sstate-sysroot-cruft.sh: Improve to use it from CI

* strip tmpdir prefix, so that we have shorter paths which aren't
  builder specific
* use '#' for regexp delimiter so that we don't need to prefix
  forward slashes in paths
* extend default whitelist to cover typical cases
* add parameter for external whitelist file
* use number of found paths as return code, so that CI can easily
  report error when new untracked files are found
* use .txt suffix for all output files, so that they can be easily
  viewed in browser
* add populate_sysroot task, because somewhere between dora and daisy
  the populate-sysroot files in sstate-control were renamed to have
  underscore instead of dash
* only few entries not covered by this default whitelist were found
  in world build (but I'll leave these for people to whitelist, because
  they are not generated in most builds)
  * [^/]*/home/builder
    home directory from meta/recipes-graphics/builder/builder_0.1.bb
  * [^/]*/usr/src/kernel/patches
  * [^/]*/usr/lib/gdk-pixbuf-2.0/.*/loaders.cache
    3 places are using this, not sure which one creates it
    meta/recipes-gnome/gdk-pixbuf/gdk-pixbuf_2.30.8.bb:
      GDK_PIXBUF_MODULE_FILE=${STAGING_LIBDIR_NATIVE}/gdk-pixbuf-2.0/${LIBV}/loaders.cache
    meta/recipes-gnome/gtk+/gtk-update-icon-cache-native_3.4.4.bb:
      GDK_PIXBUF_MODULE_FILE=${STAGING_LIBDIR_NATIVE}/gdk-pixbuf-2.0/2.10.0/loaders.cache
    scripts/postinst-intercepts/update_pixbuf_cache:
      >$GDK_PIXBUF_MODULEDIR/../loaders.cache && \
      sed -i -e "s:$D::g" $GDK_PIXBUF_MODULEDIR/../loaders.cache

(From OE-Core rev: b1bfec63949e16abe8c11c34530dfbfb176c04cd)

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

scripts/sstate-sysroot-cruft.sh

@@ -17,6 +17,15 @@ Options:
   --tmpdir=<tmpdir>
         Specify tmpdir, will use the environment variable TMPDIR if it is not specified.
 	Something like /OE/oe-core/tmp-eglibc (no / at the end).
+
+  --whitelist=<whitelist-file>
+        Text file, each line is regular expression for paths we want to ignore in resulting diff.
+        You can use diff file from the script output, if it contains only expected exceptions.
+        '#' is used as regexp delimiter, so you don't need to prefix forward slashes in paths.
+        ^ and $ is automatically added, so provide only the middle part.
+        Lines starting with '#' are ignored as comments.
+        All paths are relative to "sysroots" directory.
+        Directories don't end with forward slash.
 EOF
 }
 
@@ -33,6 +42,11 @@ while [ -n "$1" ]; do
       [ -d "$tmpdir" ] || echo_error "Invalid argument to --tmpdir"
       shift
         ;;
+    --whitelist=*)
+      fwhitelist=`echo $1 | sed -e 's#^--whitelist=##' | xargs readlink -e`
+      [ -f "$fwhitelist" ] || echo_error "Invalid argument to --whitelist"
+      shift
+        ;;
     --help|-h)
       usage
       exit 0
@@ -51,28 +65,88 @@ done
 [ -d "$tmpdir" ] || echo_error "Invalid tmpdir \"$tmpdir\""
 
