trouble understanding None

[Peter Otten]

Terry Reedy wrote:

> Unless the OP actually needs a list of lines, I think I would make the
> value corresponding to each letter one string with embedded newlines:
[...]
> This is very easy to edit (with a fixed pitch editor), aud to use.

Yes, this would be a cleaner approach.

> The only real problem is (maybe) the extra newline at the beginning of
> the file, which makes the first line line-up with the rest.  The raw
> mode input makes it impossible (as far as I know) to escape it (with
> the usual '\', which gets printed instead).  So one could either
> postprocess the dict to slice all values or delete the initial newline
> in the source code after getting the letter right.

I mixed normal and raw strings because I recalled a problem with backslashes
at the end of a raw string. I've since found out that this applies only to
a single last backslash at the end of the string:

>>> r"\\"
'\\\\'
>>> r"\"
  File "<stdin>", line 1
    r"\"
       ^
SyntaxError: EOL while scanning single-quoted string
>>>

For my Big writer class , that somewhat extends the original task,
postprocessing would be the way to go, because of other problems:

- every line of a character must be of the same length
- every character must have the same height
- if character definitions with different heights are allowed, additional
information about the baseline is required for the program to decide where
to fill in the blank lines


Peter