netcdf-c/include/netcdf_json.h

1251 lines
32 KiB
C
Raw Normal View History

Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
/* Copyright 2018, UCAR/Unidata.
See the COPYRIGHT file for more information.
*/
Add support for Zarr string type to NCZarr * re: https://github.com/Unidata/netcdf-c/pull/2278 * re: https://github.com/Unidata/netcdf-c/issues/2485 * re: https://github.com/Unidata/netcdf-c/issues/2474 This PR subsumes PR https://github.com/Unidata/netcdf-c/pull/2278. Actually is a bit an omnibus covering several issues. ## PR https://github.com/Unidata/netcdf-c/pull/2278 Add support for the Zarr string type. Zarr strings are restricted currently to be of fixed size. The primary issue to be addressed is to provide a way for user to specify the size of the fixed length strings. This is handled by providing the following new attributes special: 1. **_nczarr_default_maxstrlen** — This is an attribute of the root group. It specifies the default maximum string length for string types. If not specified, then it has the value of 64 characters. 2. **_nczarr_maxstrlen** — This is a per-variable attribute. It specifies the maximum string length for the string type associated with the variable. If not specified, then it is assigned the value of **_nczarr_default_maxstrlen**. This PR also requires some hacking to handle the existing netcdf-c NC_CHAR type, which does not exist in zarr. The goal was to choose numpy types for both the netcdf-c NC_STRING type and the netcdf-c NC_CHAR type such that if a pure zarr implementation read them, it would still work and an NC_CHAR type would be handled by zarr as a string of length 1. For writing variables and NCZarr attributes, the type mapping is as follows: * "|S1" for NC_CHAR. * ">S1" for NC_STRING && MAXSTRLEN==1 * ">Sn" for NC_STRING && MAXSTRLEN==n Note that it is a bit of a hack to use endianness, but it should be ok since for string/char, the endianness has no meaning. For reading attributes with pure zarr (i.e. with no nczarr atribute types defined), they will always be interpreted as of type NC_CHAR. ## Issue: https://github.com/Unidata/netcdf-c/issues/2474 This PR partly fixes this issue because it provided more comprehensive support for Zarr attributes that are JSON valued expressions. This PR still does not address the problem in that issue where the _ARRAY_DIMENSION attribute is incorrectly set. Than can only be fixed by the creator of the datasets. ## Issue: https://github.com/Unidata/netcdf-c/issues/2485 This PR also fixes the scalar failure shown in this issue. It generally cleans up scalar handling. It also adds a note to the documentation describing that NCZarr supports scalars while Zarr does not and also how scalar interoperability is achieved. ## Misc. Other Changes 1. Convert the nczarr special attributes and keys to be all lower case. So "_NCZARR_ATTR" now used "_nczarr_attr. Support back compatibility for the upper case names. 2. Cleanup my too-clever-by-half handling of scalars in libnczarr.
2022-08-28 10:21:13 +08:00
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
#ifndef NETCDF_JSON_H
#define NETCDF_JSON_H 1
Add support for Zarr string type to NCZarr * re: https://github.com/Unidata/netcdf-c/pull/2278 * re: https://github.com/Unidata/netcdf-c/issues/2485 * re: https://github.com/Unidata/netcdf-c/issues/2474 This PR subsumes PR https://github.com/Unidata/netcdf-c/pull/2278. Actually is a bit an omnibus covering several issues. ## PR https://github.com/Unidata/netcdf-c/pull/2278 Add support for the Zarr string type. Zarr strings are restricted currently to be of fixed size. The primary issue to be addressed is to provide a way for user to specify the size of the fixed length strings. This is handled by providing the following new attributes special: 1. **_nczarr_default_maxstrlen** — This is an attribute of the root group. It specifies the default maximum string length for string types. If not specified, then it has the value of 64 characters. 2. **_nczarr_maxstrlen** — This is a per-variable attribute. It specifies the maximum string length for the string type associated with the variable. If not specified, then it is assigned the value of **_nczarr_default_maxstrlen**. This PR also requires some hacking to handle the existing netcdf-c NC_CHAR type, which does not exist in zarr. The goal was to choose numpy types for both the netcdf-c NC_STRING type and the netcdf-c NC_CHAR type such that if a pure zarr implementation read them, it would still work and an NC_CHAR type would be handled by zarr as a string of length 1. For writing variables and NCZarr attributes, the type mapping is as follows: * "|S1" for NC_CHAR. * ">S1" for NC_STRING && MAXSTRLEN==1 * ">Sn" for NC_STRING && MAXSTRLEN==n Note that it is a bit of a hack to use endianness, but it should be ok since for string/char, the endianness has no meaning. For reading attributes with pure zarr (i.e. with no nczarr atribute types defined), they will always be interpreted as of type NC_CHAR. ## Issue: https://github.com/Unidata/netcdf-c/issues/2474 This PR partly fixes this issue because it provided more comprehensive support for Zarr attributes that are JSON valued expressions. This PR still does not address the problem in that issue where the _ARRAY_DIMENSION attribute is incorrectly set. Than can only be fixed by the creator of the datasets. ## Issue: https://github.com/Unidata/netcdf-c/issues/2485 This PR also fixes the scalar failure shown in this issue. It generally cleans up scalar handling. It also adds a note to the documentation describing that NCZarr supports scalars while Zarr does not and also how scalar interoperability is achieved. ## Misc. Other Changes 1. Convert the nczarr special attributes and keys to be all lower case. So "_NCZARR_ATTR" now used "_nczarr_attr. Support back compatibility for the upper case names. 2. Cleanup my too-clever-by-half handling of scalars in libnczarr.
2022-08-28 10:21:13 +08:00
/*
WARNING:
If you modify this file,
then you need to got to
the include/ directory
and do the command:
make makepluginjson
*/
/* Inside libnetcdf and for plugins, export the json symbols */
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
#ifndef DLLEXPORT
#ifdef _WIN32
#define DLLEXPORT __declspec(dllexport)
#else
#define DLLEXPORT
#endif
#endif
/* Override for plugins */
#ifdef NETCDF_JSON_H
#define OPTEXPORT static
#else
#define OPTEXPORT DLLEXPORT
#endif /*NETCDF_JSON_H*/
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
/**************************************************/
/* Json object sorts (note use of term sort rather than e.g. type or discriminant) */
#define NCJ_UNDEF 0
#define NCJ_STRING 1
#define NCJ_INT 2
#define NCJ_DOUBLE 3
#define NCJ_BOOLEAN 4
#define NCJ_DICT 5
#define NCJ_ARRAY 6
#define NCJ_NULL 7
#define NCJ_NSORTS 8
/* No flags are currently defined, but the argument is a placeholder */
/* Define a struct to store primitive values as unquoted
strings. The sort will provide more info. Do not bother with
a union since the amount of saved space is minimal.
*/
typedef struct NCjson {
int sort; /* of this object */
char* string; /* sort != DICT|ARRAY */
struct NCjlist {
int len;
struct NCjson** contents;
} list; /* sort == DICT|ARRAY */
} NCjson;
/* Structure to hold result of convertinf one json sort to value of another type;
don't use union so we can know when to reclaim sval
*/
struct NCJconst {int bval; long long ival; double dval; char* sval;};
Add support for Zarr string type to NCZarr * re: https://github.com/Unidata/netcdf-c/pull/2278 * re: https://github.com/Unidata/netcdf-c/issues/2485 * re: https://github.com/Unidata/netcdf-c/issues/2474 This PR subsumes PR https://github.com/Unidata/netcdf-c/pull/2278. Actually is a bit an omnibus covering several issues. ## PR https://github.com/Unidata/netcdf-c/pull/2278 Add support for the Zarr string type. Zarr strings are restricted currently to be of fixed size. The primary issue to be addressed is to provide a way for user to specify the size of the fixed length strings. This is handled by providing the following new attributes special: 1. **_nczarr_default_maxstrlen** — This is an attribute of the root group. It specifies the default maximum string length for string types. If not specified, then it has the value of 64 characters. 2. **_nczarr_maxstrlen** — This is a per-variable attribute. It specifies the maximum string length for the string type associated with the variable. If not specified, then it is assigned the value of **_nczarr_default_maxstrlen**. This PR also requires some hacking to handle the existing netcdf-c NC_CHAR type, which does not exist in zarr. The goal was to choose numpy types for both the netcdf-c NC_STRING type and the netcdf-c NC_CHAR type such that if a pure zarr implementation read them, it would still work and an NC_CHAR type would be handled by zarr as a string of length 1. For writing variables and NCZarr attributes, the type mapping is as follows: * "|S1" for NC_CHAR. * ">S1" for NC_STRING && MAXSTRLEN==1 * ">Sn" for NC_STRING && MAXSTRLEN==n Note that it is a bit of a hack to use endianness, but it should be ok since for string/char, the endianness has no meaning. For reading attributes with pure zarr (i.e. with no nczarr atribute types defined), they will always be interpreted as of type NC_CHAR. ## Issue: https://github.com/Unidata/netcdf-c/issues/2474 This PR partly fixes this issue because it provided more comprehensive support for Zarr attributes that are JSON valued expressions. This PR still does not address the problem in that issue where the _ARRAY_DIMENSION attribute is incorrectly set. Than can only be fixed by the creator of the datasets. ## Issue: https://github.com/Unidata/netcdf-c/issues/2485 This PR also fixes the scalar failure shown in this issue. It generally cleans up scalar handling. It also adds a note to the documentation describing that NCZarr supports scalars while Zarr does not and also how scalar interoperability is achieved. ## Misc. Other Changes 1. Convert the nczarr special attributes and keys to be all lower case. So "_NCZARR_ATTR" now used "_nczarr_attr. Support back compatibility for the upper case names. 2. Cleanup my too-clever-by-half handling of scalars in libnczarr.
