Fix dresden parser

[hesstobi]

Studentenwerk Dresden launched a redesign of their webpage, thus the parser was not working anymore.

[hesstobi]

I don't like failing tests...therefore fixed marburg

[klemens]

Thanks for the work! It does look quite good already! But please, don't use exceptions for control flow if not necessary. 😉

I will review the marburg parser separately.

[klemens]

There are some minor issues, but overall this also looks very good! I did not add comments for "exception as control flow", but there are definitely 5 too many AttributeErrors in this parser. 😉

I also notices, that the side dishes are sometimes formatted strangely, eg:

<span class="neo-menu-single-title">
                          , Schokoladenpudding mit Vanillesauce (<abbr title="Milch"><abbr title="Milch"><abbr title="Milch">22</abbr></abbr></abbr>) , Joghurtdessert (<abbr title="Milch"><abbr title="Milch"><abbr title="Milch">22</abbr></abbr></abbr>) ,Waldbeeren-Sahnequark (<abbr title="Milch"><abbr title="Milch"><abbr title="Milch">22</abbr></abbr></abbr>)                        </span>

Which results in:

<category name="Beilagen: Dessert">
  <meal>
    <name>, Schokoladenpudding mit Vanillesauce , Joghurtdessert ,Waldbeeren-Sahnequark </name>
    <note>Milch</note>
  </meal>
</category>

Ideally, we would want something like Schokoladenpudding mit Vanillesauce, Joghurtdessert, Waldbeeren-Sahnequark, so maybe strip(", ") and then re.sub(r'\s*,\s*', ", ", name)? We can also add this later.

[klemens]

Thanks! I squashed the commits into one for each parser and merged them. There seems to be a problem with hannover (I will take a look at it) and #89 is still a problem, so it might take a while before the changes are visible unfortunately.

[klemens]

So, hannover is fixed and I now host the parsers myself. So dresden and marburg should start working again tomorrow.