script to find incorrect tests on unsigneds

[Roel Kluin <12o3l@tiscali.nl>]

A bash script to find incorrect tests on unsigned values. for instance:

unsigned int i;
...
i = neg_ret_function();
...
if (i < 0) ...

#!/bin/bash
# (c) roel kluin 2008 GPL v2
#
# TODO: make this working also for 
#                              ... $unsigned_var = $signed_var ...
#                              ... $unsigned_var = \(-POSITIVE_DEF\|NEGATIVE_DEF\) ...
#                              ... ( $unsigned_var [+*/%^&~-]?= ... ) < 0 ...

# a number
int="[0-9]"
hex="[a-f0-9]"
hEx="[A-Fa-f0-9]"
HEX="[A-F0-9]"
upp="[A-Z]"
up_="[A-Z_]"
low="[a-z0-9]"
lo_="[a-z0-9_]"
alp="[A-Za-z]"
al_="[A-Za-z_]"
ALN="[A-Z0-9]"
AN_="[A-Z0-9_]"
aln="[A-Za-z0-9]"
an_="[A-Za-z0-9_]"
# to match something like 1ul, floats or hexes as well:
D="$int*\.\?$int\+x\?$hex*[uUlL]\{0,3\}[fF]\?"

# more strict and catches it (costs one backreference for (git-)grep)
d="\($int\+[uUlLfF]\?\|$int\+[uU]\?[lL][lL]\?\|0x$hex\+\|0x$HEX\+\|$i\+[lL][lL][uU]\|$i*\.$i\+[fF]\?\)"

# capital: can be used to catch a definition or config option
K="$UP_\+$AN_*";

# can be used for a variable/function name:
V="$an_\+$an_*"

# works the same as above, but also for members and arrays: one backreference
# is more strict
W="$V\(\[$s$V$s\]\|\[$s$D$s\]\|\.$V\|->$V\)*"
# catches it at once (less strict)
w="\($V\|${V}\[$s$an_*${s}\]\|$V\.\|$V->\)\+"

# seperators:
s="[[:space:]]*";
S="[[:space:]]\+"

# useful to delimit a variable name:
Q="[^[:alnum:]_]"

# match the end of the line, including comments: one backreference (but at eol)
cendl="$s\(\/[\*\/].*\)\?$"

# match something that is not comment, string or character: 2 backreferences
ccode="\([^\/\"']*\|\/[^\*\/]\|\/\*\([^\*]*\|\**[^\/\*]\)*\*\+\/\|'[^']*'\|\"[^\"]*\"\)*"

# resp function open and end (only works when indentation is correct.
fo="^[\{]$cendl"
fe="^[\}]$cendl"
se="^[\}];$cendl"

# to match nested parentheses
nps="[^()]*"
nstdps="\(($nps\(($nps\(($nps\(($nps\(($nps)$nps\)*)$nps\)*)$nps\)*)$nps\)*)$nps\)*"


# first determine unsigned typedefs
arr="\(\[[^\]]*\]$s\)*"
attr="__attribute__$s(([^;]*))"
utype="${s}typedef${S}unsigned$S\($V$S\)*"
uns="unsigned$(
git-grep "^$utype\($V$s$arr\|$attr$S$V$s$arr\|$V$s$arr$S$attr\)$s;$cendl" | sed -n "s/^[^.]*\.[hc]:$utype\(\($V\)$s$arr\|$attr$S\($V\)$s$arr\|\($V\)$s$arr$S$attr\)$s;$cendl/\\\\|\3\5\7/p" | sort | uniq | tr -d "\n")"

# define left and right operators
# to decrease the number backrefences, these are assigned in loops
oplfor op in "[;,|^?:(]" "[\!+*/%&|~^-]=" ">>=" "<<=" "\[" "&&" "$an_$s&"; do 
 opl="$opl$op\|$op$s++\|$op$s--\|"
done
opl="\(${opl%|})"
oprfor op in "[;,&|^?:)]" "[\!+*/%&|~^<>-]=" ">>=" "<<=" ">[^>]" "<[^<]" "\]"; do 
 opr="$opr$op\|$op$s++\|$op$s--\|"
done
opr="\(${opr%|})"

# string catches invalid comparison
q1="$opl$s\($w$s\(>=${s}0\|<${s}0\|[><\!=]=$s-$s$D\|[<>]$s-$s$D\)\|\(0$s>\|0$s<=\|-$s$D${s}[><\!=]=\|-$s$D${s}[<>]\)$s$w\)$s$opr"

start=0
end=$(echo $uns | tr -cd "|" | wc -c)

# main function
while [ $start -lt $end ]; do
  # we match 30 typedefs at a time
  tuns="$(echo $uns | cut -d "\\" -f$start-$(($start+29)))"

  # catch candidate files 
  for f in $(git-grep -l "^\(${ccode}[,;]$s\)\?\($tuns\)$S" | grep "[^.]*\.[ch]" | xargs grep -l "$q1"); do
    for n in $(sed -n "/^.*$q1/=" $f); do # lines
      for v in $(sed -n "${n}s/^.*$q1.*$/\3\6/p" $f); do
        # n2 = wine there
        head -n$n $f | tac | sed -n "/^[{]/q; /^\(.*$Q\)\?\($tuns\)$S\($V$S\)*$v$Q.*$/=" | while read n2;        
        do
          echo "# --- invalid test on unsigned variable '$v' --- #"
          echo "vi $f +$n2 # unsigned declaration"
          echo "vi $f +$n # invalid test"
        done
      done
    done
  done
  start=$(($start+30))
done | less