2022-08-28 10:21:13 +08:00
#define NCJconst_empty {0,0,0.0,NULL}
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
/**************************************************/
/* Extended API */
/* Return 0 if ok else -1 */
#if defined(__cplusplus)
extern "C" {
#endif
/* Parse a string to NCjson*/
OPTEXPORT int NCJparse(const char* text, unsigned flags, NCjson** jsonp);
/* Parse a counted string to NCjson*/
OPTEXPORT int NCJparsen(size_t len, const char* text, unsigned flags, NCjson** jsonp);
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
/* Reclaim a JSON tree */
OPTEXPORT void NCJreclaim(NCjson* json);
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
/* Create a new JSON node of a given sort */
OPTEXPORT int NCJnew(int sort, NCjson** objectp);
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
/* Create new json object with given string content */
OPTEXPORT int NCJnewstring(int sort, const char* value, NCjson** jsonp);
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
/* Create new json object with given counted string content */
OPTEXPORT int NCJnewstringn(int sort, size_t len, const char* value, NCjson** jsonp);
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
/* Get dict key value by name */
OPTEXPORT int NCJdictget(const NCjson* dict, const char* key, NCjson** valuep);
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
/* Convert one json sort to value of another type; don't use union so we can know when to reclaim sval */
OPTEXPORT int NCJcvt(const NCjson* value, int outsort, struct NCJconst* output);
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
/* Insert an atomic value to an array or dict object. */
OPTEXPORT int NCJaddstring(NCjson* json, int sort, const char* s);
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
/* Append value to an array or dict object. */
OPTEXPORT int NCJappend(NCjson* object, NCjson* value);
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
/* Insert key-value pair into a dict object. key will be copied */
OPTEXPORT int NCJinsert(NCjson* object, char* key, NCjson* value);
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
/* Unparser to convert NCjson object to text in buffer */
OPTEXPORT int NCJunparse(const NCjson* json, unsigned flags, char** textp);
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
/* Deep clone a json object */
OPTEXPORT int NCJclone(const NCjson* json, NCjson** clonep);
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
#ifndef NETCDF_JSON_H
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
/* dump NCjson* object to output file */
OPTEXPORT void NCJdump(const NCjson* json, unsigned flags, FILE*);
Add support for Zarr string type to NCZarr * re: https://github.com/Unidata/netcdf-c/pull/2278 * re: https://github.com/Unidata/netcdf-c/issues/2485 * re: https://github.com/Unidata/netcdf-c/issues/2474 This PR subsumes PR https://github.com/Unidata/netcdf-c/pull/2278. Actually is a bit an omnibus covering several issues. ## PR https://github.com/Unidata/netcdf-c/pull/2278 Add support for the Zarr string type. Zarr strings are restricted currently to be of fixed size. The primary issue to be addressed is to provide a way for user to specify the size of the fixed length strings. This is handled by providing the following new attributes special: 1. **_nczarr_default_maxstrlen** — This is an attribute of the root group. It specifies the default maximum string length for string types. If not specified, then it has the value of 64 characters. 2. **_nczarr_maxstrlen** — This is a per-variable attribute. It specifies the maximum string length for the string type associated with the variable. If not specified, then it is assigned the value of **_nczarr_default_maxstrlen**. This PR also requires some hacking to handle the existing netcdf-c NC_CHAR type, which does not exist in zarr. The goal was to choose numpy types for both the netcdf-c NC_STRING type and the netcdf-c NC_CHAR type such that if a pure zarr implementation read them, it would still work and an NC_CHAR type would be handled by zarr as a string of length 1. For writing variables and NCZarr attributes, the type mapping is as follows: * "|S1" for NC_CHAR. * ">S1" for NC_STRING && MAXSTRLEN==1 * ">Sn" for NC_STRING && MAXSTRLEN==n Note that it is a bit of a hack to use endianness, but it should be ok since for string/char, the endianness has no meaning. For reading attributes with pure zarr (i.e. with no nczarr atribute types defined), they will always be interpreted as of type NC_CHAR. ## Issue: https://github.com/Unidata/netcdf-c/issues/2474 This PR partly fixes this issue because it provided more comprehensive support for Zarr attributes that are JSON valued expressions. This PR still does not address the problem in that issue where the _ARRAY_DIMENSION attribute is incorrectly set. Than can only be fixed by the creator of the datasets. ## Issue: https://github.com/Unidata/netcdf-c/issues/2485 This PR also fixes the scalar failure shown in this issue. It generally cleans up scalar handling. It also adds a note to the documentation describing that NCZarr supports scalars while Zarr does not and also how scalar interoperability is achieved. ## Misc. Other Changes 1. Convert the nczarr special attributes and keys to be all lower case. So "_NCZARR_ATTR" now used "_nczarr_attr. Support back compatibility for the upper case names. 2. Cleanup my too-clever-by-half handling of scalars in libnczarr.
2022-08-28 10:21:13 +08:00
/* convert NCjson* object to output string */
OPTEXPORT const char* NCJtotext(const NCjson* json);
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
#endif
#if defined(__cplusplus)
}
#endif
/* Getters */
#define NCJsort(x) ((x)->sort)
#define NCJstring(x) ((x)->string)
#define NCJlength(x) ((x)==NULL ? 0 : (x)->list.len)
#define NCJcontents(x) ((x)->list.contents)
#define NCJith(x,i) ((x)->list.contents[i])
/* Setters */
#define NCJsetsort(x,s) (x)->sort=(s)
#define NCJsetstring(x,y) (x)->string=(y)
#define NCJsetcontents(x,c) (x)->list.contents=(c)
#define NCJsetlength(x,l) (x)->list.len=(l)
/* Misc */
#define NCJisatomic(j) ((j)->sort != NCJ_ARRAY && (j)->sort != NCJ_DICT && (j)->sort != NCJ_NULL && (j)->sort != NCJ_UNDEF)
/**************************************************/
#endif /*NETCDF_JSON_H*/
/* Copyright 2018, UCAR/Unidata.
See the COPYRIGHT file for more information.
*/
/*
TODO: make utf8 safe
*/
Add support for Zarr string type to NCZarr * re: https://github.com/Unidata/netcdf-c/pull/2278 * re: https://github.com/Unidata/netcdf-c/issues/2485 * re: https://github.com/Unidata/netcdf-c/issues/2474 This PR subsumes PR https://github.com/Unidata/netcdf-c/pull/2278. Actually is a bit an omnibus covering several issues. ## PR https://github.com/Unidata/netcdf-c/pull/2278 Add support for the Zarr string type. Zarr strings are restricted currently to be of fixed size. The primary issue to be addressed is to provide a way for user to specify the size of the fixed length strings. This is handled by providing the following new attributes special: 1. **_nczarr_default_maxstrlen** — This is an attribute of the root group. It specifies the default maximum string length for string types. If not specified, then it has the value of 64 characters. 2. **_nczarr_maxstrlen** — This is a per-variable attribute. It specifies the maximum string length for the string type associated with the variable. If not specified, then it is assigned the value of **_nczarr_default_maxstrlen**. This PR also requires some hacking to handle the existing netcdf-c NC_CHAR type, which does not exist in zarr. The goal was to choose numpy types for both the netcdf-c NC_STRING type and the netcdf-c NC_CHAR type such that if a pure zarr implementation read them, it would still work and an NC_CHAR type would be handled by zarr as a string of length 1. For writing variables and NCZarr attributes, the type mapping is as follows: * "|S1" for NC_CHAR. * ">S1" for NC_STRING && MAXSTRLEN==1 * ">Sn" for NC_STRING && MAXSTRLEN==n Note that it is a bit of a hack to use endianness, but it should be ok since for string/char, the endianness has no meaning. For reading attributes with pure zarr (i.e. with no nczarr atribute types defined), they will always be interpreted as of type NC_CHAR. ## Issue: https://github.com/Unidata/netcdf-c/issues/2474 This PR partly fixes this issue because it provided more comprehensive support for Zarr attributes that are JSON valued expressions. This PR still does not address the problem in that issue where the _ARRAY_DIMENSION attribute is incorrectly set. Than can only be fixed by the creator of the datasets. ## Issue: https://github.com/Unidata/netcdf-c/issues/2485 This PR also fixes the scalar failure shown in this issue. It generally cleans up scalar handling. It also adds a note to the documentation describing that NCZarr supports scalars while Zarr does not and also how scalar interoperability is achieved. ## Misc. Other Changes 1. Convert the nczarr special attributes and keys to be all lower case. So "_NCZARR_ATTR" now used "_nczarr_attr. Support back compatibility for the upper case names. 2. Cleanup my too-clever-by-half handling of scalars in libnczarr.
2022-08-28 10:21:13 +08:00
/*
WARNING:
If you modify this file,
then you need to got to
the include/ directory
and do the command:
make makenetcdfjson
*/
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <assert.h>
#undef NCJDEBUG
#ifdef NCJDEBUG
Mitigate S3 test interference + Unlimited Dimensions in NCZarr This PR started as an attempt to add unlimited dimensions to NCZarr. It did that, but this exposed significant problems with test interference. So this PR is mostly about fixing -- well mitigating anyway -- test interference. The problem of test interference is now documented in the document docs/internal.md. The solutions implemented here are also describe in that document. The solution is somewhat fragile but multiple cleanup mechanisms are provided. Note that this feature requires that the AWS command line utility must be installed. ## Unlimited Dimensions. The existing NCZarr extensions to Zarr are modified to support unlimited dimensions. NCzarr extends the Zarr meta-data for the ".zgroup" object to include netcdf-4 model extensions. This information is stored in ".zgroup" as dictionary named "_nczarr_group". Inside "_nczarr_group", there is a key named "dims" that stores information about netcdf-4 named dimensions. The value of "dims" is a dictionary whose keys are the named dimensions. The value associated with each dimension name has one of two forms Form 1 is a special case of form 2, and is kept for backward compatibility. Whenever a new file is written, it uses format 1 if possible, otherwise format 2. * Form 1: An integer representing the size of the dimension, which is used for simple named dimensions. * Form 2: A dictionary with the following keys and values" - "size" with an integer value representing the (current) size of the dimension. - "unlimited" with a value of either "1" or "0" to indicate if this dimension is an unlimited dimension. For Unlimited dimensions, the size is initially zero, and as variables extend the length of that dimension, the size value for the dimension increases. That dimension size is shared by all arrays referencing that dimension, so if one array extends an unlimited dimension, it is implicitly extended for all other arrays that reference that dimension. This is the standard semantics for unlimited dimensions. Adding unlimited dimensions required a number of other changes to the NCZarr code-base. These included the following. * Did a partial refactor of the slice handling code in zwalk.c to clean it up. * Added a number of tests for unlimited dimensions derived from the same test in nc_test4. * Added several NCZarr specific unlimited tests; more are needed. * Add test of endianness. ## Misc. Other Changes * Modify libdispatch/ncs3sdk_aws.cpp to optionally support use of the AWS Transfer Utility mechanism. This is controlled by the ```#define TRANSFER```` command in that file. It defaults to being disabled. * Parameterize both the standard Unidata S3 bucket (S3TESTBUCKET) and the netcdf-c test data prefix (S3TESTSUBTREE). * Fixed an obscure memory leak in ncdump. * Removed some obsolete unit testing code and test cases. * Uncovered a bug in the netcdf-c handling of big-endian floats and doubles. Have not fixed yet. See tst_h5_endians.c. * Renamed some nczarr_tests testcases to avoid name conflicts with nc_test4. * Modify the semantics of zmap\#ncsmap_write to only allow total rewrite of objects. * Modify the semantics of zodom to properly handle stride > 1. * Add a truncate operation to the libnczarr zmap code.
