[GRASSLIST:9843] Re: [GRASS5] Fwd: Indic support in GRASS GIS with Indix2

Markus Neteler neteler at itc.it
Sun Jan 15 16:12:03 EST 2006


Hi Swapnil Hajare,

can you send the necessary changes as CVS diffs? It is explained
in the GRASS 6.1-CVS SUBMITTING file in the main source directory
how to generate it properly.

Then we can better see what needs to be changed.

thanks

 Markus

On Thu, Jan 12, 2006 at 04:32:55PM +0530, Swapnil Hajare wrote:
> ---------- Forwarded message ----------
> From: Swapnil Hajare <dreamil at gmail.com>
> Date: Jan 5, 2006 3:02 PM
> Subject: Re: Hindi support in GRASS 6.1 - was Re: [GRASS5] Fwd: Indic
> support in GRASS GIS with Indix2
> To: Sajith VK <sajithvk at gmail.com>, grass developers list <
> grass5 at grass.itc.it>, GRASS user list <grasslist at baylor.edu>
> 
> Hi,
>   sorry for such a late response.
>   I downloaded the CVS version of GRASS GIS and compiled it on my debian
> sarge machine. I then used the commands given by you to create a sample
> vector and displayed it on monitor. For the example that you have in your
> snapshot, it works fine. But then I realised that you have displayed a
> string in devanagari/hindi which does not require any processing on font
> side. i.e. the text is displayed just similar to english text
>  using juxtapositioning of glyphs. So I tried with some conjunct text which
> requires some positioning/replacement rules to be used from the opentype
> font, it couldnot display properly. See the snapshot for details (snapshot
> uploaded on http://203.199.16.202/grass_cvs_comparison.png ). The left
> monitor show the GRASS CVS version as downloaded on 2006-01-05 and compiled
> as-it-is. The right monitor shows output of d.tect.freetype compiled with
> Indix2 libraries. The right monitor show properly rendered text while the
> left monitor show text in form of  juxtapositioned glyphs. I knew that
> because of freetype library GRASS will be able to use opentype fonts but I
> also knew that it doesn't have any rendering engine like pango/qt/indix2 to
> render complex text in indic scripts correctly.
> 
> We had tried this same thing in past without any rendering engine and
> failed. That is why I had filed this bug (
> http://intevation.de/rt/webrt?serial_num=3189 ). I must apologise for not
> following it up properly.
> 
> regards
> 
> 
> On 12/22/05, Markus Neteler <neteler at itc.it> wrote:
> >
> > Hi again,
> >
> > an update for support of Indian languages:
> >
> > - we now have messages files in Hindi available (still to be
> >   translated)
> > - we can out-of-the-box generate Hindi (or other) scripts
> >   in the GRASS monitor using TTF fonts, see screenshot at
> >   http://grass.itc.it/devel/i18n.php
> >   -> example for Hindi
> >      http://grass.itc.it/images/grass61_hindi_hi.jpg
> >
> > Note that this is a proof of concept example in Spearfish,
> > done with:
> >
> > #extraction of phrase in Hindi:
> > head -37 locale/po/grassmods_hi.po | tail -1 > /tmp/hindi
> > echo "599505|4921010|`cat /tmp/hindi | cut -d'"' -f2 `" | v.in.asciiout=hindi
> >
> > d.rast elevation.dem
> > v.db.select hindi
> > d.vect hindi icon=basic/circle
> > d.font.freetype font=/usr/share/fonts/hi/lohit_hi.ttf
> > d.vect hindi disp=attr attr=str_1 lsize=40 lcol=blue
> >
> > Works :-)
> >
> > Cheers
> >
> > Markus
> >
> > On Tue, Dec 20, 2005 at 10:38:44PM +0100, Markus Neteler wrote:
> > > Hi,
> > >
> > > I have added Hindi support to GRASS by adding two .po files.
> > > After installing
> > >  kde-i18n-hi-3.4.2-10mdk
> > >  locales-hi-2.3.5-1mdk
