netcdf-c/ncgen/cdfdata.c

478 lines
14 KiB
C
Raw Normal View History

2010-06-03 21:24:43 +08:00
/*********************************************************************
2018-12-07 06:40:43 +08:00
* Copyright 2018, UCAR/Unidata
2010-06-03 21:24:43 +08:00
* See netcdf/COPYRIGHT file for copying and redistribution conditions.
*********************************************************************/
/* $Id: cdfdata.c,v 1.4 2010/05/24 19:59:56 dmh Exp $ */
/* $Header: /upc/share/CVS/netcdf-3/ncgen/cdfdata.c,v 1.4 2010/05/24 19:59:56 dmh Exp $ */
#include "includes.h"
Primary change: add dap4 support Specific changes: 1. Add dap4 code: libdap4 and dap4_test. Note that until the d4ts server problem is solved, dap4 is turned off. 2. Modify various files to support dap4 flags: configure.ac, Makefile.am, CMakeLists.txt, etc. 3. Add nc_test/test_common.sh. This centralizes the handling of the locations of various things in the build tree: e.g. where is ncgen.exe located. See nc_test/test_common.sh for details. 4. Modify .sh files to use test_common.sh 5. Obsolete separate oc2 by moving it to be part of netcdf-c. This means replacing code with netcdf-c equivalents. 5. Add --with-testserver to configure.ac to allow override of the servers to be used for --enable-dap-remote-tests. 6. There were multiple versions of nctypealignment code. Try to centralize in libdispatch/doffset.c and include/ncoffsets.h 7. Add a unit test for the ncuri code because of its complexity. 8. Move the findserver code out of libdispatch and into a separate, self contained program in ncdap_test and dap4_test. 9. Move the dispatch header files (nc{3,4}dispatch.h) to .../include because they are now shared by modules. 10. Revamp the handling of TOPSRCDIR and TOPBUILDDIR for shell scripts. 11. Make use of MREMAP if available 12. Misc. minor changes e.g. - #include <config.h> -> #include "config.h" - Add some no-install headers to /include - extern -> EXTERNL and vice versa as needed - misc header cleanup - clean up checking for misc. unix vs microsoft functions 13. Change copyright decls in some files to point to LICENSE file. 14. Add notes to RELEASENOTES.md
2017-03-09 08:01:10 +08:00
#include "ncoffsets.h"
2010-06-03 21:24:43 +08:00
#ifdef ENABLE_BINARY
/**************************************************/
/* Code for generating binary data lists*/
/**************************************************/
struct Vlendata {
char* data;
unsigned long count;
}* vlendata;
/* Forward*/
static void genbin_data(Symbol* sym, Datasrc*, Datalist*, Bytebuffer*);
static void genbin_arraydatar(Symbol* basetype,
Datasrc* src,
Odometer* odom,
int index,
int checkpoint,
Bytebuffer* code);
static void genbin_primdata(Symbol*, Datasrc*, Datalist*, Bytebuffer*);
static void genbin_fieldarray(Symbol*, Datasrc*, Dimset*, int, Bytebuffer*);
static void alignbuffer(Constant* prim, Bytebuffer* buf, int base);
static void alignto(int,ByteBuffer*, int base);
2010-06-03 21:24:43 +08:00
/* Datalist rules: see the rules on the man page */
