2017-03-09 08:01:10 +08:00
|
|
|
/*********************************************************************
|
2018-12-07 05:24:28 +08:00
|
|
|
* Copyright 2018, UCAR/Unidata
|
2017-03-09 08:01:10 +08:00
|
|
|
* See netcdf/COPYRIGHT file for copying and redistribution conditions.
|
|
|
|
*********************************************************************/
|
|
|
|
|
2024-05-16 08:46:25 +08:00
|
|
|
#include "config.h"
|
2017-03-09 08:01:10 +08:00
|
|
|
#include "ncdispatch.h"
|
|
|
|
#include "ncd4dispatch.h"
|
|
|
|
#include "d4includes.h"
|
|
|
|
#include "d4curlfunctions.h"
|
2023-11-25 02:20:52 +08:00
|
|
|
#include <stddef.h>
|
2017-03-09 08:01:10 +08:00
|
|
|
|
2017-07-06 00:03:48 +08:00
|
|
|
#ifdef _MSC_VER
|
2017-03-09 08:01:10 +08:00
|
|
|
#include <process.h>
|
|
|
|
#include <direct.h>
|
|
|
|
#endif
|
|
|
|
|
|
|
|
#ifdef HAVE_SYS_STAT_H
|
|
|
|
#include <sys/stat.h>
|
|
|
|
#endif
|
|
|
|
|
|
|
|
/**************************************************/
|
|
|
|
/* Forward */
|
|
|
|
|
|
|
|
static int constrainable(NCURI*);
|
|
|
|
static void freeCurl(NCD4curl*);
|
2022-11-14 04:15:11 +08:00
|
|
|
static int fragmentcheck(NCD4INFO*, const char* key, const char* subkey);
|
|
|
|
static const char* getfragment(NCD4INFO* info, const char* key);
|
|
|
|
static const char* getquery(NCD4INFO* info, const char* key);
|
2017-03-09 08:01:10 +08:00
|
|
|
static int set_curl_properties(NCD4INFO*);
|
2021-01-15 12:39:08 +08:00
|
|
|
static int makesubstrate(NCD4INFO* d4info);
|
2017-03-09 08:01:10 +08:00
|
|
|
|
|
|
|
/**************************************************/
|
|
|
|
/* Constants */
|
|
|
|
|
|
|
|
static const char* checkseps = "+,:;";
|
|
|
|
|
2023-10-09 09:59:28 +08:00
|
|
|
/*Define the set of protocols known to be constrainable */
|
|
|
|
static const char* constrainableprotocols[] = {"http", "https",NULL};
|
|
|
|
|
2017-03-09 08:01:10 +08:00
|
|
|
/**************************************************/
|
|
|
|
int
|
|
|
|
NCD4_open(const char * path, int mode,
|
2018-09-23 09:22:34 +08:00
|
|
|
int basepe, size_t *chunksizehintp,
|
2019-08-02 04:30:20 +08:00
|
|
|
void *mpidata, const NC_Dispatch *dispatch, int ncid)
|
2017-03-09 08:01:10 +08:00
|
|
|
{
|
|
|
|
int ret = NC_NOERR;
|
|
|
|
NCD4INFO* d4info = NULL;
|
|
|
|
const char* value;
|
2019-08-02 02:46:12 +08:00
|
|
|
NC* nc;
|
2021-01-15 12:39:08 +08:00
|
|
|
size_t len = 0;
|
|
|
|
void* contents = NULL;
|
2023-10-09 09:59:28 +08:00
|
|
|
NCD4response* dmrresp = NULL;
|
|
|
|
|
2017-03-09 08:01:10 +08:00
|
|
|
if(path == NULL)
|
|
|
|
return THROW(NC_EDAPURL);
|
|
|
|
|
|
|
|
assert(dispatch != NULL);
|
|
|
|
|
2019-08-02 02:46:12 +08:00
|
|
|
/* Find pointer to NC struct for this file. */
|
2019-08-02 04:30:20 +08:00
|
|
|
ret = NC_check_id(ncid,&nc);
|
2019-08-02 02:46:12 +08:00
|
|
|
if(ret != NC_NOERR) {goto done;}
|
|
|
|
|
2017-03-09 08:01:10 +08:00
|
|
|
/* Setup our NC and NCDAPCOMMON state*/
|
|
|
|
|
2023-10-09 09:59:28 +08:00
|
|
|
if((ret=NCD4_newInfo(&d4info))) goto done;
|
2017-03-09 08:01:10 +08:00
|
|
|
nc->dispatchdata = d4info;
|
|
|
|
nc->int_ncid = nc__pseudofd(); /* create a unique id */
|
|
|
|
d4info->controller = (NC*)nc;
|
|
|
|
|
|
|
|
/* Parse url and params */
|
2023-10-09 09:59:28 +08:00
|
|
|
if(ncuriparse(nc->path,&d4info->dmruri))
|
2017-03-09 08:01:10 +08:00
|
|
|
{ret = NC_EDAPURL; goto done;}
|
|
|
|
|
2017-09-01 04:19:56 +08:00
|
|
|
/* Load auth info from rc file */
|
2023-10-09 09:59:28 +08:00
|
|
|
if((ret = NC_authsetup(&d4info->auth, d4info->dmruri)))
|
2017-09-01 04:19:56 +08:00
|
|
|
goto done;
|
|
|
|
NCD4_curl_protocols(d4info);
|
|
|
|
|
2023-10-09 09:59:28 +08:00
|
|
|
if(!constrainable(d4info->dmruri))
|
2017-03-09 08:01:10 +08:00
|
|
|
SETFLAG(d4info->controls.flags,NCF_UNCONSTRAINABLE);
|
|
|
|
|
|
|
|
/* fail if we are unconstrainable but have constraints */
|
|
|
|
if(FLAGSET(d4info->controls.flags,NCF_UNCONSTRAINABLE)) {
|
2023-10-09 09:59:28 +08:00
|
|
|
if(d4info->dmruri != NULL) {
|
|
|
|
const char* ce = ncuriquerylookup(d4info->dmruri,DAP4CE); /* Look for dap4.ce */
|
2022-11-14 04:15:11 +08:00
|
|
|
if(ce != NULL) {
|
|
|
|
nclog(NCLOGWARN,"Attempt to constrain an unconstrainable data source: %s=%s",
|
|
|
|
DAP4CE,ce);
|
|
|
|
ret = THROW(NC_EDAPCONSTRAINT);
|
|
|
|
goto done;
|
|
|
|
}
|
2017-03-09 08:01:10 +08:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2022-11-14 04:15:11 +08:00
|
|
|
/* process control client fragment parameters */
|
|
|
|
NCD4_applyclientfragmentcontrols(d4info);
|
2017-03-09 08:01:10 +08:00
|
|
|
|
|
|
|
/* Use libsrc4 code (netcdf-4) for storing metadata */
|
|
|
|
{
|
|
|
|
char tmpname[NC_MAX_NAME];
|
|
|
|
|
|
|
|
/* Create fake file name: exact name must be unique,
|
|
|
|
but is otherwise irrelevant because we are using NC_DISKLESS
|
|
|
|
*/
|
|
|
|
if(strlen(d4info->controls.substratename) > 0)
|
|
|
|
snprintf(tmpname,sizeof(tmpname),"%s",d4info->controls.substratename);
|
|
|
|
else
|
|
|
|
snprintf(tmpname,sizeof(tmpname),"tmp_%d",nc->int_ncid);
|
|
|
|
|
2021-01-15 12:39:08 +08:00
|
|
|
/* Compute the relevant names for the substrate file */
|
|
|
|
d4info->substrate.filename = strdup(tmpname);
|
|
|
|
if(d4info->substrate.filename == NULL)
|
|
|
|
{ret = NC_ENOMEM; goto done;}
|
2017-03-09 08:01:10 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/* Turn on logging; only do this after oc_open*/
|
2023-10-09 09:59:28 +08:00
|
|
|
if((value = ncurifragmentlookup(d4info->dmruri,"log")) != NULL) {
|
2017-03-09 08:01:10 +08:00
|
|
|
ncloginit();
|
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
|
|
|
ncsetloglevel(NCLOGNOTE);
|
2017-03-09 08:01:10 +08:00
|
|
|
}
|
|
|
|
|
2021-05-30 11:30:33 +08:00
|
|
|
/* Check env values */
|
|
|
|
if(getenv("CURLOPT_VERBOSE") != NULL)
|
|
|
|
d4info->auth->curlflags.verbose = 1;
|
|
|
|
|
|
|
|
|
2017-03-09 08:01:10 +08:00
|
|
|
/* Setup a curl connection */
|
|
|
|
{
|
|
|
|
