2010-06-03 21:24:43 +08:00
|
|
|
/*********************************************************************
|
2018-12-07 05:13:56 +08:00
|
|
|
* Copyright 2018, UCAR/Unidata
|
2010-06-03 21:24:43 +08:00
|
|
|
* See netcdf/COPYRIGHT file for copying and redistribution conditions.
|
2012-08-20 05:12:35 +08:00
|
|
|
* $Header$
|
2010-06-03 21:24:43 +08:00
|
|
|
*********************************************************************/
|
|
|
|
#ifndef NCHASHMAP_H
|
2018-02-09 10:53:40 +08:00
|
|
|
#define NCHASHMAP_H
|
|
|
|
|
2018-11-26 19:31:23 +08:00
|
|
|
#if HAVE_CONFIG_H
|
|
|
|
#include "config.h"
|
|
|
|
#endif
|
|
|
|
#include <stdint.h>
|
|
|
|
|
2018-02-09 10:53:40 +08:00
|
|
|
/*
|
2018-03-17 01:46:18 +08:00
|
|
|
This hashmap is optimized to assume null-terminated strings as the
|
|
|
|
key.
|
|
|
|
|
|
|
|
Data is presumed to be an index into some other table Assume it
|
|
|
|
can be compared using simple == The key is some hash of some
|
|
|
|
null terminated string.
|
|
|
|
|
|
|
|
One problem here is that we need to do a final equality check on
|
|
|
|
the name string to avoid an accidental hash collision. It would
|
|
|
|
be nice if we had a large enough hashkey that was known to have
|
|
|
|
an extremely low probability of collisions so we could compare
|
|
|
|
the hashkeys to determine exact match. A quick internet search
|
|
|
|
indicates that this is rather more tricky than just using
|
|
|
|
e.g. crc64 or such. Needs some thought.
|
2018-02-09 10:53:40 +08:00
|
|
|
*/
|
|
|
|
|
|
|
|
/*! Hashmap-related structs.
|
|
|
|
NOTES:
|
2018-03-17 01:46:18 +08:00
|
|
|
1. 'data' is the an arbitrary uintptr_t integer or void* pointer.
|
|
|
|
2. hashkey is a crc32 hash of key
|
|
|
|
|
|
|
|
WARNINGS:
|
|
|
|
1. It is critical that |uintptr_t| == |void*|
|
2018-02-09 10:53:40 +08:00
|
|
|
*/
|
2018-03-17 01:46:18 +08:00
|
|
|
|
2020-11-20 08:01:04 +08:00
|
|
|
/** The type and # bits in a hashkey */
|
|
|
|
#ifndef nchashkey_t
|
|
|
|
#define nchashkey_t unsigned
|
|
|
|
#define NCHASHKEYBITS (sizeof(nchashkey_t)*8)
|
|
|
|
#endif
|
|
|
|
|
2018-02-09 10:53:40 +08:00
|
|
|
typedef struct NC_hentry {
|
|
|
|
int flags;
|
2018-03-17 01:46:18 +08:00
|
|
|
uintptr_t data;
|
2020-11-20 08:01:04 +08:00
|
|
|
nchashkey_t hashkey; /* Hash id (crc)*/
|
2018-03-17 01:46:18 +08:00
|
|
|
size_t keysize;
|
|
|
|
char* key; /* copy of the key string; kept as unsigned char */
|
2018-02-09 10:53:40 +08:00
|
|
|
} NC_hentry;
|
|
|
|
|
|
|
|
/*
|
|
|
|
The hashmap object must give us the hash table (table),
|
|
|
|
the |table| size, and the # of defined entries in the table
|
|
|
|
*/
|
|
|
|
typedef struct NC_hashmap {
|
2018-03-17 01:46:18 +08:00
|
|
|
size_t alloc; /* allocated # of entries */
|
|
|
|
size_t active; /* # of active entries */
|
2018-02-09 10:53:40 +08:00
|
|
|
NC_hentry* table;
|
|
|
|
} NC_hashmap;
|
|
|
|
|
2018-03-17 01:46:18 +08:00
|
|
|
/* defined in nchashmap.c */
|
|
|
|
|
|
|
|
/*
|
|
|
|
There are two "kinds" of functions:
|
|
|
|
1. those that take the key+size -- they compute the hashkey internally.
