Additional
Information relating to the Editor
Companion
Applications
Local Language Browser
- LB
This browser can be
used for viewing Multilingual documents prepared by using the Editor -
IITMfcEd. The language in which the document is desired to be viewed can
be selected from the dialog box, which appears when the utility is invoked.
As the name suggests, local language files can only be viewed and no editing
operations are allowed in this browser. Files can be printed from the browser
following the same procedure as for printing from the Editor. The browser
uses the same support files as the multilingual editor but requires a separate
lb.ini file in the c:\windows directory. The .ini file will be installed
during setup. The local language browser lb.exe may be setup as an external
viewer to view .llf files served from a web server. Email attachments may
also include a .llf file and the contents automatically displayed when
mail is seen, if lb.exe is specified as the viewer to be invoked by the
email client when an attachment of a .llf file is present in the received
mail.
Transliterated text Viewer
- TVIEW
For several years
now the transliteration method using Roman letters has been used for representing
letters or aksharas of our scripts. Quite a lot of data entry using the
transliteration approach has been effected and there are many websites
offering texts of important works in Indian languages in transliterated
form. The IITMadras Software includes a program called tview (as well as
tconvert) to allow transliterated text to be viewed directly in the appropriate
scripts.
"tview" is essentially
a conversion program which converts the Roman into the .llf format (temporarily)
which is viewed by automatic invocation of "lb" browser. Thus all the benefits
of 'lb' are available for tview. To identify the language/script that should
be used in displaying the text, tview looks for the following information
with the very first line of text in the file
(we call this the header)
% <language_name> <method>
where the % symbol should
occupy the first column of the string. <language_name> specifies the
language or script in which tview should display the converted text while
<method> refers to one of the following transliteration schemes
ITRANS - Popular Transliteration
method for Devanagari.
ADHAWIN -- Tamil Transliteration
scheme
RIT -- Telugu Transliteration
scheme
MYLAI -- Another Tamil Transliteration
scheme.
If the entered text
incorporates this header as the first line, tview would bring up the appropriate
display. If the header is missing, tview will not make a default assumption
about the languages as well as the scheme and will display an error message
and quit. The tconvert program is identical to the tview program in functionality
but will prompt the user for the missing information if a header is not
present. Also tconvert will save the converted file in a .llf format the
user may edit later. Tconvert is not included in the distribution. Those
interested are requested to send an e-mail message to the address specified in the contact page
and ask for a copy. Currently versions for Linux (ELF binaries) and Microsoft
Windows(98/Me/2000/XP) are distributed (April 2006).
Note: tview and tconvert
will have to be invoked from the command line since they were written to
be compatible with other systems. Under windows one does this easily be
invoking the same from the Dos prompt. More recent versions of tview support
GUI features. Please refer to the concerned page discussing these utilities.
Local Language file to
HTML file Converter - LLF2HTML
This converter is
a utility that converts a file which you have stored in the Local Language
Format (.llf) into Hypertext Markup Language format (HTML) thereby providing
you with an option to put up your local language files on the web. This
may also be used to convert the .llf files into RTF (.rtf files). The resulting
.rtf files can be viewed with appropriate utilities Under Win95, Linux
and other Unix systems. The conversion program also supports the use of
Unicode fonts so that the local language files may be displayed on web
browsers which correctly handle Unicode fonts.
Top
Some Limitations
The following may be viewed
as some of the inadequacies and difficulties experienced with the Editor.
Magnification and Italicization
of fonts have not been incorporated at present in the system. The required
magnification can however be set in the .ini file but once set, it cannot
be changed during the operation of the Editor.
Arbitrary Samyuktaksharas
cannot be keyed in. Some four and five consonant syllables in regular use
will have to be handled specifically with the "Half Form" based data entry.
The text typed into the editor
window will not automatically wrap around in the Windows version. Essentially,
it works as a typewriter where at the end of the line, a return must be
pressed. The editor will sound the bell when data entry has reached the
end of the line and there is space for only a couple of more aksharas.
The Editor is a text Editor
and so lacks word processing features. Such features are not really very
essential in the context of document preparation since the text prepared
by the Editor may be taken to Word or other applications and formatted
quickly. Spelling Check features are not supported.
There is no Undelete feature
in the Editor.
Cut/Copy Paste operations
work on whole lines and not a portion of text on a line. Also the option
of cut and paste is limited to a single file, one cannot cut and paste
across .llf files.
The standard Editor has no
provision for scripts written right to left, such as Urdu, Arabic and Hebrew.
A separate version of the Editor must be used if data entry is to be done
for Urdu. Also, since the lexical ordering of the characters in Urdu is
different from that of other Indian languages, transliteration between
Urdu and Indian scripts will not be strictly correct.
IIT Madras has developed
a special version of the Editor for Urdu, Arabic, Hebrew and Avesta. Please
visit the web site of the lab for additional information. (May 2001)
General Recommendations
for using the Editor
Though the system can handle
any number of languages, it is recommended that users type in only two
or three languages in a file.
Languages like Sanskrit may
have complicated combination of different conjuncts for a single character.
There are quite a few conjuncts, which have not been included in the IITM
set of supported conjuncts. In such cases you might have to split up the
character into individual conjuncts where each conjunct is stored as a
character by itself. Alternately you may use the special half form data
entry described above.
It is recommended that the maximum
number of characters per line be limited to typically 40 local language
characters. This is based on the observation that in normal printouts,
one sees about 30-40 characters in a line of text.. The Editor will accept
upto a hundred aksharas per line but then horizontal scrolling will come
into effect.
Minimal Troubleshooting
guide
Some of the commonly observed
problems and the sources of the errors are given below.
Only rectangular boxes or junk
characters appear on the screen as you type in.
Caused by the absence of
fonts. Please install the fonts for the language/script chosen.
Bindu, Chandra Bindu and some
special characters (such as Anudatam ) do not appear as they are typed
in.
This is a consequence of
the Editor having to display these on the previously typed in akshara.
The symbols will appear if the screen is redrawn say by minimizing and
then maximizing the window. The text will appear properly when you open
the file again.
Using the carat key in Find
and Replace.
When entering a search string,
the Ctrl key will not work. During the entry the search string is entered
as a sequence of Roman letters and with the Ctrl key the key the sequence
will be wrongly interpreted. Use the caret key '^' to indicate that conjuncts
are being formed. Read the section under Search and Replace.
At times after randomly switching
between English and the Local language data entry, the keyboard may misbehave.
Just save the file and restart
the Editor. Open the file again and continue editing. On some rare occasions,
data entered in the Local language might appear in English. This portion
may be cut out or edited out.
On some systems with Accelerated
Graphic display adapters, some of the matras may not be seen during data
entry, especially those, which are written on top of a character or below.
The internally stored text will be proper however. When you open the file
again, the text should be seen properly.
In systems running WindowsXP/2000
some kind of screen lockup is seen when an invalid akshara is formed. A
long beep accompanies this and the editor returns to the normal state after
a long time, typically a minute or longer. Also on WinXP systems, gaps
will be observed between aksharas. The screen lockup problem has been sorted
out in the latest version of the Editor (April 2006). The gap in the aksharas
will not be seen if the more recently introduced IITM Fonts are used.
Top