On 2016-03-19 12:24, BartC wrote:
> So a string that looks like:
>
> "ññññññññññññññññññññññññññññññññññññññññññññññññññ"
>
> can have 2**50 different representations? And occupy somewhere
> between 50 and 200 bytes? Or is that 400?
And moreover, they're all distinct if you don't normalize them.
Which certain environments such as CSS & HTML don't do. So you can
have
css = """
<style>
.r\N{LATIN SMALL LETTER E WITH ACUTE}sum\N{LATIN SMALL LETTER E WITH ACUTE}
{color: red;}
.r\N{LATIN SMALL LETTER E WITH ACUTE}sume\N{COMBINING ACUTE ACCENT}
{color: blue;}
.re\N{COMBINING ACUTE ACCENT}sum\N{LATIN SMALL LETTER E WITH ACUTE}
{color: purple;}
.re\N{COMBINING ACUTE ACCENT}sume\N{COMBINING ACUTE ACCENT}
{color: purple;}
<style>
"""
html_fragment = """
<ul>
<li class="r\N{LATIN SMALL LETTER E WITH ACUTE}sum\N{LATIN SMALL LETTER E
WITH ACUTE}">One
<li class="r\N{LATIN SMALL LETTER E WITH ACUTE}sume\N{COMBINING ACUTE
ACCENT}">Two
<li class="re\N{COMBINING ACUTE ACCENT}sum\N{LATIN SMALL LETTER E WITH
ACUTE}">Three
<li class="re\N{COMBINING ACUTE ACCENT}sume\N{COMBINING ACUTE ACCENT}">Four
</ul>
"""
which will all appear visually identical in the source code, but
each is unique according to the DOM/CSS parser.
-tkc
--
https://mail.python.org/mailman/listinfo/python-list
