void CVDBMgr::x_Init(void)
{
if ( rc_t rc = VDBManagerMakeRead(x_InitPtr(), 0) ) {
*x_InitPtr() = 0;
NCBI_THROW2(CSraException, eInitFailed,
"Cannot open VDBManager", rc);
}
uint32_t sdk_ver;
if ( rc_t rc = VDBManagerVersion(*this, &sdk_ver) ) {
NCBI_THROW2(CSraException, eInitFailed,
"Cannot get VDBManager version", rc);
}
CKNSManager kns_mgr(CVFSManager(*this));
CNcbiOstrstream str;
CNcbiApplication* app = CNcbiApplication::Instance();
if ( app ) {
str << app->GetAppName() << ": " << app->GetVersion().Print() << "; ";
}
#if NCBI_PACKAGE
str << "Package: " << NCBI_PACKAGE_NAME << ' ' <<
NCBI_PACKAGE_VERSION << "; ";
#endif
str << "C++ ";
#ifdef NCBI_PRODUCTION_VER
str << NCBI_PRODUCTION_VER << "/";
#endif
#ifdef NCBI_DEVELOPMENT_VER
str << NCBI_DEVELOPMENT_VER;
#endif
string prefix = CNcbiOstrstreamToString(str);
KNSManagerSetUserAgent(kns_mgr, "%s; SRA Toolkit %V",
prefix.c_str(),
sdk_ver);
// redirect VDB log to C++ Toolkit
if ( s_GetDiagHandler() ) {
KLogInit();
KLogLevelSet(klogDebug);
KLogLibHandlerSet(VDBLogWriter, 0);
}
if ( app ) {
string host = app->GetConfig().GetString("CONN", "HTTP_PROXY_HOST", kEmptyStr);
int port = app->GetConfig().GetInt("CONN", "HTTP_PROXY_PORT", 0);
if ( !host.empty() && port != 0 ) {
if ( rc_t rc = KNSManagerSetHTTPProxyPath(kns_mgr, "%s:%d", host.c_str(), port) ) {
NCBI_THROW2(CSraException, eInitFailed,
"Cannot set KNSManager proxy parameters", rc);
}
KNSManagerSetHTTPProxyEnabled(kns_mgr, true);
}
}
}
void CExecAndParseStructuredOutput::ThrowUnexpectedCharError()
{
size_t position = m_Ch - m_NSOutput.data() + 1;
if (*m_Ch == '\0') {
NCBI_THROW2(CStringException, eFormat,
"Unexpected end of NetSchedule output", position);
} else {
NCBI_THROW2(CStringException, eFormat,
"Unexpected character in NetSchedule output", position);
}
}
void CSraPath::x_Init(void)
{
CSraMgr::RegisterFunctions();
if ( rc_t rc = SRAPathMake(x_InitPtr(), 0) ) {
*x_InitPtr() = 0;
NCBI_THROW2(CSraException, eInitFailed,
"Cannot make SRAPath", rc);
}
if ( rc_t rc = SRAPathClear(*this) ) {
NCBI_THROW2(CSraException, eInitFailed,
"Cannot clear SRAPath", rc);
}
}
void CKConfig::Commit() const
{
if ( rc_t rc = KConfigCommit(const_cast<KConfig*>(GetPointer())) ) {
NCBI_THROW2(CSraException, eOtherError,
"CKConfig: Cannot commit config changes", rc);
}
}
void CVDBMgr::DeleteCacheOlderThan(Uint4 days)
{
if ( rc_t rc = VDBManagerDeleteCacheOlderThan(*this, days) ) {
NCBI_THROW2(CSraException, eOtherError,
"CVDBMgr: Cannot delete old cache files", rc);
}
}
void CVFSManager::x_InitNew(void)
{
if ( rc_t rc = VFSManagerMake(x_InitPtr()) ) {
*x_InitPtr() = 0;
NCBI_THROW2(CSraException, eInitFailed,
"Cannot create VFSManager", rc);
}
}
CKDBManager::CKDBManager(const CVDBMgr& mgr)
{
if ( rc_t rc = VDBManagerGetKDBManagerRead(mgr, x_InitPtr()) ) {
*x_InitPtr() = 0;
NCBI_THROW2(CSraException, eInitFailed,
"Cannot get KDBManager", rc);
}
}
CKNSManager::CKNSManager(const CVFSManager& mgr)
{
if ( rc_t rc = VFSManagerGetKNSMgr(mgr, x_InitPtr()) ) {
*x_InitPtr() = 0;
NCBI_THROW2(CSraException, eInitFailed,
"Cannot get KNSManager", rc);
}
}
void CVDBMgr::SetCacheRoot(const string& path)
{
