Thread: [Indic-computing-devel] Testing OpenType support
Status: Alpha
Brought to you by:
jkoshy
Menu
▾
▴
indic-computing-devel
|
From: Arun M <ar...@fr...> - 2003-02-19 15:59:42
|
Hi All, I've been working with Eric Manders of ICU project to get Malayalam work better. Except for Devnagari, Tamil and Malayalam not much work seems to be happening to test the code. In ICU same function is being used to handle 9 scripts. A fix in one language can break others. We can address this if a team form this list to do the testing for each indian languages. A bug free version of ICU will help FreeSoftware localisation as most of the many projects goes to ICU for OpenType support (OpenOffice, Mozilla, GTK). I've some selfish interest too ;) some fixes for Malayalam hasnt gone to 1.2 branch (available in CVS HEAD) because test has to be done to see how the changes affect other indic languages Regards, Arun. PS: If any one wants to test Malayalam write to me, i've some changes which is not yet in the pango/ICU cvs. |
|
From: Guntupalli K. <kar...@fr...> - 2003-02-19 16:11:59
|
On 19 Feb 2003 21:41:03 +0530 Arun M <ar...@fr...> wrote: > Hi All, > > I've been working with Eric Manders of ICU project to get > Malayalam > work better. Except for Devnagari, Tamil and Malayalam not much work > seems to be happening to test the code. > > In ICU same function is being used to handle 9 scripts. A fix in one > language can break others. We can address this if a team form this > list to do the testing for each indian languages. A bug free version > of ICU will help FreeSoftware localisation as most of the many > projects goes to ICU for OpenType support (OpenOffice, Mozilla, > GTK). > > I've some selfish interest too ;) some fixes for Malayalam hasnt > gone to 1.2 branch (available in CVS HEAD) because test has to be > done to see how the changes affect other indic languages > I am doing some testing for Telugu, using Pothana2000 font, on Gnome 2.2 release (pango 1.2.1) Conjuncts work properly, but not intermediate halant forms also consonants not getting reduced to dead form when vowel sign is attached to it (this is only for vowelsigns coming on top, like I II AA etc). Regards, Karunakar -- Hating people is like burning down your house to get rid of a rat - Anon --------------------------- * Indian Linux project * * http://www.indlinux.org * --------------------------- |
|
From: Arun M <ar...@fr...> - 2003-02-19 16:45:47
|
> > I've some selfish interest too ;) some fixes for Malayalam hasnt > > gone to 1.2 branch (available in CVS HEAD) because test has to be > > done to see how the changes affect other indic languages > > > I am doing some testing for Telugu, using Pothana2000 font, on Gnome Is it a Free Font ? Can i get it ? regards, arun |
|
From: G K. <kar...@fr...> - 2003-02-19 18:55:13
|
Arun M wrote: >>>I've some selfish interest too ;) some fixes for Malayalam hasnt >>>gone to 1.2 branch (available in CVS HEAD) because test has to be >>>done to see how the changes affect other indic languages >>> >> >> I am doing some testing for Telugu, using Pothana2000 font, on Gnome > > > Is it a Free Font ? Can i get it ? > Free as in 'free beer' , available from http://www.kavya-nandanam.com/Pothana2k-95.exe Regards, Karunakar |
|
From: Sayamindu D. <unm...@So...> - 2003-02-19 20:07:48
|
On Wed, 2003-02-19 at 21:41, Arun M wrote: > Hi All, > > I've been working with Eric Manders of ICU project to get Malayalam > work better. Except for Devnagari, Tamil and Malayalam not much work > seems to be happening to test the code. > > In ICU same function is being used to handle 9 scripts. A fix in one > language can break others. We can address this if a team form this list > to do the testing for each indian languages. A bug free version of ICU > will help FreeSoftware localisation as most of the many projects goes to > ICU for OpenType support (OpenOffice, Mozilla, GTK). Wrt Bangla, we at bengalinux.org are willing to help How do we go about it?? -sayamindu- -- Sayamindu Dasgupta [ http://www.peacefulaction.org/sayamindu/ ] ========================================= Speak out on social and cultural issues at PeacefulAction.Org http://www.peacefulaction.org ***************************************** Due to circumstances beyond your control, you are master of your fate and captain of your soul. |
|
From: Arun M <ar...@fr...> - 2003-02-20 02:55:19
|
