in Naskh,there do have calligraphic works written in Nastaliq as Uyghur culture was strongly influenced by Persian.）

Screenshot

Fonts

NotoNastaliqUrdu-Regular.ttf
Version 1.02uh
Came from: https://noto-website-2.storage.googleapis.com/pkgs/NotoNastaliqUrdu-unhinted.zip
Date: 2019-09-14
README says this:

Built on 2017-10-24 from the following noto repository:

Repo: noto-fonts
Tag: v2017-10-24-phase3-second-cleanup
Date: 2017-10-24 12:10:34 GMT
Commit: 8ef14e6c606a7a0ef3943b9ca01fd49445620d79

NotoNastaliqUrdu-Regular.ttf
Version 2.000
Came from: https://github.com/googlefonts/noto-fonts/blob/master/unhinted/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.ttf
Date: 2019-09-14

Issue

In v1.02uh, U+0600 ARABIC NUMBER SIGN has two shapes:

• Shape 1: when it is isolated, or followed by a single digit
• Shape 2: when it is followed by two or more digits
  

In v2.000, this is corrected and U+0600 has a single shape (Shape 2), no matter the context of its usage:

I personally prefer Shape 1, and think that it matches the rest of the font design better than Shape 2. Shape 1 also closely resembles the words نمره or نمبر (from which U+0600 was most probably derived.)

So here's my request: can we please bring Shape 1 back for U+0600? If Shape 2 must be the default, then maybe Shape 1 can be made available under a stylistic set or as a character variant?

Urdu Wikipedia uses a Kashida variant of a Nastaliq font, "Jameel Noori Kasheeda", instead making the default font, "Jameel Noori Nastaleeq", italic. Kashida variant on here means making text occupy twice as space they do in normal typography [may even more] and this is a technique to avoid making Italic text which is not familiar concept for Arabic script and specially Nastaliq.

In order to help Urdu Wikipedia on adapting Noto Urdu Nastaliq font, it would be nice if Noto font could provide Kashida variant by supporting a OpenType feature (say, a feature that could be enabled font-feature-settings: "ss01" 1; on CSS) or making another variant of the font specially for making Kashida applied text. Thank you.

Font

NotoNastaliqUrdu-Regular.otf

Where the font came from, and when

Site: https://github.com/googlefonts/noto-fonts/blob/97e5bf0bbaf85c14498696b408d60b517b0e9e70/unhinted/otf/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.otf
Date: 2022-02-08

Font version

Version 3.002

Issue

U+25CC DOTTED CIRCLE lacks anchors.

Character data

◌ِّ
U+25CC DOTTED CIRCLE
U+0650 ARABIC KASRA
U+0651 ARABIC SHADDA

Screenshot

Fonts

NotoNastaliqUrdu-Regular.ttf
Came from: https://github.com/googlefonts/noto-fonts/blob/ad82c41a1b0277ec45081beb6abedba788598074/unhinted/ttf/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.ttf
Version: 2.000

NotoNastaliqUrdu-Regular.ttf
Came from: https://github.com/googlefonts/noto-fonts/blob/f01b460a7125a4025988a59e0616973ce9eb91c5/unhinted/ttf/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.ttf
Version: 3.002

Issue

This is like #36, with the exception that these sequences (listed below) have been rendering incorrectly since v2.001 (the screenshot below uses v3.002 to show the incorrect rendering). The last version which handled them correctly was v2.000.

Strings

بجھتے
بجھنے
بجھئے
بچپنے
بچکتے
بچکنے
بچکئے
بچلتے
بچلنے
بچلئے
بچھتے
بچھنے
بچھئے
پچکتے
پچکئے
پچھتے
پچھلے
پچھنے
پچھئے
پیچھے
جیٹھے
چیختے
چیخئے
چینگے
ہچکتے
ہچکئے
چبھتے
چبھئے
چکلتے
چکلئے
چہیتے

Screenshot

In v2.000, dots are correctly positioned.
In v3.002, they are incorrectly positioned and/or colliding with each other.
(The last five strings are not as bad as the rest, but their dots and hook positioning can still be made better.)