CVPath vpath(CVFSManager(*this), path, CVPath::eSys);
if ( rc_t rc = VDBManagerSetCacheRoot(*this, vpath) ) {
NCBI_THROW2(CSraException, eOtherError,
"CVDBMgr: Cannot set cache root", rc);
}
}
void CSraMgr::x_DoInit(void)
{
if ( rc_t rc = SRAMgrMakeRead(x_InitPtr()) ) {
*x_InitPtr() = 0;
NCBI_THROW2(CSraException, eInitFailed,
"Cannot open SRAMgr", rc);
}
}
CVResolver::CVResolver(const CVFSManager& mgr, const CKConfig& cfg)
: m_Mgr(mgr)
{
if ( rc_t rc = VFSManagerMakeResolver(mgr, x_InitPtr(), cfg) ) {
*x_InitPtr() = 0;
NCBI_THROW2(CSraException, eInitFailed,
"Cannot create VResolver", rc);
}
}
CKConfig::CKConfig(const CVDBMgr& mgr)
{
*x_InitPtr() = VFSManagerGetConfig(CVFSManager(mgr));
if ( rc_t rc = KConfigAddRef(*this) ) {
*x_InitPtr() = 0;
NCBI_THROW2(CSraException, eInitFailed,
"Cannot get reference to KConfig", rc);
}
}
CKConfig::CKConfig(void)
{
KConfig* cfg;
if ( rc_t rc = KConfigMake(&cfg, 0) ) {
*x_InitPtr() = 0;
NCBI_THROW2(CSraException, eInitFailed,
"Cannot create KConfig", rc);
}
*x_InitPtr() = cfg;
}
string CVPath::ToString(EType type) const
{
const String* str = 0;
if (type == eSys) {
if (rc_t rc = VPathMakeSysPath(*this, &str)) {
NCBI_THROW2(CSraException, eOtherError,
"Cannot get path from VPath", rc);
}
}
else {
if (rc_t rc = VPathMakeString(*this, &str)) {
NCBI_THROW2(CSraException, eOtherError,
"Cannot get path from VPath", rc);
}
}
string ret(str->addr, str->size);
StringWhack(str);
return ret;
}
void CVDBCursor::CloseRow(void)
{
if ( !RowIsOpened() ) {
return;
}
if ( rc_t rc = VCursorCloseRow(*this) ) {
NCBI_THROW2(CSraException, eInitFailed,
"Cannot close VDB cursor row", rc);
}
m_RowOpened = false;
}
void CVDBCursor::Init(const CVDBTable& table)
{
if ( *this ) {
NCBI_THROW2(CSraException, eInvalidState,
"Cannot init VDB cursor again",
RC(rcApp, rcCursor, rcConstructing, rcSelf, rcOpen));
}
if ( rc_t rc = VTableCreateCursorRead(table, x_InitPtr()) ) {
*x_InitPtr() = 0;
NCBI_THROW2(CSraException, eInitFailed,
"Cannot create VDB cursor", rc);
}
if ( rc_t rc = VCursorPermitPostOpenAdd(*this) ) {
NCBI_THROW2(CSraException, eInitFailed,
"Cannot allow VDB cursor post open column add", rc);
}
if ( rc_t rc = VCursorOpen(*this) ) {
NCBI_THROW2(CSraException, eInitFailed,
"Cannot open VDB cursor", rc);
}
m_Table = table;
}
string CVDBMgr::GetCacheRoot() const
{
const VPath* ret;
if ( rc_t rc = VDBManagerGetCacheRoot(*this, &ret) ) {
if ( GetRCObject(rc) == RCObject(rcPath) &&
GetRCState(rc) == rcNotFound ) {
return kEmptyStr;
}
NCBI_THROW2(CSraException, eOtherError,
"CVDBMgr: Cannot get cache root", rc);
}
return CVPath(ret).ToString(CVPath::eSys);
}
// ----------------------------------------------------------------------------
void CMultiReaderApp::xProcessDefault(
const CArgs& args,
CNcbiIstream& istr,
CNcbiOstream& ostr)
// ----------------------------------------------------------------------------
{
auto_ptr<CReaderBase> pReader(CReaderBase::GetReader(m_uFormat, m_iFlags));
if (!pReader.get()) {
NCBI_THROW2(CObjReaderParseException, eFormat,
"File format not supported", 0);
}
CRef<CSerialObject> object = pReader->ReadObject(istr, m_pErrors);
xWriteObject(*object, ostr);
}