> Wrt Bangla, we at bengalinux.org are willing to help > How do we go about it?? We can do the testing regularly and post the issues in this list. Then try to fix the stuff ourselves and submit to ICU. If we are not able to fix we can file the bug report. arun. |
|
From: Arun M <ar...@fr...> - 2003-02-20 03:01:58
|
> >> I am doing some testing for Telugu, using Pothana2000 font, on Gnome > > > > > > Is it a Free Font ? Can i get it ? > > > Free as in 'free beer' , available from > http://www.kavya-nandanam.com/Pothana2k-95.exe Any idea about Free as in 'freedom' Telugu OpenType font ? It would be nice if some one start working on converting Akruti Telugu font to OpenType. We should try to make one indian language OTF font. (Tamil, Malayalam and Bengla seems to be ready) |
|
From: Arun M <ar...@fr...> - 2003-02-21 15:02:19
|
Hi all, In Devanagari what should be the output of a sequence like <consonant><virama><ZWNJ> Should we apply 'haln' feature of OpenType here ? regards, arun. |
|
From: Arun M <ar...@fr...> - 2003-02-23 04:18:14
|
> In Devanagari what should be the output of a sequence like > > <consonant><virama><ZWNJ> > > Should we apply 'haln' feature of OpenType here ? No response :( Is it because i was not clear or on one know how devanagari works ? ICU and Pango renderer allows 'haln' form of the consonant when a sequence like one above comes. For Malayalam its not the right thing. So just wanted to know if its special case for Malayalam. To Koshy/Tapan/Other organisers As i wrote earlier we need a dedicated team which can work together providing information on every Indian language. For each language there should be one responsible. Our group is too loose with out people taking clear responsibilities. (Just my feeling looking back, may be i am wrong). Regards, Arun. |
|
From: Andy W. <And...@bt...> - 2003-02-23 21:58:54
|
Arun M wrote: > > In Devanagari what should be the output of a sequence like > > > > <consonant><virama><ZWNJ> > > > > Should we apply 'haln' feature of OpenType here ? > > > No response :( > > Is it because i was not clear or on one know how devanagari works ? I did not answer - I was sure that someone else would! (The answer is yes) > ICU and Pango renderer allows 'haln' form of the consonant > when a sequence like one above comes. For Malayalam its not > the right thing. > So just wanted to know if its special case for Malayalam. I am confused. Why do you think that this is not the right thing for Malayalam? According to the Unicode encoding model, consonant + virama + ZWNJ should render a visible Halant. In Uniscribe, ICU and Pango, this causes lookups in feature 'haln' to be applied. Therefore the feature 'haln' should only be used to create visible Halant forms of consonants. Do you think that is wrong for Malayalam? Andy |
|
From: Arun M <ar...@fr...> - 2003-02-24 02:43:43
|
> I am confused. Why do you think that this is not the right thing for > Malayalam? > According to the Unicode encoding model, consonant + virama + ZWNJ > should render a visible Halant. In Uniscribe, ICU and Pango, this causes > lookups in feature 'haln' to be applied. Therefore the feature 'haln' > should only be used to create visible Halant forms of consonants. Do you > think that is wrong for Malayalam? My knowledge of OpenType is very limited. Here is the issue. http://www.microsoft.com/typography/otfntdev/indicot/features.htm It says for some consonants (eg NNA) halant form will be represented by chillaksharam. My understanding is <NNA><virama> will form chillu NNA. <NNA><virama><ZWNJ> will make virama visible as it is and consonant will be in base form. How does NNA+virama and NNA+virama+ZWNJ should differ ? thanks for the response, regards, arun. |
|
From: Keyur S. <key...@ya...> - 2003-02-24 06:13:54
|
--- Arun M <ar...@fr...> wrote: > > My knowledge of OpenType is very limited. Here is the issue. > http://www.microsoft.com/typography/otfntdev/indicot/features.htm > > It says for some consonants (eg NNA) halant form will be represented by > chillaksharam. My understanding is <NNA><virama> will form chillu NNA. > <NNA><virama><ZWNJ> will make virama visible as it is and consonant will > be in base form. > > How does NNA+virama and NNA+virama+ZWNJ should differ ? > This has been the issue pending since long ago. Unfortunately, Unicode remains almost silent about this confusion :-(. I think some clarification is required in Unicode Indic-FAQ when some consensus has been reached. We want use virama sign in three different cases. (A) chillu