Unicode support in Interchange

2005-08-06 16:11

Interchange is an open-source web-application server, that has grown out of an internet shop software. At some point in time it was developed and maintained by RedHat, but now there is an independent group of developers, who took it over.

Interchange is written in Perl (mainly).

Interchange includes some machinery for internationlization (i18n) and localization (l10n), but not Unicode support. When I thought of using it for a project, I looked into building it (the Unicode support) into IC and tried.

The patch

Get it here: ic5.2.0-unicode.patch.

This is a patch for Interchange version 5.2.0. When you apply it, Interchange will:

generate HTML pages in Unicode UTF-8 encoding;
expect the template files to be in UTF-8 encoding;
store your data (like product names and descriptions) in UTF-8 in databases;
expect data from CSV database files, GDBM, DB_File and in-memory databases to be in UTF-8 encoding.

The patch is not complete. It does not fix:

encoding for data coming from SQL (DBI, including MySQL) and LDAP databases — I didn’t try that, expect no problem;
it does not fix encoding for data coming from forms with method='POST' and ENCTYPE="multipart/form-data" — I tried to do that, but failed.

Also:

it is not well-tested;
it requires Perl 5.8 or newer;
it will probably break the existing internationalization schemes.

But it may be useful as an example, a starting point, that shows how to procede if you want to finish it.

* * *

The interchange-users mailing list
My message on the list, where I announced my work on July, 12, 2005.

* * *

My other Unicode related writings: