first stab at re-adding nc_def_var_szip()

include/nc4internal.h

@@ -210,7 +210,7 @@ typedef struct NC_VAR_INFO
     nc_bool_t dimscale;          /**< True if var is a dimscale */
     nc_bool_t *dimscale_attached;  /**< Array of flags that are true if dimscale is attached for that dim index */
     nc_bool_t deflate;           /**< True if var has deflate filter applied */
-    int deflate_level;
+    int deflate_level;           /**< If deflate is true, this is the deflate level, between 1 and 9. */
     nc_bool_t shuffle;           /**< True if var has shuffle filter applied */
     nc_bool_t fletcher32;        /**< True if var has fletcher32 filter applied */
     size_t chunk_cache_size, chunk_cache_nelems;

include/netcdf.h

@@ -317,6 +317,9 @@ there. */
 #define NC_MIN_DEFLATE_LEVEL 0 /**< Minimum deflate level. */
 #define NC_MAX_DEFLATE_LEVEL 9 /**< Maximum deflate level. */
 
+#define NC_SZIP_EC 4  /**< Selects entropy coding method for szip. */
+#define NC_SZIP_NN 32 /**< Selects nearest neighbor coding method for szip. */
+
 /** The netcdf version 3 functions all return integer error status.
  * These are the possible values, in addition to certain values from
  * the system errno.h.
@@ -850,6 +853,10 @@ EXTERNL int
 nc_inq_var_deflate(int ncid, int varid, int *shufflep,
                    int *deflatep, int *deflate_levelp);
 
+/* Set szip compression for a variable. */
+EXTERNL int
+nc_def_var_szip(int ncid, int varid, int options_mask, int pixels_per_block);
+
 /* Find out szip settings of a var. */
 EXTERNL int
 nc_inq_var_szip(int ncid, int varid, int *options_maskp, int *pixels_per_blockp);

libhdf5/hdf5var.c

@@ -25,6 +25,12 @@
 /** Number of bytes in 64 KB. */
 #define SIXTY_FOUR_KB (65536)
 
+/** Number of parameters needed when turning on szip filter. */
+#define NUM_SZIP_PARAM 2
+
+/** The HDF5 ID for the szip filter. */
+#define HDF5_FILTER_SZIP 4
+
 #ifdef LOGGING
 /**
  * Report the chunksizes selected for a variable.
@@ -661,8 +667,7 @@ nc_def_var_extra(int ncid, int varid, int *shuffle, int *deflate,
         return NC_ENOTVAR;
     assert(var && var->hdr.id == varid);
 
-    /* Can't turn on parallel and deflate/fletcher32/szip/shuffle
-     * before HDF5 1.10.2. */
+    /* Can't turn on parallel and any filters before HDF5 1.10.2. */
 #ifndef HDF5_SUPPORTS_PAR_FILTERS
     if (h5->parallel == NC_TRUE)
         if (deflate || fletcher32 || shuffle)
@@ -872,8 +877,8 @@ nc_def_var_extra(int ncid, int varid, int *shuffle, int *deflate,
 }
 
 /**
- * @internal Set compression settings on a variable. This is called by
- * nc_def_var_deflate().
+ * @internal Set zlib compression settings on a variable. This is
+ * called by nc_def_var_deflate().
  *
  * @param ncid File ID.
  * @param varid Variable ID.
@@ -900,6 +905,104 @@ NC4_def_var_deflate(int ncid, int varid, int shuffle, int deflate,
                             &deflate_level, NULL, NULL, NULL, NULL, NULL, NULL);
 }
 
+/**
+ * Set szip compression settings on a variable. Szip is an
+ * implementation of the extended-Rice lossless compression
+ * algorithm. Szip is reported to provide fast and effective
+ * compression.
+ *
+ * SZIP compression cannot be applied to variables with any
+ * user-defined type.
+ *
+ * @note The options_mask parameter may be either NC_SZIP_EC (entropy
+ * coding) or NC_SZIP_NN (nearest neighbor):
+ * * The entropy coding method is best suited for data that has been
+ * processed. The EC method works best for small numbers.
+ * * The nearest neighbor coding method preprocesses the data then the
+ * applies EC method as above.
+ *
+ * @param ncid File ID.
+ * @param varid Variable ID.
+ * @param options_mask The options mask. Can be NC_SZIP_EC or
+ * NC_SZIP_NN.
+ * @param pixels_per_block Pixels per block. Must be even and not
+ * greater than 32, with typical values being 8, 10, 16, or 32. This
+ * parameter affects compression ratio; the more pixel values vary,
+ * the smaller this number should be to achieve better performance. If
+ * pixels_per_block is bigger than the total number of elements in a
+ * dataset chunk, H5Pset_szip will succeed but the subsequent call to
+ * H5Dcreate will fail; the conflict can be detected only when the
+ * property list is used.
+ *
+ * @returns ::NC_NOERR No error.
+ * @returns ::NC_ENOTBUILT This HDF5 install was not built with szip.
+ * @returns ::NC_EBADID Bad ncid.
+ * @returns ::NC_ENOTVAR Invalid variable ID.
+ * @returns ::NC_ENOTNC4 Attempting netcdf-4 operation on file that is
+ * not netCDF-4/HDF5.
+ * @returns ::NC_ELATEDEF Too late to change settings for this variable.
+ * @returns ::NC_ENOTINDEFINE Not in define mode.
+ * @returns ::NC_EINVAL Invalid input
+ * @author Ed Hartnett
+ */
+int
+nc_def_var_szip(int ncid, int varid, int options_mask, int pixels_per_block)
+{
+    NC_GRP_INFO_T *grp;
+    NC_FILE_INFO_T *h5;
+    NC_VAR_INFO_T *var;
+    unsigned int szip_params[NUM_SZIP_PARAM]; /* [0]=options_mask [1]=pixels_per_block */
+    int built = 0;
+    int ret;
+
+    LOG((2, "%s: ncid 0x%x varid %d", __func__, ncid, varid));
+
+    /* If HDF5 was not built with szip, then return error. */
+#ifdef HAVE_H5Z_SZIP
+    built = 1;
+#endif /* HAVE_H5Z_SZIP */
+    if (!built)
+        return NC_ENOTBUILT;
+
+    /* Find info for this file and group, and set pointer to each. */
+    if ((ret = nc4_find_nc_grp_h5(ncid, NULL, &grp, &h5)))
+        return ret;
+    assert(grp && h5);
+
+    /* Trying to write to a read-only file? No way, Jose! */
+    if (h5->no_write)
+        return NC_EPERM;
+
+    /* Can't turn on parallel and szip before HDF5 1.10.2. */
+#ifdef USE_PARALLEL
+#ifndef HDF5_SUPPORTS_PAR_FILTERS
+    if (h5->parallel == NC_TRUE)
+        return NC_EINVAL;
+#endif /* HDF5_SUPPORTS_PAR_FILTERS */
+#endif /* USE_PARALLEL */
+
+    /* Find the var. */
+    if (!(var = (NC_VAR_INFO_T *)ncindexith(grp->vars, varid)))
+        return NC_ENOTVAR;
+    assert(var && var->hdr.id == varid);
+
+#ifdef USE_PARALLEL
+    /* Switch to collective access. HDF5 requires collevtive access
+     * for filter use with parallel I/O. */
+    if (h5->parallel)
+        var->parallel_access = NC_COLLECTIVE;
+#endif /* USE_PARALLEL */
+
+    /* This will cause H5Pset_szip to be called when the var is created. */
+    szip_params[0] = options_mask == NC_SZIP_EC ? H5_SZIP_EC_OPTION_MASK:
+        H5_SZIP_NN_OPTION_MASK;
+    szip_params[1] = pixels_per_block;
+    if ((ret = nc_def_var_filter(ncid, varid, HDF5_FILTER_SZIP, 2, szip_params)))
+        return ret;
+
+    return NC_NOERR;
+}
+
 /**
  * @internal Set checksum on a variable. This is called by
  * nc_def_var_fletcher32().