CURL* curl = NULL; /* curl handle*/
|
|
|
|
d4info->curl = (NCD4curl*)calloc(1,sizeof(NCD4curl));
|
|
|
|
if(d4info->curl == NULL)
|
|
|
|
{ret = NC_ENOMEM; goto done;}
|
|
|
|
/* create the connection */
|
|
|
|
if((ret=NCD4_curlopen(&curl))!= NC_NOERR) goto done;
|
|
|
|
d4info->curl->curl = curl;
|
2018-08-27 07:04:46 +08:00
|
|
|
/* Load misc rc properties */
|
|
|
|
NCD4_get_rcproperties(d4info);
|
2019-03-06 05:18:22 +08:00
|
|
|
if((ret=set_curl_properties(d4info))!= NC_NOERR) goto done;
|
2017-03-09 08:01:10 +08:00
|
|
|
/* Set the one-time curl flags */
|
|
|
|
if((ret=NCD4_set_flags_perlink(d4info))!= NC_NOERR) goto done;
|
|
|
|
#if 1 /* temporarily make per-link */
|
|
|
|
if((ret=NCD4_set_flags_perfetch(d4info))!= NC_NOERR) goto done;
|
|
|
|
#endif
|
|
|
|
}
|
|
|
|
|
|
|
|
d4info->curl->packet = ncbytesnew();
|
|
|
|
ncbytessetalloc(d4info->curl->packet,DFALTPACKETSIZE); /*initial reasonable size*/
|
|
|
|
|
2021-01-15 12:39:08 +08:00
|
|
|
/* Reset the substrate */
|
|
|
|
if((ret=makesubstrate(d4info))) goto done;
|
2017-03-09 08:01:10 +08:00
|
|
|
|
2023-10-09 09:59:28 +08:00
|
|
|
/* Always start by reading the whole DMR only */
|
2021-01-15 12:39:08 +08:00
|
|
|
/* reclaim substrate.metadata */
|
2023-10-09 09:59:28 +08:00
|
|
|
NCD4_resetInfoForRead(d4info);
|
2021-05-22 10:46:56 +08:00
|
|
|
/* Rebuild metadata */
|
2023-10-09 09:59:28 +08:00
|
|
|
if((ret = NCD4_newMeta(d4info,&d4info->dmrmetadata))) goto done;
|
|
|
|
|
|
|
|
/* Capture response */
|
|
|
|
if((dmrresp = (NCD4response*)calloc(1,sizeof(NCD4response)))==NULL)
|
|
|
|
{ret = NC_ENOMEM; goto done;}
|
|
|
|
dmrresp->controller = d4info;
|
2017-03-09 08:01:10 +08:00
|
|
|
|
2024-03-25 18:18:59 +08:00
|
|
|
if((ret=NCD4_readDMR(d4info, d4info->dmruri, dmrresp))) goto done;
|
2017-03-09 08:01:10 +08:00
|
|
|
|
2021-05-22 10:46:56 +08:00
|
|
|
/* set serial.rawdata */
|
2021-01-15 12:39:08 +08:00
|
|
|
len = ncbyteslength(d4info->curl->packet);
|
|
|
|
contents = ncbytesextract(d4info->curl->packet);
|
2023-10-09 09:59:28 +08:00
|
|
|
assert(dmrresp != NULL);
|
|
|
|
dmrresp->raw.size = len;
|
|
|
|
dmrresp->raw.memory = contents;
|
2017-03-09 08:01:10 +08:00
|
|
|
|
2023-10-09 09:59:28 +08:00
|
|
|
/* process checksum parameters */
|
|
|
|
NCD4_applychecksumcontrols(d4info,dmrresp);
|
2017-03-09 08:01:10 +08:00
|
|
|
|
|
|
|
/* Infer the mode */
|
2023-10-09 09:59:28 +08:00
|
|
|
if((ret=NCD4_infermode(dmrresp))) goto done;
|
2017-03-09 08:01:10 +08:00
|
|
|
|
2022-11-14 04:15:11 +08:00
|
|
|
/* Process the dmr part */
|
2023-10-09 09:59:28 +08:00
|
|
|
if((ret=NCD4_dechunk(dmrresp))) goto done;
|
2022-11-14 04:15:11 +08:00
|
|
|
|
2017-03-09 08:01:10 +08:00
|
|
|
#ifdef D4DUMPDMR
|
|
|
|
{
|
|
|
|
fprintf(stderr,"=============\n");
|
|
|
|
fputs(d4info->substrate.metadata->serial.dmr,stderr);
|
|
|
|
fprintf(stderr,"\n=============\n");
|
|
|
|
fflush(stderr);
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
2023-11-05 06:42:36 +08:00
|
|
|
if((ret = NCD4_parse(d4info->dmrmetadata,dmrresp,0))) goto done;
|
2021-01-15 12:39:08 +08:00
|
|
|
|
2017-03-09 08:01:10 +08:00
|
|
|
#ifdef D4DEBUGMETA
|
|
|
|
{
|
2023-11-05 06:42:36 +08:00
|
|
|
meta = d4info->dmrmetadata;
|
2017-03-09 08:01:10 +08:00
|
|
|
fprintf(stderr,"\n/////////////\n");
|
|
|
|
NCbytes* buf = ncbytesnew();
|
2023-10-09 09:59:28 +08:00
|
|
|
NCD4_print(meta,buf);
|
2017-03-09 08:01:10 +08:00
|
|
|
ncbytesnull(buf);
|
|
|
|
fputs(ncbytescontents(buf),stderr);
|
|
|
|
ncbytesfree(buf);
|
|
|
|
fprintf(stderr,"\n/////////////\n");
|
|
|
|
fflush(stderr);
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
2021-01-15 12:39:08 +08:00
|
|
|
/* Build the substrate metadata */
|
2023-11-05 10:23:55 +08:00
|
|
|
ret = NCD4_metabuild(d4info->dmrmetadata,d4info->dmrmetadata->ncid);
|
2020-05-31 07:36:25 +08:00
|
|
|
if(ret != NC_NOERR && ret != NC_EVARSIZE) goto done;
|
2017-03-09 08:01:10 +08:00
|
|
|
|
2023-10-09 09:59:28 +08:00
|
|
|
/* Remember the response */
|
|
|
|
nclistpush(d4info->responses,dmrresp);
|
|
|
|
|
|
|
|
/* Avoid duplicate reclaims */
|
|
|
|
dmrresp = NULL;
|
|
|
|
d4info = NULL;
|
|
|
|
|
2017-03-09 08:01:10 +08:00
|
|
|
done:
|
2023-10-09 09:59:28 +08:00
|
|
|
NCD4_reclaimResponse(dmrresp);
|
|
|
|
NCD4_reclaimInfo(d4info);
|
2017-03-09 08:01:10 +08:00
|
|
|
if(ret) {
|
|
|
|
nc->dispatchdata = NULL;
|
|
|
|
}
|
|
|
|
return THROW(ret);
|
|
|
|
}
|
|
|
|
|
|
|
|
int
|
2018-02-26 12:45:31 +08:00
|
|
|
NCD4_close(int ncid, void* ignore)
|
2017-03-09 08:01:10 +08:00
|
|
|
{
|
|
|
|
int ret = NC_NOERR;
|
|
|
|
NC* nc;
|
|
|
|
NCD4INFO* d4info;
|
|
|
|
int substrateid;
|
|
|
|
|
|
|
|
ret = NC_check_id(ncid, (NC**)&nc);
|
|
|
|
if(ret != NC_NOERR) goto done;
|
|
|
|
d4info = (NCD4INFO*)nc->dispatchdata;
|
|
|
|
substrateid = makenc4id(nc,ncid);
|
|
|
|
|
|
|
|
/* We call abort rather than close to avoid trying to write anything,
|
|
|
|
except if we are debugging
|
|
|
|
*/
|
|
|
|
if(FLAGSET(d4info->controls.debugflags,NCF_DEBUG_COPY)) {
|
|
|
|
/* Dump the data into the substrate */
|
|
|
|
if((ret = NCD4_debugcopy(d4info)))
|
|
|
|
goto done;
|
|
|
|
ret = nc_close(substrateid);
|
|
|
|
} else {
|
|
|
|
ret = nc_abort(substrateid);
|
|
|
|
}
|
|
|
|
|
2023-10-09 09:59:28 +08:00
|
|
|
NCD4_reclaimInfo(d4info);
|
2017-03-09 08:01:10 +08:00
|
|
|
|
|
|
|
done:
|
|
|
|
return THROW(ret);
|
|
|
|
}
|
|
|
|
|
|
|
|
int
|
|
|
|
NCD4_abort(int ncid)
|
|
|
|
{
|
2018-02-26 12:45:31 +08:00
|
|
|
return NCD4_close(ncid,NULL);
|
2017-03-09 08:01:10 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
|
|
|
constrainable(NCURI* durl)
|
|
|
|
{
|
2018-10-02 05:51:43 +08:00
|
|
|
const char** protocol = constrainableprotocols;
|
2017-03-09 08:01:10 +08:00
|
|
|
for(;*protocol;protocol++) {
|
|
|
|
if(strcmp(durl->protocol,*protocol)==0)
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
Set curl properties for link based on rc files etc.