CExprValue::CExprValue(Uint8 value)
: ival(0)
, m_sval("")
, m_Var(NULL)
, m_Pos(0)
, m_Tag(eINT)
{
if (static_cast<Uint8>(numeric_limits<Int8>::max()) < value) {
NCBI_THROW2(CExprParserException,
eTypeConversionError,
"Value too big to fit in the 8-byte signed integer type",
m_Pos);
}
ival = static_cast<Int8>(value);
}
Int8 CNetScheduleStructuredOutputParser::ParseInt(size_t len)
{
Int8 val = NStr::StringToInt8(CTempString(m_Ch, len));
if (*m_Ch == '-') {
++m_Ch;
--len;
}
if (*m_Ch == '0' && len > 1) {
NCBI_THROW2(CStringException, eFormat,
"Leading zeros are not allowed", GetPosition());
}
m_Ch += len;
return val;
}
CVDBTable::CVDBTable(const CVDBMgr& mgr,
const string& acc_or_path,
EMissing missing)
: m_Name(acc_or_path)
{
*x_InitPtr() = 0;
VSchema *schema;
DECLARE_SDK_GUARD();
if ( rc_t rc = SRASchemaMake(&schema, mgr) ) {
NCBI_THROW2(CSraException, eInitFailed,
"Cannot make default SRA schema", rc);
}
string path = CVPath::ConvertAccOrSysPathToPOSIX(acc_or_path);
if ( rc_t rc = VDBManagerOpenTableRead(mgr, x_InitPtr(), schema, "%.*s",
int(path.size()), path.data()) ) {
*x_InitPtr() = 0;
VSchemaRelease(schema);
if ( (GetRCObject(rc) == RCObject(rcDirectory) ||
GetRCObject(rc) == RCObject(rcPath)) &&
GetRCState(rc) == rcNotFound ) {
// no SRA accession
if ( missing != eMissing_Throw ) {
return;
}
NCBI_THROW2_FMT(CSraException, eNotFoundTable,
"Cannot open SRA table: "<<acc_or_path, rc);
}
else if ( GetRCObject(rc) == RCObject(rcDatabase) &&
GetRCState(rc) == rcIncorrect ) {
// invalid SRA database
NCBI_THROW2_FMT(CSraException, eDataError,
"Cannot open SRA table: "<<acc_or_path, rc);
}
else {
// other errors
NCBI_THROW2_FMT(CSraException, eOtherError,
"Cannot open SRA table: "<<acc_or_path, rc);
}
}
VSchemaRelease(schema);
}
CJsonNode CNetScheduleStructuredOutputParser::ParseValue()
{
size_t max_len = GetRemainder();
size_t len = 0;
switch (*m_Ch) {
/* Array */
case '[':
++m_Ch;
return ParseArray(']');
/* Object */
case '{':
++m_Ch;
return ParseObject('}');
/* String */
case '\'':
case '"':
return CJsonNode::NewStringNode(ParseString(max_len));
/* Number */
case '-':
// Check that there's at least one digit after the minus sign.
if (max_len <= 1 || !isdigit((unsigned char) m_Ch[1])) {
++m_Ch;
break;
}
len = 1;
case '0': case '1': case '2': case '3': case '4':
case '5': case '6': case '7': case '8': case '9':
// Skim through the integer part.
do
if (++len >= max_len)
return CJsonNode::NewIntegerNode(ParseInt(len));
while (isdigit((unsigned char) m_Ch[len]));
// Stumbled upon a non-digit character -- check
// if it's a fraction part or an exponent part.
switch (m_Ch[len]) {
case '.':
if (++len == max_len || !isdigit((unsigned char) m_Ch[len])) {
NCBI_THROW2(CStringException, eFormat,
"At least one digit after the decimal "
"point is required", GetPosition());
}
for (;;) {
if (++len == max_len)
return CJsonNode::NewDoubleNode(ParseDouble(len));
if (!isdigit((unsigned char) m_Ch[len])) {
if (m_Ch[len] == 'E' || m_Ch[len] == 'e')
break;
return CJsonNode::NewDoubleNode(ParseDouble(len));
}
}
/* FALL THROUGH */
case 'E':
case 'e':
if (++len == max_len ||
(m_Ch[len] == '-' || m_Ch[len] == '+' ?
