build and tests: curl_10char_object_name() shell function

lib/objnames.inc provides definition of curl_10char_object_name() shell
function. The intended purpose of this function is to transliterate a
(*.c) source file name that may be longer than 10 characters, or not,
into a string with at most 10 characters which may be used as an OS/400
object name.

Test case 1221 does unit testng of this function and also verifies
that it is possible to generate distinct short object names for all
curl and libcurl *.c source file names.

lib/objnames-test.sh is the shell script used for test case 1221.

tests/runtests.pl modified to accept shell script test cases.

More details inside lib/objnames.inc and lib/objnames-test.sh

3 files changed, 298 insertions, 1 deletions

diff --git a/lib/Makefile.am b/lib/Makefile.am
index 91790267a..39fb4ff75 100644
--- a/lib/Makefile.am
+++ b/lib/Makefile.am
@@ -38,7 +38,8 @@ EXTRA_DIST = Makefile.b32 Makefile.m32 Makefile.vc6 $(DSP)              \
  config-win32ce.h config-os400.h setup-os400.h config-symbian.h		\
  Makefile.Watcom config-tpf.h $(DOCS) $(VCPROJ) mk-ca-bundle.pl		\
  mk-ca-bundle.vbs firefox-db2pem.sh $(CMAKE_DIST) config-vxworks.h	\
- Makefile.vxworks config-vms.h checksrc.pl
+ Makefile.vxworks config-vms.h checksrc.pl objnames-test.sh		\
+ objnames.inc
 
 CLEANFILES = $(DSP) $(VCPROJ)
 
diff --git a/lib/objnames-test.sh b/lib/objnames-test.sh
new file mode 100755
index 000000000..fc1ce437f
--- /dev/null
+++ b/lib/objnames-test.sh
@@ -0,0 +1,211 @@
+#!/bin/sh
+# ***************************************************************************
+# *                                  _   _ ____  _
+# *  Project                     ___| | | |  _ \| |
+# *                             / __| | | | |_) | |
+# *                            | (__| |_| |  _ <| |___
+# *                             \___|\___/|_| \_\_____|
+# *
+# * Copyright (C) 2012, Daniel Stenberg, <daniel@haxx.se>, et al.
+# *
+# * This software is licensed as described in the file COPYING, which
+# * you should have received as part of this distribution. The terms
+# * are also available at http://curl.haxx.se/docs/copyright.html.
+# *
+# * You may opt to use, copy, modify, merge, publish, distribute and/or sell
+# * copies of the Software, and permit persons to whom the Software is
+# * furnished to do so, under the terms of the COPYING file.
+# *
+# * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
+# * KIND, either express or implied.
+# *
+# ***************************************************************************
+
+#
+# This Bourne shell script file is used by test case 1221 to do
+# unit testing of curl_10char_object_name() shell function which
+# is defined in file objnames.inc and sourced by this file and
+# any other shell script that may use it.
+#
+
+#
+# argument validation
+#
+
+if ! test $# -eq 1; then
+  echo "Usage: ${0} srcdir"
+  exit 1
+fi
+
+if ! test -f "${1}/runtests.pl"; then
+  echo "${0}: Wrong srcdir"
+  exit 1
+fi
+
+srcdir=${1}
+
+if ! test -f "$srcdir/../lib/objnames.inc"; then
+  echo "$0: Missing objnames.inc"
+  exit 1
+fi
+
+#
+# Some variables
+#
+
+logdir=log
+tstnum=1221
+
+list_c=$logdir/${tstnum}_list_c
+list_obj=$logdir/${tstnum}_list_obj
+list_obj_c=$logdir/${tstnum}_list_obj_c
+list_obj_uniq=$logdir/${tstnum}_list_obj_uniq
+
+
+#
+# Source curl_10char_object_name() function definition
+#
+
+. $srcdir/../lib/objnames.inc
+
+#
+# Some curl_10char_object_name() unit tests
+#
+
+echo 'Testing curl_10char_object_name...'
+echo ""
+
+argstr=123__678__ABC__FGH__KLM__PQRSTUV
+expect=16AFKPQRST
+outstr=`curl_10char_object_name $argstr`
+echo "result: $outstr expected: $expect input: $argstr"
+
+argstr=123__678__ABC__FGH__KLM__PQ.S.UV
+expect=16AFKPQ
+outstr=`curl_10char_object_name $argstr`
+echo "result: $outstr expected: $expect input: $argstr"
+
+argstr=123__678__ABC..FGH..KLM..PQRSTUV
+expect=16ABC
+outstr=`curl_10char_object_name $argstr`
+echo "result: $outstr expected: $expect input: $argstr"
+
+argstr=123__678_.ABC._FGH__KLM__PQRSTUV
+expect=16
+outstr=`curl_10char_object_name $argstr`
+echo "result: $outstr expected: $expect input: $argstr"
+
+argstr=123.567.90ABCDEFGHIJKLMNOPQRSTUV
+expect=123
+outstr=`curl_10char_object_name $argstr`
+echo "result: $outstr expected: $expect input: $argstr"
+
+argstr=1234567.90A.CDEFGHIJKLMNOPQRSTUV
+expect=1234567
+outstr=`curl_10char_object_name $argstr`
+echo "result: $outstr expected: $expect input: $argstr"
+
+argstr=1234567890.BCD.FGHIJKLMNOPQRSTUV
+expect=1234567890
+outstr=`curl_10char_object_name $argstr`
+echo "result: $outstr expected: $expect input: $argstr"
+
+argstr=12=45-78+0AB.DE.GHIJKLMNOPQRSTUV
+expect=1470AB
+outstr=`curl_10char_object_name $argstr`
+echo "result: $outstr expected: $expect input: $argstr"
+
+argstr=1234567890ABCDEFGHIJKLMNOPQRSTUV
+expect=1234567890
+outstr=`curl_10char_object_name $argstr`
+echo "result: $outstr expected: $expect input: $argstr"
+
+argstr=123_567_90A_CDE_GHIJKLMNOPQRSTUV
+expect=159CGHIJKL
+outstr=`curl_10char_object_name $argstr`
+echo "result: $outstr expected: $expect input: $argstr"
+
+argstr=123_567_90A_CDEFGHIJKLMNOPQRSTUV
+expect=159CDEFGHI
+outstr=`curl_10char_object_name $argstr`
+echo "result: $outstr expected: $expect input: $argstr"
+
+argstr=123_567_90ABCDEFGHIJKLMNOPQRSTUV
+expect=1590ABCDEF
+outstr=`curl_10char_object_name $argstr`
+echo "result: $outstr expected: $expect input: $argstr"
+
+argstr=123_567890ABCDEFGHIJKLMNOPQRSTUV
+expect=1567890ABC
+outstr=`curl_10char_object_name $argstr`
+echo "result: $outstr expected: $expect input: $argstr"
+
+argstr=1234567890ABCDEFGHIJKLMNOPQRSTUV
+expect=1234567890
+outstr=`curl_10char_object_name $argstr`
+echo "result: $outstr expected: $expect input: $argstr"
+
+#
+# Verify that generated object name is distinct for
+# all *.c source files in lib and src subdirectories.
+#
+
+ls $srcdir/../lib/*.c > $list_c
+ls $srcdir/../src/*.c >> $list_c
+
+rm -f $list_obj
+
+for c_fname in `cat $list_c`; do
+  obj_name=`curl_10char_object_name $c_fname`
+  echo "$obj_name" >> $list_obj
+done
+
+sort -u $list_obj > $list_obj_uniq
+
+cnt_c=`cat $list_c | wc -l`
+cnt_u=`cat $list_obj_uniq | wc -l`
+
+echo ""
+echo ""
+echo ""
+if test $cnt_c -eq $cnt_u; then
