In Sphinx reST, you can't put spaces at the beginning or end of
certain literals and similar things. I think this is more to
help catch or avoid mistakes than a real technical requirement.

With three changes to the source code, you can apparently get
around this restriction, but I have no idea if it always works
or if it's safe, since I barely tested it and just messed
around in the Sphinx code without knowing what I was doing.

Here are the details:

The following four literals are usually not allowed in Sphinx
reST source because they start or end with a space:

abc `` def`` ghi third change

jkl ``mno `` pqr first change

stu :literal:` vwx` yza third change

bcd :literal:`efg ` hij second change

But if you make these three changes to the Sphinx
"docutils/parsers/rst/states.py" file, it works
(no guarantees, try it at your own risk!):

First change

Original code

literal=re.compile(self.non_whitespace_before + '(``)' + end_string_suffix),

Changed code

literal=re.compile( '(``)' + end_string_suffix),

Second change

Original code

non_unescaped_whitespace_escape_before = r'(?<!(?<!\x00)[\s\x00])'

Changed code

non_unescaped_whitespace_escape_before = r'(?<!(?<!\x00)[\x00])'

Third change

Original code

non_whitespace_after = r'(?!\s)'

Changed code

non_whitespace_after = r'()'



--- MBSE BBS v1.1.1 (Linux-x86_64)
* Origin: Stefan Ram (3:633/280.2@fidonet)

ram@zedat.fu-berlin.de (Stefan Ram) wrote or quoted:

But if you make these three changes to the Sphinx >"docutils/parsers/rst/states.py" file, it works
(no guarantees, try it at your own risk!):

What are those spaces at the start or end of a literal even
supposed to be good for? They might not be the best idea if you
think about what HTML or LaTeX could do with them (line break!).
But I wanted to generate reST, and it made me uneasy to think
that certain characters just could not be in certain spots . . .



--- MBSE BBS v1.1.1 (Linux-x86_64)
* Origin: Stefan Ram (3:633/280.2@fidonet)