form, (B) virama (halant) form, and (C) ligature (or consonant conjuct). A scheme should be able to support all 3 possible kind of rendering when using halant in Malayalam. It seems that there are two approaches to select alternate forms at present; using ZWJ and/or using ZWNJ. It is general belief (I may be wrong ;-)) that ZWJ becomes part of the syllable and ZWNJ forces termination of a syllable. Approach-1: ---------- A1. <C1><virama> will produce Chillu form (if any). A2. <C1><virama><ZWNJ> will produce halant form of C1. A3. <C1><virama><C2> will produce a ligature if <C1> doesn't have chillu form. A4. <C1><virama><ZWJ><C2> will produce a ligature if <C1> has chillu form. A4 has been given to answer the question: "How to produce <C1-chillu><C2> sequence if <C1><virama><C2> produce a ligature". As a result the rule A3 should be used when there is no chillu form of <C1> and A4 should be used when there is a chillu form of the consonant <C1>. The issue with this approach is that we have to use ZWJ in conjunct formation when <C1> has chillu form. Approach-2: ---------- B1. <C1><virama><ZWJ> will produce Chillu form (if any). B2. <C1><virama> will produce halant form if <C1> is not part of a ligature. B3. <C1><virama><ZWNJ> will also produce halant form of <C1>. B4. <C1><virama><C2> will produce a ligature (or consonant conjunct). The issue here is that people generally believe that ZWJ is used to join the consonant preceding it with the consonant immediately following ZWJ. Thus, ZWJ virtually becomes part of the syllable. Here we are using it to terminate the syllable. However, I am not a Malayalam expert :-) My observations are based on my discussion with a Malayalam font designer. It is better if you once again raise this issue on Unicode list. - Keyur __________________________________________________ Do you Yahoo!? Yahoo! Tax Center - forms, calculators, tips, more http://taxes.yahoo.com/ |
|
From: Andy W. <And...@bt...> - 2003-02-24 07:16:18
|
Keyur Shroff wrote: > > Approach-1: > ---------- > > A1. <C1><virama> will produce Chillu form (if any). > A2. <C1><virama><ZWNJ> will produce halant form of C1. > A3. <C1><virama><C2> will produce a ligature if <C1> doesn't > have chillu form. > A4. <C1><virama><ZWJ><C2> will produce a > ligature if <C1> has chillu form. The problem with A4 here is that the current specification of 'half' does not allow a ligatures of C1C2. IMHO a new feature would need to be specified. (feature 'cillu' ?). > A4 has been given to answer the question: > "How to produce <C1-chillu><C2> sequence if <C1><virama><C2> > produce a ligature". As a result the rule A3 should be used > when there is no chillu form of <C1> and A4 should be used > when there is a chillu form of the consonant <C1>. > > The issue with this approach is that we have to use ZWJ in > conjunct formation when <C1> has chillu form. Why not just always put a ZWJ whenever a chillu is desired? So C1chillu = C1+ZWJ (always) > > Approach-2: > ---------- > > B1. <C1><virama><ZWJ> will produce Chillu form (if any). > B2. <C1><virama> will produce halant form if <C1> is not part > of a ligature. > B3. <C1><virama><ZWNJ> will also produce > halant form of <C1>. > B4. <C1><virama><C2> will produce a > ligature (or consonant conjunct). > > The issue here is that people generally believe that ZWJ is > used to join the consonant preceding it with the consonant > immediately following ZWJ. Thus, ZWJ virtually becomes part > of the syllable. Here we are using it to terminate the syllable. Keyur, I agree with you. Both of the approaches above make sense. However, Approach-2 is more compatible with the current state of the situation. It would need less work to get it to work and is less likely to break any existing implementations (IMHO). Approach- 2 does not explain how to create a chillu+consonant conjunct. Assuming your answer is <C1><virama><ZWJ><C2>, then again, either a new feature definition is required, or the existing feature 'half' will need respecifying. Andy |
|
From: Arun M <ar...@fr...> - 2003-02-24 11:15:48
|