/* Specialty wrappers for genbin_data */
void
genbin_attrdata(Symbol* asym, Bytebuffer* memory)
{
Datasrc* src;
int chartype = (asym->typ.basetype->typ.typecode == NC_CHAR);
if(asym->data == NULL) return;
if(chartype) {gen_charattr(asym,memory); return;}
src = datalist2src(asym->data);
while(srcmore(src)) {
genbin_data(asym->typ.basetype,src,NULL,memory);
}
}
#if 0 /* Apparently not used */
2010-06-03 21:24:43 +08:00
void
genbin_scalardata(Symbol* vsym, Bytebuffer* memory)
{
Datasrc* src;
if(vsym->data == NULL) return;
src = datalist2src(vsym->data);
genbin_data(vsym->typ.basetype,src,
vsym->var.special._Fillvalue,memory);
if(srcmore(src)) {
semerror(srcline(src),"Extra data at end of datalist");
}
}
*/
2010-06-03 21:24:43 +08:00
void
genbin_datalist(struct Symbol* sym, Datalist* list, Bytebuffer* memory)
{
Datasrc* src = datalist2src(list);
genbin_data(sym,src,NULL,memory);
}
static void
genbin_data(Symbol* tsym, Datasrc* datasrc, Datalist* fillsrc,
Bytebuffer* memory)
{
int usecmpd;
Constant* con = srcpeek(datasrc);
if(con == NULL || con->nctype == NC_FILLVALUE) {
srcnext(datasrc);
genbin_fillvalue(tsym,fillsrc,datasrc,memory);
return;
}
switch (tsym->subclass) {
case NC_ENUM:
case NC_OPAQUE:
case NC_PRIM:
if(issublist(datasrc)) {
semerror(srcline(datasrc),"Expected primitive found {..}");
}
genbin_primdata(tsym,datasrc,fillsrc,memory);
break;
case NC_COMPOUND:
genbin_compound(tsym,datasrc,fillsrc,memory);
break;
2010-06-03 21:24:43 +08:00
case NC_VLEN: {
Constant* cp;
nc_vlen_t ptr;
if(!issublist(datasrc)) {
semerror(srcline(datasrc),"Vlen data must be enclosed in {..}");
}
cp = srcnext(datasrc);
/* generate the nc_vlen_t instance*/
ptr.p = vlendata[cp->value.compoundv->vlen.uid].data;
ptr.len = vlendata[cp->value.compoundv->vlen.uid].count;
bbAppendn(memory,(char*)&ptr,sizeof(ptr));
} break;
case NC_FIELD:
/* enclose in braces if and only if field is an array */
usecmpd = (issublist(datasrc) && tsym->typ.dimset.ndims > 0);
if(usecmpd) srcpush(datasrc);
if(tsym->typ.dimset.ndims > 0) {
genbin_fieldarray(tsym->typ.basetype,datasrc,&tsym->typ.dimset,0,memory);
} else {
genbin_data(tsym->typ.basetype,datasrc,NULL,memory);
}
if(usecmpd) srcpop(datasrc);
break;
default: PANIC1("genbin_data: unexpected subclass %d",tsym->subclass);
}
}
/* Used for compound instances */
static void
genbin_compound(Symbol* tsym, Datasrc* datasrc, Datalist* fillsrc, Bytebuffer* memory)
{
int i;
int base = bblength(memory);
if(!issublist(datasrc)) {
semerror(srcline(datasrc),"Compound data must be enclosed in {..}");
}
/* Use this datasrc list to get values for compound fields */
srcpush(datasrc);
for(i=0;i<listlength(tsym->subnodes);i++) {
Symbol* field = (Symbol*)listget(tsym->subnodes,i);
if(!srcmore(datasrc)) { /* generate a fill value*/
Datalist* fillsrc = getfiller(tsym);
genbin_data(field,datasrc,fillsrc,memory);
} else
genbin_data(field,datasrc,NULL,memory);
}
srcpop(datasrc);
/* Re: github issue 323: we may need to pad the end of the structure
to make its size be a multiple of the largest alignment.