++len == max_len ||
!isdigit((unsigned char) m_Ch[len]) :
!isdigit((unsigned char) m_Ch[len]))) {
m_Ch += len;
NCBI_THROW2(CStringException, eFormat,
"Invalid exponent specification", GetPosition());
}
while (++len < max_len && isdigit((unsigned char) m_Ch[len]))
;
return CJsonNode::NewDoubleNode(ParseDouble(len));
default:
return CJsonNode::NewIntegerNode(ParseInt(len));
}
/* Constant */
case 'F': case 'f': case 'N': case 'n':
case 'T': case 't': case 'Y': case 'y':
while (len <= max_len && isalpha((unsigned char) m_Ch[len]))
++len;
{
CTempString val(m_Ch, len);
m_Ch += len;
return val == "null" ? CJsonNode::NewNullNode() :
CJsonNode::NewBooleanNode(NStr::StringToBool(val));
}
}
INVALID_FORMAT_ERROR();
}
CDecimal& CDecimal::operator=(const CTempStringEx& s)
{
int ptr = 0;
char c = s[ptr++];
// skip leading blanks
while ( isspace((unsigned char)c) ) {
c = s[ptr++];
}
m_Sign = 0;
if ( c == '-' ) {
m_Sign = -1;
c = s[ptr++];
}
else if ( c == '+' ) {
m_Sign = +1;
c = s[ptr++];
}
bool dot = false, expn = false, anydigits = false;
int dot_position = 0;
m_Mantissa.erase();
// up to exponent
for ( ; ; c = s[ptr++] ) {
if (c >= '0' && c <= '9') {
// digits: accumulate
anydigits = true;
if ( m_Mantissa.empty() ) {
if ( c != '0' ) {
m_Mantissa += c;
}
else {
if ( dot ) {
--dot_position;
}
}
}
else {
m_Mantissa += c;
}
}
else if (c == '.') {
// dot
// if second dot, stop
if (dot) {
--ptr;
break;
}
dot_position = m_Mantissa.size();
dot = true;
}
else if (c == 'e' || c == 'E') {
// if exponent, stop
if (!anydigits) {
--ptr;
break;
}
expn = true;
break;
}
else if (!c) {
--ptr;
break;
}
else {
--ptr;
NCBI_THROW2(CStringException, eConvert,
"Cannot convert '"+string(s)+"'", ptr);
}
}
// if no digits, stop now - error
if (!anydigits) {
NCBI_THROW2(CStringException, eConvert,
"Cannot convert '"+string(s)+"'", ptr);
}
int exponent = dot ? dot_position - m_Mantissa.size() : 0;
// read exponent
if (expn && s[ptr]) {
int expvalue = 0;
bool expsign = false, expnegate= false;
int expdigits= 0;
for( ; ; ++ptr) {
c = s[ptr];
// sign: should be no digits at this point
if (c == '-' || c == '+') {
// if there was sign or digits, stop
if (expsign || expdigits) {
break;
}
expsign = true;
expnegate = c == '-';
}
// digits: accumulate
else if (c >= '0' && c <= '9') {
++expdigits;
int newexpvalue = expvalue*10 + (c-'0');
if (newexpvalue > expvalue) {
expvalue = newexpvalue;
}
}
else {
break;
}
}
// if no digits, rollback
if (!expdigits) {
// rollback sign
if (expsign) {
--ptr;
}
// rollback exponent
if (expn) {
--ptr;
}
}
else {
exponent = expnegate ? exponent - expvalue : exponent + expvalue;
}
}
m_Exponent = exponent+m_Mantissa.size();
if ( !m_Sign && !m_Mantissa.empty() ) {
m_Sign = 1;
}
Normalize();
return *this;
}
void CAlnReader::Read(bool guess, bool generate_local_ids)
{
if (m_ReadDone) {
return;
}
// make a SSequenceInfo corresponding to our CSequenceInfo argument
SSequenceInfo info;
info.alphabet = const_cast<char *>(m_Alphabet.c_str());
info.beginning_gap = const_cast<char *>(m_BeginningGap.c_str());
info.end_gap = const_cast<char *>(m_EndGap.c_str());;
info.middle_gap = const_cast<char *>(m_MiddleGap.c_str());
info.missing = const_cast<char *>(m_Missing.c_str());
info.match = const_cast<char *>(m_Match.c_str());
// read the alignment stream
TAlignmentFilePtr afp;
m_Errors.clear();
afp = ReadAlignmentFile2(s_ReadLine, (void *) &m_IS,
s_ReportError, &(m_Errors), &info,
(generate_local_ids ? eTrue : eFalse));
if (!afp) {
NCBI_THROW2(CObjReaderParseException, eFormat,
"Error reading alignment", 0);
}
int first_len = strlen (afp->sequences[0]);
for (int i = 1; i < afp->num_sequences; i++) {
if (strlen (afp->sequences[i]) != first_len) {
AlignmentFileFree (afp);
NCBI_THROW2(CObjReaderParseException, eFormat,
"Error reading alignment: Not all sequences have same length", 0);
}
}
// if we're trying to guess whether this is an alignment file,
// and no tell-tale alignment format lines were found,
// check to see if any of the lines contain gaps.
