Merge pull request #1669 from Unidata/gh1668.wif

Add test to #1668

RELEASE_NOTES.md

@@ -7,6 +7,7 @@ This file contains a high-level description of this package's evolution. Release
 
 ## 4.7.4 - TBD
 
+* [Bug Fix] Correct behavior for the command line utilities when directly accessing a directory using utf8 characters. See [Github #1669](https://github.com/Unidata/netcdf-c/issues/1669), [Github #1668](https://github.com/Unidata/netcdf-c/issues/1668) and [Github #1666](https://github.com/Unidata/netcdf-c/issues/1666) for more information.
 * [Bug Fix] Attempts to set filters or chunked storage on scalar vars will now return NC_EINVAL. Scalar vars cannot be chunked, and only chunked vars can have filters. Previously the library ignored these attempts, and always storing scalars as contiguous storage. See [https://github.com/Unidata/netcdf-c/issues/1644].
 * [Enhancement] Support has been added for multiple filters per variable.  See [https://github.com/Unidata/netcdf-c/issues/1584].
 * [Enhancement] Now nc_inq_var_szip retuns 0 for parameter values if szip is not in use for var. See [https://github.com/Unidata/netcdf-c/issues/1618].

libdispatch/dfile.c

@@ -1851,8 +1851,8 @@ NC_create(const char *path0, int cmode, size_t initialsz,
 
     {
         /* Skip past any leading whitespace in path */
-        const char* p;
-        for(p=(char*)path0;*p;p++) {if(*p > ' ') break;}
+        const unsigned char* p;
+        for(p=(const unsigned char*)path0;*p;p++) {if(*p > ' ') break;}
 #ifdef WINPATH
         /* Need to do path conversion */
         path = NCpathcvt(p);
@@ -1999,8 +1999,8 @@ NC_open(const char *path0, int omode, int basepe, size_t *chunksizehintp,
 
     {
         /* Skip past any leading whitespace in path */
-        const char* p;
-        for(p=(char*)path0;*p;p++) {if(*p > ' ') break;}
+        const unsigned char* p;
+        for(p=(const unsigned char*)path0;*p;p++) {if(*p > ' ') break;}
 #ifdef WINPATH
         /* Need to do path conversion */
         path = NCpathcvt(p);

ncdump/CMakeLists.txt

@@ -151,6 +151,7 @@ ENDIF(MSVC)
     add_sh_test(ncdump tst_fileinfo)
     add_sh_test(ncdump tst_hdf5_offset)
   ENDIF(USE_NETCDF4)
+  add_sh_test(ncdump test_unicode_directory)
 
   add_sh_test(ncdump tst_null_byte_padding)
   IF(USE_STRICT_NULL_BYTE_HEADER_PADDING)

ncdump/Makefile.am

@@ -34,7 +34,7 @@ utils.h utils.c dimmap.h dimmap.c list.c list.h
 # netcdf-3 validator program
 # (https://github.com/Parallel-NetCDF/PnetCDF/blob/master/src/utils/ncvalidator/ncvalidator.c)
 # that prints out the structure of a netcdf-3 file.
-# This program is built but not installed. 
+# This program is built but not installed.
 noinst_PROGRAMS += ncvalidator
 ncvalidator_SOURCES = ncvalidator.c
 
@@ -61,7 +61,7 @@ bom tst_dimsizes nctrunc
 # Tests for classic and 64-bit offset files.
 TESTS = tst_inttags.sh run_tests.sh tst_64bit.sh ref_ctest	\
 ref_ctest64 tst_output.sh tst_lengths.sh tst_calendars.sh	\
-run_utf8_tests.sh tst_nccopy3.sh tst_nccopy3_subset.sh		\
+run_utf8_tests.sh test_unicode_directory.sh tst_nccopy3.sh tst_nccopy3_subset.sh		\
 tst_charfill.sh tst_iter.sh tst_formatx3.sh tst_bom.sh		\
 tst_dimsizes.sh run_ncgen_tests.sh tst_ncgen4_classic.sh test_radix.sh
 
@@ -149,7 +149,9 @@ tst_ncgen4.sh tst_ncgen4_classic.sh tst_ncgen4_diff.sh			\
 tst_ncgen4_cycle.sh tst_null_byte_padding.sh				\
 ref_null_byte_padding_test.nc ref_tst_irish_rover.nc ref_provenance_v1.nc \
 ref_tst_radix.cdl tst_radix.cdl test_radix.sh                           \
-ref_nccopy_w.cdl tst_nccopy_w3.sh tst_nccopy_w4.sh ref_no_ncproperty.nc
+ref_nccopy_w.cdl tst_nccopy_w3.sh tst_nccopy_w4.sh ref_no_ncproperty.nc \
+test_unicode_directory.sh
+
 
 # The L512.bin file is file containing exactly 512 bytes each of value 0.
 # It is used for creating hdf5 files with varying offsets for testing.

ncdump/test_unicode_directory.sh

@@ -0,0 +1,33 @@
+#!/bin/sh
+#
+# Test to make sure ncdump works with a subdirectory which starts
+# with a unicode character.
+# See https://github.com/Unidata/netcdf-c/issues/1666 for more information.
+# Ward Fisher
+
+if test "x$srcdir" = x ; then srcdir=`pwd`; fi
+. ../test_common.sh
+
+
+ERR() {
+    RES=$?
+    if [ $RES -ne 0 ]; then
+        echo "Error found: $RES"
+        exit $RES
+    fi
+}
+
+UNISTRING=$(echo '\xe6\xb5\xb7')
+
+echo ""
+echo "Creating Unicode String Directory ${UNISTRING}"
+mkdir -p ${UNISTRING}; ERR
+
+echo "*** Generating binary file ${UNISTRING}/tst_utf.nc..."
+${NCGEN} -b -o "${UNISTRING}/tst_utf.nc" "${srcdir}/ref_tst_utf8.cdl"; ERR
+echo "*** Accessing binary file ${UNISTRING}/tst_utf.nc..."
+${NCDUMP} -h "${UNISTRING}/tst_utf.nc"; ERR
+
+echo "Test Passed. Cleaning up."
+rm "${UNISTRING}/tst_utf.nc"; ERR
+rmdir "${UNISTRING}"; ERR