*/
alignto(tsym->cmpdalignment,buf,base);
}
2010-06-03 21:24:43 +08:00
/* Used only for structure field arrays*/
static void
genbin_fieldarray(Symbol* basetype, Datasrc* src, Dimset* dimset, int index,
Bytebuffer* memory)
{
int i;
Symbol* dim = dimset->dimsyms[index];
unsigned int size = dim->dim.size;
int lastdim = (index == (dimset->ndims - 1)); /* last dimension*/
int chartype = (basetype->typ.typecode == NC_CHAR);
if(chartype) {
/* Collect the char field in a separate buffer */
Bytebuffer* fieldbuf = bbNew();
gen_charfield(src,dimset,index,fieldbuf);
bbAppendn(memory,bbContents(fieldbuf),bbLength(fieldbuf));
bbFree(fieldbuf);
} else {
ASSERT(size != 0);
for(i=0;i<size;i++) {
if(lastdim) {
genbin_data(basetype,src,NULL,memory);
} else { /* !lastdim*/
genbin_fieldarray(basetype,src,dimset,index+1,memory);
}
}
}
}
static void
genbin_primdata(Symbol* basetype, Datasrc* src, Datalist* fillsrc,
Bytebuffer* memory)
{
Constant* prim;
Constant target;
prim = srcnext(src);
if(prim == NULL || prim->nctype == NC_FILLVALUE) {
genbin_fillvalue(basetype,fillsrc,src,memory);
return;
}
target.nctype = basetype->typ.typecode;
if(prim == NULL) {
#ifdef GENFILL
/* generate a fill value*/
nc_getfill(&target);
/* fall thru*/
#else
return;
#endif
}
ASSERT(prim->nctype != NC_COMPOUND);
if(target.nctype != NC_ECONST) {
convert1(prim,&target);
alignbuffer(&target,memory);
}
switch (target.nctype) {
case NC_ECONST:
if(basetype->subclass != NC_ENUM) {
semerror(prim->lineno,"Conversion to enum not supported (yet)");
} else {
Datalist* econ = builddatalist(1);
srcpushlist(src,econ);
dlappend(econ,&prim->value.enumv->typ.econst);
genbin_primdata(prim->value.enumv->typ.basetype,src,
fillsrc,memory);
srcpop(src);
}
break;
case NC_OPAQUE: {
unsigned char* bytes;
size_t len;
setprimlength(&target,basetype->typ.size*2);
bytes=makebytestring(target.value.opaquev.stringv,&len);
bbAppendn(memory,(void*)bytes,len);
} break;
case NC_CHAR:
bbAppendn(memory,&target.value.charv,sizeof(target.value.charv));
break;
case NC_BYTE:
bbAppendn(memory,(void*)&target.value.int8v,sizeof(target.value.int8v));
break;
case NC_SHORT:
bbAppendn(memory,(void*)&target.value.int16v,sizeof(target.value.int16v));
break;
case NC_INT:
bbAppendn(memory,(void*)&target.value.int32v,sizeof(target.value.int32v));
break;
case NC_FLOAT:
bbAppendn(memory,(void*)&target.value.floatv,sizeof(target.value.floatv));
break;
case NC_DOUBLE:
bbAppendn(memory,(void*)&target.value.doublev,sizeof(target.value.doublev));
break;
case NC_UBYTE:
bbAppendn(memory,(void*)&target.value.uint8v,sizeof(target.value.uint8v));
break;
case NC_USHORT:
bbAppendn(memory,(void*)&target.value.uint16v,sizeof(target.value.uint16v));
break;
case NC_UINT:
bbAppendn(memory,(void*)&target.value.uint32v,sizeof(target.value.uint32v));
break;
case NC_INT64: {
union SI64 { char ch[8]; long long i64;} si64;
si64.i64 = target.value.int64v;
bbAppendn(memory,(void*)si64.ch,sizeof(si64.ch));
} break;
case NC_UINT64: {
union SU64 { char ch[8]; unsigned long long i64;} su64;
su64.i64 = target.value.uint64v;
bbAppendn(memory,(void*)su64.ch,sizeof(su64.ch));
} break;
case NC_STRING: {
if(usingclassic) {
bbAppendn(memory,target.value.stringv.stringv,target.value.stringv.len);
} else if(target.nctype == NC_CHAR) {
bbAppendn(memory,target.value.stringv.stringv,target.value.stringv.len);
} else {
char* ptr;
int len = (size_t)target.value.stringv.len;
ptr = poolalloc(len+1); /* CAREFUL: this has short lifetime*/
memcpy(ptr,target.value.stringv.stringv,len);
ptr[len] = '\0';
bbAppendn(memory,(void*)&ptr,sizeof(ptr));
}
} break;
default: PANIC1("genbin_primdata: unexpected type: %d",target.nctype);
}
}
void
genbin_fillvalue(Symbol* tsym, Datalist* fillsrc, Datasrc* src,
Bytebuffer* memory)
{
Datalist* list = NULL;
ASSERT(tsym->objectclass == NC_TYPE);
list = fillsrc;
if(list == NULL) list = getfiller(tsym);
srcpushlist(src,list);
genbin_data(tsym,src,NULL,memory);
srcpop(src);
}
/*
This walk of the data lists collects
vlen sublists and constructs separate C constants
for each of them. The "id" of each list is then
recorded in the containing datalist.
