Bug newsreader content download / content downloads

[antoinedavid]

Hi,

We can't use content download or content downloads in a news because the file are corrupt when downloading.

A part of the template are injected in the header and we don't have the end part of the file.

I found this issue on the demo site too. I've not test on Contao 4.0 but i've the same issue since Contao 4.1

Header of a PDF file for exemple:

<div class="layout_full block" itemscope itemtype="http://schema.org/Article">

  <h1 itemprop="name">Hello world</h1>

      <p class="info"><time datetime="2017-01-30T17:27:00+01:00" itemprop="datePublished">2017-01-30 17:27</time> par Admin  </p>

      %PDF-1.5
%âãÏÓ
10 0 obj
<</Linearized 1/L 134289/O 12/E 129957/N 1/T 133984/H [ 777 289]>>
endobj

90 0 obj
<</DecodeParms<</Columns 5/Predictor 12>>/Filter/FlateDecode/ID[<E182E5868FCF0E01A64FE1F579ECD28B><BC0DE23470124040AC4DE59087625AF3>]/Index[10 135]/Info 9 0 R/Length 209/Prev 133985/Root 11 0 R/Size 145/Type/XRef/W[1 3 1]>>stream
hÞbbd```b``‘‘~ ’qˆä¬‘<A$×S°H˜¼   "ÙÝ@$·ˆäð
³Áê¹[@¤À°9µ`2DƒH¦n°-ý@’1ì'Xl;㙑 ‹$Ho0»Lšƒõž³ÁzçÉû`RL
!±1eÑU2ÌER#„ ™„ÐEzÀì{`R˜!âßÍ›€ä&°    í@øÓ^dóÙH$"ÿY˜€±qã(I2ùŸùÏ€  Øã)
endstream
endobj
startxref
0
%%EOF...............

Best

[leofeyer]

See contao/core-bundle#696.