Hello. Thank you for providing us with great fonts.
I’d like to report what a native Urdu speaker hopes to have updated in Noto Nastaliq Urdu.

It applies to all Noto Nastaliq Urdu fonts:
https://github.com/notofonts/nastaliq/releases/download/NotoNastaliqUrdu-v3.007/NotoNastaliqUrdu-v3.007.zip

It is hope for the character form for Arabic Thousands Separator (U+066C) to stay as it is, and lower to the baseline, similar to the English comma.
By the way, Noto Naskh Arabic has the design this way.
https://github.com/notofonts/arabic/releases/download/NotoNaskhArabic-v2.016/NotoNaskhArabic-v2.016.zip

Please find below the links of a few pages where you can find some examples:
https://akhbar-e-jehan.com/detail/42535/executive-director-of-commission-renowned-bureaucrat-and-former-ambassador-dr-muhammad-nafees-zakaria

https://akhbar-e-jehan.com/detail/42618/referendum-for-indigenous-parliamentary-rights-in-australia

https://www.express.pk/story/761623/?__cf_chl_tk=y3PGyKn4m2nbjYD4wGJOkif.fcS0SiL3vDo1N1CSFfY-1682502157-0-gaNycGzNEjs

https://akhbar-e-jehan.com/detail/42764/why-fifteen-minute-cities-are-becoming-popular-in-the-world

Thank you for taking in this request into consideration.

Some letters in Noto Nastaliq Urdu Regular have no contextual forms. Such letters do not connect properly.
For example:

Comparison with Noto Nastaliq Urdu Bold:

Fonts

NotoNastaliqUrdu-Regular.ttf and NotoNastaliqUrdu-Bold.ttf
Came from: https://github.com/googlefonts/noto-fonts/tree/ad82c41a1b0277ec45081beb6abedba788598074/unhinted/ttf/NotoNastaliqUrdu
Version: 2.000

NotoNastaliqUrdu-Regular.ttf and NotoNastaliqUrdu-Bold.ttf
Came from: https://github.com/googlefonts/noto-fonts/tree/d5aa70fe8b425321411bb92d5cc901f87ac727b9/unhinted/ttf/NotoNastaliqUrdu
Version: 3.003

Issue

Some two-lettered sequences that end in U+0648 ARABIC LETTER WAW (و) have joining issues that were not present in v2.000. (As far as I can tell, these issues first popped up in v3.001.)

First, the more obvious ones:

کو لو
U+06A9 U+0648  U+0644 U+0648

In v2.000, ک and ل have a nice bend when joining with و. In v3.003, they are bumping right into و without a bend:

Next, a subtler issue:

بو
U+0628 U+0648

In v2.000, bottom edge of ب is aligned with the bottom of و's head. In v3.000, ب is slightly raised upwards.

(The screenshot shows بو, but the same base glyph is used for پو تو ٹو ثو نو یو too and has the same issue in these sequences.)

A comparison of regular and bold weights, with these strings: بو کو لو

It is interesting to note that for the bold weight, the "bumping right into" behaviour for کو and لو is also present in v2.000, and probably makes sense due to the thickness of glyphs. However, a slight bend can still be seen in the bold weight of بو, which may also be adapted for کو and لو.

(Please have a look at the images in their original sizes to view the issues/comparisons in detail.)

NotoNastaliqUrdu-Bold.ttf weight 700 outside font capability 400..400, NotoSerifMyanmar-Bold.ttf weight 700 outside font capability 400..400

I am amending the test to expect this (so it can pass again); remove that expectation once fixed

Font

NotoNastaliqUrdu-Regular.otf

Where the font came from, and when

Site: https://github.com/googlefonts/noto-fonts/blob/f7556efcaf940cf6c2f8ae0c3386d651e930b5cc/unhinted/otf/NotoNastaliqUrdu/NotoNastaliqUrdu-Regular.otf
Date: 2022-02-03

Font version

Version 3.002

Issue

Noto Nastaliq Urdu version 3.002 (not to be confused with yesterday’s build, also called 3.002) ligates some <lam, lam, heh> sequences which are clearly unrelated to the word “اللّٰه” and should not be ligated. If the heh is not really a heh but another letter that the font converts into a heh glyph, the ligature should not be formed. If the diacritics are incompatible with the pronunciation “llah”, the ligature should not be formed.

My rationale is that other <lam, lam, heh> sequences are not ligated; e.g. inserting a kasra, or ijam on a lam, blocks the ligature. I conclude that this ligature depends on the meaning of the word and not just the shapes of the glyphs. That’s why I include the second example below, though ae and heh are otherwise indistinguishable in final position.

Character data

للّٰة
U+0644 ARABIC LETTER LAM
U+0644 ARABIC LETTER LAM
U+0651 ARABIC SHADDA
U+0670 ARABIC LETTER SUPERSCRIPT ALEF
U+0629 ARABIC LETTER TEH MARBUTA

للّٰە
U+0644 ARABIC LETTER LAM
U+0644 ARABIC LETTER LAM
U+0651 ARABIC SHADDA
U+0670 ARABIC LETTER SUPERSCRIPT ALEF
U+06D5 ARABIC LETTER AE

لَله
U+0644 ARABIC LETTER LAM
U+064E ARABIC FATHA
U+0644 ARABIC LETTER LAM
U+0647 ARABIC LETTER HEH

Screenshots

The letters U+077A and U+077B disappear when set in Noto Nastaliq Urdu Bold.
Regular:

Bold:

(Blank)

Font

NotoNastaliqUrdu-Regular.ttf
Version 2.000; ttfautohint (v1.8.3) -l 8 -r 50 -G 200 -x 14 -D arab -f none -a qsq -X ""
Site: https://github.com/googlefonts/noto-fonts/hinted/ttf/NotoNastaliqUrdu/
Date: 2020-12-31

OS name and version

macOS Mojave (same in Big Sur)

Application name and version

Firefox, Chrome

Issue

ێ [U+06CE ARABIC LETTER YEH WITH SMALL V] and ٚ [U+065A ARABIC VOWEL SIGN SMALL V ABOVE] are used to write the sound 'e' in Kashmiri, but they cause joining problems in the Nastaliq Urdu font.

Examples (beb and babe):

In position 1 in the image the YEH WITH V should be a medial shaped YEH with V above.
In position 2 the second BEH should join with the YEH BARREE and the V should appear above the BEH.

Codepoints:
ب U+0628 ARABIC LETTER BEH
ێ U+06CE ARABIC LETTER YEH WITH SMALL V
ب U+0628 ARABIC LETTER BEH
U+0020 SPACE
ب U+0628 ARABIC LETTER BEH
ب U+0628 ARABIC LETTER BEH
ٚ U+065A ARABIC VOWEL SIGN SMALL V ABOVE
ے U+06D2 ARABIC LETTER YEH BARREE

Live test: https://r12a.github.io/pickers/arab-ks/?text=%D8%A8%DB%8E%D8%A8%20%D8%A8%D8%A8%D9%9A%DB%92

Same results on both Chrome, Edge, and Firefox. (Safari font support has fundamental problems, so not considering it here.)

Rationale

I'm aware of https://github.com/googlefonts/noto-fonts/issues/652 but would like to argue Noto should support the Kashmiri language as a high priority, since Kashmiri is spoken by around 7 million people, and as of this year is an official language of Jammu and Kashmir, and it is among the 22 scheduled languages of India. The predominantly muslim population uses the nastaliq orthography.

SIL has the Awami nastaliq font, but although it covers around 25 Pakistani languages it doesn't appear to support this Kashmiri letter either.

It may be that this is a rendering issue rather than a font issue.

PS

There is also an occasional problem with ۆ U+06C6 ARABIC LETTER OE joining. See https://github.com/googlefonts/noto-fonts/issues/1353