From 4a4edb5807d094f22788c7fbd23b05a29ef98bdc Mon Sep 17 00:00:00 2001
From: Thibault Kruse <thibault.kruse@gmx.de>
Date: Fri, 13 May 2016 23:14:52 +0900
Subject: [PATCH 1/2] add hunspell spell checks, fix bugs in generation of
 plaintext file

---
 .travis.yml                 |   5 +
 scripts/Makefile            |  15 +
 scripts/hunspell/isocpp.dic | 539 ++++++++++++++++++++++++++++++++++++
 scripts/python/md-split.py  |  16 +-
 4 files changed, 566 insertions(+), 9 deletions(-)
 create mode 100644 scripts/hunspell/isocpp.dic

diff --git a/.travis.yml b/.travis.yml
index 222ad98..64550f2 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -9,6 +9,11 @@ language: c++
 # alternatives: gcc, clang, or both (as yaml list)
 compiler: clang
 
+addons:
+  apt:
+    packages:
+    - hunspell
+
 install:
  -
 
diff --git a/scripts/Makefile b/scripts/Makefile
index ada1344..133737e 100644
--- a/scripts/Makefile
+++ b/scripts/Makefile
@@ -17,6 +17,7 @@ all: \
 check-markdown \
 check-references \
 check-notabs \
+hunspell-check \
 cpplint-all \
 check-badchars
 
@@ -78,6 +79,17 @@ check-badchars: $(SOURCEPATH) $(BUILD_DIR) Makefile
 	if [ -s $(BUILD_DIR)/CppCoreGuidelines.md.badchars ]; then echo 'Warning: Undesired chars (–’‘“”¸…¦) found, use straight quotes instead:'; cat $(BUILD_DIR)/CppCoreGuidelines.md.badchars; false; fi;
 
 
+.PHONY: hunspell-check
+hunspell-check: $(BUILD_DIR)/plain-nohtml.txt
+	hunspell -p hunspell/isocpp.dic -u < build/plain-nohtml.txt > $(BUILD_DIR)/hunspell-report.txt
+	if [ -s $(BUILD_DIR)/hunspell-report.txt ]; then echo 'Warning: Spellcheck failed, fix words or add to dictionary:'; cat $(BUILD_DIR)/hunspell-report.txt; false; fi;
+
+# only list words that are not in dict
+# to include all add them to bottom of hunspell/isocpp.dict, and run
+# cat hunspell/isocpp.dic | sort | uniq > hunspell/isocpp.dic2; mv hunspell/isocpp.dic2 hunspell/isocpp.dic
+.PHONY: hunspell-list
+hunspell-list: $(BUILD_DIR)/plain.txt
+	hunspell -p hunspell/isocpp.dic -l < build/plain-nohtml.txt
 