|
|
|
|
*/
|
|
|
|
static int
|
|
|
|
set_curl_properties(NCD4INFO* d4info)
|
|
|
|
{
|
|
|
|
int ret = NC_NOERR;
|
|
|
|
|
2020-11-20 08:01:04 +08:00
|
|
|
if(d4info->auth->curlflags.useragent == NULL) {
|
2017-11-24 01:55:24 +08:00
|
|
|
char* agent;
|
|
|
|
size_t len = strlen(DFALTUSERAGENT) + strlen(VERSION);
|
|
|
|
len++; /*strlcat nul*/
|
|
|
|
agent = (char*)malloc(len+1);
|
2017-03-09 08:01:10 +08:00
|
|
|
strncpy(agent,DFALTUSERAGENT,len);
|
2017-11-24 01:55:24 +08:00
|
|
|
strlcat(agent,VERSION,len);
|
2020-11-20 08:01:04 +08:00
|
|
|
d4info->auth->curlflags.useragent = agent;
|
2017-03-09 08:01:10 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/* Some servers (e.g. thredds and columbia) appear to require a place
|
|
|
|
to put cookies in order for some security functions to work
|
|
|
|
*/
|
2020-11-20 08:01:04 +08:00
|
|
|
if(d4info->auth->curlflags.cookiejar != NULL
|
|
|
|
&& strlen(d4info->auth->curlflags.cookiejar) == 0) {
|
|
|
|
free(d4info->auth->curlflags.cookiejar);
|
|
|
|
d4info->auth->curlflags.cookiejar = NULL;
|
2017-03-09 08:01:10 +08:00
|
|
|
}
|
|
|
|
|
2020-11-20 08:01:04 +08:00
|
|
|
if(d4info->auth->curlflags.cookiejar == NULL) {
|
2017-03-09 08:01:10 +08:00
|
|
|
/* If no cookie file was defined, define a default */
|
2017-07-06 00:03:48 +08:00
|
|
|
char* path = NULL;
|
2017-09-04 05:09:10 +08:00
|
|
|
char* newpath = NULL;
|
2023-11-25 02:20:52 +08:00
|
|
|
size_t len;
|
2017-03-09 08:01:10 +08:00
|
|
|
errno = 0;
|
2022-01-30 06:27:52 +08:00
|
|
|
NCglobalstate* globalstate = NC_getglobalstate();
|
2019-03-31 04:06:20 +08:00
|
|
|
|
2017-03-09 08:01:10 +08:00
|
|
|
/* Create the unique cookie file name */
|
2017-07-06 00:03:48 +08:00
|
|
|
len =
|
2019-03-31 04:06:20 +08:00
|
|
|
strlen(globalstate->tempdir)
|
2017-07-06 00:03:48 +08:00
|
|
|
+ 1 /* '/' */
|
|
|
|
+ strlen("ncd4cookies");
|
|
|
|
path = (char*)malloc(len+1);
|
|
|
|
if(path == NULL) return NC_ENOMEM;
|
2019-03-31 04:06:20 +08:00
|
|
|
snprintf(path,len,"%s/nc4cookies",globalstate->tempdir);
|
2017-07-06 00:03:48 +08:00
|
|
|
/* Create the unique cookie file name */
|
2017-09-04 05:09:10 +08:00
|
|
|
newpath = NC_mktmp(path);
|
2017-07-06 00:03:48 +08:00
|
|
|
free(path);
|
2017-09-04 05:09:10 +08:00
|
|
|
if(newpath == NULL) {
|
2017-03-09 08:01:10 +08:00
|
|
|
fprintf(stderr,"Cannot create cookie file\n");
|
|
|
|
goto fail;
|
|
|
|
}
|
2020-11-20 08:01:04 +08:00
|
|
|
d4info->auth->curlflags.cookiejar = newpath;
|
|
|
|
d4info->auth->curlflags.cookiejarcreated = 1;
|
2017-03-09 08:01:10 +08:00
|
|
|
errno = 0;
|
|
|
|
}
|
2020-11-20 08:01:04 +08:00
|
|
|
assert(d4info->auth->curlflags.cookiejar != NULL);
|
2017-03-09 08:01:10 +08:00
|
|
|
|
|
|
|
/* Make sure the cookie jar exists and can be read and written */
|
|
|
|
{
|
|
|
|
FILE* f = NULL;
|
2020-11-20 08:01:04 +08:00
|
|
|
char* fname = d4info->auth->curlflags.cookiejar;
|
2017-03-09 08:01:10 +08:00
|
|
|
/* See if the file exists already */
|
2022-02-09 11:53:30 +08:00
|
|
|
f = NCfopen(fname,"r");
|
2017-03-09 08:01:10 +08:00
|
|
|
if(f == NULL) {
|
|
|
|
/* Ok, create it */
|
2022-02-09 11:53:30 +08:00
|
|
|
f = NCfopen(fname,"w+");
|
2017-03-09 08:01:10 +08:00
|
|
|
if(f == NULL) {
|
|
|
|
fprintf(stderr,"Cookie file cannot be read and written: %s\n",fname);
|
|
|
|
{ret= NC_EPERM; goto fail;}
|
|
|
|
}
|
|
|
|
} else { /* test if file can be written */
|
|
|
|
fclose(f);
|
2022-02-09 11:53:30 +08:00
|
|
|
f = NCfopen(fname,"r+");
|
2017-03-09 08:01:10 +08:00
|
|
|
if(f == NULL) {
|
|
|
|
fprintf(stderr,"Cookie file is cannot be written: %s\n",fname);
|
|
|
|
{ret = NC_EPERM; goto fail;}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if(f != NULL) fclose(f);
|
|
|
|
}
|
|
|
|
|
|
|
|
return THROW(ret);
|
|
|
|
|
|
|
|
fail:
|
|
|
|
return THROW(ret);
|
|
|
|
}
|
|
|
|
|
2018-10-31 10:48:12 +08:00
|
|
|
void
|
2022-11-14 04:15:11 +08:00
|
|
|
NCD4_applyclientfragmentcontrols(NCD4INFO* info)
|
2017-03-09 08:01:10 +08:00
|
|
|
{
|
|
|
|
const char* value;
|
|
|
|
|
2022-11-14 04:15:11 +08:00
|
|
|
/* clear all flags */
|
2023-01-19 10:47:29 +08:00
|
|
|
CLRFLAG(info->controls.flags,ALL_NCF_FLAGS);
|
2017-03-09 08:01:10 +08:00
|
|
|
|
|
|
|
/* Turn on any default on flags */
|
|
|
|
SETFLAG(info->controls.flags,DFALT_ON_FLAGS);
|
2023-01-19 10:47:29 +08:00
|
|
|
|
|
|
|
/* Set these also */
|
2017-03-09 08:01:10 +08:00
|
|
|
SETFLAG(info->controls.flags,(NCF_NC4|NCF_NCDAP));
|
|
|
|
|
2022-11-14 04:15:11 +08:00
|
|
|
if(fragmentcheck(info,"show","fetch"))
|
2017-03-09 08:01:10 +08:00
|
|
|
SETFLAG(info->controls.flags,NCF_SHOWFETCH);
|
|
|
|
|
2022-11-14 04:15:11 +08:00
|
|
|
if(fragmentcheck(info,"translate","nc4"))
|
2017-03-09 08:01:10 +08:00
|
|
|
info->controls.translation = NCD4_TRANSNC4;
|
|
|
|
|
|
|
|
/* Look at the debug flags */
|
2022-11-14 04:15:11 +08:00
|
|
|
if(fragmentcheck(info,"debug","copy"))
|
2017-03-09 08:01:10 +08:00
|
|
|
SETFLAG(info->controls.debugflags,NCF_DEBUG_COPY); /* => close */
|
|
|
|
|
2022-11-14 04:15:11 +08:00
|
|
|
value = getfragment(info,"substratename");
|
2017-03-09 08:01:10 +08:00
|
|
|
if(value != NULL)
|
2019-03-06 05:18:22 +08:00
|
|
|
strncpy(info->controls.substratename,value,(NC_MAX_NAME-1));
|
2017-03-09 08:01:10 +08:00
|
|
|
|
2018-10-31 10:48:12 +08:00
|
|
|
info->controls.opaquesize = DFALTOPAQUESIZE;
|
2022-11-14 04:15:11 +08:00
