Spaces inside and around links are concatenated

[alir3z4]

Spaces inside and around links are concatenated

[Alir3z4]

I don't think this is a big bug.
Spaces and half-spaces should be preserved for some formatting reasons.

I'll leave this issue here for now, but It shouldn't be considered as a bug.
Although formatting options/flags are welcome.

half-spaces: In Persian based languages such as Arabic, Farsi, Dari, etc is a must. Words like می باشد and می‌باشد are such examples.

[theSage21]

I do not understand the issue? Is the preservation of spaces the bug in question?

[aykevl]

I had to work aroud this bug in another project, as the text version was showing two spaces where the HTML version (in an email) was showing only one space as intended.

As an additional argument, when the spaces are at one side of the link tag, html2text only outputs one space:

>>> h.handle('foo  <a href="/">bar</a>')
u'foo [bar](/)\n\n'
>>> h.handle('foo<a href="/">  bar</a>')
u'foo[ bar](/)\n\n'
>>> h.ignore_links = True
>>> h.handle('foo  <a href="/">bar</a>')
u'foo bar\n\n'
>>> h.handle('foo<a href="/">  bar</a>')
u'foo bar\n\n'

Just like browsers do:

foo bar
foo bar

Note that the space in the second example is part of the link (in Chrome).

I don't know how non-Latin-based languages do this formatting.

@theSage21 the problem is that there are two spaces rendered where there should be (in my opinion and as browsers indicate) only one.

[theSage21]

A browser renders one space irrespective of how many spaces are there between two words.
Creating a file sample.html or testing here with the following html

<p>arjoonn sharma
<br>
arjoonn                     sharma</p>

shows that the spaces do not matter and only one space is rendered even though the html is evidently not the same.

@aykevl My point being that html2text preserves information from the original html and translates that to text.

edit
Maybe we should be stripping the link texts? @Alir3z4 your take on this?

[Alir3z4]

As @mcepl once mentioned can we transform the data to its original form or we gonna loose the format of the original.

We're fine with any implementations as long as the text can keep it's original form after parsing to html by python markdown parser.

@mcepl do you have any input on this?

[Alir3z4]

Not a bug and based on #58 (comment) I'll close this.