*/
void
genbin_vlenconstants(List* vlenconstants)
{
int i,nvlen;
Datasrc* vlensrc;
Bytebuffer* memory = bbNew();
/* Prepare a place to store vlen constants */
nvlen = listlength(vlenconstants);
if(nvlen == 0) return;
2017-10-31 05:52:08 +08:00
vlendata = (struct Vlendata*)ecalloc(sizeof(struct Vlendata)*nvlen+1);
2010-06-03 21:24:43 +08:00
memset((void*)vlendata,0,sizeof(struct Vlendata)*nvlen+1);
for(i=0;i<nvlen;i++) {
Constant* cmpd = (Constant*)listget(vlenconstants,i);
int chartype;
Symbol* tsym = cmpd->value.compoundv->vlen.schema;
unsigned long uid = cmpd->value.compoundv->vlen.uid;
unsigned long count;
ASSERT(tsym != NULL);
chartype = (tsym->typ.basetype->typ.typecode == NC_CHAR);
vlensrc = datalist2src(cmpd->value.compoundv);
bbClear(memory);
count = 0;
if(chartype) {
Fix more memory leaks in netcdf-c library This is a follow up to PR https://github.com/Unidata/netcdf-c/pull/1173 Sorry that it is so big, but leak suppression can be complex. This PR fixes all remaining memory leaks -- as determined by -fsanitize=address, and with the exceptions noted below. Unfortunately. there remains a significant leak that I cannot solve. It involves vlens, and it is unclear if the leak is occurring in the netcdf-c library or the HDF5 library. I have added a check_PROGRAM to the ncdump directory to show the problem. The program is called tst_vlen_demo.c To exercise it, build the netcdf library with -fsanitize=address enabled. Then go into ncdump and do a "make clean check". This should build tst_vlen_demo without actually executing it. Then do the command "./tst_vlen_demo" to see the output of the memory checker. Note the the lost malloc is deep in the HDF5 library (in H5Tvlen.c). I am temporarily working around this error in the following way. 1. I modified several test scripts to not execute known vlen tests that fail as described above. 2. Added an environment variable called NC_VLEN_NOTEST. If set, then those specific tests are suppressed. This should mean that the --disable-utilities option to ./configure should not need to be set to get a memory leak clean build. This should allow for detection of any new leaks. Note: I used an environment variable rather than a ./configure option to control the vlen tests. This is because it is temporary (I hope) and because it is a bit tricky for shell scripts to access ./configure options. Finally, as before, this only been tested with netcdf-4 and hdf5 support.