2023-09-27 06:56:48 +08:00
/* Warning: do not evaluate err more than once */
#define NCJTHROW(err) ncjbreakpoint(err)
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
static int ncjbreakpoint(int err) {return err;}
#else
#define NCJTHROW(err) (err)
#endif
/**************************************************/
#define NCJ_OK 0
#define NCJ_ERR (-1)
#define NCJ_EOF -2
#define NCJ_LBRACKET '['
#define NCJ_RBRACKET ']'
#define NCJ_LBRACE '{'
#define NCJ_RBRACE '}'
#define NCJ_COLON ':'
#define NCJ_COMMA ','
#define NCJ_QUOTE '"'
#define NCJ_ESCAPE '\\'
#define NCJ_TAG_TRUE "true"
#define NCJ_TAG_FALSE "false"
#define NCJ_TAG_NULL "null"
/* JSON_WORD Subsumes Number also */
#define JSON_WORD "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789_$+-."
/**************************************************/
typedef struct NCJparser {
char* text;
char* pos;
size_t yylen; /* |yytext| */
char* yytext; /* string or word */
long long num;
int tf;
int status; /* NCJ_ERR|NCJ_OK */
} NCJparser;
typedef struct NCJbuf {
int len; /* |text|; does not include nul terminator */
char* text; /* NULL || nul terminated */
} NCJbuf;
/**************************************************/
#if defined(_WIN32) && !defined(__MINGW32__)
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
#define strdup _strdup
#define strcasecmp _stricmp
#else
#include <strings.h>
#endif
#ifndef nullfree
#define nullfree(x) {if(x)free(x);}
#endif
#ifndef nulldup
#define nulldup(x) ((x)?strdup(x):(x))
#endif
#ifdef NCJDEBUG
static char* tokenname(int token);
#endif
/**************************************************/
/* Forward */
static int NCJparseR(NCJparser* parser, NCjson**);
static int NCJparseArray(NCJparser* parser, struct NCjlist* array);
static int NCJparseDict(NCJparser* parser, struct NCjlist* dict);
static int testbool(const char* word);
static int testint(const char* word);
static int testdouble(const char* word);
static int testnull(const char* word);
static int NCJlex(NCJparser* parser);
static int NCJyytext(NCJparser*, char* start, size_t pdlen);
static void NCJreclaimArray(struct NCjlist*);
static void NCJreclaimDict(struct NCjlist*);
static int NCJunescape(NCJparser* parser);
static int unescape1(int c);
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
static int listappend(struct NCjlist* list, NCjson* element);
static int NCJcloneArray(const NCjson* array, NCjson** clonep);
static int NCJcloneDict(const NCjson* dict, NCjson** clonep);
static int NCJunparseR(const NCjson* json, NCJbuf* buf, unsigned flags);
static int bytesappendquoted(NCJbuf* buf, const char* s);
static int bytesappend(NCJbuf* buf, const char* s);
static int bytesappendc(NCJbuf* bufp, const char c);
Add support for Zarr string type to NCZarr * re: https://github.com/Unidata/netcdf-c/pull/2278 * re: https://github.com/Unidata/netcdf-c/issues/2485 * re: https://github.com/Unidata/netcdf-c/issues/2474 This PR subsumes PR https://github.com/Unidata/netcdf-c/pull/2278. Actually is a bit an omnibus covering several issues. ## PR https://github.com/Unidata/netcdf-c/pull/2278 Add support for the Zarr string type. Zarr strings are restricted currently to be of fixed size. The primary issue to be addressed is to provide a way for user to specify the size of the fixed length strings. This is handled by providing the following new attributes special: 1. **_nczarr_default_maxstrlen** &mdash; This is an attribute of the root group. It specifies the default maximum string length for string types. If not specified, then it has the value of 64 characters. 2. **_nczarr_maxstrlen** &mdash; This is a per-variable attribute. It specifies the maximum string length for the string type associated with the variable. If not specified, then it is assigned the value of **_nczarr_default_maxstrlen**. This PR also requires some hacking to handle the existing netcdf-c NC_CHAR type, which does not exist in zarr. The goal was to choose numpy types for both the netcdf-c NC_STRING type and the netcdf-c NC_CHAR type such that if a pure zarr implementation read them, it would still work and an NC_CHAR type would be handled by zarr as a string of length 1. For writing variables and NCZarr attributes, the type mapping is as follows: * "|S1" for NC_CHAR. * ">S1" for NC_STRING && MAXSTRLEN==1 * ">Sn" for NC_STRING && MAXSTRLEN==n Note that it is a bit of a hack to use endianness, but it should be ok since for string/char, the endianness has no meaning. For reading attributes with pure zarr (i.e. with no nczarr atribute types defined), they will always be interpreted as of type NC_CHAR. ## Issue: https://github.com/Unidata/netcdf-c/issues/2474 This PR partly fixes this issue because it provided more comprehensive support for Zarr attributes that are JSON valued expressions. This PR still does not address the problem in that issue where the _ARRAY_DIMENSION attribute is incorrectly set. Than can only be fixed by the creator of the datasets. ## Issue: https://github.com/Unidata/netcdf-c/issues/2485 This PR also fixes the scalar failure shown in this issue. It generally cleans up scalar handling. It also adds a note to the documentation describing that NCZarr supports scalars while Zarr does not and also how scalar interoperability is achieved. ## Misc. Other Changes 1. Convert the nczarr special attributes and keys to be all lower case. So "_NCZARR_ATTR" now used "_nczarr_attr. Support back compatibility for the upper case names. 2. Cleanup my too-clever-by-half handling of scalars in libnczarr.
2022-08-28 10:21:13 +08:00
/* Hide everything for plugins */
#ifdef NETCDF_JSON_H
#define OPTSTATIC static
static int NCJparsen(size_t len, const char* text, unsigned flags, NCjson** jsonp);
static int NCJnew(int sort, NCjson** objectp);
static int NCJnewstring(int sort, const char* value, NCjson** jsonp);
static int NCJnewstringn(int sort, size_t len, const char* value, NCjson** jsonp);
static int NCJclone(const NCjson* json, NCjson** clonep);
static int NCJaddstring(NCjson* json, int sort, const char* s);
static int NCJinsert(NCjson* object, char* key, NCjson* jvalue);
static int NCJappend(NCjson* object, NCjson* value);
static int NCJunparse(const NCjson* json, unsigned flags, char** textp);
#else /*!NETCDF_JSON_H*/
#define OPTSTATIC
#endif /*NETCDF_JSON_H*/
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
/**************************************************/
OPTSTATIC int
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
NCJparse(const char* text, unsigned flags, NCjson** jsonp)
{
return NCJparsen(strlen(text),text,flags,jsonp);
}
OPTSTATIC int
NCJparsen(size_t len, const char* text, unsigned flags, NCjson** jsonp)
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
{
int stat = NCJ_OK;
NCJparser* parser = NULL;
NCjson* json = NULL;
parser = calloc(1,sizeof(NCJparser));
if(parser == NULL)
{stat = NCJTHROW(NCJ_ERR); goto done;}
parser->text = (char*)malloc(len+1+1);
if(parser->text == NULL)
{stat = NCJTHROW(NCJ_ERR); goto done;}
memcpy(parser->text,text,len);
Add support for Zarr string type to NCZarr * re: https://github.com/Unidata/netcdf-c/pull/2278 * re: https://github.com/Unidata/netcdf-c/issues/2485 * re: https://github.com/Unidata/netcdf-c/issues/2474 This PR subsumes PR https://github.com/Unidata/netcdf-c/pull/2278. Actually is a bit an omnibus covering several issues. ## PR https://github.com/Unidata/netcdf-c/pull/2278 Add support for the Zarr string type. Zarr strings are restricted currently to be of fixed size. The primary issue to be addressed is to provide a way for user to specify the size of the fixed length strings. This is handled by providing the following new attributes special: 1. **_nczarr_default_maxstrlen** &mdash; This is an attribute of the root group. It specifies the default maximum string length for string types. If not specified, then it has the value of 64 characters. 2. **_nczarr_maxstrlen** &mdash; This is a per-variable attribute. It specifies the maximum string length for the string type associated with the variable. If not specified, then it is assigned the value of **_nczarr_default_maxstrlen**. This PR also requires some hacking to handle the existing netcdf-c NC_CHAR type, which does not exist in zarr. The goal was to choose numpy types for both the netcdf-c NC_STRING type and the netcdf-c NC_CHAR type such that if a pure zarr implementation read them, it would still work and an NC_CHAR type would be handled by zarr as a string of length 1. For writing variables and NCZarr attributes, the type mapping is as follows: * "|S1" for NC_CHAR. * ">S1" for NC_STRING && MAXSTRLEN==1 * ">Sn" for NC_STRING && MAXSTRLEN==n Note that it is a bit of a hack to use endianness, but it should be ok since for string/char, the endianness has no meaning. For reading attributes with pure zarr (i.e. with no nczarr atribute types defined), they will always be interpreted as of type NC_CHAR. ## Issue: https://github.com/Unidata/netcdf-c/issues/2474 This PR partly fixes this issue because it provided more comprehensive support for Zarr attributes that are JSON valued expressions. This PR still does not address the problem in that issue where the _ARRAY_DIMENSION attribute is incorrectly set. Than can only be fixed by the creator of the datasets. ## Issue: https://github.com/Unidata/netcdf-c/issues/2485 This PR also fixes the scalar failure shown in this issue. It generally cleans up scalar handling. It also adds a note to the documentation describing that NCZarr supports scalars while Zarr does not and also how scalar interoperability is achieved. ## Misc. Other Changes 1. Convert the nczarr special attributes and keys to be all lower case. So "_NCZARR_ATTR" now used "_nczarr_attr. Support back compatibility for the upper case names. 2. Cleanup my too-clever-by-half handling of scalars in libnczarr.
2022-08-28 10:21:13 +08:00
/* trim trailing whitespace */
if(len > 0) {
char* p;
for(p=parser->text+(len-1);p >= parser->text;p--) {
if(*p > ' ') break;
}
len = (size_t)((p - parser->text) + 1);
}
if(len == 0)
{stat = NCJTHROW(NCJ_ERR); goto done;}
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
parser->text[len] = '\0';
parser->text[len+1] = '\0';
parser->pos = &parser->text[0];
parser->status = NCJ_OK;
#ifdef NCJDEBUG
fprintf(stderr,"json: |%s|\n",parser->text);
#endif
if((stat=NCJparseR(parser,&json))==NCJ_ERR) goto done;
Add support for Zarr string type to NCZarr * re: https://github.com/Unidata/netcdf-c/pull/2278 * re: https://github.com/Unidata/netcdf-c/issues/2485 * re: https://github.com/Unidata/netcdf-c/issues/2474 This PR subsumes PR https://github.com/Unidata/netcdf-c/pull/2278. Actually is a bit an omnibus covering several issues. ## PR https://github.com/Unidata/netcdf-c/pull/2278 Add support for the Zarr string type. Zarr strings are restricted currently to be of fixed size. The primary issue to be addressed is to provide a way for user to specify the size of the fixed length strings. This is handled by providing the following new attributes special: 1. **_nczarr_default_maxstrlen** &mdash; This is an attribute of the root group. It specifies the default maximum string length for string types. If not specified, then it has the value of 64 characters. 2. **_nczarr_maxstrlen** &mdash; This is a per-variable attribute. It specifies the maximum string length for the string type associated with the variable. If not specified, then it is assigned the value of **_nczarr_default_maxstrlen**. This PR also requires some hacking to handle the existing netcdf-c NC_CHAR type, which does not exist in zarr. The goal was to choose numpy types for both the netcdf-c NC_STRING type and the netcdf-c NC_CHAR type such that if a pure zarr implementation read them, it would still work and an NC_CHAR type would be handled by zarr as a string of length 1. For writing variables and NCZarr attributes, the type mapping is as follows: * "|S1" for NC_CHAR. * ">S1" for NC_STRING && MAXSTRLEN==1 * ">Sn" for NC_STRING && MAXSTRLEN==n Note that it is a bit of a hack to use endianness, but it should be ok since for string/char, the endianness has no meaning. For reading attributes with pure zarr (i.e. with no nczarr atribute types defined), they will always be interpreted as of type NC_CHAR. ## Issue: https://github.com/Unidata/netcdf-c/issues/2474 This PR partly fixes this issue because it provided more comprehensive support for Zarr attributes that are JSON valued expressions. This PR still does not address the problem in that issue where the _ARRAY_DIMENSION attribute is incorrectly set. Than can only be fixed by the creator of the datasets. ## Issue: https://github.com/Unidata/netcdf-c/issues/2485 This PR also fixes the scalar failure shown in this issue. It generally cleans up scalar handling. It also adds a note to the documentation describing that NCZarr supports scalars while Zarr does not and also how scalar interoperability is achieved. ## Misc. Other Changes 1. Convert the nczarr special attributes and keys to be all lower case. So "_NCZARR_ATTR" now used "_nczarr_attr. Support back compatibility for the upper case names. 2. Cleanup my too-clever-by-half handling of scalars in libnczarr.
2022-08-28 10:21:13 +08:00
/* Must consume all of the input */
if(parser->pos != (parser->text+len)) {stat = NCJ_ERR; goto done;}
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
*jsonp = json;
json = NULL;
done:
if(parser != NULL) {
nullfree(parser->text);
nullfree(parser->yytext);
free(parser);
}
(void)NCJreclaim(json);
return NCJTHROW(stat);
}
/*
Simple recursive descent
intertwined with dict and list parsers.
Invariants:
1. The json argument is provided by caller and filled in by NCJparseR.
2. Each call pushed back last unconsumed token
*/
static int
NCJparseR(NCJparser* parser, NCjson** jsonp)
{
int stat = NCJ_OK;
int token = NCJ_UNDEF;
NCjson* json = NULL;
if(jsonp == NULL)
{stat = NCJTHROW(NCJ_ERR); goto done;}
if((token = NCJlex(parser)) == NCJ_UNDEF)
{stat = NCJTHROW(NCJ_ERR); goto done;}
switch (token) {
case NCJ_EOF:
break;
case NCJ_NULL:
if((stat = NCJnew(NCJ_NULL,&json))==NCJ_ERR) goto done;
break;
case NCJ_BOOLEAN:
if((stat = NCJnew(NCJ_BOOLEAN,&json))==NCJ_ERR) goto done;
json->string = strdup(parser->yytext);
break;
case NCJ_INT:
if((stat = NCJnew(NCJ_INT,&json))==NCJ_ERR) goto done;
json->string = strdup(parser->yytext);
break;
case NCJ_DOUBLE:
if((stat = NCJnew(NCJ_DOUBLE,&json))==NCJ_ERR) goto done;
json->string = strdup(parser->yytext);
break;
case NCJ_STRING:
if((stat = NCJnew(NCJ_STRING,&json))==NCJ_ERR) goto done;
json->string = strdup(parser->yytext);
break;
case NCJ_LBRACE:
if((stat = NCJnew(NCJ_DICT,&json))==NCJ_ERR) goto done;
if((stat = NCJparseDict(parser, &json->list))==NCJ_ERR) goto done;
break;
case NCJ_LBRACKET:
if((stat = NCJnew(NCJ_ARRAY,&json))==NCJ_ERR) goto done;
if((stat = NCJparseArray(parser, &json->list))==NCJ_ERR) goto done;
break;
case NCJ_RBRACE: /* We hit end of the dict we are parsing */
parser->pos--; /* pushback so NCJparseArray will catch */
json = NULL;
break;
case NCJ_RBRACKET:
parser->pos--; /* pushback so NCJparseDict will catch */
json = NULL;
break;
default:
stat = NCJTHROW(NCJ_ERR);
break;
}
if(jsonp && json) {*jsonp = json; json = NULL;}
done:
NCJreclaim(json);
return NCJTHROW(stat);
}
static int
NCJparseArray(NCJparser* parser, struct NCjlist* arrayp)
{
int stat = NCJ_OK;
int token = NCJ_UNDEF;
NCjson* element = NULL;
int stop = 0;
/* [ ^e1,e2, ...en] */
while(!stop) {
/* Recurse to get the value ei (might be null) */
if((stat = NCJparseR(parser,&element))==NCJ_ERR) goto done;
token = NCJlex(parser); /* Get next token */
/* Next token should be comma or rbracket */
switch(token) {
case NCJ_RBRACKET:
if(element != NULL) listappend(arrayp,element);
element = NULL;
stop = 1;
break;
case NCJ_COMMA:
/* Append the ei to the list */
if(element == NULL) {stat = NCJTHROW(NCJ_ERR); goto done;} /* error */
listappend(arrayp,element);
element = NULL;
break;
case NCJ_EOF:
case NCJ_UNDEF:
default:
stat = NCJTHROW(NCJ_ERR);
goto done;
}
}
done:
if(element != NULL)
NCJreclaim(element);
return NCJTHROW(stat);
}
static int
NCJparseDict(NCJparser* parser, struct NCjlist* dictp)
{
int stat = NCJ_OK;
int token = NCJ_UNDEF;
NCjson* value = NULL;
NCjson* key = NULL;
int stop = 0;
/* { ^k1:v1,k2:v2, ...kn:vn] */
while(!stop) {
/* Get the key, which must be a word of some sort */
token = NCJlex(parser);
switch(token) {
case NCJ_STRING:
case NCJ_BOOLEAN:
case NCJ_INT: case NCJ_DOUBLE: {
if((stat=NCJnewstring(token,parser->yytext,&key))==NCJ_ERR) goto done;
} break;
case NCJ_RBRACE: /* End of containing Dict */
stop = 1;
continue; /* leave loop */
case NCJ_EOF: case NCJ_UNDEF:
default:
stat = NCJTHROW(NCJ_ERR);
goto done;
}
/* Next token must be colon*/
switch((token = NCJlex(parser))) {
case NCJ_COLON: break;
case NCJ_UNDEF: case NCJ_EOF:
default: stat = NCJTHROW(NCJ_ERR); goto done;
}
/* Get the value */
if((stat = NCJparseR(parser,&value))==NCJ_ERR) goto done;
/* Next token must be comma or RBRACE */
switch((token = NCJlex(parser))) {
case NCJ_RBRACE:
stop = 1;
/* fall thru */
case NCJ_COMMA:
/* Insert key value into dict: key first, then value */
listappend(dictp,key);
key = NULL;
listappend(dictp,value);
value = NULL;
break;
case NCJ_EOF:
case NCJ_UNDEF:
default:
stat = NCJTHROW(NCJ_ERR);
goto done;
}
}
done:
if(key != NULL)
NCJreclaim(key);
if(value != NULL)
NCJreclaim(value);
return NCJTHROW(stat);
}
static int
NCJlex(NCJparser* parser)
{
int c;
int token = NCJ_UNDEF;
char* start;
size_t count;
while(token == 0) { /* avoid need to goto when retrying */
c = *parser->pos;
if(c == '\0') {
token = NCJ_EOF;
} else if(c <= ' ' || c == '\177') {/* ignore whitespace */
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
parser->pos++;
continue;
} else if(c == NCJ_ESCAPE) {
parser->pos++;
c = *parser->pos;
*parser->pos = unescape1(c);
continue;
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
} else if(strchr(JSON_WORD, c) != NULL) {
start = parser->pos;
for(;;) {
c = *parser->pos++;
if(c == '\0' || strchr(JSON_WORD,c) == NULL) break; /* end of word */
}
/* Pushback c */
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
parser->pos--;
count = ((parser->pos) - start);
if(NCJyytext(parser,start,count)) goto done;
/* Discriminate the word string to get the proper sort */
if(testbool(parser->yytext) == NCJ_OK)
token = NCJ_BOOLEAN;
/* do int test first since double subsumes int */
else if(testint(parser->yytext) == NCJ_OK)
token = NCJ_INT;
else if(testdouble(parser->yytext) == NCJ_OK)
token = NCJ_DOUBLE;
else if(testnull(parser->yytext) == NCJ_OK)
token = NCJ_NULL;
else
token = NCJ_STRING;
} else if(c == NCJ_QUOTE) {
parser->pos++;
start = parser->pos;
for(;;) {
c = *parser->pos++;
if(c == NCJ_ESCAPE) parser->pos++;
else if(c == NCJ_QUOTE || c == '\0') break;
}
if(c == '\0') {
parser->status = NCJ_ERR;
token = NCJ_UNDEF;
goto done;
}
count = ((parser->pos) - start) - 1; /* -1 for trailing quote */
if(NCJyytext(parser,start,count)==NCJ_ERR) goto done;
if(NCJunescape(parser)==NCJ_ERR) goto done;
token = NCJ_STRING;
} else { /* single char token */
if(NCJyytext(parser,parser->pos,1)==NCJ_ERR) goto done;
token = *parser->pos++;
}
#ifdef NCJDEBUG
fprintf(stderr,"%s(%d): |%s|\n",tokenname(token),token,parser->yytext);
#endif
} /*for(;;)*/
done:
if(parser->status == NCJ_ERR)
token = NCJ_UNDEF;
return token;
}
static int
testnull(const char* word)
{
if(strcasecmp(word,NCJ_TAG_NULL)==0)
return NCJTHROW(NCJ_OK);
return NCJTHROW(NCJ_ERR);
}
static int
testbool(const char* word)
{
if(strcasecmp(word,NCJ_TAG_TRUE)==0
|| strcasecmp(word,NCJ_TAG_FALSE)==0)
return NCJTHROW(NCJ_OK);
return NCJTHROW(NCJ_ERR);
}
static int
testint(const char* word)
{
int ncvt;
long long i;
int count = 0;
/* Try to convert to number */
ncvt = sscanf(word,"%lld%n",&i,&count);
return NCJTHROW((ncvt == 1 && strlen(word)==count ? NCJ_OK : NCJ_ERR));
}
static int
testdouble(const char* word)
{
int ncvt;
double d;
int count = 0;
/* Check for Nan and Infinity */
Improve S3 Documentation and Support ## Improvements to S3 Documentation * Create a new document *quickstart_paths.md* that give a summary of the legal path formats used by netcdf-c. This includes both file paths and URL paths. * Modify *nczarr.md* to remove most of the S3 related text. * Move the S3 text from *nczarr.md* to a new document *cloud.md*. * Add some S3-related text to the *byterange.md* document. Hopefully, this will make it easier for users to find the information they want. ## Rebuild NCZarr Testing In order to avoid problems with running make check in parallel, two changes were made: 1. The *nczarr_test* test system was rebuilt. Now, for each test. any generated files are kept in a test-specific directory, isolated from all other test executions. 2. Similarly, since the S3 test bucket is shared, any generated S3 objects are isolated using a test-specific key path. ## Other S3 Related Changes * Add code to ensure that files created on S3 are reclaimed at end of testing. * Used the bash "trap" command to ensure S3 cleanup even if the test fails. * Cleanup the S3 related configure.ac flag set since S3 is used in several places. So now one should use the option *--enable-s3* instead of *--enable-nczarr-s3*, although the latter is still kept as a deprecated alias for the former. * Get some of the github actions yml to work with S3; required fixing various test scripts adding a secret to access the Unidata S3 bucket. * Cleanup S3 portion of libnetcdf.settings.in and netcdf_meta.h.in and test_common.in. * Merge partial S3 support into dhttp.c. * Create an experimental s3 access library especially for use with Windows. It is enabled by using the options *--enable-s3-internal* (automake) or *-DENABLE_S3_INTERNAL=ON* (CMake). Also add a unit-test for it. * Move some definitions from ncrc.h to ncs3sdk.h ## Other Changes * Provide a default implementation of strlcpy and move this and similar defaults into *dmissing.c*.