// no gaps plus no alignment indicators -> don't guess alignment
if (guess && !afp->align_format_found) {
bool found_gap = false;
for (int i = 0; i < afp->num_sequences && !found_gap; i++) {
if (strchr (afp->sequences[i], '-') != NULL) {
found_gap = true;
}
}
if (!found_gap) {
AlignmentFileFree (afp);
NCBI_THROW2(CObjReaderParseException, eFormat,
"Error reading alignment", 0);
}
}
// build the CAlignment
m_Seqs.resize(afp->num_sequences);
m_Ids.resize(afp->num_sequences);
for (int i = 0; i < afp->num_sequences; ++i) {
m_Seqs[i] = afp->sequences[i];
m_Ids[i] = afp->ids[i];
}
m_Organisms.resize(afp->num_organisms);
for (int i = 0; i < afp->num_organisms; ++i) {
if (afp->organisms[i]) {
m_Organisms[i] = afp->organisms[i];
} else {
m_Organisms[i].erase();
}
}
m_Deflines.resize(afp->num_deflines);
for (int i = 0; i < afp->num_deflines; ++i) {
if (afp->deflines[i]) {
m_Deflines[i] = afp->deflines[i];
} else {
m_Deflines[i].erase();
}
}
AlignmentFileFree(afp);
{{
m_Dim = m_Ids.size();
}}
m_ReadDone = true;
return;
}
CRef<CSeq_align> CAlnReader::GetSeqAlign()
{
if (m_Aln) {
return m_Aln;
} else if ( !m_ReadDone ) {
NCBI_THROW2(CObjReaderParseException, eFormat,
"CAlnReader::GetSeqAlign(): "
"Seq_align is not available until after Read()", 0);
}
typedef CDense_seg::TNumseg TNumseg;
typedef CDense_seg::TDim TNumrow;
m_Aln = new CSeq_align();
m_Aln->SetType(CSeq_align::eType_not_set);
m_Aln->SetDim(m_Dim);
CDense_seg& ds = m_Aln->SetSegs().SetDenseg();
ds.SetDim(m_Dim);
CDense_seg::TIds& ids = ds.SetIds();
CDense_seg::TStarts& starts = ds.SetStarts();
//CDense_seg::TStrands& strands = ds.SetStrands();
CDense_seg::TLens& lens = ds.SetLens();
ids.resize(m_Dim);
// get the length of the alignment
TSeqPos aln_stop = m_Seqs[0].size();
for (TNumrow row_i = 1; row_i < m_Dim; row_i++) {
if (m_Seqs[row_i].size() > aln_stop) {
aln_stop = m_Seqs[row_i].size();
}
}
for (TNumrow row_i = 0; row_i < m_Dim; row_i++) {
CBioseq::TId xid;
if (CSeq_id::ParseFastaIds(xid, m_Ids[row_i], true) > 0) {
ids[row_i] = xid.front();
} else {
ids[row_i] = new CSeq_id(CSeq_id::e_Local, m_Ids[row_i]);
}
}
m_SeqVec.resize(m_Dim);
for (TNumrow row_i = 0; row_i < m_Dim; row_i++) {
m_SeqVec[row_i].resize(m_Seqs[row_i].length(), 0);
}
m_SeqLen.resize(m_Dim, 0);
vector<bool> is_gap; is_gap.resize(m_Dim, true);
vector<bool> prev_is_gap; prev_is_gap.resize(m_Dim, true);
vector<TSignedSeqPos> next_start; next_start.resize(m_Dim, 0);
int starts_i = 0;
TSeqPos prev_aln_pos = 0, prev_len = 0;
bool new_seg = true;
TNumseg numseg = 0;
for (TSeqPos aln_pos = 0; aln_pos < aln_stop; aln_pos++) {
for (TNumrow row_i = 0; row_i < m_Dim; row_i++) {
if (aln_pos >= m_Seqs[row_i].length()) {
if (!is_gap[row_i]) {
is_gap[row_i] = true;
new_seg = true;
}
} else {
string residue = m_Seqs[row_i].substr(aln_pos, 1);
NStr::ToUpper(residue);
if (NStr::Find(m_MiddleGap, residue) == string::npos &&
NStr::Find(m_EndGap, residue) == string::npos &&
NStr::Find(m_BeginningGap, residue) == string::npos) {
if (is_gap[row_i]) {
is_gap[row_i] = false;
new_seg = true;
}
// add to the sequence vector
m_SeqVec[row_i][m_SeqLen[row_i]++] = residue.c_str()[0];
} else {
if ( !is_gap[row_i] ) {
is_gap[row_i] = true;
new_seg = true;
}
}
}
}
if (new_seg) {
if (numseg) { // if not the first seg
lens.push_back(prev_len = aln_pos - prev_aln_pos);
for (TNumrow row_i = 0; row_i < m_Dim; row_i++) {
if ( !prev_is_gap[row_i] ) {
next_start[row_i] += prev_len;
}
}
}
starts.resize(starts_i + m_Dim);
for (TNumrow row_i = 0; row_i < m_Dim; row_i++) {
if (is_gap[row_i]) {
starts[starts_i++] = -1;
} else {
starts[starts_i++] = next_start[row_i];;
}
prev_is_gap[row_i] = is_gap[row_i];
}
prev_aln_pos = aln_pos;
numseg++;
new_seg = false;
}
}
for (TNumrow row_i = 0; row_i < m_Dim; row_i++) {
m_SeqVec[row_i].resize(m_SeqLen[row_i]); // resize down to actual size
}
lens.push_back(aln_stop - prev_aln_pos);
//strands.resize(numseg * m_Dim, eNa_strand_plus);
_ASSERT(lens.size() == numseg);
ds.SetNumseg(numseg);
#if _DEBUG
m_Aln->Validate(true);
#endif
return m_Aln;
}
CRef<CSeq_entry> CAlnReader::GetSeqEntry()
{
if (m_Entry) {
return m_Entry;
} else if ( !m_ReadDone ) {
NCBI_THROW2(CObjReaderParseException, eFormat,
"CAlnReader::GetSeqEntry(): "
"Seq_entry is not available until after Read()", 0);
}
m_Entry = new CSeq_entry();
CRef<CSeq_annot> seq_annot (new CSeq_annot);
seq_annot->SetData().SetAlign().push_back(GetSeqAlign());
m_Entry->SetSet().SetClass(CBioseq_set::eClass_pop_set);
m_Entry->SetSet().SetAnnot().push_back(seq_annot);
CBioseq_set::TSeq_set& seq_set = m_Entry->SetSet().SetSeq_set();
typedef CDense_seg::TDim TNumrow;
for (TNumrow row_i = 0; row_i < m_Dim; row_i++) {
const string& seq_str = m_SeqVec[row_i];
const size_t& seq_str_len = seq_str.size();
CRef<CSeq_entry> seq_entry (new CSeq_entry);
// seq-id(s)
CBioseq::TId& ids = seq_entry->SetSeq().SetId();
CSeq_id::ParseFastaIds(ids, m_Ids[row_i], true);
if (ids.empty()) {
ids.push_back(CRef<CSeq_id>(new CSeq_id(CSeq_id::e_Local,
m_Ids[row_i])));
}
// mol
CSeq_inst::EMol mol = CSeq_inst::eMol_not_set;
CSeq_id::EAccessionInfo ai = ids.front()->IdentifyAccession();
if (ai & CSeq_id::fAcc_nuc) {
mol = CSeq_inst::eMol_na;
} else if (ai & CSeq_id::fAcc_prot) {
mol = CSeq_inst::eMol_aa;
} else {
switch (CFormatGuess::SequenceType(seq_str.data(), seq_str_len)) {
case CFormatGuess::eNucleotide: mol = CSeq_inst::eMol_na; break;
case CFormatGuess::eProtein: mol = CSeq_inst::eMol_aa; break;
default: break;
}
}
// seq-inst
CRef<CSeq_inst> seq_inst (new CSeq_inst);
seq_entry->SetSeq().SetInst(*seq_inst);
seq_set.push_back(seq_entry);
// repr
seq_inst->SetRepr(CSeq_inst::eRepr_raw);
// mol
seq_inst->SetMol(mol);
// len
_ASSERT(seq_str_len == m_SeqLen[row_i]);
seq_inst->SetLength(seq_str_len);
// data
CSeq_data& data = seq_inst->SetSeq_data();
if (mol == CSeq_inst::eMol_aa) {
data.SetIupacaa().Set(seq_str);
} else {
data.SetIupacna().Set(seq_str);
CSeqportUtil::Pack(&data);
}
}
return m_Entry;
}
CStringUTF8 CHTMLHelper::HTMLDecode(const string& str, EEncoding encoding,
THTMLDecodeFlags* result_flags)
{
CStringUTF8 ustr;
THTMLDecodeFlags result = 0;
if (encoding == eEncoding_Unknown) {
encoding = CUtf8::GuessEncoding(str);
if (encoding == eEncoding_Unknown) {
NCBI_THROW2(CStringException, eBadArgs,
"Unable to guess the source string encoding", 0);
}
}
// wild guess...
ustr.reserve(str.size());
string::const_iterator i, e = str.end();
char ch;
TUnicodeSymbol uch;
for (i = str.begin(); i != e;) {
ch = *(i++);
//check for HTML entities and character references
if (i != e && ch == '&') {
string::const_iterator itmp, end_of_entity, start_of_entity;
itmp = end_of_entity = start_of_entity = i;
bool ent, dec, hex, parsed=false;
ent = isalpha((unsigned char)(*itmp)) != 0;
dec = !ent && *itmp == '#' && ++itmp != e &&
isdigit((unsigned char)(*itmp)) != 0;
hex = !dec && itmp != e &&
(*itmp == 'x' || *itmp == 'X') && ++itmp != e &&
isxdigit((unsigned char)(*itmp)) != 0;
start_of_entity = itmp;
if (itmp != e && (ent || dec || hex)) {
// do not look too far
for (int len=0; len<16 && itmp != e; ++len, ++itmp) {
if (*itmp == '&' || *itmp == '#') {
break;
}
if (*itmp == ';') {
end_of_entity = itmp;
break;
}
ent = ent && isalnum( (unsigned char)(*itmp)) != 0;
dec = dec && isdigit( (unsigned char)(*itmp)) != 0;
hex = hex && isxdigit((unsigned char)(*itmp)) != 0;
}
if (end_of_entity != i && (ent || dec || hex)) {
uch = 0;
if (ent) {
string entity(start_of_entity,end_of_entity);
const struct tag_HtmlEntities* p = s_HtmlEntities;
for ( ; p->u != 0; ++p) {
if (entity.compare(p->s) == 0) {
uch = p->u;
parsed = true;
result |= fCharRef_Entity;
break;
}
}
} else {
parsed = true;
result |= fCharRef_Numeric;
for (itmp = start_of_entity;
itmp != end_of_entity; ++itmp) {
TUnicodeSymbol ud = *itmp;
if (dec) {
uch = 10 * uch + (ud - '0');
} else if (hex) {
if (ud >='0' && ud <= '9') {
ud -= '0';
} else if (ud >='a' && ud <= 'f') {
ud -= 'a';
ud += 10;
} else if (ud >='A' && ud <= 'F') {
ud -= 'A';
ud += 10;
}
uch = 16 * uch + ud;
}
}
}
if (parsed) {
ustr += CUtf8::AsUTF8(&uch,1);
i = ++end_of_entity;
continue;
}
}
}
}
// no entity - append as is
if (encoding == eEncoding_UTF8 || encoding == eEncoding_Ascii) {
ustr.append( 1, ch );
} else {
result |= fEncoding;
ustr += CUtf8::AsUTF8(CTempString(&ch,1), encoding);
}
}
if (result_flags) {
*result_flags = result;
}
return ustr;
}
void CDate_std::GetDate(string* label, const string& format) const
{
static const char* const kMonths[] = {
"0", "January", "February", "March", "April", "May", "June",
"July", "August", "September", "October", "November", "December"
};
static const int kNumMonths = sizeof (kMonths) / sizeof (char*);
if (!label) {
return;
}
unsigned int depth = 0;
vector<pair<SIZE_TYPE, SIZE_TYPE> > starts;
starts.push_back(make_pair(label->size(), (SIZE_TYPE)0));
ITERATE (string, it, format) {
if (*it != '%') {
*label += *it;
continue;
}
if (++it == format.end()) {
NCBI_THROW2(CGeneralParseException, eFormat,
"CDate_std::GetDate(): incomplete % expression",
it - format.begin());
}
// Check for things that can only immediately follow %
if (*it == '%') {
*label += '%';
continue;
}
else if (*it == '{') {
depth++;
starts.push_back(make_pair(label->size(),
SIZE_TYPE(it - format.begin())));
continue;
} else if (*it == '}') {
if (depth == 0) {
NCBI_THROW2(CGeneralParseException, eFormat,
"CDate_std::GetDate(): unbalanced %}",
it - format.begin());
}
depth--;
starts.pop_back();
continue;
} else if (*it == '|') {
// We survived, so just look for the appropriate %}.
