[GHC] #9230: Make -fwarn-tabs the default

[GHC]

#9230: Make -fwarn-tabs the default
-------------------------------------+-------------------------------------
              Reporter:  dfeuer      |            Owner:  mlen
                  Type:  feature     |           Status:  closed
  request                            |        Milestone:  7.10.1
              Priority:  normal      |          Version:  7.8.2
             Component:  Compiler    |         Keywords:
            Resolution:  fixed       |     Architecture:  Unknown/Multiple
      Operating System:              |       Difficulty:  Easy (less than 1
  Unknown/Multiple                   |  hour)
       Type of failure:              |       Blocked By:
  None/Unknown                       |  Related Tickets:
             Test Case:              |
              Blocking:              |
Differential Revisions:  Phab:D255,  |
  Phab:D399                          |
-------------------------------------+-------------------------------------

Comment (by isaacdupree):

 re: Polarina: Let's try describing a rule that would catch all suspect
 uses of tabs.

 Tabs may not be preceded by a non-tab character: that's a start.

 Let's add: If two adjacent lines have different numbers of tabs, the line
 with fewer tabs must have a non-whitespace character following the last
 tab.  This way the layout algorithm will never compare the width of a
 space and a tab (I think).  Except that is not enough when layout is begun
 on the same line as the layout-inducing keyword:

 {{{
 f x = z where y = x + x
               z = y + y
 }}}

 (Personally I consider it bad style to do that even with spaces.  If you
 change the first line before the layout keyword, you need to change the
 whole layout.  Also if you dare to use a non-monospace font it will look
 wrong.  I prefer starting the layout on a new line.  [I wish I had a
 warning for this.])

 I think there is a way to warn exactly when the tab width affects layout,
 but it would have to happen after lexing, and I'm not volunteering to
 implement it.

--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/9230#comment:31>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler