netcdf-c/libdap4/d4read.c
Dennis Heimbigner df3636b959 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.
2023-09-26 16:56:48 -06:00

377 lines
10 KiB
C

/*! \file
Copyright 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002,
2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012, 2013, 2014,
2015, 2016, 2017, 2018
University Corporation for Atmospheric Research/Unidata.
See \ref copyright file for more info.
*/
#include "d4includes.h"
#include "d4curlfunctions.h"
#ifdef HAVE_FCNTL_H
#include <fcntl.h>
#endif
#include "ncpathmgr.h"
/* Do conversion if this code was compiled via Vis. Studio or Mingw */
/*Forward*/
static int readpacket(NCD4INFO* state, NCURI*, NCbytes*, NCD4mode, NCD4format, long*);
static int readfile(NCD4INFO* state, const NCURI* uri, NCD4mode dxx, NCD4format fxx, NCbytes* packet);
static int readfiletofile(NCD4INFO* state, const NCURI* uri, NCD4mode dxx, NCD4format fxx, FILE* stream, d4size_t* sizep);
static int readfileDAPDMR(NCD4INFO* state, const NCURI* uri, NCbytes* packet);
#ifdef HAVE_GETTIMEOFDAY
static double
deltatime(struct timeval time0,struct timeval time1)
{
double t0, t1;
t0 = ((double)time0.tv_sec);
t0 += ((double)time0.tv_usec) / 1000000.0;
t1 = ((double)time1.tv_sec);
t1 += ((double)time1.tv_usec) / 1000000.0;
return (t1 - t0);
}
#endif
int
NCD4_readDMR(NCD4INFO* state, int flags)
{
int stat = NC_NOERR;
long lastmod = -1;
if((flags & NCF_ONDISK) == 0) {
ncbytesclear(state->curl->packet);
stat = readpacket(state,state->uri,state->curl->packet,NCD4_DMR,NCD4_FORMAT_XML,&lastmod);
if(stat == NC_NOERR)
state->data.dmrlastmodified = lastmod;
} else { /*((flags & NCF_ONDISK) != 0) */
NCURI* url = state->uri;
int fileprotocol = (strcmp(url->protocol,"file")==0);
if(fileprotocol) {
stat = readfiletofile(state, url, NCD4_DMR, NCD4_FORMAT_XML, state->data.ondiskfile, &state->data.datasize);
} else {
char* readurl = NULL;
int flags = 0;
if(!fileprotocol) flags |= NCURIQUERY;
flags |= NCURIENCODE;
flags |= NCURIPWD;
#ifdef FIX
ncurisetconstraints(url,state->constraint);
#endif
readurl = ncuribuild(url,NULL,".dmr.xml",NCURISVC);
if(readurl == NULL)
return THROW(NC_ENOMEM);
stat = NCD4_fetchurl_file(state->curl, readurl, state->data.ondiskfile,
&state->data.datasize, &lastmod);
nullfree(readurl);
if(stat == NC_NOERR)
state->data.dmrlastmodified = lastmod;
}
}
return THROW(stat);
}
int
NCD4_readDAP(NCD4INFO* state, int flags)
{
int stat = NC_NOERR;
long lastmod = -1;
if((flags & NCF_ONDISK) == 0) {
ncbytesclear(state->curl->packet);
stat = readpacket(state,state->uri,state->curl->packet,NCD4_DAP,NCD4_FORMAT_NONE,&lastmod);
if(stat) {
NCD4_seterrormessage(state->substrate.metadata, nclistlength(state->curl->packet), nclistcontents(state->curl->packet));
goto done;
} else
state->data.daplastmodified = lastmod;
} else { /*((flags & NCF_ONDISK) != 0) */
NCURI* url = state->uri;
int fileprotocol = (strcmp(url->protocol,"file")==0);
if(fileprotocol) {
stat = readfiletofile(state, url, NCD4_DAP, NCD4_FORMAT_NONE, state->data.ondiskfile, &state->data.datasize);
} else {
char* readurl = NULL;
int flags = 0;
if(!fileprotocol) flags |= NCURIQUERY;
flags |= NCURIENCODE;
flags |= NCURIPWD;
#ifdef FIX
ncurisetconstraints(url,state->constraint);
#endif
readurl = ncuribuild(url,NULL,".dap",NCURISVC);
if(readurl == NULL)
return THROW(NC_ENOMEM);
stat = NCD4_fetchurl_file(state->curl, readurl, state->data.ondiskfile,
&state->data.datasize, &lastmod);
nullfree(readurl);
if(stat == NC_NOERR)
state->data.daplastmodified = lastmod;
}
}
done:
return THROW(stat);
}
static const char*
dxxextension(int dxx)
{
switch(dxx) {
case NCD4_DMR: return ".dmr";
case NCD4_DAP: return ".dap";
default: break;
}
return NULL;
}
static const char*
dxxformat(int fxx, int dxx)
{
switch(dxx) {
case NCD4_DMR:
switch(fxx) {
case NCD4_FORMAT_XML:return ".xml";
default: break;
}
break;
case NCD4_DAP:
switch(fxx) {
case NCD4_FORMAT_NONE:return "";
default: break;
}
break;
default: break;
}
return NULL;
}
static int
readpacket(NCD4INFO* state, NCURI* url, NCbytes* packet, NCD4mode dxx, NCD4format fxx, long* lastmodified)
{
int stat = NC_NOERR;
int fileprotocol = 0;
CURL* curl = state->curl->curl;
#ifdef HAVE_GETTIMEOFDAY
struct timeval time0;
struct timeval time1;
#endif
char suffix[256];
suffix[0] = '\0';
strlcat(suffix,dxxextension(dxx),sizeof(suffix));
strlcat(suffix,dxxformat(fxx,dxx),sizeof(suffix));
fileprotocol = (strcmp(url->protocol,"file")==0);
if(fileprotocol) {
/* Short circuit file://... urls*/
/* We do this because the test code always needs to read files*/
stat = readfile(state, url, dxx, fxx, packet);
} else {
char* fetchurl = NULL;
int flags = NCURIBASE;
if(!fileprotocol) flags |= NCURIQUERY;
flags |= NCURIENCODE;
fetchurl = ncuribuild(url,NULL,suffix,flags);
MEMCHECK(fetchurl);
if(FLAGSET(state->controls.flags,NCF_SHOWFETCH)) {
nclog(NCLOGDEBUG,"fetch url=%s",fetchurl);
#ifdef HAVE_GETTIMEOFDAY
gettimeofday(&time0,NULL);
#endif
}
stat = NCD4_fetchurl(curl,fetchurl,packet,lastmodified,&state->substrate.metadata->error.httpcode);
nullfree(fetchurl);
if(stat) goto fail;
if(FLAGSET(state->controls.flags,NCF_SHOWFETCH)) {
double secs = 0;
#ifdef HAVE_GETTIMEOFDAY
gettimeofday(&time1,NULL);
secs = deltatime(time0,time1);
#endif
nclog(NCLOGDEBUG,"fetch complete: %0.3f",secs);
}
}
#ifdef D4DEBUG
{
fprintf(stderr,"readpacket: packet.size=%lu\n",
(unsigned long)ncbyteslength(packet));
}
#endif
fail:
return THROW(stat);
}
static int
readfiletofile(NCD4INFO* state, const NCURI* uri, NCD4mode dxx, NCD4format fxx, FILE* stream, d4size_t* sizep)
{
int stat = NC_NOERR;
NCbytes* packet = ncbytesnew();
size_t len;
stat = readfile(state, uri, dxx, fxx, packet);
#ifdef D4DEBUG
fprintf(stderr,"readfiletofile: packet.size=%lu\n",
(unsigned long)ncbyteslength(packet));
#endif
if(stat != NC_NOERR) goto unwind;
len = nclistlength(packet);
if(stat == NC_NOERR) {
size_t written;
fseek(stream,0,SEEK_SET);
written = fwrite(ncbytescontents(packet),1,len,stream);
if(written != len) {
#ifdef D4DEBUG
fprintf(stderr,"readfiletofile: written!=length: %lu :: %lu\n",
(unsigned long)written,(unsigned long)len);
#endif
stat = NC_EIO;
}
}
if(sizep != NULL) *sizep = len;
unwind:
ncbytesfree(packet);
return THROW(stat);
}
static int
readfile(NCD4INFO* state, const NCURI* uri, NCD4mode dxx, NCD4format fxx, NCbytes* packet)
{
int stat = NC_NOERR;
NCbytes* tmp = ncbytesnew();
