Preserve   entity

[nltesown]

The non-breaking space entity ( ) should not be by a normal space character.
More generally, it may be useful to let the user decide how HTML entities should be dealt with (converted or preserved).

[myakura]

The non-breaking space entity ( ) should not be by a normal space character.

is it really a "normal" space character? i mean, does   return a character whose code point is U+0020 not U+00A0?

[nltesown]

I think it doesn't (I may be wrong). Even if it did, the main problem remains: when you see a white space, there's no way to tell them apart, so preserving the   entity would be at least useful.

[myakura]

so preserving the   entity would be at least useful.

perhaps. I'm not sure if it's possible to "preserve" entities as jsdom converts all named (and numeric) references to their matching character(s).

[wizardforcel]

It's because node.childNodes[i].data (index.js, line 133) will parse HTML entity.

So I think it should be changed into node.childNodes[i].data.replace(/</g, '&lt;').replace(/>/g, '&gt;'), or some better way.

[Tyriar]

Another problem with not preserving  , it can lead to an invalid format conversion:

<i>foo&nbsp;</i>bar

Converts to:

_foo _bar

Because the _ is no longer wrapping the phrase it does not end the italics style.

[kadishmal]

In my case the following HTML is incorrectly parsed. The output Markdown doesn't have the trailing space.

<p>I have it:&nbsp;</p>

Is converted into I have it: without the trailing space.

[rabeesh]

I tried to convert

<em>ds &nbsp; &nbsp; </em><strong>fs&nbsp;</strong>

Which is parsed into markdown as

_ds    _ **fs **

Need to keep html entity space  

[martincizek]

Just for reference - a through analysis on this topic: https://github.com/orchitech/turndown/wiki/Whitespace

A PR will come soon. :)