fijam / md-rec Goto Github PK

I have one MZ-R90 which supports the katakana character set.

Charset layout is: lowercase+common+numbers+common+katakana+common+uppercase

The katakana set has the complete 48 characters + diacritics + characters used in digraphs (?).

I'd appreciate help in identifying all the characters.

Currently input_string function only inputs letters from their own character set. For example, to enter 'Z' it will enter uppercase_set and press SearchRight 26 times. It would be faster to enter lowercase_set and press SearchLeft 6 times instead.

A naive way to do it would be to split each character set roughly in half.

A more fancy way to do it is have the entire combined character set and 'entry points' for each character set such as index position of 'A', 'a' and '0'. Then using modulo artithmetic find the closest distance between index of the letter we want to input and one of the entry points.

This should improve entry time some 30% at the cost of increased complexity.

Add a mode where the script processes new line terminated strings from stdin and nothing else. This should ease integration with other custom scripts people have.

It may be useful to have a mode for labelling discs which were already recorded and have track marks but lack track names.

Have the user navigate to the track they want to name, enter labelling mode, take the name from input, save. Ask the user if they want to do another track.

md-rec will insert a TMark at the end of a track and the recorder may insert another if it detects long enough silence resulting in duplicates. Two ways to deal with that:

1. Just delete the extra TMarks manually later.
2. Add a 2s silence track between all items on the playlist and let the player handle TMarks.

#2 needs a mode where no TMarks are entered programatically and empty track names are not entered.