|
|
|
value = getfragment(info,"opaquesize");
|
2018-10-31 10:48:12 +08:00
|
|
|
if(value != NULL) {
|
|
|
|
long long len = 0;
|
|
|
|
if(sscanf(value,"%lld",&len) != 1 || len == 0)
|
|
|
|
nclog(NCLOGWARN,"bad [opaquesize] tag: %s",value);
|
|
|
|
else
|
2019-03-06 05:18:22 +08:00
|
|
|
info->controls.opaquesize = (size_t)len;
|
2018-10-31 10:48:12 +08:00
|
|
|
}
|
2019-03-06 05:18:22 +08:00
|
|
|
|
2022-11-14 04:15:11 +08:00
|
|
|
value = getfragment(info,"fillmismatch");
|
|
|
|
if(value != NULL) {
|
2018-10-02 05:51:43 +08:00
|
|
|
SETFLAG(info->controls.flags,NCF_FILLMISMATCH);
|
2022-11-14 04:15:11 +08:00
|
|
|
CLRFLAG(info->controls.debugflags,NCF_FILLMISMATCH_FAIL);
|
|
|
|
}
|
|
|
|
value = getfragment(info,"nofillmismatch");
|
|
|
|
if(value != NULL) {
|
2018-10-02 05:51:43 +08:00
|
|
|
CLRFLAG(info->controls.debugflags,NCF_FILLMISMATCH);
|
2022-11-14 04:15:11 +08:00
|
|
|
SETFLAG(info->controls.debugflags,NCF_FILLMISMATCH_FAIL);
|
|
|
|
}
|
2017-03-09 08:01:10 +08:00
|
|
|
}
|
|
|
|
|
2023-10-09 09:59:28 +08:00
|
|
|
/* Checksum controls are found both in the query and fragment
|
|
|
|
parts of a URL.
|
|
|
|
*/
|
2022-11-14 04:15:11 +08:00
|
|
|
void
|
2023-10-09 09:59:28 +08:00
|
|
|
NCD4_applychecksumcontrols(NCD4INFO* info, NCD4response* resp)
|
2017-03-09 08:01:10 +08:00
|
|
|
{
|
2023-01-19 10:47:29 +08:00
|
|
|
const char* value = getquery(info,DAP4CSUM);
|
|
|
|
if(value == NULL) {
|
2023-10-09 09:59:28 +08:00
|
|
|
resp->querychecksumming = DEFAULT_CHECKSUM_STATE;
|
2023-01-19 10:47:29 +08:00
|
|
|
} else {
|
|
|
|
if(strcasecmp(value,"false")==0) {
|
2023-10-09 09:59:28 +08:00
|
|
|
resp->querychecksumming = 0;
|
2023-01-19 10:47:29 +08:00
|
|
|
} else if(strcasecmp(value,"true")==0) {
|
2023-10-09 09:59:28 +08:00
|
|
|
resp->querychecksumming = 1;
|
2023-01-19 10:47:29 +08:00
|
|
|
} else {
|
|
|
|
nclog(NCLOGWARN,"Unknown checksum mode: %s ; using default",value);
|
2023-10-09 09:59:28 +08:00
|
|
|
resp->querychecksumming = DEFAULT_CHECKSUM_STATE;
|
2023-01-19 10:47:29 +08:00
|
|
|
}
|
2017-03-09 08:01:10 +08:00
|
|
|
}
|
2023-10-09 09:59:28 +08:00
|
|
|
value = getfragment(info,"hyrax");
|
|
|
|
if(value != NULL) {
|
|
|
|
resp->checksumignore = 1; /* Assume checksum, but ignore */
|
|
|
|
}
|
2017-03-09 08:01:10 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/* Search for substring in value of param. If substring == NULL; then just
|
2022-11-14 04:15:11 +08:00
|
|
|
check if fragment is defined.
|
2017-03-09 08:01:10 +08:00
|
|
|
*/
|
|
|
|
static int
|
2022-11-14 04:15:11 +08:00
|
|
|
fragmentcheck(NCD4INFO* info, const char* key, const char* subkey)
|
2017-03-09 08:01:10 +08:00
|
|
|
{
|
|
|
|
const char* value;
|
|
|
|
char* p;
|
|
|
|
|
2022-11-14 04:15:11 +08:00
|
|
|
value = getfragment(info, key);
|
2017-03-09 08:01:10 +08:00
|
|
|
if(value == NULL)
|
|
|
|
return 0;
|
|
|
|
if(subkey == NULL) return 1;
|
|
|
|
p = strstr(value,subkey);
|
|
|
|
if(p == NULL) return 0;
|
|
|
|
p += strlen(subkey);
|
|
|
|
if(*p != '\0' && strchr(checkseps,*p) == NULL) return 0;
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2022-11-14 04:15:11 +08:00
|
|
|
Given a fragment key, return its value or NULL if not defined.
|
2017-03-09 08:01:10 +08:00
|
|
|
*/
|
|
|
|
static const char*
|
2022-11-14 04:15:11 +08:00
|
|
|
getfragment(NCD4INFO* info, const char* key)
|
2017-03-09 08:01:10 +08:00
|
|
|
{
|
|
|
|
const char* value;
|
|
|
|
|
|
|
|
if(info == NULL || key == NULL) return NULL;
|
2023-10-09 09:59:28 +08:00
|
|
|
if((value=ncurifragmentlookup(info->dmruri,key)) == NULL)
|
2017-03-09 08:01:10 +08:00
|
|
|
return NULL;
|
|
|
|
return value;
|
|
|
|
}
|
2021-01-15 12:39:08 +08:00
|
|
|
|
2022-11-14 04:15:11 +08:00
|
|
|
/*
|
|
|
|
Given a query key, return its value or NULL if not defined.
|
|
|
|
*/
|
|
|
|
static const char*
|
|
|
|
getquery(NCD4INFO* info, const char* key)
|
|
|
|
{
|
|
|
|
const char* value;
|
|
|
|
|
|
|
|
if(info == NULL || key == NULL) return NULL;
|
2023-10-09 09:59:28 +08:00
|
|
|
if((value=ncuriquerylookup(info->dmruri,key)) == NULL)
|
2022-11-14 04:15:11 +08:00
|
|
|
return NULL;
|
|
|
|
return value;
|
|
|
|
}
|
|
|
|
|
2021-01-15 12:39:08 +08:00
|
|
|
/**************************************************/
|
|
|
|
|
|
|
|
static int
|
|
|
|
makesubstrate(NCD4INFO* d4info)
|
|
|
|
{
|
|
|
|
int ret = NC_NOERR;
|
|
|
|
int new = NC_NETCDF4;
|
|
|
|
int old = 0;
|
|
|
|
int ncid = 0;
|
|
|
|
int ncflags = NC_NETCDF4|NC_CLOBBER;
|
|
|
|
|
Fix various problem around VLEN's
re: https://github.com/Unidata/netcdf-c/issues/541
re: https://github.com/Unidata/netcdf-c/issues/1208
re: https://github.com/Unidata/netcdf-c/issues/2078
re: https://github.com/Unidata/netcdf-c/issues/2041
re: https://github.com/Unidata/netcdf-c/issues/2143
For a long time, there have been known problems with the
management of complex types containing VLENs. This also
involves the string type because it is stored as a VLEN of
chars.
This PR (mostly) fixes this problem. But note that it adds new
functions to netcdf.h (see below) and this may require bumping
the .so number. These new functions can be removed, if desired,
in favor of functions in netcdf_aux.h, but netcdf.h seems the
better place for them because they are intended as alternatives
to the nc_free_vlen and nc_free_string functions already in
netcdf.h.