2023-04-26 07:15:06 +08:00
if(0==(int)strcasecmp("nan",word)) return NCJTHROW(NCJ_OK);
if(0==(int)strcasecmp("infinity",word)) return NCJTHROW(NCJ_OK);
if(0==(int)strcasecmp("-infinity",word)) return NCJTHROW(NCJ_OK);
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
/* Allow the XXXf versions as well */
Improve S3 Documentation and Support ## Improvements to S3 Documentation * Create a new document *quickstart_paths.md* that give a summary of the legal path formats used by netcdf-c. This includes both file paths and URL paths. * Modify *nczarr.md* to remove most of the S3 related text. * Move the S3 text from *nczarr.md* to a new document *cloud.md*. * Add some S3-related text to the *byterange.md* document. Hopefully, this will make it easier for users to find the information they want. ## Rebuild NCZarr Testing In order to avoid problems with running make check in parallel, two changes were made: 1. The *nczarr_test* test system was rebuilt. Now, for each test. any generated files are kept in a test-specific directory, isolated from all other test executions. 2. Similarly, since the S3 test bucket is shared, any generated S3 objects are isolated using a test-specific key path. ## Other S3 Related Changes * Add code to ensure that files created on S3 are reclaimed at end of testing. * Used the bash "trap" command to ensure S3 cleanup even if the test fails. * Cleanup the S3 related configure.ac flag set since S3 is used in several places. So now one should use the option *--enable-s3* instead of *--enable-nczarr-s3*, although the latter is still kept as a deprecated alias for the former. * Get some of the github actions yml to work with S3; required fixing various test scripts adding a secret to access the Unidata S3 bucket. * Cleanup S3 portion of libnetcdf.settings.in and netcdf_meta.h.in and test_common.in. * Merge partial S3 support into dhttp.c. * Create an experimental s3 access library especially for use with Windows. It is enabled by using the options *--enable-s3-internal* (automake) or *-DENABLE_S3_INTERNAL=ON* (CMake). Also add a unit-test for it. * Move some definitions from ncrc.h to ncs3sdk.h ## Other Changes * Provide a default implementation of strlcpy and move this and similar defaults into *dmissing.c*.
2023-04-26 07:15:06 +08:00
if(0==(int)strcasecmp("nanf",word)) return NCJTHROW(NCJ_OK);
if(0==(int)strcasecmp("infinityf",word)) return NCJTHROW(NCJ_OK);
if(0==(int)strcasecmp("-infinityf",word)) return NCJTHROW(NCJ_OK);
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
/* Try to convert to number */
ncvt = sscanf(word,"%lg%n",&d,&count);
return NCJTHROW((ncvt == 1 && strlen(word)==count ? NCJ_OK : NCJ_ERR));
}
static int
NCJyytext(NCJparser* parser, char* start, size_t pdlen)
{
size_t len = (size_t)pdlen;
if(parser->yytext == NULL) {
parser->yytext = (char*)malloc(len+1);
parser->yylen = len;
} else if(parser->yylen <= len) {
parser->yytext = (char*) realloc(parser->yytext,len+1);
parser->yylen = len;
}
if(parser->yytext == NULL) return NCJTHROW(NCJ_ERR);
memcpy(parser->yytext,start,len);
parser->yytext[len] = '\0';
return NCJTHROW(NCJ_OK);
}
/**************************************************/
OPTSTATIC void
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
NCJreclaim(NCjson* json)
{
if(json == NULL) return;
switch(json->sort) {
case NCJ_INT:
case NCJ_DOUBLE:
case NCJ_BOOLEAN:
case NCJ_STRING:
nullfree(json->string);
break;
case NCJ_DICT:
NCJreclaimDict(&json->list);
break;
case NCJ_ARRAY:
NCJreclaimArray(&json->list);
break;
default: break; /* nothing to reclaim */
}
free(json);
}
static void
NCJreclaimArray(struct NCjlist* array)
{
int i;
for(i=0;i<array->len;i++) {
NCJreclaim(array->contents[i]);
}
nullfree(array->contents);
array->contents = NULL;
}
static void
NCJreclaimDict(struct NCjlist* dict)
{
NCJreclaimArray(dict);
}
/**************************************************/
/* Build Functions */
OPTSTATIC int
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
NCJnew(int sort, NCjson** objectp)
{
int stat = NCJ_OK;
NCjson* object = NULL;
if((object = (NCjson*)calloc(1,sizeof(NCjson))) == NULL)
{stat = NCJTHROW(NCJ_ERR); goto done;}
NCJsetsort(object,sort);
switch (sort) {
case NCJ_INT:
case NCJ_DOUBLE:
case NCJ_BOOLEAN:
case NCJ_STRING:
case NCJ_NULL:
break;
case NCJ_DICT:
case NCJ_ARRAY:
break;
default:
stat = NCJTHROW(NCJ_ERR);
goto done;
}
if(objectp) {*objectp = object; object = NULL;}
done:
if(stat) NCJreclaim(object);
return NCJTHROW(stat);
}
OPTSTATIC int
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
NCJnewstring(int sort, const char* value, NCjson** jsonp)
{
return NCJTHROW(NCJnewstringn(sort,strlen(value),value,jsonp));
}
OPTSTATIC int
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
NCJnewstringn(int sort, size_t len, const char* value, NCjson** jsonp)
{
int stat = NCJ_OK;
NCjson* json = NULL;
if(jsonp) *jsonp = NULL;
if(value == NULL)
{stat = NCJTHROW(NCJ_ERR); goto done;}
if((stat = NCJnew(sort,&json))==NCJ_ERR)
goto done;
if((json->string = (char*)malloc(len+1))==NULL)
{stat = NCJTHROW(NCJ_ERR); goto done;}
memcpy(json->string,value,len);
json->string[len] = '\0';
if(jsonp) *jsonp = json;
json = NULL; /* avoid memory errors */
done:
NCJreclaim(json);
return NCJTHROW(stat);
}
OPTSTATIC int
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
NCJdictget(const NCjson* dict, const char* key, NCjson** valuep)
{
int i,stat = NCJ_OK;
if(dict == NULL || dict->sort != NCJ_DICT)
{stat = NCJTHROW(NCJ_ERR); goto done;}
if(valuep) {*valuep = NULL;}
for(i=0;i<NCJlength(dict);i+=2) {
NCjson* jkey = NCJith(dict,i);
if(jkey->string != NULL && strcmp(jkey->string,key)==0) {
if(valuep) {*valuep = NCJith(dict,i+1); break;}
}
}
done:
return NCJTHROW(stat);
}
/* Unescape the text in parser->yytext; can
do in place because unescaped string will
always be shorter */
static int
NCJunescape(NCJparser* parser)
{
char* p = parser->yytext;
char* q = p;
int c;
for(;(c=*p++);) {
if(c == NCJ_ESCAPE) {
c = *p++;
switch (c) {
case 'b': c = '\b'; break;
case 'f': c = '\f'; break;
case 'n': c = '\n'; break;
case 'r': c = '\r'; break;
case 't': c = '\t'; break;
case NCJ_QUOTE: c = c; break;
case NCJ_ESCAPE: c = c; break;
default: c = c; break;/* technically not Json conformant */
}
}
*q++ = c;
}
*q = '\0';
return NCJTHROW(NCJ_OK);
}
/* Unescape a single character */
static int
unescape1(int c)
{
switch (c) {
case 'b': c = '\b'; break;
case 'f': c = '\f'; break;
case 'n': c = '\n'; break;
case 'r': c = '\r'; break;
case 't': c = '\t'; break;
default: c = c; break;/* technically not Json conformant */
}
return c;
}
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
#ifdef NCJDEBUG
static char*
tokenname(int token)
{
switch (token) {
case NCJ_STRING: return ("NCJ_STRING");
case NCJ_INT: return ("NCJ_INT");
case NCJ_DOUBLE: return ("NCJ_DOUBLE");
case NCJ_BOOLEAN: return ("NCJ_BOOLEAN");
case NCJ_DICT: return ("NCJ_DICT");
case NCJ_ARRAY: return ("NCJ_ARRAY");
case NCJ_NULL: return ("NCJ_NULL");
default:
if(token > ' ' && token <= 127) {
static char s[4];
s[0] = '\'';
s[1] = (char)token;
s[2] = '\'';
s[3] = '\0';
return (s);
} else
break;
}
return ("NCJ_UNDEF");
}
#endif
/* Convert a JSON value to an equivalent value of a specified sort */
OPTSTATIC int
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
NCJcvt(const NCjson* jvalue, int outsort, struct NCJconst* output)
{
int stat = NCJ_OK;
if(output == NULL) goto done;
#undef CASE
#define CASE(t1,t2) ((t1)<<4 | (t2)) /* the shift constant must be larger than log2(NCJ_NSORTS) */
switch (CASE(jvalue->sort,outsort)) {
case CASE(NCJ_BOOLEAN,NCJ_BOOLEAN):
if(strcasecmp(jvalue->string,NCJ_TAG_FALSE)==0) output->bval = 0; else output->bval = 1;
break;
case CASE(NCJ_BOOLEAN,NCJ_INT):
if(strcasecmp(jvalue->string,NCJ_TAG_FALSE)==0) output->ival = 0; else output->ival = 1;
break;
case CASE(NCJ_BOOLEAN,NCJ_DOUBLE):
if(strcasecmp(jvalue->string,NCJ_TAG_FALSE)==0) output->dval = 0.0; else output->dval = 1.0;
break;
case CASE(NCJ_BOOLEAN,NCJ_STRING):
