…lasses

Floating point numbers used for the 2D map room and exit-line sizes were
included in the XML file format twice and being read as integers when one,
at least, is a decimal number in range 0.1 to 1.1 so was being set to zero
when read for 9 out 11 cases.

The miss-spelt duplicate "commandSeperator" element is no longer written as
the parsing code can handle an element missing when reading the Host part
of the XML file.

The XMLimport class will silently ignore the above duplicate details so as
to avoid producing qDebug() messages about them when reading them in XML
files produced from older code.

I have taken the opportunity to clean up the code layout in these two
classes, the XMLexport had some code block indentation issues and the
XMLimport class some pointless if() code.  I also have added in error
detection/reporting code in the export code so that if errors arise in
method function calls they are reported up the caller chain and the
XMLWriter aborts as soon as an error is detected.  I have also added
QDebug() reporting for all instances where lua scripts are inserting into
the Mudlet items as they are loaded where they were only previously being
done for some types of item.

It was also possible to eliminate the redundant (QString) XMLexport::mType
member and the unneeded XMLexport(Host *, bool) C'tor.

I also renamed:
* (void) XMLimport::readMapList(QMap<QString, QStringList> & )
 ==> (void) XMLimport::readModulesDetailsMap(QMap<QString, QStringList> & )

to reduce any confusion over what it applies to (a QMap not the profile's
Map!)

I have held off on applying QStringLiteral(...) wrappers around all the
string literals in these classes as that would just make the change-set
even larger on this occasion but it is something to be done at a later
stage IMHO.

Signed-off-by: Stephen Lyons <slysven@virginmedia.com>

…e_30"

I note I had some details incorrect or not completely done when porting the
bug correction bits to the other "release_30" active branch.  Also some
comments were worth revising so that they match ones left in that branch.

* Removed misspelt "commandSeperator" sub-element of the "Host" element
  and the associated (QString) Host::mCommandSeperator which was unused
  and duplicated the (QString) Host::mCommandSeperator member...!

* Tweaked some QDebug() message texts for a uniform style and to correct
  some amendments missed when they were copied from one method to another.

* Added an argument to the XMLimport::readIntegerList(...) to pass the name
  of the parent trigger for use if/when the error condition of having an
  invalid (not a number) as an element that encodes the type of the trigger
  condition for a particular condition.  This is justified because under
  those conditions qFatal() is used {probably because it is the only QDebug
  class method that WILL be detectable on a WINDOWS platform for a RELEASE
  build} unfortunately it will KILL the Mudlet application - so it behoves
  Mudlet to report what killed it when it does die through this!

* Tweaked some comments to match versions in the release_30 bug_fix only
  version of this Pull Request.

* Rename the argument to void XMLimport::readHostPackage(Host *) from pT
  to pHost as the latter name is more prevalent elsewhere in the
  application.

Signed-off-by: Stephen Lyons <slysven@virginmedia.com>

…format

Had used QString argument replacement specifier (%1..%99) instead of C
printf() one (%s) for error message.

Also used clang-format on the class as per Vadim's suggestion.  I also
added the already present ./src/.clang-format file to the Qt project file
so that it is visible within Qt Creator environment.  I was unhappy with
some of the settings and revised/amended them:

It was necessary to use "//clang-format off" and "//clang-format on" around
the special pre_guard.h and post_guard.h headers needed to allow the use
of some memory checking facilities available on the Windows Platform using
the MSVC suite as without them these headers that themselves provide a
barrier around some DECLARATIONS and other needed bits within those two
files are re-positioned to BOTH be before the needed Qt libraries headers
files whereas they need to straddle just THOSE headers...

* ColumnLimit: 200 ==> 120
  I try to ensure error messages that are sent to the user in the Lua
  environment are limited to 80 characters which usually needs more than
  80 columns to lay multi-line messages out within the code as they would
  on screen (using a hanging indent for the code that leads up to a string
  itself, however 200 characters is wider than I would like to go even on
  my widescreen monitors equipped development platform.  I choose 120 as a
  reasonable compromise...

* ConstructorInitializerIndentWidth: 4 ==> 0
  0 was what we had used in the past with manual formatting and it does not
  seem necessary to change that.

* BreakBeforeBraces: Linux ==> Stroustrup
  "Linux" compacts the "else" to be on just one line with the preceding
  AND the succeeding braces on the same line, I had got used to the closing
  brace being followed with a line break with the "else" and the following
  opening brace on the next line and in line with the related "if" which I
  thought we had adopted.

* SpacesInParentheses: false ==> true
  pacesInSquareBrackets: {omitted} ==> true
  Although it takes up more columns I prefer to have white space between
  braces/brackets and the things they are wrapped around, it makes it
  a little easier to visually pick out the parameter/arguments enclosed
  - especially if I do not have the right spectacles on...!

* BinPackArguments: {omitted} ==> false
  If we are adding or removing arguments from a function call it is easier
  to "diff" pick out items if they can be laid out individually on their
  own line (it can also leave space for short comments about arguments).

* PointerAlignment: Left ==> Middle
  Whilst the left alignment emphasises that the type being declared is a
  pointer or reference, the right more closely matches the case where the
  variable is used.  I prefer the middle case which favours neither but
  flags that the item is not a plain variable... I realise that this is one
  that perhaps the most contentious change I have made here!

* AllowShortCaseLabelsOnASingleLine: {omitted} ==> true
  I am aware of several cases, particularly in the map drawing code
  where single line case label and statements make the choosing one
  set of settings from a group of them easier on the eye (and take up
  less lines on screen) with an increase readability IMHO as a result.

* AlignConsecutiveDeclarations: {omitted} ==> true
  Improves readability in header files IMHO.

Signed-off-by: Stephen Lyons <slysven@virginmedia.com>

…mport

The modifications I made to the file used to control the action of
clang-format needed to be undone and then used for a re-run on the files
affected (the XMLimport class ones).

By comparison to the release_30 (the Mudlet 3.0 release) branch this
branch stores the same config file in the ./src sub-directory compared to
the parent directory that the other branch uses.  A revision to the
QMake project file ./src/src.pro was needed to display the file in the
Qt Creator IDE as I had put in the location for the release_30 version in
error in the previous commit.

Signed-off-by: Stephen Lyons <slysven@virginmedia.com>