The term complex type refers to any type that directly or
transitively references a VLEN type. So an array of VLENS, a
compound with a VLEN field, and so on.
In order to properly handle instances of these complex types, it
is necessary to have function that can recursively walk
instances of such types to perform various actions on them. The
term "deep" is also used to mean recursive.
At the moment, the two operations needed by the netcdf library are:
* free'ing an instance of the complex type
* copying an instance of the complex type.
The current library does only shallow free and shallow copy of
complex types. This means that only the top level is properly
free'd or copied, but deep internal blocks in the instance are
not touched.
Note that the term "vector" will be used to mean a contiguous (in
memory) sequence of instances of some type. Given an array with,
say, dimensions 2 X 3 X 4, this will be stored in memory as a
vector of length 2*3*4=24 instances.
The use cases are primarily these.
## nc_get_vars
Suppose one is reading a vector of instances using nc_get_vars
(or nc_get_vara or nc_get_var, etc.). These functions will
return the vector in the top-level memory provided. All
interior blocks (form nested VLEN or strings) will have been
dynamically allocated.
After using this vector of instances, it is necessary to free
(aka reclaim) the dynamically allocated memory, otherwise a
memory leak occurs. So, the recursive reclaim function is used
to walk the returned instance vector and do a deep reclaim of
the data.
Currently functions are defined in netcdf.h that are supposed to
handle this: nc_free_vlen(), nc_free_vlens(), and
nc_free_string(). Unfortunately, these functions only do a
shallow free, so deeply nested instances are not properly
handled by them.
Note that internally, the provided data is immediately written so
there is no need to copy it. But the caller may need to reclaim the
data it passed into the function.
## nc_put_att
Suppose one is writing a vector of instances as the data of an attribute
using, say, nc_put_att.
Internally, the incoming attribute data must be copied and stored
so that changes/reclamation of the input data will not affect
the attribute.
Again, the code inside the netcdf library does only shallow copying
rather than deep copy. As a result, one sees effects such as described
in Github Issue https://github.com/Unidata/netcdf-c/issues/2143.
Also, after defining the attribute, it may be necessary for the user
to free the data that was provided as input to nc_put_att().
## nc_get_att
Suppose one is reading a vector of instances as the data of an attribute
using, say, nc_get_att.
Internally, the existing attribute data must be copied and returned
to the caller, and the caller is responsible for reclaiming
the returned data.
Again, the code inside the netcdf library does only shallow copying
rather than deep copy. So this can lead to memory leaks and errors
because the deep data is shared between the library and the user.
# Solution
The solution is to build properly recursive reclaim and copy
functions and use those as needed.
These recursive functions are defined in libdispatch/dinstance.c
and their signatures are defined in include/netcdf.h.
For back compatibility, corresponding "ncaux_XXX" functions
are defined in include/netcdf_aux.h.
````
int nc_reclaim_data(int ncid, nc_type xtypeid, void* memory, size_t count);
int nc_reclaim_data_all(int ncid, nc_type xtypeid, void* memory, size_t count);
int nc_copy_data(int ncid, nc_type xtypeid, const void* memory, size_t count, void* copy);
int nc_copy_data_all(int ncid, nc_type xtypeid, const void* memory, size_t count, void** copyp);
````
There are two variants. The first two, nc_reclaim_data() and
nc_copy_data(), assume the top-level vector is managed by the
caller. For reclaim, this is so the user can use, for example, a
statically allocated vector. For copy, it assumes the user
provides the space into which the copy is stored.
The second two, nc_reclaim_data_all() and
nc_copy_data_all(), allows the functions to manage the
top-level. So for nc_reclaim_data_all, the top level is
assumed to be dynamically allocated and will be free'd by
nc_reclaim_data_all(). The nc_copy_data_all() function
will allocate the top level and return a pointer to it to the
user. The user can later pass that pointer to
nc_reclaim_data_all() to reclaim the instance(s).
# Internal Changes
The netcdf-c library internals are changed to use the proper
reclaim and copy functions. It turns out that the places where
these functions are needed is quite pervasive in the netcdf-c
library code. Using these functions also allows some
simplification of the code since the stdata and vldata fields of
NC_ATT_INFO are no longer needed. Currently this is commented
out using the SEPDATA \#define macro. When any bugs are largely
fixed, all this code will be removed.
# Known Bugs
1. There is still one known failure that has not been solved.
All the failures revolve around some variant of this .cdl file.
The proximate cause of failure is the use of a VLEN FillValue.
````
netcdf x {
types:
float(*) row_of_floats ;
dimensions:
m = 5 ;
variables:
row_of_floats ragged_array(m) ;
row_of_floats ragged_array:_FillValue = {-999} ;
data:
ragged_array = {10, 11, 12, 13, 14}, {20, 21, 22, 23}, {30, 31, 32},
{40, 41}, _ ;
}
````
When a solution is found, I will either add it to this PR or post a new PR.
# Related Changes
* Mark nc_free_vlen(s) as deprecated in favor of ncaux_reclaim_data.
* Remove the --enable-unfixed-memory-leaks option.
* Remove the NC_VLENS_NOTEST code that suppresses some vlen tests.
* Document this change in docs/internal.md
* Disable the tst_vlen_data test in ncdump/tst_nccopy4.sh.
* Mark types as fixed size or not (transitively) to optimize the reclaim
and copy functions.
# Misc. Changes
* Make Doxygen process libdispatch/daux.c
* Make sure the NC_ATT_INFO_T.container field is set.
2022-01-09 09:30:00 +08:00
|
|
|
if(d4info->substrate.nc4id != 0) {
|
2021-01-15 12:39:08 +08:00
|
|
|
/* reset the substrate */
|
|
|
|
nc_abort(d4info->substrate.nc4id);
|
|
|
|
d4info->substrate.nc4id = 0;
|
|
|
|
}
|
|
|
|
/* Create the hidden substrate netcdf file.
|
|
|
|
We want this hidden file to always be NC_NETCDF4, so we need to
|
|
|
|
force default format temporarily in case user changed it.
|
|
|
|
Since diskless is enabled, create file in-memory.
|
|
|
|
*/
|
|
|
|
ncflags |= NC_DISKLESS;
|
|
|
|
if(FLAGSET(d4info->controls.debugflags,NCF_DEBUG_COPY)) {
|
|
|
|
/* Cause data to be dumped to real file */
|
|
|
|
ncflags |= NC_WRITE;
|
|
|
|
ncflags &= ~(NC_DISKLESS); /* use real file */
|
|
|
|
}
|
|
|
|
nc_set_default_format(new,&old); /* save and change */
|
|
|
|
ret = nc_create(d4info->substrate.filename,ncflags,&ncid);
|
|
|
|
nc_set_default_format(old,&new); /* restore */
|
|
|
|
/* Avoid fill on the substrate */
|
|
|
|
nc_set_fill(ncid,NC_NOFILL,NULL);
|
|
|
|
d4info->substrate.nc4id = ncid;
|
|
|
|
return THROW(ret);
|
|
|
|
}
|
|
|
|
|
Improve performance of the nc_reclaim_data and nc_copy_data functions.
re: Issue https://github.com/Unidata/netcdf-c/issues/2685
re: PR https://github.com/Unidata/netcdf-c/pull/2179
As noted in PR https://github.com/Unidata/netcdf-c/pull/2179,
the old code did not allow for reclaiming instances of types,
nor for properly copying them. That PR provided new functions
capable of reclaiming/copying instances of arbitrary types.
However, as noted by Issue https://github.com/Unidata/netcdf-c/issues/2685, using these
most general functions resulted in a significant performance
degradation, even for common cases.
This PR attempts to mitigate the cost of using the general
reclaim/copy functions in two ways.
First, the previous functions operating at the top level by
using ncid and typeid arguments. These functions were augmented
with equivalent versions that used the netcdf-c library internal
data structures to allow direct access to needed information.
These new functions are used internally to the library.
The second mitigation involves optimizing the internal functions
by providing early tests for common cases. This avoids
unnecessary recursive function calls.
The overall result is a significant improvement in speed by a
factor of roughly twenty -- your mileage may vary. These
optimized functions are still not as fast as the original (more
limited) functions, but they are getting close. Additional optimizations are
possible. But the cost is a significant "uglification" of the
code that I deemed a step too far, at least for now.