output->sval = nulldup(jvalue->string);
break;
case CASE(NCJ_INT,NCJ_BOOLEAN):
sscanf(jvalue->string,"%lldd",&output->ival);
output->bval = (output->ival?1:0);
break;
case CASE(NCJ_INT,NCJ_INT):
sscanf(jvalue->string,"%lld",&output->ival);
break;
case CASE(NCJ_INT,NCJ_DOUBLE):
sscanf(jvalue->string,"%lld",&output->ival);
output->dval = (double)output->ival;
break;
case CASE(NCJ_INT,NCJ_STRING):
output->sval = nulldup(jvalue->string);
break;
case CASE(NCJ_DOUBLE,NCJ_BOOLEAN):
sscanf(jvalue->string,"%lf",&output->dval);
output->bval = (output->dval == 0?0:1);
break;
case CASE(NCJ_DOUBLE,NCJ_INT):
sscanf(jvalue->string,"%lf",&output->dval);
output->ival = (long long)output->dval;
break;
case CASE(NCJ_DOUBLE,NCJ_DOUBLE):
sscanf(jvalue->string,"%lf",&output->dval);
break;
case CASE(NCJ_DOUBLE,NCJ_STRING):
output->sval = nulldup(jvalue->string);
break;
case CASE(NCJ_STRING,NCJ_BOOLEAN):
if(strcasecmp(jvalue->string,NCJ_TAG_FALSE)==0) output->bval = 0; else output->bval = 1;
break;
case CASE(NCJ_STRING,NCJ_INT):
sscanf(jvalue->string,"%lld",&output->ival);
break;
case CASE(NCJ_STRING,NCJ_DOUBLE):
sscanf(jvalue->string,"%lf",&output->dval);
break;
case CASE(NCJ_STRING,NCJ_STRING):
output->sval = nulldup(jvalue->string);
break;
default:
stat = NCJTHROW(NCJ_ERR);
break;
}
done:
return NCJTHROW(stat);
}
static int
listappend(struct NCjlist* list, NCjson* json)
{
int stat = NCJ_OK;
NCjson** newcontents = NULL;
assert(list->len == 0 || list->contents != NULL);
if(json == NULL)
{stat = NCJTHROW(NCJ_ERR); goto done;}
if(list->len == 0) {
nullfree(list->contents);
list->contents = (NCjson**)calloc(2,sizeof(NCjson*));
if(list->contents == NULL)
{stat = NCJTHROW(NCJ_ERR); goto done;}
list->contents[0] = json;
list->len++;
} else {
if((newcontents = (NCjson**)calloc((2*list->len)+1,sizeof(NCjson*)))==NULL)
{stat = NCJTHROW(NCJ_ERR); goto done;}
memcpy(newcontents,list->contents,list->len*sizeof(NCjson*));
newcontents[list->len] = json;
list->len++;
free(list->contents);
list->contents = newcontents; newcontents = NULL;
}
done:
nullfree(newcontents);
return NCJTHROW(stat);
}
/**************************************************/
OPTSTATIC int
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
NCJclone(const NCjson* json, NCjson** clonep)
{
int stat = NCJ_OK;
NCjson* clone = NULL;
if(json == NULL) goto done;
switch(NCJsort(json)) {
case NCJ_INT:
case NCJ_DOUBLE:
case NCJ_BOOLEAN:
case NCJ_STRING:
if((stat=NCJnew(NCJsort(json),&clone))==NCJ_ERR) goto done;
if((NCJstring(clone) = strdup(NCJstring(json))) == NULL)
{stat = NCJTHROW(NCJ_ERR); goto done;}
break;
case NCJ_NULL:
if((stat=NCJnew(NCJsort(json),&clone))==NCJ_ERR) goto done;
break;
case NCJ_DICT:
if((stat=NCJcloneDict(json,&clone))==NCJ_ERR) goto done;
break;
case NCJ_ARRAY:
if((stat=NCJcloneArray(json,&clone))==NCJ_ERR) goto done;
break;
default: break; /* nothing to clone */
}
done:
if(stat == NCJ_OK && clonep) {*clonep = clone; clone = NULL;}
NCJreclaim(clone);
return NCJTHROW(stat);
}
static int
NCJcloneArray(const NCjson* array, NCjson** clonep)
{
int i, stat=NCJ_OK;
NCjson* clone = NULL;
if((stat=NCJnew(NCJ_ARRAY,&clone))==NCJ_ERR) goto done;
for(i=0;i<NCJlength(array);i++) {
NCjson* elem = NCJith(array,i);
NCjson* elemclone = NULL;
if((stat=NCJclone(elem,&elemclone))==NCJ_ERR) goto done;
NCJappend(clone,elemclone);
}
done:
if(stat == NCJ_OK && clonep) {*clonep = clone; clone = NULL;}
NCJreclaim(clone);
return stat;
}
static int
NCJcloneDict(const NCjson* dict, NCjson** clonep)
{
int i, stat=NCJ_OK;
NCjson* clone = NULL;
if((stat=NCJnew(NCJ_DICT,&clone))==NCJ_ERR) goto done;
for(i=0;i<NCJlength(dict);i++) {
NCjson* elem = NCJith(dict,i);
NCjson* elemclone = NULL;
if((stat=NCJclone(elem,&elemclone))==NCJ_ERR) goto done;
NCJappend(clone,elemclone);
}
done:
if(stat == NCJ_OK && clonep) {*clonep = clone; clone = NULL;}
NCJreclaim(clone);
return NCJTHROW(stat);
}
OPTSTATIC int
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
NCJaddstring(NCjson* json, int sort, const char* s)
{
int stat = NCJ_OK;
NCjson* jtmp = NULL;
if(NCJsort(json) != NCJ_DICT && NCJsort(json) != NCJ_ARRAY)
{stat = NCJTHROW(NCJ_ERR); goto done;}
if((stat = NCJnewstring(sort, s, &jtmp))==NCJ_ERR) goto done;
if((stat = NCJappend(json,jtmp))==NCJ_ERR) goto done;
jtmp = NULL;
done:
NCJreclaim(jtmp);
return NCJTHROW(stat);
}
/* Insert key-value pair into a dict object. key will be strdup'd */
OPTSTATIC int
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
NCJinsert(NCjson* object, char* key, NCjson* jvalue)
{
int stat = NCJ_OK;
NCjson* jkey = NULL;
if(object == NULL || object->sort != NCJ_DICT || key == NULL || jvalue == NULL)
{stat = NCJTHROW(NCJ_ERR); goto done;}
if((stat = NCJnewstring(NCJ_STRING,key,&jkey))==NCJ_ERR) goto done;
if((stat = NCJappend(object,jkey))==NCJ_ERR) goto done;
if((stat = NCJappend(object,jvalue))==NCJ_ERR) goto done;
done:
return NCJTHROW(stat);
}
/* Append value to an array or dict object. */
OPTSTATIC int
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
NCJappend(NCjson* object, NCjson* value)
{
if(object == NULL || value == NULL)
return NCJTHROW(NCJ_ERR);
switch (object->sort) {
case NCJ_ARRAY:
case NCJ_DICT:
listappend(&object->list,value);
break;
default:
return NCJTHROW(NCJ_ERR);
}
return NCJTHROW(NCJ_OK);
}
/**************************************************/
/* Unparser to convert NCjson object to text in buffer */
OPTSTATIC int
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
NCJunparse(const NCjson* json, unsigned flags, char** textp)
{
int stat = NCJ_OK;
NCJbuf buf = {0,NULL};
if((stat = NCJunparseR(json,&buf,flags))==NCJ_ERR)
goto done;
if(textp) {*textp = buf.text; buf.text = NULL; buf.len = 0;}
done:
nullfree(buf.text);
return NCJTHROW(stat);
}
static int
NCJunparseR(const NCjson* json, NCJbuf* buf, unsigned flags)
{
int stat = NCJ_OK;
int i;
switch (NCJsort(json)) {
case NCJ_STRING:
bytesappendquoted(buf,json->string);
break;
case NCJ_INT:
case NCJ_DOUBLE:
case NCJ_BOOLEAN:
bytesappend(buf,json->string);
break;
case NCJ_DICT:
bytesappendc(buf,NCJ_LBRACE);
if(json->list.len > 0 && json->list.contents != NULL) {
int shortlist = 0;
for(i=0;!shortlist && i < json->list.len;i+=2) {
if(i > 0) {bytesappendc(buf,NCJ_COMMA);bytesappendc(buf,' ');};
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
NCJunparseR(json->list.contents[i],buf,flags); /* key */
bytesappendc(buf,NCJ_COLON);
bytesappendc(buf,' ');
/* Allow for the possibility of a short dict entry */
if(json->list.contents[i+1] == NULL) { /* short */
bytesappendc(buf,'?');
shortlist = 1;
} else {
NCJunparseR(json->list.contents[i+1],buf,flags);
}
}
}
bytesappendc(buf,NCJ_RBRACE);
break;
case NCJ_ARRAY:
bytesappendc(buf,NCJ_LBRACKET);
if(json->list.len > 0 && json->list.contents != NULL) {
for(i=0;i < json->list.len;i++) {
if(i > 0) bytesappendc(buf,NCJ_COMMA);
NCJunparseR(json->list.contents[i],buf,flags);
}
}
bytesappendc(buf,NCJ_RBRACKET);
break;
case NCJ_NULL:
bytesappend(buf,"null");
break;
default:
stat = NCJTHROW(NCJ_ERR); goto done;
}
done:
return NCJTHROW(stat);
}
/* Escape a string and append to buf */
static int
escape(const char* text, NCJbuf* buf)
{
const char* p = text;
int c;
for(;(c=*p++);) {
char replace = 0;
switch (c) {
case '\b': replace = 'b'; break;
case '\f': replace = 'f'; break;
case '\n': replace = 'n'; break;
case '\r': replace = 'r'; break;
case '\t': replace = 't'; break;
case NCJ_QUOTE: replace = '\"'; break;
