Review Suggested Edit

You can't approve or reject suggested edits because you haven't yet earned the Edit Posts ability.

~~## C, 59 byte~~
```c
h(n){return n&4?n&2?5:n&1?1e3:n&8?50:500:n&2?100:n&1?1:10;}
```
~~### Old version, 60 byte:~~
```c
h(n){return n&4?n&2?5:n&1?1000:n&8?50:500:n&2?100:n&1?1:10;}
```
Not very creative, there is probably a smaller version. Takes a ASCII character as argument and returns the value.
`n&4` separates `'V'`/`5`, `'L'`/`50`, `'D'`/`500` and `'M'`/`1000`, which have bit 2 set and end up in this part: `n&2?5:n&1?1000:n&8?50:500`, and `'I'`/`1`, `'X'`/`10`, `'C'`/`100` , which have bit 2 cleared and end up in this part: `n&2?100:n&1?1:10`.
A similar thing is done after that, of `'V', 'L', 'D', 'M'`, only `'V'` has bit 1 set, so if `n&2` is true, the value is 5. If not, we check bit 0 with `n&1` for `'M'` or 1000, bit 3 for `'L'` or `50` and if it is none of them it must be `'D'` or `500`. A similar thing is done for the numbers `100`, `1` and `10`. This has the nice side effect that it works work lower and upper case letters, since it ignores all bits >3.
~~Tried a version for EBCDIC-Encoding but that is longer (because some characters only differ at bit 4, which requires a `n&16`, which is 1 byte longer).~~

## C, 59 bytes
```c
h(n){return n&4?n&2?5:n&1?1e3:n&8?50:500:n&2?100:n&1?1:10;}
```
### Old version, 60 bytes:
```c
h(n){return n&4?n&2?5:n&1?1000:n&8?50:500:n&2?100:n&1?1:10;}
```
Not very creative, there is probably a smaller version. Takes a ASCII character as argument and returns the value.
`n&4` separates `'V'`/`5`, `'L'`/`50`, `'D'`/`500` and `'M'`/`1000`, which have bit 2 set and end up in this part: `n&2?5:n&1?1000:n&8?50:500`, and `'I'`/`1`, `'X'`/`10`, `'C'`/`100` , which have bit 2 cleared and end up in this part: `n&2?100:n&1?1:10`.
A similar thing is done after that, of `'V', 'L', 'D', 'M'`, only `'V'` has bit 1 set, so if `n&2` is true, the value is 5. If not, we check bit 0 with `n&1` for `'M'` or 1000, bit 3 for `'L'` or `50` and if it is none of them it must be `'D'` or `500`. A similar thing is done for the numbers `100`, `1` and `10`. This has the nice side effect that it works work lower and upper case letters, since it ignores all bits >3.
Tried a version for EBCDIC-Encoding but that is longer (because some characters only differ at bit 4, which requires a `n&16`, which is 1 byte longer).