+  echo "10-characters-or-less generated object names are unique."
+  obj_name_clash="no"
+else
+  echo "10-characters-or-less generated object names are clashing..."
+  obj_name_clash="yes"
+fi
+
+if test $obj_name_clash = "yes"; then
+  #
+  # Show clashing object names and respective source file names
+  #
+  echo ""
+  paste $list_obj $list_c | sort > $list_obj_c
+  prev_match="no"
+  prev_line="unknown"
+  prev_obj_name="unknown"
+  while read this_line; do
+    obj_name=`echo "$this_line" | cut -f1`
+    if test "x$obj_name" = "x$prev_obj_name"; then
+      if test "x$prev_match" != "xyes"; then
+        echo "$prev_line"
+        echo "$this_line"
+        prev_match="yes"
+      else
+        echo "$this_line"
+      fi
+    else
+      prev_match="no"
+    fi
+    prev_line=$this_line
+    prev_obj_name=$obj_name
+  done < $list_obj_c
+fi
+
+rm -f $list_c
+rm -f $list_obj
+rm -f $list_obj_c
+rm -f $list_obj_uniq
+
+# end of objnames-test.sh
diff --git a/lib/objnames.inc b/lib/objnames.inc
new file mode 100644
index 000000000..2ec8915b3
--- /dev/null
+++ b/lib/objnames.inc
@@ -0,0 +1,85 @@
+# ***************************************************************************
+# *                                  _   _ ____  _
+# *  Project                     ___| | | |  _ \| |
+# *                             / __| | | | |_) | |
+# *                            | (__| |_| |  _ <| |___
+# *                             \___|\___/|_| \_\_____|
+# *
+# * Copyright (C) 2012, Daniel Stenberg, <daniel@haxx.se>, et al.
+# *
+# * This software is licensed as described in the file COPYING, which
+# * you should have received as part of this distribution. The terms
+# * are also available at http://curl.haxx.se/docs/copyright.html.
+# *
+# * You may opt to use, copy, modify, merge, publish, distribute and/or sell
+# * copies of the Software, and permit persons to whom the Software is
+# * furnished to do so, under the terms of the COPYING file.
+# *
+# * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
+# * KIND, either express or implied.
+# *
+# ***************************************************************************
+
+#
+# This file is sourced from curl/packages/OS400/initscript.sh and
+# other Bourne shell scripts. Keep it as portable as possible.
+#
+
+#
+# curl_10char_object_name
+#
+# This shell function accepts a single string argument with unspecified
+# length representing a (*.c) source file name and returns a string which
+# is a transformation of given argument.
+#
+# The intended purpose of this function is to transliterate a (*.c) source
+# file name that may be longer than 10 characters, or not, into a string
+# with at most 10 characters which may be used as an OS/400 object name.
+#
+# This function might not be universally usefull, nor we care about it.
+#
+# It is intended to be used with libcurl's (*.c) source file names, so
+# dependency on libcurl's source file naming scheme is acceptable and
+# good enough for its intended use. Specifically it makes use of the fact
+# that libcurl's (*.c) source file names which may be longer than 10 chars
+# are conformed with underscore '_' separated substrings, or separated by
+# other character which does not belong to the [0-9], [a-z] or [A-Z] sets.
+#
+# This allows repeatable and automatic short object name generation with
+# no need for a hardcoded mapping table.
+#
+# Transformation is done in the following way:
+#
+# 1) Leading directory components are removed.
+# 2) Leftmost dot character and any other char following it are removed.
+# 3) Lowercase characters are transliterated to uppercase.
+# 4) Characters not in [A-Z] or [0-9] are transliterated to underscore '_'.
+# 5) Every sequence of one or more underscores is replaced with a single one.
+# 6) Five leftmost substrings which end in an underscore character are
+#    replaced by the first character of each substring, while retaining
+#    the rest of the string.
+# 7) Finally the result is truncated to 10 characters.
+#
+# Resulting object name may be shorter than 10 characters.
+#
+# Test case 1221 does unit testng of this function and also verifies
+# that it is possible to generate distinct short object names for all
+# curl and libcurl *.c source file names.
+#
+
+curl_10char_object_name() {
+  echo "${1}" | \
+  sed -e 's:.*/::' \
+   -e 's:[.].*::' \
+   -e 'y:abcdefghijklmnopqrstuvwxyz:ABCDEFGHIJKLMNOPQRSTUVWXYZ:' \
+   -e 's:[^ABCDEFGHIJKLMNOPQRSTUVWXYZ1234567890_]:_:g' \
+   -e 's:__*:_:g' \
+   -e 's:\([^_]\)[^_]*_\(.*\):\1\2:' \
+   -e 's:\([^_]\)\([^_]\)[^_]*_\(.*\):\1\2\3:' \
+   -e 's:\([^_]\)\([^_]\)\([^_]\)[^_]*_\(.*\):\1\2\3\4:' \
+   -e 's:\([^_]\)\([^_]\)\([^_]\)\([^_]\)[^_]*_\(.*\):\1\2\3\4\5:' \
+   -e 's:\([^_]\)\([^_]\)\([^_]\)\([^_]\)\([^_]\)[^_]*_\(.*\):\1\2\3\4\5\6:' \
+   -e 's:^\(..........\).*:\1:'
+}
+
+# end of objectname.inc