html_entity_decode() clears the DOM entities

[techi602]

Although biggest issue with foreign encoding was fixed in #131 the problem is that some e-mail clients such as thunderbird have issues with displaying entities as characters. Therefore is safer to present them UTF-8 characters instead of entities.

[stof]

this is broken, because some of them cannot be decoded (for instance, decoding < can change the HTML a lot, opening vulnerabilities in the HTML if it is followed by something under the control of the user)

[techi602]

@stof thanks for noticing security issue.
i added some nasty preg_replace hack to avoid lt; gt; being translated

I have noticed similar hack used in #138

[stof]

@techi602 decoding & also changes issue. If the original code was < (to display a literal < in an explanation about entities for instance), you will turn it into <, which displays <

[techi602]

@stof I dont think so - so i added testSpecialCharactersExplicit()

[stof]

line wrapping is useless. You should keep things on the same line

[techi602]

well the reason is to be human readable so you can see that two lines are the same
Or would you prefer to have only one variable for $actual and $expected?

[stof]

I don't find this code more readable actually. It just looks weird

[techi602]

so how about now? 😄

[techi602]

Fixed in #154