Merge pull request #1264 from Unidata/pr-aggregation.wif

YAA (Yet Another Aggregation)

configure.ac

@@ -550,13 +550,12 @@ if test "x$enable_large_file_tests" = xyes; then
 fi
 
 # Does the user want to run benchmarks?
-AC_MSG_CHECKING([whether benchmarks should be run (experimental)])
+AC_MSG_CHECKING([whether benchmarks should be run])
 AC_ARG_ENABLE([benchmarks],
               [AS_HELP_STRING([--enable-benchmarks],
-                              [Run benchmarks. This is an experimental feature. You must fetch
-                              sample data files from the Unidata ftp site to use these benchmarks.
-                              The benchmarks are a bunch of extra tests, which are timed. We use these
-			      tests to check netCDF performance.])])
+                              [Run benchmarks. This will cause sample data files from the Unidata ftp
+                              site to be fetched. The benchmarks are a bunch of extra tests, which
+                              are timed. We use these tests to check netCDF performance.])])
 test "x$enable_benchmarks" = xyes || enable_benchmarks=no
 AC_MSG_RESULT($enable_benchmarks)
 AM_CONDITIONAL(BUILD_BENCHMARKS, [test x$enable_benchmarks = xyes])

debug/README.txt

@@ -1,5 +1,6 @@
 This directory contains various scripts for debugging by Dennis
 Heimbigner @ Unidata.
+DO NOT DELETE.
 USE AT YOUR OWN PERIL.
 
 

debug/cf.cmake

@@ -97,7 +97,7 @@ NCLIB="${NCLIB}/build/liblib"
 #G="-GUnix Makefiles"
 #T="--trace-expand"
 cmake "${G}" $FLAGS ..
-if test "x$NOBUILD" == x ; then
+if test "x$NOBUILD" = x ; then
 make all
 make test
 fi

include/hdf5internal.h

@@ -105,7 +105,6 @@ int hdf5_set_log_level();
 int rec_detach_scales(NC_GRP_INFO_T *grp, int dimid, hid_t dimscaleid);
 int rec_reattach_scales(NC_GRP_INFO_T *grp, int dimid, hid_t dimscaleid);
 int delete_existing_dimscale_dataset(NC_GRP_INFO_T *grp, int dimid, NC_DIM_INFO_T *dim);
-int nc4_rec_match_dimscales(NC_GRP_INFO_T *grp);
 
 /* Write metadata. */
 int nc4_rec_write_metadata(NC_GRP_INFO_T *grp, nc_bool_t bad_coord_order);
@@ -137,7 +136,6 @@ extern hid_t NC4_image_init(NC_FILE_INFO_T* h5);
 extern void NC4_image_finalize(void*);
 
 /* These functions are internal to the libhdf5 directory. */
-int nc4_detect_preserve_dimids(NC_GRP_INFO_T *grp, nc_bool_t *bad_coord_orderp);
 int nc4_get_fill_value(NC_FILE_INFO_T *h5, NC_VAR_INFO_T *var, void **fillp);
 
 

include/nc4internal.h

@@ -141,10 +141,10 @@ typedef struct NC_VAR_INFO
 {
    NC_OBJ hdr;
    char *hdf5_name; /* used if different from name */
-   struct NC_GRP_INFO* container; /* containing group */
+   struct NC_GRP_INFO *container; /* containing group */
    size_t ndims;
    int *dimids;
-   NC_DIM_INFO_T** dim;
+   NC_DIM_INFO_T **dim;
    nc_bool_t is_new_var;        /* True if variable is newly created */
    nc_bool_t was_coord_var;     /* True if variable was a coordinate var, but either the dim or var has been renamed */
    nc_bool_t became_coord_var;  /* True if variable _became_ a coordinate var, because either the dim or var has been renamed */
@@ -154,7 +154,7 @@ typedef struct NC_VAR_INFO
    nc_bool_t written_to;        /* True if variable has data written to it */
    struct NC_TYPE_INFO *type_info;
    int atts_not_read;           /* If true, the atts have not yet been read. */
-   NCindex* att; 		/* NCindex<NC_ATT_INFO_T*> */
+   NCindex *att; 		/* NCindex<NC_ATT_INFO_T*> */
    nc_bool_t no_fill;           /* True if no fill value is defined for var */
    void *fill_value;
    size_t *chunksizes;
@@ -172,7 +172,7 @@ typedef struct NC_VAR_INFO
    /* Stuff for arbitrary filters */
    unsigned int filterid;
    size_t nparams;
-   unsigned int* params;
+   unsigned int *params;
 } NC_VAR_INFO_T;
 
 typedef struct NC_FIELD_INFO

include/nc_provenance.h

@@ -6,8 +6,6 @@
  * info and/or displaying provenance info.
  *
  * @author Dennis Heimbigner, Ward Fisher
- */
-/**
 
 It has come to pass that we can't guarantee that this information is
 contained only within netcdf4 files.  As a result, we need

libhdf5/hdf5file.c

@@ -75,6 +75,84 @@ NC_findreserved(const char* name)
    return NULL;
 }
 
+/**
+ * @internal Recursively determine if there is a mismatch between
+ * order of coordinate creation and associated dimensions in this
+ * group or any subgroups, to find out if we have to handle that
+ * situation.  Also check if there are any multidimensional coordinate
+ * variables defined, which require the same treatment to fix a
+ * potential bug when such variables occur in subgroups.
+ *
+ * @param grp Pointer to group info struct.
+ * @param bad_coord_orderp Pointer that gets 1 if there is a bad
+ * coordinate order.
+ *
+ * @returns NC_NOERR No error.
+ * @returns NC_EHDFERR HDF5 returned an error.
+ * @author Ed Hartnett
+ */
+static int
+detect_preserve_dimids(NC_GRP_INFO_T *grp, nc_bool_t *bad_coord_orderp)
+{
+   NC_VAR_INFO_T *var;
+   NC_GRP_INFO_T *child_grp;
+   int last_dimid = -1;
+   int retval;
+   int i;
+
+   /* Iterate over variables in this group */
+   for (i=0; i < ncindexsize(grp->vars); i++)
+   {
+      var = (NC_VAR_INFO_T*)ncindexith(grp->vars,i);
+      if (var == NULL) continue;
+      /* Only matters for dimension scale variables, with non-scalar dimensionality */
+      if (var->dimscale && var->ndims)
+      {
+         /* If the user writes coord vars in a different order then he
+          * defined their dimensions, then, when the file is reopened, the
+          * order of the dimids will change to match the order of the coord
+          * vars. Detect if this is about to happen. */
+         if (var->dimids[0] < last_dimid)
+         {
+            LOG((5, "%s: %s is out of order coord var", __func__, var->hdr.name));
+            *bad_coord_orderp = NC_TRUE;
+            return NC_NOERR;
+         }
+         last_dimid = var->dimids[0];
+
+         /* If there are multidimensional coordinate variables defined, then
+          * it's also necessary to preserve dimension IDs when the file is
+          * reopened ... */
+         if (var->ndims > 1)
+         {
+            LOG((5, "%s: %s is multidimensional coord var", __func__, var->hdr.name));
+            *bad_coord_orderp = NC_TRUE;
+            return NC_NOERR;
+         }
+
+         /* Did the user define a dimension, end define mode, reenter define
+          * mode, and then define a coordinate variable for that dimension?
+          * If so, dimensions will be out of order. */
+         if (var->is_new_var || var->became_coord_var)
+         {
+            LOG((5, "%s: coord var defined after enddef/redef", __func__));
+            *bad_coord_orderp = NC_TRUE;
+            return NC_NOERR;
+         }
+      }
+   }
+
+   /* If there are any child groups, check them also for this condition. */
+   for (i = 0; i < ncindexsize(grp->children); i++)
+   {
+      if (!(child_grp = (NC_GRP_INFO_T *)ncindexith(grp->children, i)))
+         continue;
+      if ((retval = detect_preserve_dimids(child_grp, bad_coord_orderp)))
+         return retval;
+   }
+   return NC_NOERR;
+}
+
 /**
  * @internal This function will write all changed metadata and flush
  * HDF5 file to disk.
@@ -127,7 +205,7 @@ sync_netcdf4_file(NC_FILE_INFO_T *h5)
       /* Check to see if the coordinate order is messed up. If
        * detected, propagate to all groups to consistently store
        * dimids. */
-      if ((retval = nc4_detect_preserve_dimids(h5->root_grp, &bad_coord_order)))
+      if ((retval = detect_preserve_dimids(h5->root_grp, &bad_coord_order)))
          return retval;
 
       /* Write all the metadata. */