 #### Cpplint
 
@@ -96,6 +108,9 @@ $(BUILD_DIR)/codeblocks: splitfile
 
 $(BUILD_DIR)/plain.txt: splitfile
 
+$(BUILD_DIR)/plain-nohtml.txt: $(BUILD_DIR)/plain.txt
+	sed 's;<a \(name\|href\)=".*</a>;;g' $(BUILD_DIR)/plain.txt > $(BUILD_DIR)/plain-nohtml.txt
+
 .PHONY: splitfile
 splitfile: $(SOURCEPATH) ./python/md-split.py
 	python ./python/md-split.py $(SOURCEPATH) $(BUILD_DIR)/plain.txt $(BUILD_DIR)/codeblocks
diff --git a/scripts/hunspell/isocpp.dic b/scripts/hunspell/isocpp.dic
new file mode 100644
index 0000000..aad6238
--- /dev/null
+++ b/scripts/hunspell/isocpp.dic
@@ -0,0 +1,539 @@
+'
+0xFF0000
+'14
+2D
+2K
+2ndEdition
+2RDU00001
+3rdEdition
+78e
+86xWVb4XIyE
+98's
+à
+a1
+A1
+a2
+A2
+aa
+ABA
+abi
+ABI
+ABIs
+abstr
+accessor
+ack
+addressof
+adl
+ADL
+Adve
+Alexandrescu
+Alexandrescu01
+algo
+alloc
+alloc0
+ap
+API
+APIs
+archetypical
+arg
+argh
+args
+arr2
+arrayindex
+ASIC
+AST
+async
+BDE
+behaviorless
+BigPOD
+Bjarne
+Bloomberg
+Boehm
+bool
+buf
+bufmax
+C1
+C2
+callees
+callers'
+call's
+camelCase
+CamelCase
+CaMelcAse
+CaMeLcAsEvArIaBlE
+Cargill
+Cargill92
+CComPtr
+cerr
+chrono
+cin
+class'
+clib
+Cline99
+ClosePort
+CommonMark
+composability
+composable
+conceptsTS
+cond
+const
+Const
+constcast
+constexpr
+constness
+copy2
+CORBA
+cout
+CP
+cplusplus
+Cplusplus
+Cplusplus03
+CplusplusCS
+cpp
+cpp98
+CppCon
+CRTP
+cstdarg
+cstring
+cstylecast
+ctor
+ctors
+cxx
+cyclomatic
+Cyclomatic
+czstring
+d1
+D1
+d1's
+D2
+d2's
+dag
+dcl
+dd
+de
+Dechev
+default0
+default00
+defop
+del
+derived1
+derived2
+destructors
+Destructors
+detatch
+Dewhurst
+Dewhurst03
+disambiguator
+draw2
+dtor
+dtors
+dyn
+dynarray
+ECBS
+endl
+enum
+Enum
+enums
+eq
+EqualityComparable
+errno
+expr
+f1
+f2
+f3
+f4
+fac
+Facebook
+fallthrough
+fallthroughs
+faq
+fclose
+fct
+fib10
+file1
+file2
+file3
+flag1
+fmt
+fn
+fo
+foo
+Foo
+foobar
+Foobar
+FOOBAR
+fopen
+fs
+func
+func1
+fx
+g1
+g2
+GCC
+Geosoft
+getx
+GFM
+Girou
+github
+GitHub
+gp
+GPLv3
+gsl
+GSL
+gx
+handcoded
+Henricson
+Henricson97
+hh
+hier
+hierclass
+hnd
+homebrew
+HPL
+href
+Hyslop
+IDE
+IDEs
+IEC
+ifdef
+iff
+ifstream
+impactful
+Impl
+incompleat
+increment1
+Incrementable
+indices
+ing
+init
+inkorrekt
+inline
+inlined
+inlining
+inout
+InputIterator
+int32
+int64
+ints
+io
+iostream
+Iostream
+iso
+isocpp
+ISORC
+istream
+Iter
+Jiangang
+join's
+JSF
+Juhl
+knr
+Koenig97
+l
+Lakos
+Lakos96
+Lavavej
+LCSD05
+lifecycle
+llvm
+lockfree
+Lomow
+LSP
+lst
+lvalue
+lvalues
+m1
+m2
+macros2
+malloc
+mallocfree
+'many'
+Mathematizing
+maul2
+md
+memberinit
+members'
+memcmp
+memmove
+memoization
+memoized
+memset
+metameta
+metaprogram
+metaprogramming
+Metaprogramming
+Meyers01
+Meyers05
+Meyers15
+Meyers96
+Meyers97
+microbenchmarks
+modify1
+modify2
+moredata
+msgsl
+mtx
+Murray93
+mutex
+mutexes
+myMap
+MyMap
+myset
+myX
+n'
+namespace
+namespaces
+Namespaces
+NaN
+nargs
+Naumann
+ness
+newdelete
+nh
+NL
+noexcept
+nondependent
+nonexported
+nongeneric
+nonlocally
+nonprivate
+nonreusable
+nonvirtual
+nonvirtually
+nothrow
+NR
+nullptr
+NVI
+ok
+oo
+OO
+OOP
+OOPSLA'09
+oper
+O'Reilly
+org
+ostream
+overabstract
+overconstrain
+overconstrained
+overridable
+p1
+p2
+p3
+pµÃoorly
+Pardoe
+parens
+passthrough
+pb
+pb1
+pb2
+pc
+performant
+pessimization
+PIMPL
+Pirkelbauer
+PL4
+PLDI
+Poco
+PODs
+poly
+polymorphically
+POPL
+PortHandle
+PostInitialize
+PPP
+pre
+Pre
+precomputation
+prefetcher
+printf
+printf's
+Proc
+productinfo
+Productinfo
+proto
+ps
+ptr
+Ptr
+ptr2
+ptr's
+q2
+qqq
+R0
+r2
+raii
+RAII
+Rc
+rcon
+Rcon
+Rconc
+Rconst
+Rcpl
+Rec2
+refactor
+refactored
+refcount
+regex
+Regex
+RegularFunction
+reimplement
+reinterpretcast
+Reis
+Renum
+reseat
+reseating
+reseats
+resizable
+retryable
+reusability
+Reusability
+Ri
+Rl
+rnd
+Rnr
+Ro
+Rouquette
+Rp
+Rper
+Rr
+RRconc
+Rsl
+RTTI
+rvalue
+rvalues
+RVO
+'s
+s1
+s1's
+s2
+Sarkar
+scanf
+Sd
+SEI
+Semiregular
+SemiRegular
+Sergey
+Sewell
+SFINAE
+sharedness
+sharedptrparam
+'sharedptrparam'
+SignedIntegral
+simpleFunc
+'size'
+sizeof
+sl
+SL
+smartptrconcepts
+smartptrget
+smartptrparam
+smartptrs
+SMS
+SomeLargeType
+specialization2
+spinlock
+splonk
+splunk
+SScp
+stdarg
+stdlib
+Stepanov
+stl
+STL
+stmt
+str
+strdup
+strlen
+Stroustrup
+Stroustrup00
+Stroustrup05
+Stroustrup13
+Stroustrup14
+Stroustrup's
+struct
+suboperations
+subsetting
+sum1
+sum2
+supertype
+Susmit
+SuttAlex05
+Sutter
+Sutter00
+Sutter02
+Sutter04
+Sutter's
+SuttHysl04b
+sz
+T0
+Taligent94
+Taligent's
+TBD
+templated
+Templating
+templatize
+templatized
+thread1
+thread2
+Tjark
+tmp
+TMP
+TODO
+toolchains
+TotallyOrdered
+TP
+tradeoff
+TSs
+tt
+typeid
+typename
+typesafe
+UB
+unaliased
+uncompromised
+unenforcable
+uninit
+uniqueptrparam
+unnamed2
+use1
+users'
+util
+v1
+va
+ValueType
+vararg
+varargs
+variables'
+variadic
+Variadic
+vbase
+vd1
+vec
+Vector0
+Vector1
+Vector2
+vid
+virtuality
+virtuals
+VLAs
+volatile2
+vr
+vtbls
+vv
+w0
+webby
+Webcolor
+webcolors
+WG21
+'widen'
+x1
+x2
+xmax
+xor
+Xs
+years'
+yy
+Zhuang
+zstring
+Zubkov
+zz
diff --git a/scripts/python/md-split.py b/scripts/python/md-split.py
index 7b8768f..8968153 100755
--- a/scripts/python/md-split.py
+++ b/scripts/python/md-split.py
@@ -63,7 +63,7 @@ def main():
                     code_block_index += 1
                 # reach here either line was not code, or was code
                 # and we dealt with n code lines
-                if not is_code(line, indent_depth):
+                if indent_depth < 4 or not is_code(line, indent_depth):
                     # store header id for codeblock
                     section_id = get_marker(line)
                     if section_id is not None:
@@ -80,7 +80,7 @@ def process_code(read_filehandle, text_filehandle, line, linenum, sourcefile, co
         try:
             line = read_filehandle.next()
             linenum += 1
-            text_filehandle.write('')
+            text_filehandle.write('\n')
         except StopIteration:
             return ('', linenum)
     start_linenum = linenum
@@ -97,14 +97,13 @@ def process_code(read_filehandle, text_filehandle, line, linenum, sourcefile, co
             if (not has_actual_code
                     and not line.strip().startswith('//')
                     and not line.strip().startswith('???')
-                    and not line.strip() ==''):
+                    and not line.strip() == ''):
                 has_actual_code = True
         else:
             # write empty line so line numbers stay stable
-            text_filehandle.write('')
+            text_filehandle.write('\n')
 