## Misc. Changes
1. Added a test case to check the proper reclamation/copy of complex types.
2. Found and fixed some places where nc_reclaim/copy should have been used.
3. Replaced, in the netcdf-c library, (almost all) occurrences of nc_reclaim_copy with calls to NC_reclaim/copy. This plus the optimizations is the primary speed-up mechanism.
4. In DAP4, the metadata is held in a substrate in-memory file; this required some changes so that the reclaim/copy code accessed that substrate dispatcher rather than the DAP4 dispatcher.
5. Re-factored and isolated the code that computes if a type is (transitively) variable-sized or not.
6. Clean up the reclamation code in ncgen; adding the use of nc_reclaim exposed some memory problems.
2023-05-21 07:11:25 +08:00
|
|
|
/* This function breaks the abstraction, but is necessary for
|
|
|
|
code that accesses the underlying netcdf-4 metadata objects.
|
|
|
|
Used in: NC_reclaim_data[_all]
|
|
|
|
NC_copy_data[_all]
|
|
|
|
*/
|
|
|
|
|
|
|
|
NC*
|
|
|
|
NCD4_get_substrate(NC* nc)
|
Fix various problem around VLEN's
re: https://github.com/Unidata/netcdf-c/issues/541
re: https://github.com/Unidata/netcdf-c/issues/1208
re: https://github.com/Unidata/netcdf-c/issues/2078
re: https://github.com/Unidata/netcdf-c/issues/2041
re: https://github.com/Unidata/netcdf-c/issues/2143
For a long time, there have been known problems with the
management of complex types containing VLENs. This also
involves the string type because it is stored as a VLEN of
chars.
This PR (mostly) fixes this problem. But note that it adds new
functions to netcdf.h (see below) and this may require bumping
the .so number. These new functions can be removed, if desired,
in favor of functions in netcdf_aux.h, but netcdf.h seems the
better place for them because they are intended as alternatives
to the nc_free_vlen and nc_free_string functions already in
netcdf.h.
The term complex type refers to any type that directly or
transitively references a VLEN type. So an array of VLENS, a
compound with a VLEN field, and so on.
In order to properly handle instances of these complex types, it
is necessary to have function that can recursively walk
instances of such types to perform various actions on them. The
term "deep" is also used to mean recursive.
At the moment, the two operations needed by the netcdf library are:
* free'ing an instance of the complex type
* copying an instance of the complex type.
The current library does only shallow free and shallow copy of
complex types. This means that only the top level is properly
free'd or copied, but deep internal blocks in the instance are
not touched.
Note that the term "vector" will be used to mean a contiguous (in
memory) sequence of instances of some type. Given an array with,
say, dimensions 2 X 3 X 4, this will be stored in memory as a
vector of length 2*3*4=24 instances.
The use cases are primarily these.
## nc_get_vars
Suppose one is reading a vector of instances using nc_get_vars
(or nc_get_vara or nc_get_var, etc.). These functions will
return the vector in the top-level memory provided. All
interior blocks (form nested VLEN or strings) will have been
dynamically allocated.
After using this vector of instances, it is necessary to free
(aka reclaim) the dynamically allocated memory, otherwise a
memory leak occurs. So, the recursive reclaim function is used
to walk the returned instance vector and do a deep reclaim of
the data.
Currently functions are defined in netcdf.h that are supposed to
handle this: nc_free_vlen(), nc_free_vlens(), and
nc_free_string(). Unfortunately, these functions only do a
shallow free, so deeply nested instances are not properly
handled by them.
Note that internally, the provided data is immediately written so
there is no need to copy it. But the caller may need to reclaim the
data it passed into the function.
## nc_put_att
Suppose one is writing a vector of instances as the data of an attribute
using, say, nc_put_att.
Internally, the incoming attribute data must be copied and stored
so that changes/reclamation of the input data will not affect
the attribute.
Again, the code inside the netcdf library does only shallow copying
rather than deep copy. As a result, one sees effects such as described
in Github Issue https://github.com/Unidata/netcdf-c/issues/2143.
Also, after defining the attribute, it may be necessary for the user
to free the data that was provided as input to nc_put_att().
## nc_get_att
Suppose one is reading a vector of instances as the data of an attribute
using, say, nc_get_att.
Internally, the existing attribute data must be copied and returned
to the caller, and the caller is responsible for reclaiming
the returned data.
Again, the code inside the netcdf library does only shallow copying
rather than deep copy. So this can lead to memory leaks and errors
because the deep data is shared between the library and the user.
# Solution
The solution is to build properly recursive reclaim and copy
functions and use those as needed.
These recursive functions are defined in libdispatch/dinstance.c
and their signatures are defined in include/netcdf.h.
For back compatibility, corresponding "ncaux_XXX" functions
are defined in include/netcdf_aux.h.
````
int nc_reclaim_data(int ncid, nc_type xtypeid, void* memory, size_t count);
int nc_reclaim_data_all(int ncid, nc_type xtypeid, void* memory, size_t count);
int nc_copy_data(int ncid, nc_type xtypeid, const void* memory, size_t count, void* copy);
int nc_copy_data_all(int ncid, nc_type xtypeid, const void* memory, size_t count, void** copyp);
````
There are two variants. The first two, nc_reclaim_data() and
nc_copy_data(), assume the top-level vector is managed by the
caller. For reclaim, this is so the user can use, for example, a
statically allocated vector. For copy, it assumes the user
provides the space into which the copy is stored.
The second two, nc_reclaim_data_all() and
nc_copy_data_all(), allows the functions to manage the
top-level. So for nc_reclaim_data_all, the top level is
assumed to be dynamically allocated and will be free'd by
nc_reclaim_data_all(). The nc_copy_data_all() function
will allocate the top level and return a pointer to it to the
user. The user can later pass that pointer to
nc_reclaim_data_all() to reclaim the instance(s).
# Internal Changes
The netcdf-c library internals are changed to use the proper
reclaim and copy functions. It turns out that the places where
these functions are needed is quite pervasive in the netcdf-c
library code. Using these functions also allows some
simplification of the code since the stdata and vldata fields of
NC_ATT_INFO are no longer needed. Currently this is commented
out using the SEPDATA \#define macro. When any bugs are largely
fixed, all this code will be removed.
# Known Bugs
1. There is still one known failure that has not been solved.
All the failures revolve around some variant of this .cdl file.
The proximate cause of failure is the use of a VLEN FillValue.
````
netcdf x {
types:
float(*) row_of_floats ;
dimensions:
m = 5 ;
variables:
row_of_floats ragged_array(m) ;
row_of_floats ragged_array:_FillValue = {-999} ;
data:
ragged_array = {10, 11, 12, 13, 14}, {20, 21, 22, 23}, {30, 31, 32},
{40, 41}, _ ;
}
````
When a solution is found, I will either add it to this PR or post a new PR.
# Related Changes
* Mark nc_free_vlen(s) as deprecated in favor of ncaux_reclaim_data.
* Remove the --enable-unfixed-memory-leaks option.
* Remove the NC_VLENS_NOTEST code that suppresses some vlen tests.
* Document this change in docs/internal.md
* Disable the tst_vlen_data test in ncdump/tst_nccopy4.sh.
* Mark types as fixed size or not (transitively) to optimize the reclaim
and copy functions.
# Misc. Changes
* Make Doxygen process libdispatch/daux.c
* Make sure the NC_ATT_INFO_T.container field is set.
2022-01-09 09:30:00 +08:00
|
|
|
{
|
Improve performance of the nc_reclaim_data and nc_copy_data functions.
re: Issue https://github.com/Unidata/netcdf-c/issues/2685
re: PR https://github.com/Unidata/netcdf-c/pull/2179
As noted in PR https://github.com/Unidata/netcdf-c/pull/2179,
the old code did not allow for reclaiming instances of types,
nor for properly copying them. That PR provided new functions
capable of reclaiming/copying instances of arbitrary types.
However, as noted by Issue https://github.com/Unidata/netcdf-c/issues/2685, using these
most general functions resulted in a significant performance
degradation, even for common cases.
This PR attempts to mitigate the cost of using the general
reclaim/copy functions in two ways.
First, the previous functions operating at the top level by
using ncid and typeid arguments. These functions were augmented
with equivalent versions that used the netcdf-c library internal
data structures to allow direct access to needed information.
These new functions are used internally to the library.
The second mitigation involves optimizing the internal functions
by providing early tests for common cases. This avoids
unnecessary recursive function calls.