Add filter support to NCZarr Filter support has three goals: 1. Use the existing HDF5 filter implementations, 2. Allow filter metadata to be stored in the NumCodecs metadata format used by Zarr, 3. Allow filters to be used even when HDF5 is disabled Detailed usage directions are define in docs/filters.md. For now, the existing filter API is left in place. So filters are defined using ''nc_def_var_filter'' using the HDF5 style where the id and parameters are unsigned integers. This is a big change since filters affect many parts of the code. In the following, the terms "compressor" and "filter" and "codec" are generally used synonomously. ### Filter-Related Changes: * In order to support dynamic loading of shared filter libraries, a new library was added in the libncpoco directory; it helps to isolate dynamic loading across multiple platforms. * Provide a json parsing library for use by plugins; this is created by merging libdispatch/ncjson.c with include/ncjson.h. * Add a new _Codecs attribute to allow clients to see what codecs are being used; let ncdump -s print it out. * Provide special headers to help support compilation of HDF5 filters when HDF5 is not enabled: netcdf_filter_hdf5_build.h and netcdf_filter_build.h. * Add a number of new test to test the new nczarr filters. * Let ncgen parse _Codecs attribute, although it is ignored. ### Plugin directory changes: * Add support for the Blosc compressor; this is essential because it is the most common compressor used in Zarr datasets. This also necessitated adding a CMake FindBlosc.cmake file * Add NCZarr support for the big-four filters provided by HDF5: shuffle, fletcher32, deflate (zlib), and szip * Add a Codec defaulter (see docs/filters.md) for the big four filters. * Make plugins work with windows by properly adding __declspec declaration. ### Misc. Non-Filter Changes * Replace most uses of USE_NETCDF4 (deprecated) with USE_HDF5. * Improve support for caching * More fixes for path conversion code * Fix misc. memory leaks * Add new utility -- ncdump/ncpathcvt -- that does more or less the same thing as cygpath. * Add a number of new test to test the non-filter fixes. * Update the parsers * Convert most instances of '#ifdef _MSC_VER' to '#ifdef _WIN32'
2021-09-03 07:04:26 +08:00
case NCJ_ESCAPE: replace = '\\'; break;
default: break;
}
if(replace) {
bytesappendc(buf,NCJ_ESCAPE);
bytesappendc(buf,replace);
} else
bytesappendc(buf,c);
}
return NCJTHROW(NCJ_OK);
}
static int
bytesappendquoted(NCJbuf* buf, const char* s)
{
bytesappend(buf,"\"");
escape(s,buf);
bytesappend(buf,"\"");
return NCJTHROW(NCJ_OK);
}
static int
bytesappend(NCJbuf* buf, const char* s)
{
int stat = NCJ_OK;
char* newtext = NULL;
if(buf == NULL)
{stat = NCJTHROW(NCJ_ERR); goto done;}
if(s == NULL) s = "";
if(buf->len == 0) {
assert(buf->text == NULL);
buf->text = strdup(s);
if(buf->text == NULL)
{stat = NCJTHROW(NCJ_ERR); goto done;}
buf->len = strlen(s);
} else {
size_t slen = strlen(s);
size_t newlen = buf->len + slen + 1;
if((newtext = (char*)malloc(newlen))==NULL)
{stat = NCJTHROW(NCJ_ERR); goto done;}
strcpy(newtext,buf->text);
strcat(newtext,s);
free(buf->text); buf->text = NULL;
buf->text = newtext; newtext = NULL;
buf->len = newlen;
}
done:
nullfree(newtext);
return NCJTHROW(stat);
}
static int
bytesappendc(NCJbuf* bufp, const char c)
{
char s[2];
s[0] = c;
s[1] = '\0';
return bytesappend(bufp,s);
}
OPTSTATIC void
NCJdump(const NCjson* json, unsigned flags, FILE* out)
{
char* text = NULL;
(void)NCJunparse(json,0,&text);
if(out == NULL) out = stderr;
fprintf(out,"%s\n",text);
fflush(out);
nullfree(text);
}
Add support for Zarr string type to NCZarr * re: https://github.com/Unidata/netcdf-c/pull/2278 * re: https://github.com/Unidata/netcdf-c/issues/2485 * re: https://github.com/Unidata/netcdf-c/issues/2474 This PR subsumes PR https://github.com/Unidata/netcdf-c/pull/2278. Actually is a bit an omnibus covering several issues. ## PR https://github.com/Unidata/netcdf-c/pull/2278 Add support for the Zarr string type. Zarr strings are restricted currently to be of fixed size. The primary issue to be addressed is to provide a way for user to specify the size of the fixed length strings. This is handled by providing the following new attributes special: 1. **_nczarr_default_maxstrlen** &mdash; This is an attribute of the root group. It specifies the default maximum string length for string types. If not specified, then it has the value of 64 characters. 2. **_nczarr_maxstrlen** &mdash; This is a per-variable attribute. It specifies the maximum string length for the string type associated with the variable. If not specified, then it is assigned the value of **_nczarr_default_maxstrlen**. This PR also requires some hacking to handle the existing netcdf-c NC_CHAR type, which does not exist in zarr. The goal was to choose numpy types for both the netcdf-c NC_STRING type and the netcdf-c NC_CHAR type such that if a pure zarr implementation read them, it would still work and an NC_CHAR type would be handled by zarr as a string of length 1. For writing variables and NCZarr attributes, the type mapping is as follows: * "|S1" for NC_CHAR. * ">S1" for NC_STRING && MAXSTRLEN==1 * ">Sn" for NC_STRING && MAXSTRLEN==n Note that it is a bit of a hack to use endianness, but it should be ok since for string/char, the endianness has no meaning. For reading attributes with pure zarr (i.e. with no nczarr atribute types defined), they will always be interpreted as of type NC_CHAR. ## Issue: https://github.com/Unidata/netcdf-c/issues/2474 This PR partly fixes this issue because it provided more comprehensive support for Zarr attributes that are JSON valued expressions. This PR still does not address the problem in that issue where the _ARRAY_DIMENSION attribute is incorrectly set. Than can only be fixed by the creator of the datasets. ## Issue: https://github.com/Unidata/netcdf-c/issues/2485 This PR also fixes the scalar failure shown in this issue. It generally cleans up scalar handling. It also adds a note to the documentation describing that NCZarr supports scalars while Zarr does not and also how scalar interoperability is achieved. ## Misc. Other Changes 1. Convert the nczarr special attributes and keys to be all lower case. So "_NCZARR_ATTR" now used "_nczarr_attr. Support back compatibility for the upper case names. 2. Cleanup my too-clever-by-half handling of scalars in libnczarr.
2022-08-28 10:21:13 +08:00
OPTSTATIC const char*
NCJtotext(const NCjson* json)
{
static char outtext[4096];
char* text = NULL;
if(json == NULL) {strcpy(outtext,"<null>"); goto done;}
(void)NCJunparse(json,0,&text);
Improve S3 Documentation and Support ## Improvements to S3 Documentation * Create a new document *quickstart_paths.md* that give a summary of the legal path formats used by netcdf-c. This includes both file paths and URL paths. * Modify *nczarr.md* to remove most of the S3 related text. * Move the S3 text from *nczarr.md* to a new document *cloud.md*. * Add some S3-related text to the *byterange.md* document. Hopefully, this will make it easier for users to find the information they want. ## Rebuild NCZarr Testing In order to avoid problems with running make check in parallel, two changes were made: 1. The *nczarr_test* test system was rebuilt. Now, for each test. any generated files are kept in a test-specific directory, isolated from all other test executions. 2. Similarly, since the S3 test bucket is shared, any generated S3 objects are isolated using a test-specific key path. ## Other S3 Related Changes * Add code to ensure that files created on S3 are reclaimed at end of testing. * Used the bash "trap" command to ensure S3 cleanup even if the test fails. * Cleanup the S3 related configure.ac flag set since S3 is used in several places. So now one should use the option *--enable-s3* instead of *--enable-nczarr-s3*, although the latter is still kept as a deprecated alias for the former. * Get some of the github actions yml to work with S3; required fixing various test scripts adding a secret to access the Unidata S3 bucket. * Cleanup S3 portion of libnetcdf.settings.in and netcdf_meta.h.in and test_common.in. * Merge partial S3 support into dhttp.c. * Create an experimental s3 access library especially for use with Windows. It is enabled by using the options *--enable-s3-internal* (automake) or *-DENABLE_S3_INTERNAL=ON* (CMake). Also add a unit-test for it. * Move some definitions from ncrc.h to ncs3sdk.h ## Other Changes * Provide a default implementation of strlcpy and move this and similar defaults into *dmissing.c*.