char* filename = NULL;
char suffix[256];
#ifdef HAVE_GETTIMEOFDAY
struct timeval time0;
struct timeval time1;
#endif
suffix[0] = '\0';
strlcat(suffix,dxxextension(dxx),sizeof(suffix));
#if 0
not needed strlcat(suffix,dxxformat(fxx,dxx),sizeof(suffix));
#endif
ncbytescat(tmp,uri->path);
ncbytescat(tmp,suffix);
ncbytesnull(tmp);
filename = ncbytesextract(tmp);
ncbytesfree(tmp);
nullfree(state->fileproto.filename);
state->fileproto.filename = filename; /* filename is alloc'd here anyway */
if(FLAGSET(state->controls.flags,NCF_SHOWFETCH)) {
char* surl = NULL;
#ifdef HAVE_GETTIMEOFDAY
gettimeofday(&time0,NULL);
#endif
surl = ncuribuild((NCURI*)uri,NULL,NULL,NCURIALL);
nclog(NCLOGDEBUG,"fetch uri=%s file=%s",surl,filename);
}
switch (dxx) {
case NCD4_DMR:
if((stat = NC_readfile(filename,packet))) {
/* See if we can get the same thing from the .dap file */
stat = readfileDAPDMR(state,uri,packet);
}
break;
case NCD4_DAP:
case NCD4_DSR:
stat = NC_readfile(filename,packet);
break;
default: stat = NC_EDAP; break;
}
if(FLAGSET(state->controls.flags,NCF_SHOWFETCH)) {
double secs;
#ifdef HAVE_GETTIMEOFDAY
gettimeofday(&time1,NULL);
secs = deltatime(time0,time1);
#endif
nclog(NCLOGDEBUG,"%s fetch complete: %0.3f",suffix,secs);
}
return THROW(stat);
}
/* Extract the DMR from a DAP file */
static int
readfileDAPDMR(NCD4INFO* state, const NCURI* uri, NCbytes* packet)
{
int stat = NC_NOERR;
NCbytes* tmp = ncbytesnew();
char* filename = NULL;
NCD4HDR hdr;
#ifdef HAVE_GETTIMEOFDAY
struct timeval time0;
struct timeval time1;
#endif
ncbytescat(tmp,uri->path);
ncbytescat(tmp,".dap");
ncbytesnull(tmp);
filename = ncbytesextract(tmp);
ncbytesfree(tmp);
nullfree(state->fileproto.filename);
state->fileproto.filename = filename; /* filename is alloc'd here anyway */
if(FLAGSET(state->controls.flags,NCF_SHOWFETCH)) {
char* surl = NULL;
#ifdef HAVE_GETTIMEOFDAY
gettimeofday(&time0,NULL);
#endif
surl = ncuribuild((NCURI*)uri,NULL,".dap",NCURIALL);
nclog(NCLOGDEBUG,"fetch uri=%s file=%s",surl,filename);
}
stat = NC_readfile(filename,packet);
if(FLAGSET(state->controls.flags,NCF_SHOWFETCH)) {
double secs;
#ifdef HAVE_GETTIMEOFDAY
gettimeofday(&time1,NULL);
secs = deltatime(time0,time1);
#endif
nclog(NCLOGDEBUG,"fetch complete: %0.3f",secs);
}
if(stat != NC_NOERR) goto done;
/* Extract the DMR from the dap */
NCD4_getheader(ncbytescontents(packet),&hdr,NCD4_isLittleEndian());
if(hdr.count == 0 || (hdr.flags & NCD4_ERR_CHUNK))
return THROW(NC_EDMR);
/* patch up the packet */
{
int i;
size_t newlen;
for(i=0;i<4;i++)
ncbytesremove(packet,0); /* remove the hdr */
/* null terminate */
ncbytessetlength(packet,hdr.count-1);
ncbytesnull(packet);
/* Suppress nuls */
newlen = NCD4_elidenuls(ncbytescontents(packet),ncbyteslength(packet));
/* reset packet length */
ncbytessetlength(packet,newlen);
}
done:
return THROW(stat);
}
/* Extract packet as error message; assume httpcode set */
int
NCD4_seterrormessage(NCD4meta* metadata, size_t len, char* msg)
{
metadata->error.message = (char*)d4alloc(len+1);
if(metadata->error.message == NULL)
return THROW(NC_ENOMEM);
memcpy(metadata->error.message,msg,len);
metadata->error.message[len] = '\0';
return THROW(NC_ENODATA); /* slight lie */
}