The overall result is a significant improvement in speed by a
factor of roughly twenty -- your mileage may vary. These
optimized functions are still not as fast as the original (more
limited) functions, but they are getting close. Additional optimizations are
possible. But the cost is a significant "uglification" of the
code that I deemed a step too far, at least for now.
## Misc. Changes
1. Added a test case to check the proper reclamation/copy of complex types.
2. Found and fixed some places where nc_reclaim/copy should have been used.
3. Replaced, in the netcdf-c library, (almost all) occurrences of nc_reclaim_copy with calls to NC_reclaim/copy. This plus the optimizations is the primary speed-up mechanism.
4. In DAP4, the metadata is held in a substrate in-memory file; this required some changes so that the reclaim/copy code accessed that substrate dispatcher rather than the DAP4 dispatcher.
5. Re-factored and isolated the code that computes if a type is (transitively) variable-sized or not.
6. Clean up the reclamation code in ncgen; adding the use of nc_reclaim exposed some memory problems.
2023-05-21 07:11:25 +08:00
|
|
|
NC* subnc = NULL;
|
|
|
|
/* Iff nc->dispatch is the DAP4 dispatcher, then do a level of indirection */
|
|
|
|
if(USED4INFO(nc)) {
|
|
|
|
NCD4INFO* d4 = (NCD4INFO*)nc->dispatchdata;
|
|
|
|
/* Find pointer to NC struct for this file. */
|
|
|
|
(void)NC_check_id(d4->substrate.nc4id,&subnc);
|
|
|
|
} else subnc = nc;
|
|
|
|
return subnc;
|
Fix various problem around VLEN's
re: https://github.com/Unidata/netcdf-c/issues/541
re: https://github.com/Unidata/netcdf-c/issues/1208
re: https://github.com/Unidata/netcdf-c/issues/2078
re: https://github.com/Unidata/netcdf-c/issues/2041
re: https://github.com/Unidata/netcdf-c/issues/2143
For a long time, there have been known problems with the
management of complex types containing VLENs. This also
involves the string type because it is stored as a VLEN of
chars.
This PR (mostly) fixes this problem. But note that it adds new
functions to netcdf.h (see below) and this may require bumping
the .so number. These new functions can be removed, if desired,
in favor of functions in netcdf_aux.h, but netcdf.h seems the
better place for them because they are intended as alternatives
to the nc_free_vlen and nc_free_string functions already in
netcdf.h.
The term complex type refers to any type that directly or
transitively references a VLEN type. So an array of VLENS, a
compound with a VLEN field, and so on.
In order to properly handle instances of these complex types, it
is necessary to have function that can recursively walk
instances of such types to perform various actions on them. The
term "deep" is also used to mean recursive.
At the moment, the two operations needed by the netcdf library are:
* free'ing an instance of the complex type
* copying an instance of the complex type.
The current library does only shallow free and shallow copy of
complex types. This means that only the top level is properly
free'd or copied, but deep internal blocks in the instance are
not touched.
Note that the term "vector" will be used to mean a contiguous (in
memory) sequence of instances of some type. Given an array with,
say, dimensions 2 X 3 X 4, this will be stored in memory as a
vector of length 2*3*4=24 instances.
The use cases are primarily these.
## nc_get_vars
Suppose one is reading a vector of instances using nc_get_vars
(or nc_get_vara or nc_get_var, etc.). These functions will
return the vector in the top-level memory provided. All
interior blocks (form nested VLEN or strings) will have been
dynamically allocated.
After using this vector of instances, it is necessary to free
(aka reclaim) the dynamically allocated memory, otherwise a
memory leak occurs. So, the recursive reclaim function is used
to walk the returned instance vector and do a deep reclaim of
the data.
Currently functions are defined in netcdf.h that are supposed to
handle this: nc_free_vlen(), nc_free_vlens(), and
nc_free_string(). Unfortunately, these functions only do a
shallow free, so deeply nested instances are not properly
handled by them.
Note that internally, the provided data is immediately written so
there is no need to copy it. But the caller may need to reclaim the
data it passed into the function.
## nc_put_att
Suppose one is writing a vector of instances as the data of an attribute
using, say, nc_put_att.
Internally, the incoming attribute data must be copied and stored
so that changes/reclamation of the input data will not affect
the attribute.
Again, the code inside the netcdf library does only shallow copying
rather than deep copy. As a result, one sees effects such as described
in Github Issue https://github.com/Unidata/netcdf-c/issues/2143.
Also, after defining the attribute, it may be necessary for the user
to free the data that was provided as input to nc_put_att().
## nc_get_att
Suppose one is reading a vector of instances as the data of an attribute
using, say, nc_get_att.
Internally, the existing attribute data must be copied and returned
to the caller, and the caller is responsible for reclaiming
the returned data.
Again, the code inside the netcdf library does only shallow copying
rather than deep copy. So this can lead to memory leaks and errors
because the deep data is shared between the library and the user.
# Solution
The solution is to build properly recursive reclaim and copy
functions and use those as needed.
These recursive functions are defined in libdispatch/dinstance.c
and their signatures are defined in include/netcdf.h.
For back compatibility, corresponding "ncaux_XXX" functions
are defined in include/netcdf_aux.h.
````
int nc_reclaim_data(int ncid, nc_type xtypeid, void* memory, size_t count);
int nc_reclaim_data_all(int ncid, nc_type xtypeid, void* memory, size_t count);
int nc_copy_data(int ncid, nc_type xtypeid, const void* memory, size_t count, void* copy);
int nc_copy_data_all(int ncid, nc_type xtypeid, const void* memory, size_t count, void** copyp);
````
There are two variants. The first two, nc_reclaim_data() and
nc_copy_data(), assume the top-level vector is managed by the
caller. For reclaim, this is so the user can use, for example, a
statically allocated vector. For copy, it assumes the user
provides the space into which the copy is stored.
The second two, nc_reclaim_data_all() and
nc_copy_data_all(), allows the functions to manage the
top-level. So for nc_reclaim_data_all, the top level is
assumed to be dynamically allocated and will be free'd by
nc_reclaim_data_all(). The nc_copy_data_all() function
will allocate the top level and return a pointer to it to the
user. The user can later pass that pointer to
nc_reclaim_data_all() to reclaim the instance(s).
# Internal Changes
The netcdf-c library internals are changed to use the proper
reclaim and copy functions. It turns out that the places where
these functions are needed is quite pervasive in the netcdf-c
library code. Using these functions also allows some
simplification of the code since the stdata and vldata fields of
NC_ATT_INFO are no longer needed. Currently this is commented
out using the SEPDATA \#define macro. When any bugs are largely
fixed, all this code will be removed.
# Known Bugs
1. There is still one known failure that has not been solved.
All the failures revolve around some variant of this .cdl file.
The proximate cause of failure is the use of a VLEN FillValue.
````
netcdf x {
types:
float(*) row_of_floats ;
dimensions:
m = 5 ;
variables:
row_of_floats ragged_array(m) ;
row_of_floats ragged_array:_FillValue = {-999} ;
data:
ragged_array = {10, 11, 12, 13, 14}, {20, 21, 22, 23}, {30, 31, 32},
{40, 41}, _ ;
}
````
When a solution is found, I will either add it to this PR or post a new PR.
# Related Changes
* Mark nc_free_vlen(s) as deprecated in favor of ncaux_reclaim_data.
* Remove the --enable-unfixed-memory-leaks option.
* Remove the NC_VLENS_NOTEST code that suppresses some vlen tests.
* Document this change in docs/internal.md
* Disable the tst_vlen_data test in ncdump/tst_nccopy4.sh.
* Mark types as fixed size or not (transitively) to optimize the reclaim
and copy functions.
# Misc. Changes
* Make Doxygen process libdispatch/daux.c
* Make sure the NC_ATT_INFO_T.container field is set.