2023-04-26 07:15:06 +08:00
strncpy(outtext,text,sizeof(outtext));
Add support for Zarr string type to NCZarr * re: https://github.com/Unidata/netcdf-c/pull/2278 * re: https://github.com/Unidata/netcdf-c/issues/2485 * re: https://github.com/Unidata/netcdf-c/issues/2474 This PR subsumes PR https://github.com/Unidata/netcdf-c/pull/2278. Actually is a bit an omnibus covering several issues. ## PR https://github.com/Unidata/netcdf-c/pull/2278 Add support for the Zarr string type. Zarr strings are restricted currently to be of fixed size. The primary issue to be addressed is to provide a way for user to specify the size of the fixed length strings. This is handled by providing the following new attributes special: 1. **_nczarr_default_maxstrlen** &mdash; This is an attribute of the root group. It specifies the default maximum string length for string types. If not specified, then it has the value of 64 characters. 2. **_nczarr_maxstrlen** &mdash; This is a per-variable attribute. It specifies the maximum string length for the string type associated with the variable. If not specified, then it is assigned the value of **_nczarr_default_maxstrlen**. This PR also requires some hacking to handle the existing netcdf-c NC_CHAR type, which does not exist in zarr. The goal was to choose numpy types for both the netcdf-c NC_STRING type and the netcdf-c NC_CHAR type such that if a pure zarr implementation read them, it would still work and an NC_CHAR type would be handled by zarr as a string of length 1. For writing variables and NCZarr attributes, the type mapping is as follows: * "|S1" for NC_CHAR. * ">S1" for NC_STRING && MAXSTRLEN==1 * ">Sn" for NC_STRING && MAXSTRLEN==n Note that it is a bit of a hack to use endianness, but it should be ok since for string/char, the endianness has no meaning. For reading attributes with pure zarr (i.e. with no nczarr atribute types defined), they will always be interpreted as of type NC_CHAR. ## Issue: https://github.com/Unidata/netcdf-c/issues/2474 This PR partly fixes this issue because it provided more comprehensive support for Zarr attributes that are JSON valued expressions. This PR still does not address the problem in that issue where the _ARRAY_DIMENSION attribute is incorrectly set. Than can only be fixed by the creator of the datasets. ## Issue: https://github.com/Unidata/netcdf-c/issues/2485 This PR also fixes the scalar failure shown in this issue. It generally cleans up scalar handling. It also adds a note to the documentation describing that NCZarr supports scalars while Zarr does not and also how scalar interoperability is achieved. ## Misc. Other Changes 1. Convert the nczarr special attributes and keys to be all lower case. So "_NCZARR_ATTR" now used "_nczarr_attr. Support back compatibility for the upper case names. 2. Cleanup my too-clever-by-half handling of scalars in libnczarr.
2022-08-28 10:21:13 +08:00
nullfree(text);
done:
return outtext;
}
/* Hack to avoid static unused warning */
Add support for Zarr string type to NCZarr * re: https://github.com/Unidata/netcdf-c/pull/2278 * re: https://github.com/Unidata/netcdf-c/issues/2485 * re: https://github.com/Unidata/netcdf-c/issues/2474 This PR subsumes PR https://github.com/Unidata/netcdf-c/pull/2278. Actually is a bit an omnibus covering several issues. ## PR https://github.com/Unidata/netcdf-c/pull/2278 Add support for the Zarr string type. Zarr strings are restricted currently to be of fixed size. The primary issue to be addressed is to provide a way for user to specify the size of the fixed length strings. This is handled by providing the following new attributes special: 1. **_nczarr_default_maxstrlen** &mdash; This is an attribute of the root group. It specifies the default maximum string length for string types. If not specified, then it has the value of 64 characters. 2. **_nczarr_maxstrlen** &mdash; This is a per-variable attribute. It specifies the maximum string length for the string type associated with the variable. If not specified, then it is assigned the value of **_nczarr_default_maxstrlen**. This PR also requires some hacking to handle the existing netcdf-c NC_CHAR type, which does not exist in zarr. The goal was to choose numpy types for both the netcdf-c NC_STRING type and the netcdf-c NC_CHAR type such that if a pure zarr implementation read them, it would still work and an NC_CHAR type would be handled by zarr as a string of length 1. For writing variables and NCZarr attributes, the type mapping is as follows: * "|S1" for NC_CHAR. * ">S1" for NC_STRING && MAXSTRLEN==1 * ">Sn" for NC_STRING && MAXSTRLEN==n Note that it is a bit of a hack to use endianness, but it should be ok since for string/char, the endianness has no meaning. For reading attributes with pure zarr (i.e. with no nczarr atribute types defined), they will always be interpreted as of type NC_CHAR. ## Issue: https://github.com/Unidata/netcdf-c/issues/2474 This PR partly fixes this issue because it provided more comprehensive support for Zarr attributes that are JSON valued expressions. This PR still does not address the problem in that issue where the _ARRAY_DIMENSION attribute is incorrectly set. Than can only be fixed by the creator of the datasets. ## Issue: https://github.com/Unidata/netcdf-c/issues/2485 This PR also fixes the scalar failure shown in this issue. It generally cleans up scalar handling. It also adds a note to the documentation describing that NCZarr supports scalars while Zarr does not and also how scalar interoperability is achieved. ## Misc. Other Changes 1. Convert the nczarr special attributes and keys to be all lower case. So "_NCZARR_ATTR" now used "_nczarr_attr. Support back compatibility for the upper case names. 2. Cleanup my too-clever-by-half handling of scalars in libnczarr.
2022-08-28 10:21:13 +08:00
static void
netcdf_supresswarnings(void)
{
void* ignore;
Add support for Zarr string type to NCZarr * re: https://github.com/Unidata/netcdf-c/pull/2278 * re: https://github.com/Unidata/netcdf-c/issues/2485 * re: https://github.com/Unidata/netcdf-c/issues/2474 This PR subsumes PR https://github.com/Unidata/netcdf-c/pull/2278. Actually is a bit an omnibus covering several issues. ## PR https://github.com/Unidata/netcdf-c/pull/2278 Add support for the Zarr string type. Zarr strings are restricted currently to be of fixed size. The primary issue to be addressed is to provide a way for user to specify the size of the fixed length strings. This is handled by providing the following new attributes special: 1. **_nczarr_default_maxstrlen** &mdash; This is an attribute of the root group. It specifies the default maximum string length for string types. If not specified, then it has the value of 64 characters. 2. **_nczarr_maxstrlen** &mdash; This is a per-variable attribute. It specifies the maximum string length for the string type associated with the variable. If not specified, then it is assigned the value of **_nczarr_default_maxstrlen**. This PR also requires some hacking to handle the existing netcdf-c NC_CHAR type, which does not exist in zarr. The goal was to choose numpy types for both the netcdf-c NC_STRING type and the netcdf-c NC_CHAR type such that if a pure zarr implementation read them, it would still work and an NC_CHAR type would be handled by zarr as a string of length 1. For writing variables and NCZarr attributes, the type mapping is as follows: * "|S1" for NC_CHAR. * ">S1" for NC_STRING && MAXSTRLEN==1 * ">Sn" for NC_STRING && MAXSTRLEN==n Note that it is a bit of a hack to use endianness, but it should be ok since for string/char, the endianness has no meaning. For reading attributes with pure zarr (i.e. with no nczarr atribute types defined), they will always be interpreted as of type NC_CHAR. ## Issue: https://github.com/Unidata/netcdf-c/issues/2474 This PR partly fixes this issue because it provided more comprehensive support for Zarr attributes that are JSON valued expressions. This PR still does not address the problem in that issue where the _ARRAY_DIMENSION attribute is incorrectly set. Than can only be fixed by the creator of the datasets. ## Issue: https://github.com/Unidata/netcdf-c/issues/2485 This PR also fixes the scalar failure shown in this issue. It generally cleans up scalar handling. It also adds a note to the documentation describing that NCZarr supports scalars while Zarr does not and also how scalar interoperability is achieved. ## Misc. Other Changes 1. Convert the nczarr special attributes and keys to be all lower case. So "_NCZARR_ATTR" now used "_nczarr_attr. Support back compatibility for the upper case names. 2. Cleanup my too-clever-by-half handling of scalars in libnczarr.
2022-08-28 10:21:13 +08:00
ignore = (void*)netcdf_supresswarnings;
ignore = (void*)NCJinsert;
ignore = (void*)NCJaddstring;
ignore = (void*)NCJcvt;
ignore = (void*)NCJdictget;
ignore = (void*)NCJparse;
Add support for Zarr string type to NCZarr * re: https://github.com/Unidata/netcdf-c/pull/2278 * re: https://github.com/Unidata/netcdf-c/issues/2485 * re: https://github.com/Unidata/netcdf-c/issues/2474 This PR subsumes PR https://github.com/Unidata/netcdf-c/pull/2278. Actually is a bit an omnibus covering several issues. ## PR https://github.com/Unidata/netcdf-c/pull/2278 Add support for the Zarr string type. Zarr strings are restricted currently to be of fixed size. The primary issue to be addressed is to provide a way for user to specify the size of the fixed length strings. This is handled by providing the following new attributes special: 1. **_nczarr_default_maxstrlen** &mdash; This is an attribute of the root group. It specifies the default maximum string length for string types. If not specified, then it has the value of 64 characters. 2. **_nczarr_maxstrlen** &mdash; This is a per-variable attribute. It specifies the maximum string length for the string type associated with the variable. If not specified, then it is assigned the value of **_nczarr_default_maxstrlen**. This PR also requires some hacking to handle the existing netcdf-c NC_CHAR type, which does not exist in zarr. The goal was to choose numpy types for both the netcdf-c NC_STRING type and the netcdf-c NC_CHAR type such that if a pure zarr implementation read them, it would still work and an NC_CHAR type would be handled by zarr as a string of length 1. For writing variables and NCZarr attributes, the type mapping is as follows: * "|S1" for NC_CHAR. * ">S1" for NC_STRING && MAXSTRLEN==1 * ">Sn" for NC_STRING && MAXSTRLEN==n Note that it is a bit of a hack to use endianness, but it should be ok since for string/char, the endianness has no meaning. For reading attributes with pure zarr (i.e. with no nczarr atribute types defined), they will always be interpreted as of type NC_CHAR. ## Issue: https://github.com/Unidata/netcdf-c/issues/2474 This PR partly fixes this issue because it provided more comprehensive support for Zarr attributes that are JSON valued expressions. This PR still does not address the problem in that issue where the _ARRAY_DIMENSION attribute is incorrectly set. Than can only be fixed by the creator of the datasets. ## Issue: https://github.com/Unidata/netcdf-c/issues/2485 This PR also fixes the scalar failure shown in this issue. It generally cleans up scalar handling. It also adds a note to the documentation describing that NCZarr supports scalars while Zarr does not and also how scalar interoperability is achieved. ## Misc. Other Changes 1. Convert the nczarr special attributes and keys to be all lower case. So "_NCZARR_ATTR" now used "_nczarr_attr. Support back compatibility for the upper case names. 2. Cleanup my too-clever-by-half handling of scalars in libnczarr.
2022-08-28 10:21:13 +08:00
ignore = (void*)NCJdump;
ignore = (void*)NCJtotext;
ignore = ignore;
}