if (depth == 0) {
return; // Can ignore rest of format
}
unsigned int depth2 = 0;
for (;;) {
while (++it != format.end() && *it != '%')
;
if (it == format.end() || ++it == format.end()) {
NCBI_THROW2(CGeneralParseException, eFormat,
"CDate_std::GetDate(): unbalanced %{",
starts.back().second);
}
if (*it == '}') {
if (depth2 == 0) {
depth--;
starts.pop_back();
break;
} else {
depth2--;
}
} else if (*it == '{') {
depth2++;
}
}
continue;
}
unsigned int length = 0;
int value = -1;
while (isdigit((unsigned char)(*it))) {
length = length * 10 + *it - '0';
if (++it == format.end()) {
NCBI_THROW2(CGeneralParseException, eFormat,
"CDate_std::GetDate(): incomplete % expression",
it - format.begin());
}
}
switch (*it) {
case 'Y': value = GetYear(); break;
case 'M':
case 'N': value = CanGetMonth() ? GetMonth() : -1; break;
case 'D': value = CanGetDay() ? GetDay() : -1; break;
case 'S': value = CanGetSeason() ? 1 : -1; break;
case 'h': value = CanGetHour() ? GetHour() : -1; break;
case 'm': value = CanGetMinute() ? GetMinute() : -1; break;
case 's': value = CanGetSecond() ? GetSecond() : -1; break;
default:
NCBI_THROW2(CGeneralParseException, eFormat,
"CDate_std::GetDate(): unrecognized format specifier",
it - format.begin());
}
if (value >= 0) {
if (*it == 'N') { // special cases
const char* name;
if (value >= kNumMonths) {
name = "inv";
} else {
name = kMonths[value];
}
if (length > 0) {
label->append(name, length);
} else {
*label += name;
}
} else if (*it == 'S') {
if (length > 0) {
label->append(GetSeason(), 0, length);
} else {
*label += GetSeason();
}
} else { // just a number
if (length > 0) {
// We want exactly <length> digits.
CNcbiOstrstream oss;
oss << setfill('0') << setw(length) << value;
string s = CNcbiOstrstreamToString(oss);
label->append(s, s.size() > length ? s.size() - length : 0,
length);
} else {
*label += NStr::IntToString(value);
}
}
} else {
// missing...roll back label and look for alternatives, or
// throw if at top level and none found
label->erase(starts.back().first);
char request = *it;
unsigned int depth2 = 0;
for (;;) {
while (++it != format.end() && *it != '%')
;
if (it == format.end() || ++it == format.end()) {
if (depth > 0 || depth2 > 0) {
NCBI_THROW2(CGeneralParseException, eFormat,
"CDate_std::GetDate(): unbalanced %{",
starts.back().second);
} else {
NCBI_THROW2(CGeneralParseException, eFormat,
string("CDate_std::GetDate():"
" missing required field %")
+ request, it - format.begin() - 1);
}
}
if (*it == '|' && depth2 == 0) {
break;
} else if (*it == '}') {
if (depth2 == 0) {
if (depth == 0) {
NCBI_THROW2(CGeneralParseException, eFormat,
"CDate_std::GetDate(): unbalanced %}",
it - format.begin());
}
depth--;
starts.pop_back();
break;
} else {
depth2--;
}
} else if (*it == '{') {
depth2++;
}
}
}
}
}