2022-01-09 09:30:00 +08:00
|
|
|
}
|
2023-10-09 09:59:28 +08:00
|
|
|
|
|
|
|
/**************************************************/
|
|
|
|
/* Allocate/Free for various structures */
|
|
|
|
|
|
|
|
int
|
|
|
|
NCD4_newInfo(NCD4INFO** d4infop)
|
|
|
|
{
|
|
|
|
int ret = NC_NOERR;
|
|
|
|
NCD4INFO* info = NULL;
|
|
|
|
if((info = calloc(1,sizeof(NCD4INFO)))==NULL)
|
|
|
|
{ret = NC_ENOMEM; goto done;}
|
|
|
|
info->platform.hostlittleendian = NCD4_isLittleEndian();
|
2023-11-05 06:17:09 +08:00
|
|
|
info->responses = nclistnew();
|
2023-10-09 09:59:28 +08:00
|
|
|
if(d4infop) {*d4infop = info; info = NULL;}
|
|
|
|
done:
|
|
|
|
if(info) NCD4_reclaimInfo(info);
|
|
|
|
return THROW(ret);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Reclaim an NCD4INFO instance */
|
|
|
|
void
|
|
|
|
NCD4_reclaimInfo(NCD4INFO* d4info)
|
|
|
|
{
|
|
|
|
size_t i;
|
|
|
|
if(d4info == NULL) return;
|
|
|
|
d4info->controller = NULL; /* break link */
|
|
|
|
nullfree(d4info->rawdmrurltext);
|
|
|
|
nullfree(d4info->dmrurltext);
|
|
|
|
ncurifree(d4info->dmruri);
|
|
|
|
freeCurl(d4info->curl);
|
|
|
|
nullfree(d4info->fileproto.filename);
|
|
|
|
NCD4_resetInfoForRead(d4info);
|
|
|
|
nullfree(d4info->substrate.filename); /* always reclaim */
|
|
|
|
NC_authfree(d4info->auth);
|
|
|
|
nclistfree(d4info->blobs);
|
|
|
|
/* Reclaim dmr node tree */
|
|
|
|
NCD4_reclaimMeta(d4info->dmrmetadata);
|
2023-11-05 06:17:09 +08:00
|
|
|
/* Reclaim all responses */
|
2023-10-09 09:59:28 +08:00
|
|
|
for(i=0;i<nclistlength(d4info->responses);i++) {
|
|
|
|
NCD4response* resp = nclistget(d4info->responses,i);
|
|
|
|
NCD4_reclaimResponse(resp);
|
|
|
|
}
|
|
|
|
nclistfree(d4info->responses);
|
|
|
|
free(d4info);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Reset NCD4INFO instance for new read request */
|
|
|
|
void
|
|
|
|
NCD4_resetInfoForRead(NCD4INFO* d4info)
|
|
|
|
{
|
|
|
|
if(d4info == NULL) return;
|
|
|
|
if(d4info->substrate.realfile
|
|
|
|
&& !FLAGSET(d4info->controls.debugflags,NCF_DEBUG_COPY)) {
|
|
|
|
/* We used real file, so we need to delete the temp file
|
|
|
|
unless we are debugging.
|
|
|
|
Assume caller has done nc_close|nc_abort on the ncid.
|
|
|
|
Note that in theory, this should not be necessary since
|
|
|
|
AFAIK the substrate file is still in def mode, and
|
|
|
|
when aborted, it should be deleted. But that is not working
|
|
|
|
for some reason, so we delete it ourselves.
|
|
|
|
*/
|
|
|
|
if(d4info->substrate.filename != NULL) {
|
|
|
|
unlink(d4info->substrate.filename);
|
2023-11-05 06:17:09 +08:00
|
|
|
}
|
2023-10-09 09:59:28 +08:00
|
|
|
}
|
2023-11-05 06:17:09 +08:00
|
|
|
NCD4_reclaimMeta(d4info->dmrmetadata);
|
2023-10-09 09:59:28 +08:00
|
|
|
d4info->dmrmetadata = NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
|
|
|
freeCurl(NCD4curl* curl)
|
|
|
|
{
|
|
|
|
if(curl == NULL) return;
|
|
|
|
NCD4_curlclose(curl->curl);
|
|
|
|
ncbytesfree(curl->packet);
|
|
|
|
nullfree(curl->errdata.code);
|
|
|
|
nullfree(curl->errdata.message);
|
|
|
|
free(curl);
|
|
|
|
}
|
|
|
|
|
|
|
|
int
|
|
|
|
NCD4_newResponse(NCD4INFO* info, NCD4response** respp)
|
|
|
|
{
|
|
|
|
int ret = NC_NOERR;
|
|
|
|
NCD4response* resp = NULL;
|
|
|
|
NC_UNUSED(info);
|
|
|
|
if((resp = calloc(1,sizeof(NCD4response)))==NULL)
|
|
|
|
{ret = NC_ENOMEM; goto done;}
|
|
|
|
resp->controller = info;
|
|
|
|
if(respp) {*respp = resp; resp = NULL;}
|
|
|
|
done:
|
|
|
|
if(resp) NCD4_reclaimResponse(resp);
|
|
|
|
return THROW(ret);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
/* Reclaim an NCD4response instance */
|
|
|
|
void
|
|
|
|
NCD4_reclaimResponse(NCD4response* d4resp)
|
|
|
|
{
|
|
|
|
struct NCD4serial* serial = NULL;
|
|
|
|
if(d4resp == NULL) return;
|
|
|
|
serial = &d4resp->serial;
|
|
|
|
d4resp->controller = NULL; /* break link */
|
|
|
|
|
|
|
|
nullfree(d4resp->raw.memory);
|
|
|
|
nullfree(serial->dmr);
|
2023-11-05 06:17:09 +08:00
|
|
|
nullfree(serial->errdata);
|
2023-10-09 09:59:28 +08:00
|
|
|
|
2023-11-05 10:23:55 +08:00
|
|
|
/* clear all fields */
|
|
|
|
memset(serial,0,sizeof(struct NCD4serial));
|
|
|
|
|
|
|
|
nullfree(d4resp->error.parseerror);
|
|
|
|
nullfree(d4resp->error.message);
|
|
|
|
nullfree(d4resp->error.context);
|
|
|
|
nullfree(d4resp->error.otherinfo);
|
|
|
|
memset(&d4resp->error,0,sizeof(d4resp->error));
|
|
|
|
|
|
|
|
free(d4resp);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Create an empty NCD4meta object for
|
|
|
|
use in subsequent calls
|
|
|
|
(is the the right src file to hold this?)
|
|
|
|
*/
|
|
|
|
|
|
|
|
int
|
|
|
|
NCD4_newMeta(NCD4INFO* info, NCD4meta** metap)
|
|
|
|
{
|
|
|
|
int ret = NC_NOERR;
|
|
|
|
NCD4meta* meta = (NCD4meta*)calloc(1,sizeof(NCD4meta));
|
|
|
|
if(meta == NULL) return NC_ENOMEM;
|
|
|
|
meta->allnodes = nclistnew();
|
|
|
|
#ifdef D4DEBUG
|
|
|
|
meta->debuglevel = 1;
|
|
|
|
#endif
|
|
|
|
meta->controller = info;
|
|
|
|
meta->ncid = info->substrate.nc4id; /* Transfer netcdf ncid */
|
|
|
|
if(metap) {*metap = meta; meta = NULL;}
|
|
|
|
return THROW(ret);
|
|
|
|
}
|
|
|
|
|
|
|
|
void
|
|
|
|
NCD4_reclaimMeta(NCD4meta* dataset)
|
|
|
|
{
|
2023-11-27 19:36:03 +08:00
|
|
|
size_t i;
|
2023-11-05 10:23:55 +08:00
|
|
|
if(dataset == NULL) return;
|
|
|
|
|
2023-10-09 09:59:28 +08:00
|
|
|
for(i=0;i<nclistlength(dataset->allnodes);i++) {
|
|
|
|
NCD4node* node = (NCD4node*)nclistget(dataset->allnodes,i);
|
|
|
|
reclaimNode(node);
|
|
|
|
}
|
|
|
|
nclistfree(dataset->allnodes);
|
|
|
|
nclistfree(dataset->groupbyid);
|
|
|
|
nclistfree(dataset->atomictypes);
|
|
|
|
free(dataset);
|
|
|
|
}
|
|
|
|
|
2023-11-05 06:17:09 +08:00
|
|
|
#if 0
|
2023-10-09 09:59:28 +08:00
|
|
|
void
|
|
|
|
NCD4_resetMeta(NCD4meta* dataset)
|
|
|
|
{
|
|
|
|
if(dataset == NULL) return;
|
|
|
|
#if 0
|
|
|
|
for(i=0;i<nclistlength(dataset->blobs);i++) {
|
|
|
|
void* p = nclistget(dataset->blobs,i);
|
|
|
|
nullfree(p);
|
|
|
|
}
|
|
|
|
nclistfree(dataset->blobs);
|
|
|
|
#endif
|
|
|
|
}
|
2023-11-05 06:17:09 +08:00
|
|
|
#endif
|