 OUTPUT=${tmpdir}/sysroot.cruft.`date "+%s"`
-WHITELIST="\/var\/pseudo\($\|\/[^\/]*$\) \/shlibs$ \.pyc$ \.pyo$"
+
+# top level directories
+WHITELIST="[^/]*"
+
+# generated by base-passwd recipe
+WHITELIST="${WHITELIST} \
+  .*/etc/group-\? \
+  .*/etc/passwd-\? \
+"
+# generated by pseudo-native
+WHITELIST="${WHITELIST} \
+  .*/var/pseudo \
+  .*/var/pseudo/[^/]* \
+"
+
+# generated by package.bbclass:SHLIBSDIRS = "${PKGDATA_DIR}/${MLPREFIX}shlibs"
+WHITELIST="${WHITELIST} \
+  .*/shlibs \
+  .*/pkgdata \
+"
+
+# generated by python
+WHITELIST="${WHITELIST} \
+  .*\.pyc \
+  .*\.pyo \
+"
+
+# generated by sgml-common-native
+WHITELIST="${WHITELIST} \
+  .*/etc/sgml/sgml-docbook.bak \
+"
+
+# generated by toolchain
+WHITELIST="${WHITELIST} \
+  [^/]*-tcbootstrap/lib \
+"
+
+# generated by useradd.bbclass
+WHITELIST="${WHITELIST} \
+  [^/]*/home \
+  [^/]*/home/xuser \
+"
+
+SYSROOTS="`readlink -f ${tmpdir}`/sysroots/"
 
 mkdir ${OUTPUT}
-find ${tmpdir}/sstate-control -name \*.populate-sysroot\* -o -name \*.package\* | xargs cat | grep sysroots | \
+find ${tmpdir}/sstate-control -name \*.populate-sysroot\* -o -name \*.populate_sysroot\* -o -name \*.package\* | xargs cat | grep sysroots | \
   sed 's#/$##g; s#///*#/#g' | \
   # work around for paths ending with / for directories and multiplied // (e.g. paths to native sysroot)
-  sort > ${OUTPUT}/master.list.all
-sort -u ${OUTPUT}/master.list.all > ${OUTPUT}/master.list # -u because some directories are listed for more recipes
+  sort | sed "s#^${SYSROOTS}##g" > ${OUTPUT}/master.list.all.txt
+sort -u ${OUTPUT}/master.list.all.txt > ${OUTPUT}/master.list.txt # -u because some directories are listed for more recipes
 find ${tmpdir}/sysroots/ | \
-  sort > ${OUTPUT}/sysroot.list
+  sort | sed "s#^${SYSROOTS}##g" > ${OUTPUT}/sysroot.list.txt
 
-diff ${OUTPUT}/master.list.all ${OUTPUT}/master.list > ${OUTPUT}/duplicates
-diff ${OUTPUT}/master.list ${OUTPUT}/sysroot.list > ${OUTPUT}/diff.all
+diff ${OUTPUT}/master.list.all.txt ${OUTPUT}/master.list.txt > ${OUTPUT}/duplicates.txt
+diff ${OUTPUT}/master.list.txt ${OUTPUT}/sysroot.list.txt > ${OUTPUT}/diff.all.txt
 
-cp ${OUTPUT}/diff.all ${OUTPUT}/diff
+grep "^> ." ${OUTPUT}/diff.all.txt | sed 's/^> //g' > ${OUTPUT}/diff.txt
 for item in ${WHITELIST}; do
-  sed -i "/${item}/d" ${OUTPUT}/diff;
+  sed -i "\\#^${item}\$#d" ${OUTPUT}/diff.txt;
+  echo "${item}" >> ${OUTPUT}/used.whitelist.txt
 done
 
+if [ -s "$fwhitelist" ] ; then
+  cat $fwhitelist >> ${OUTPUT}/used.whitelist.txt
+  cat $fwhitelist | grep -v '^#' | while read item; do
+    sed -i "\\#^${item}\$#d" ${OUTPUT}/diff.txt;
+  done
+fi
 # too many false positives for directories
 # echo "Following files are installed in sysroot at least twice"
 # cat ${OUTPUT}/duplicates
 
-echo "Following files are installed in sysroot, but not tracked by sstate"
-cat ${OUTPUT}/diff
+RESULT=`cat ${OUTPUT}/diff.txt | wc -l`
+
+if [ "${RESULT}" != "0" ] ; then
+  echo "ERROR: ${RESULT} issues were found."
+  echo "ERROR: Following files are installed in sysroot, but not tracked by sstate:"
+  cat ${OUTPUT}/diff.txt
+else
+  echo "INFO: All files are tracked by sstate or were explicitly ignored by this script"
+fi
+
+echo "INFO: Output written in: ${OUTPUT}"
+exit ${RESULT}