Thanks, Keyur and Andy for valuable comments > > A1. <C1><virama> will produce Chillu form (if any). > > A2. <C1><virama><ZWNJ> will produce halant form of C1. > > A3. <C1><virama><C2> will produce a ligature if <C1> doesn't > > have chillu form. > > A4. <C1><virama><ZWJ><C2> will produce a > > ligature if <C1> has chillu form. > > The problem with A4 here is that the current specification of 'half' > does not allow a ligatures of C1C2. IMHO a new feature would need to be > specified. (feature 'cillu' ?). A4 is not required, based on my understanding of malayalam. > Why not just always put a ZWJ whenever a chillu is desired? > So C1chillu = C1+ZWJ (always) This is what we did with Malayalam pango module with x backend we made some time back. (following malayalam support in TeX.) http://members.tripod.com/~jhellingman/IndianScriptsUnicode.html > > Approach-2: > > ---------- > > > > B1. <C1><virama><ZWJ> will produce Chillu form (if any). > > B2. <C1><virama> will produce halant form if <C1> is not part > > of a ligature. > > B3. <C1><virama><ZWNJ> will also produce > > halant form of <C1>. > > B4. <C1><virama><C2> will produce a > > ligature (or consonant conjunct). > > > > The issue here is that people generally believe that ZWJ is > > used to join the consonant preceding it with the consonant > > immediately following ZWJ. Thus, ZWJ virtually becomes part > > of the syllable. Here we are using it to terminate the syllable. > > Keyur, I agree with you. Both of the approaches above make sense. > However, Approach-2 is more compatible with the current state of the > situation. It would need less work to get it to work and is less likely > to break any existing implementations (IMHO). > Approach- 2 does not explain how to create a chillu+consonant conjunct. > Assuming your answer is <C1><virama><ZWJ><C2>, then again, either a new > feature definition is required, or the existing feature 'half' will need > respecifying. Approach 1 works better in current situation(note that step A4 is not needed). We will just need to add chillu feature. Also it goes with the unicode suggestion that <consonant><virama><ZWNJ> will give halant form. Also what i understand from some language experts is that when we write NNA(or other consonants having chillu form) with visible halant now days, it actually represents <NNA><vowel sign U><halant> (Vowel A is removed from consonant and U is added) Approach 2 will be easier to make as we may not need any change in renderer. regards arun |
|
From: Arun M <ar...@fr...> - 2003-03-04 16:26:06
|
Hi, We have seen two ways to represent chillu. How can we decide upon and publicise one as standard. I will try to put this infront of IT department of the Kerala Govt. Arun. |
|
From: Andy W. <And...@bt...> - 2003-02-24 06:44:41
|
Arun M wrote: > My knowledge of OpenType is very limited. Here is the issue. > http://www.microsoft.com/typography/otfntdev/indicot/features.htm This document was written at a long time ago, when OpenType for Indic scripts was still in the early planning stage, it has only been partially updated. Most of it has been rewritten as separate script specific documents. Unfortunately, the Malayalam specification has been left behind. > It says for some consonants (eg NNA) halant form will be > represented by chillaksharam. My understanding is > <NNA><virama> will form chillu NNA. The form that any given isolated 'Consonant + Virama' will be subsequently rendered as, is not prescribed in the Unicode standard; so implementers are free to call any feature they like. However, all current implementations call feature 'haln' for the sequence, 'Consonant + Virama'. The sequence Consonant+ Virama+ZWNJ has been prescribed to render as a consonant with a visible virama (chandrakkala) All current implementations call feature 'haln' for this sequence also. Hence it seems that feature 'haln' should render a visible chandrakkala The Unicode sequence 'Consonant + Virama + ZWJ' is equivalent to the ISCII sequence 'Consonant + Virama + Nukta'. In ISCII, 'Consonant + Virama + Nukta' is prescribed to form the chillu forms of consonants. As Unicode is modelled on ISCII, it follows that Unicode sequence, 'Consonant + Virama + ZWJ' should also be used to render the Chillu letters. Current implementations call feature 'half' for this sequence, hence it seems that feature 'half' should deal with the formation of the chillu. <NNA><virama><ZWNJ> will make virama visible as it is and consonant will be in base form. >How does NNA+virama and NNA+virama+ZWNJ should differ ? The Unicode standard allows the freedom for 'Consonant + Virama' & 'Consonant + Virama+ ZWNJ' to differ in appearance, if it is deemed appropriate by an implementation (usually they would not differ). Andy |
×
Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.
Thanks for helping keep SourceForge clean.
X