Came across an interesting bug the other day. We were discussing the positioning of the Bengali character Ishwar along any given alphabet. Ishwar (৺) is a sign used as a prefix to the name of a deceased person and looks similar to the Chandrabindu ( ঁ), thus giving rise to an oft made mistake regarding the two signs, in casual parlance. Ishwar sign is not represented on any of the keymaps that are available with Fedora, as it is rarely used. In this case, I tried using the RAW Unicode input with the codepoint for Ishwar (U+09FA) to write it. Unfortunately, RAW Unicode keymap seemed to have been missing. Turns out that the RAW Unicode Keymap is now packaged separately as scim-rawcode. However, this package is only installed when an user takes the pain to do so. Not a good idea, if you have a similar case like me. Often for languages using non-latin scripts, the QWERTY keyboard is not sufficient to hold all possible characters for a keymap. In such cases the characters to be given a slot on the keymap are prioritized and some have to be left out. The only convenient way to input these characters is to shift to RAW Unicode input. So this package needs to be installed when any language using a non-latin script is selected for installation.
Since blog != bug, there is a bug too.