|
|
|
|
2. those that take the hashkey directly
|
|
|
|
*/
|
2018-02-09 10:53:40 +08:00
|
|
|
|
|
|
|
/** Creates a new hashmap near the given size. */
|
|
|
|
extern NC_hashmap* NC_hashmapnew(size_t startsize);
|
|
|
|
|
2018-03-17 01:46:18 +08:00
|
|
|
/** Inserts a new element into the hashmap; takes key+size */
|
|
|
|
/* key points to size bytes to convert to hash key */
|
|
|
|
extern int NC_hashmapadd(NC_hashmap*, uintptr_t data, const char* key, size_t keysize);
|
2018-02-09 10:53:40 +08:00
|
|
|
|
2018-03-17 01:46:18 +08:00
|
|
|
/** Removes the storage for the element of the key; takes key+size.
|
2018-02-09 10:53:40 +08:00
|
|
|
Return 1 if found, 0 otherwise; returns the data in datap if !null
|
|
|
|
*/
|
2018-03-17 01:46:18 +08:00
|
|
|
extern int NC_hashmapremove(NC_hashmap*, const char* key, size_t keysize, uintptr_t* datap);
|
2018-02-09 10:53:40 +08:00
|
|
|
|
2018-03-17 01:46:18 +08:00
|
|
|
/** Returns the data for the key; takes key+size.
|
2018-02-09 10:53:40 +08:00
|
|
|
Return 1 if found, 0 otherwise; returns the data in datap if !null
|
|
|
|
*/
|
2018-03-17 01:46:18 +08:00
|
|
|
extern int NC_hashmapget(NC_hashmap*, const char* key, size_t keysize, uintptr_t* datap);
|
2018-02-09 10:53:40 +08:00
|
|
|
|
2018-03-17 01:46:18 +08:00
|
|
|
/** Change the data for the specified key; takes hashkey.
|
2018-02-09 10:53:40 +08:00
|
|
|
Return 1 if found, 0 otherwise
|
|
|
|
*/
|
2018-03-17 01:46:18 +08:00
|
|
|
extern int NC_hashmapsetdata(NC_hashmap*, const char* key, size_t keylen, uintptr_t newdata);
|
2018-02-09 10:53:40 +08:00
|
|
|
|
This PR adds EXPERIMENTAL support for accessing data in the
cloud using a variant of the Zarr protocol and storage
format. This enhancement is generically referred to as "NCZarr".
The data model supported by NCZarr is netcdf-4 minus the user-defined
types and the String type. In this sense it is similar to the CDF-5
data model.
More detailed information about enabling and using NCZarr is
described in the document NUG/nczarr.md and in a
[Unidata Developer's blog entry](https://www.unidata.ucar.edu/blogs/developer/en/entry/overview-of-zarr-support-in).
WARNING: this code has had limited testing, so do use this version
for production work. Also, performance improvements are ongoing.
Note especially the following platform matrix of successful tests:
Platform | Build System | S3 support
------------------------------------
Linux+gcc | Automake | yes
Linux+gcc | CMake | yes
Visual Studio | CMake | no
Additionally, and as a consequence of the addition of NCZarr,
major changes have been made to the Filter API. NOTE: NCZarr
does not yet support filters, but these changes are enablers for
that support in the future. Note that it is possible
(probable?) that there will be some accidental reversions if the
changes here did not correctly mimic the existing filter testing.
In any case, previously filter ids and parameters were of type
unsigned int. In order to support the more general zarr filter
model, this was all converted to char*. The old HDF5-specific,
unsigned int operations are still supported but they are
wrappers around the new, char* based nc_filterx_XXX functions.
This entailed at least the following changes:
1. Added the files libdispatch/dfilterx.c and include/ncfilter.h
2. Some filterx utilities have been moved to libdispatch/daux.c
3. A new entry, "filter_actions" was added to the NCDispatch table
and the version bumped.
4. An overly complex set of structs was created to support funnelling
all of the filterx operations thru a single dispatch
"filter_actions" entry.
5. Move common code to from libhdf5 to libsrc4 so that it is accessible
to nczarr.
Changes directly related to Zarr:
1. Modified CMakeList.txt and configure.ac to support both C and C++
-- this is in support of S3 support via the awd-sdk libraries.
2. Define a size64_t type to support nczarr.
3. More reworking of libdispatch/dinfermodel.c to
support zarr and to regularize the structure of the fragments
section of a URL.
Changes not directly related to Zarr:
1. Make client-side filter registration be conditional, with default off.
2. Hack include/nc4internal.h to make some flags added by Ed be unique:
e.g. NC_CREAT, NC_INDEF, etc.
3. cleanup include/nchttp.h and libdispatch/dhttp.c.
4. Misc. changes to support compiling under Visual Studio including:
* Better testing under windows for dirent.h and opendir and closedir.
5. Misc. changes to the oc2 code to support various libcurl CURLOPT flags
and to centralize error reporting.
6. By default, suppress the vlen tests that have unfixed memory leaks; add option to enable them.
7. Make part of the nc_test/test_byterange.sh test be contingent on remotetest.unidata.ucar.edu being accessible.
Changes Left TO-DO:
1. fix provenance code, it is too HDF5 specific.
2020-06-29 08:02:47 +08:00
|
|
|
/** Returns the number of active elements. */
|
2018-02-09 10:53:40 +08:00
|
|
|
extern size_t NC_hashmapcount(NC_hashmap*);
|
|
|
|
|
|
|
|
/** Reclaims the hashmap structure. */
|
|
|
|
extern int NC_hashmapfree(NC_hashmap*);
|
2017-03-09 08:01:10 +08:00
|
|
|
|
2018-03-17 01:46:18 +08:00
|
|
|
/* Return the hash key for specified key; takes key+size*/
|
2020-11-20 08:01:04 +08:00
|
|
|
extern nchashkey_t NC_hashmapkey(const char* key, size_t size);
|
2018-03-17 01:46:18 +08:00
|
|
|
|
This PR adds EXPERIMENTAL support for accessing data in the
cloud using a variant of the Zarr protocol and storage
format. This enhancement is generically referred to as "NCZarr".
The data model supported by NCZarr is netcdf-4 minus the user-defined
types and the String type. In this sense it is similar to the CDF-5
data model.
More detailed information about enabling and using NCZarr is
described in the document NUG/nczarr.md and in a
[Unidata Developer's blog entry](https://www.unidata.ucar.edu/blogs/developer/en/entry/overview-of-zarr-support-in).
WARNING: this code has had limited testing, so do use this version
for production work. Also, performance improvements are ongoing.
Note especially the following platform matrix of successful tests:
Platform | Build System | S3 support
------------------------------------
Linux+gcc | Automake | yes
Linux+gcc | CMake | yes
Visual Studio | CMake | no
Additionally, and as a consequence of the addition of NCZarr,
major changes have been made to the Filter API. NOTE: NCZarr
does not yet support filters, but these changes are enablers for
that support in the future. Note that it is possible
(probable?) that there will be some accidental reversions if the
changes here did not correctly mimic the existing filter testing.
In any case, previously filter ids and parameters were of type
unsigned int. In order to support the more general zarr filter
model, this was all converted to char*. The old HDF5-specific,
unsigned int operations are still supported but they are
wrappers around the new, char* based nc_filterx_XXX functions.
This entailed at least the following changes:
1. Added the files libdispatch/dfilterx.c and include/ncfilter.h
2. Some filterx utilities have been moved to libdispatch/daux.c
3. A new entry, "filter_actions" was added to the NCDispatch table
and the version bumped.
4. An overly complex set of structs was created to support funnelling
all of the filterx operations thru a single dispatch
"filter_actions" entry.
5. Move common code to from libhdf5 to libsrc4 so that it is accessible
to nczarr.
Changes directly related to Zarr:
1. Modified CMakeList.txt and configure.ac to support both C and C++
-- this is in support of S3 support via the awd-sdk libraries.
2. Define a size64_t type to support nczarr.
3. More reworking of libdispatch/dinfermodel.c to
support zarr and to regularize the structure of the fragments
section of a URL.
Changes not directly related to Zarr:
1. Make client-side filter registration be conditional, with default off.
2. Hack include/nc4internal.h to make some flags added by Ed be unique:
e.g. NC_CREAT, NC_INDEF, etc.
3. cleanup include/nchttp.h and libdispatch/dhttp.c.
4. Misc. changes to support compiling under Visual Studio including:
* Better testing under windows for dirent.h and opendir and closedir.
5. Misc. changes to the oc2 code to support various libcurl CURLOPT flags
and to centralize error reporting.
6. By default, suppress the vlen tests that have unfixed memory leaks; add option to enable them.
7. Make part of the nc_test/test_byterange.sh test be contingent on remotetest.unidata.ucar.edu being accessible.
Changes Left TO-DO:
1. fix provenance code, it is too HDF5 specific.
2020-06-29 08:02:47 +08:00
|
|
|
/* Return the ith entry info:
|
|
|
|
@param map
|
|
|
|
@param i
|
|
|
|
@param entryp contains 0 if not active, otherwise the data
|
|
|
|
@param keyp contains null if not active, otherwise the key
|
|
|
|
@return NC_EINVAL if no more entries
|
|
|
|
*/
|
|
|
|
extern int NC_hashmapith(NC_hashmap* map, size_t i, uintptr_t* datap, const char** keyp);
|
|
|
|
|
2010-06-03 21:24:43 +08:00
|
|
|
#endif /*NCHASHMAP_H*/
|
|
|
|
|