         if (not line.strip() == '```'):
-
             if ('???' in no_comment_line or '...' in no_comment_line):
                 has_question_marks = True
             linebuffer.append(dedent(line) if not fenced else line)
@@ -115,9 +114,8 @@ def process_code(read_filehandle, text_filehandle, line, linenum, sourcefile, co
             line = ''
             break
     codefile = os.path.join(codedir, '%s%s.cpp' % (name, index))
-
     if fenced:
-        text_filehandle.write('')
+        text_filehandle.write('\n')
 
     if (has_actual_code and not has_question_marks):
         # add commonly used headers, so that lines can compile
@@ -139,8 +137,8 @@ using namespace std;   // by md-split
 // %s : %s
 ''' % (sourcefile, start_linenum))
             # TODO: if not toplevel code, wrap inside class
-            for line in linebuffer:
-                code_filehandle.write(line)
+            for codeline in linebuffer:
+                code_filehandle.write(codeline)
     return (line, linenum)
 
 

From 180cd89a1f3105a20e79b81964b4d388dca94040 Mon Sep 17 00:00:00 2001
From: Thibault Kruse <thibault.kruse@gmx.de>
Date: Tue, 31 May 2016 07:29:30 +0200
Subject: [PATCH 2/2] Cleanup travis file (reference check already done in
 Makefile)

---
 .travis.yml | 11 -----------
 1 file changed, 11 deletions(-)

diff --git a/.travis.yml b/.travis.yml
index 64550f2..74c3f37 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -21,16 +21,5 @@ script:
  - cd scripts; make -k
  - cd ..
 
-## find lines with tabs
-# - rm -f CppCoreGuidelines.md.tabs
-# - cat CppCoreGuidelines.md | nl -ba | sed -s 's/\(^[^\t]*\)\t/\1--/g' | grep  $'\t'  | sed -s 's/\t/\*\*\*\*/g' > CppCoreGuidelines.md.tabs
-# - if [[ -s CppCoreGuidelines.md.tabs ]]; then echo 'Tabs found'; cat CppCoreGuidelines.md.tabs; false; fi;
-
-## check references unique
- - rm -f CppCoreGuidelines.md.uniq
- - grep -oP '(?<=<a name=")[^\"]+' CppCoreGuidelines.md | uniq -d > CppCoreGuidelines.md.uniq
- - if [[ -s CppCoreGuidelines.md.uniq ]]; then echo 'Found duplicate anchors:'; cat CppCoreGuidelines.md.uniq; false; fi;
-
-
 notifications:
   email: false