> > > on my Mandriva box, I can even see the characters. With
> > > d.font.freetype it should be possible to print them on the monitor
> > > such as for Japanese and Chinese. Screenshots asap...
> > >
> > > I fetched the KDE Hindi file and autotranslated the GRASS
> > > messages, some hundred fuzzy translations resulted. Since
> > > my Hindi is a bit rusty, I would leave the verification to
> > > native speakers.
> > >
> > > You can grab the wto hindi files from here
> > >
> > > http://grass.itc.it/devel/i18n.php
> > >  -> Download current .po files from CVS
> > >     -> grasslibs_hi.po
> > >     -> grassmods_hi.po
> > >
> > > Please send updates to these files regularly, in order to
> > > stay in sync.
> > >
> > > If needed, I can add other languages are well.
> > >
> > > Cheers
> > >
> > >  Markus
> > >
> > > On Mon, Dec 19, 2005 at 07:43:53AM +0100, Swapnil Hajare wrote:
> > > > hi,
> > > >  greetings.
> > > >   not sure if the new GTK gui will solve our problem. Does the new GUI
> > means
> > > > GRASS monitors will use pango for rendering? I guess they will use the
> > same
> > > > monitor. And by the time they come up with new GUI, should we wait for
> > them
> > > > to finish and then start our localised GIS work ? My feeling is this
> > should
> > > > be seen as a patch for indic support (similar to d.text.freetype used
> > for
> > > > CJK). correct me if I am wrong.
> > > >   I am also not sure  how many indic scripts are supported well by
> > Pango/QT
> > > > and what is the quality of rendering for all these scripts. If this
> > solution
> > > > can provide a better rendering solution, it has every right to be
> > accepted
> > > > as a good alternative to Pango/Qt.
> > > >
> > > > bye
> > > >
> > > > On 12/19/05, Sajith VK <sajithvk at gmail.com> wrote:
> > > > >
> > > > > Hello all,
> > > > >
> > > > > Sorry for a much belated response.
> > > > > Indian Language support for GRASS (For GUI and
> > > > > in the map lables) will be highly benificiary.
> > > > >
> > > > > A discussion on a better GUI for GRASS is already
> > > > > going on. I think it may end up with using GTK or
> > > > > QT as toolkit. Both GTK and QT has indian language
> > > > > support.
> > > > > So this work may become irrelevent, according to the
> > > > > future plans of the GUI team. So we may need to switch
> > > > > to pango/Qt
> > > > >
> > > > > On 12/8/05, Swapnil Hajare <dreamil at gmail.com > wrote:
> > > > > > HI all,
> > > > > >    my first post. I hope I am posting this is right list.
> > > > > >    this is for your comments. If it sounds ok, can some 2-3MB code
> > of
> > > > > Indix2
> > > > > > (indic Shaping Engine) be included in grass code to claim
> > something like
> > > > > > "GRASS GIS supports 11 Indian Scripts" ?
> > > > > >
> > > > > --
> > > > > Change the rules, or the rules will change you
> > > > >      ---------Kumaranasan
> > > > >
> > > >
> > > >
> > > >
> > > > --
> > > > Swapnil Hajare
> > >
> > > --
> > > Markus Neteler     <neteler itc it>       http://mpa.itc.it
> > > ITC-irst -  Centro per la Ricerca Scientifica e Tecnologica
> > > MPBA - Predictive Models for Biol. & Environ. Data Analysis
> > > Via Sommarive, 18        -       38050 Povo (Trento), Italy
> >
> 
> 
> 
> --
> Swapnil Hajare
> 
> 
> --
> Swapnil Hajare

-- 
Markus Neteler     <neteler itc it>       http://mpa.itc.it
ITC-irst -  Centro per la Ricerca Scientifica e Tecnologica
MPBA - Predictive Models for Biol. & Environ. Data Analysis
Via Sommarive, 18        -       38050 Povo (Trento), Italy




More information about the grass-user mailing list