[Isis-users] HOW TO SETUP UNICODE

[De Smet Egbert]

Solomon,


for mixing UNICODE in ABCD (v2.0x) you need the following elements to have been set properly :

- the actab and uctab tables in either the proper database-folder within the data-folder) or in the general bases-folder (for all databases) named 'isisactab_utf8.tab' and 'isisuctab_utf8.tab'; samples are already available with the installation, containing in addition to the mandatory ANSI e.g. Arabic. and Amharic I added meanwhile also Devanagari, Sinhala, Tamil and Malayam, but probably you only need  ANSI and Arabic.

- a parameter 'UNICODE=' (0 or 1) in dr_path.def for each of the databases, this will avoid the error you got which indicates that ABCD thinks it is a unicode database while is isn't. The actab and uctab tables structure in Unicode is totatlly different from the classic tables.

ABCD will change the encoding charset according to the parameter UNICODE (in abcd.def as the default and dr_path.def for each database if different from the default). For iAH also a parameter UNICODE needs to be set in dbname.def with values 0 (no Unicode) or 1/2/3...n where each value points to another ahindx1/2/3...n.htm file in iah/scripts/en. For each alphabet this file ahindx1/2/3...n can contain the pointers as buttons on the screen, as already exemplified in the installation files where ahindx1.htm contains Amharic buttons (for Ethiopia).

Of course Unicode databases need to have been created and maintained using the correct setting, which means instead of cgi-bin/ansi/wxis (or mx) the Unicode-version cgi-bin/utf8/wxis (or mx) will be used.

The option in the Central login-screen to switch to Unicode only refers to the encoding charset of the interface. Finally there are two new utilities (in the extra-menu) allowing to change the charsets of text files (such as the PFTs, FDT, FST, .def,.par etc.) in the database-folder for the active database, either changing from ANSI to UTF8 or vice versa.

Remember however also that browsers have options to set/switch character encoding for page display, so sometimes it will be still necessary to use those settings. For Firefox e.g. there are many such add-ons.

That basically sums up the challenge.

It would have been much more simple if we would totally migrate to UTF8 (meaning all databases become Unicode), as can be expected in a modern world, but - sorry friends - Latin America, where Unicode is not needed, prefers to stick to ANSI.


Egbert de Smet
Universiteit Antwerpen


________________________________
From: isis-users <isis-users-bounces+egbert.desmet=ua.ac.be at iccisis.org> on behalf of Solomon Mutai <skmutai at gmail.com>
Sent: Friday, July 20, 2018 8:28 AM
To: isis-users at iccisis.org
Subject: [Isis-users] HOW TO SETUP UNICODE

colleagues,

I am finding a new challenge with data entry where i have some titles both in English and Arabic. My server configuration is set to ANSI (which is standard anyway for most installation) and I dont know hoe to combine this with UNICODE. When I switch encoding to UNICODE I get WXIS|execution error|actab|loadactb_utf8/max 4 columns|. How do I reconfigure this so that my system can use both? My case may be illustrated if you try to catalogue ISBN 9004125892

Thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.iccisis.org/pipermail/isis-users/attachments/20180720/0b04dfe0/attachment.html>