2018-11-16 01:00:38 +08:00
/* Collect the char sequence in a separate buffer */
gen_charseq(vlensrc,memory);
2010-06-03 21:24:43 +08:00
count = bbLength(memory);
} else {
while(srcmore(vlensrc)) {
genbin_data(tsym->typ.basetype,vlensrc,NULL,memory);
count++;
}
ASSERT(count == cmpd->value.compoundv->vlen.count);
}
vlendata[uid].data = bbDup(memory);
vlendata[uid].count = count;
}
bbFree(memory);
}
void
genbin_arraydata(Symbol* vsym, Bytebuffer* memory)
{
Datasrc* src;
Datalist* list;
int i;
int chartype = (vsym->typ.basetype->typ.typecode == NC_CHAR);
nciter_t iter;
Iterodom iterodom;
if(vsym->data == NULL) return;
if(chartype) {
gen_chararray(vsym,memory);
return;
}
list = vsym->data;
ASSERT(list->dimdata != NULL);
src = datalist2src(list);
/* Create an iterator to generate blocks of data */
nc_get_iter(vsym,databuffersize,&iter);
nc_next_iter(&iter,iterodom.start,iterodom.count);
iterodom.rank = vsym->typ.dimset.ndims;
genbin_arraydatar(vsym,memory,src,iter,&iterodom,/*index=*/0)
}
static void
genbin_arraydatar(Symbol* vsym,
Bytebuffer* memory,
nciter_t iter,
Iterodom* iterodom,
int index)
{
int i;
int rank = iterodom->rank;
int lastdim = (index == (rank - 1)); /* last dimension*/
int firstdim = (index == 0);
int declsize = odom->dims[index].declsize;
int isunlimited = (declsize == 0);
Symbol* basetype = vsym->typ.basetype;
Datalist* fillsrc = vsym->var.special._Fillvalue;
Constant* con;
ASSERT(index >= 0 && index < rank);
odom->dims[index].index = 0; /* reset*/
if(isunlimited) {
Constant* con;
if(!firstdim) {
if(!issublist(src)) {
semerror(srcline(src),"Unlimited data must be enclosed in {..}");
return;
}
srcpush(src); /* enter the unlimited data */
}
while((con=srcpeek(src))!=NULL) {
if(lastdim) {
genbin_data(basetype,src,fillsrc,memory);
} else {
genbin_arraydatar(vsym,src,odom,index+1,
checkpoint,memory);
}
odom->dims[index].index++;
if(docheckpoint) {
closure->putvar(closure,odom,memory);
}
}
odom->dims[index].datasize = odom->dims[index].index;
if(!firstdim) srcpop(src);
} else { /* !isunlimited*/
for(i=0;i<declsize;i++) {
con = srcpeek(src);
if(lastdim) {
genbin_data(basetype,src,fillsrc,memory);
} else { /* ! lastdim*/
(void)genbin_arraydatar(vsym,src,odom,
index+1,checkpoint,memory);
}
odom->dims[index].index++;
if(docheckpoint) {
closure->putvar(closure,odom,memory);
}
}
}
}
static void
alignbuffer(Constant* prim, Bytebuffer* buf, int base
2010-06-03 21:24:43 +08:00
{
int alignment,pad,offset;
if(prim->nctype == NC_ECONST)
alignment = nctypealignment(prim->value.enumv->typ.typecode);
else if(usingclassic && prim->nctype == NC_STRING)
alignment = nctypealignment(NC_CHAR);
else if(prim->nctype == NC_CHAR)
alignment = nctypealignment(NC_CHAR);
else
alignment = nctypealignment(prim->nctype);
alignto(alignment,buf,base);
}
static const char zeros[] =
"\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0";
static void
alignto(int alignment, Bytebuffer* buf, int base)
{
2010-06-03 21:24:43 +08:00
offset = bbLength(buf);
offset -= base; /* Need to actually align wrt to the base */
2010-06-03 21:24:43 +08:00
pad = getpadding(offset,alignment);
if(pad > 0) {
bbAppendn(buf,(void*)zeros,pad);
}
}
#endif /*ENABLE_BINARY*/