Open iRi-E opened 13 years ago
If ROM_KANA_RULE and KUTOUTEN_RULE (and other tables?) are configurable, many problems including yours will be solved for ever. You shouldn't add an option every time someone complains. Some people want "oh" to be "おお", "zi" to be "ずぃ", or "cha" to be "は". To satisfy all of them, the related rules have to be flexible.
But, in my opinion, GUI editor is not useful but harmful. SKKIME (an SKK implementation for Windows) has a rich editor for the rules/tables, which requires many steps of mouse-work to set up. So I suggest skk.py simply loads a python file somewhere under $HOME.
Agreed with the idea to have a configuration file. However, I think loading python file is too much this time, since ROM_KANA_RULE / KUTOUTEN_RULE are just a flat table. Perhaps json or something might be better?
I agreed with the idea of configuration file for romaji-to-kana rules (and key bindings, if possible).
However, I think the zenkaku/hankaku rule of punctuations and parenthesis is a separate issue from the romaji-to-kana rules. It's necessary to sometimes change the setting of zenkaku/hankaku according to sort of files such as LaTeX and Wiki, while romaji-to-kana table is rarely changed. I don't want to modify the config file every time I open a document file.
Could you provide settings or commands (if any) for DDSKK (or other implementation) to control that behavior, so that I can refer to the implementation as a groundwork?
Here is a sample of DDSKK's setting:
We can set initial state of zenkaku/hankaku by variable 'my-skk-paren-state' and toggle the state by command 'my-skk-toggle-paren' in each buffer.
Well, I don't know whether it's useful for this issue.
When using zenkaku symbols like "?", "!", "()", etc., I have to type Shift+l and Ctrl+j many times to change the input mode, so I'm awfully irritated.
Please add an option which specifies zenkaku/hankaku of symbols input in hiragana (or katakana) mode.