/*!
Called when FORMAT command needs to be parsed from within the DIMENSIONS block. Deals with everything after the
token FORMAT up to and including the semicolon that terminates the FORMAT command.
*/
void NxsUnalignedBlock::HandleFormat(
NxsToken & token) /* is the token used to read from `in' */
{
bool standardDataTypeAssumed = false;
bool ignoreCaseAssumed = false;
for (;;)
{
token.GetNextToken();
if (token.Equals("DATATYPE"))
{
DemandEquals(token, "after keyword DATATYPE");
// This should be one of the following: STANDARD, DNA, RNA, NUCLEOTIDE or PROTEIN
token.GetNextToken();
if (token.Equals("STANDARD"))
datatype = NxsCharactersBlock::standard;
else if (token.Equals("DNA"))
datatype = NxsCharactersBlock::dna;
else if (token.Equals("RNA"))
datatype = NxsCharactersBlock::rna;
else if (token.Equals("NUCLEOTIDE"))
datatype = NxsCharactersBlock::nucleotide;
else if (token.Equals("PROTEIN"))
datatype = NxsCharactersBlock::protein;
else
{
errormsg = token.GetToken();
errormsg += " is not a valid DATATYPE within a ";
errormsg += NCL_BLOCKTYPE_ATTR_NAME;
errormsg += " block";
throw NxsException(errormsg, token.GetFilePosition(), token.GetFileLine(), token.GetFileColumn());
}
if (standardDataTypeAssumed && datatype != NxsCharactersBlock::standard)
{
errormsg = "DATATYPE must be specified first in FORMAT command";
throw NxsException(errormsg, token.GetFilePosition(), token.GetFileLine(), token.GetFileColumn());
}
originalDatatype = datatype;
ResetSymbols();
}
else if (token.Equals("RESPECTCASE"))
{
if (ignoreCaseAssumed)
{
errormsg = "RESPECTCASE must be specified before MISSING and SYMBOLS in FORMAT command";
throw NxsException(errormsg, token.GetFilePosition(), token.GetFileLine(), token.GetFileColumn());
}
standardDataTypeAssumed = true;
respectingCase = true;
}
else if (token.Equals("MISSING"))
{
DemandEquals(token, "after keyword MISSING");
// This should be the missing data symbol (single character)
token.GetNextToken();
if (token.GetTokenLength() != 1)
{
errormsg = "MISSING symbol should be a single character, but ";
errormsg += token.GetToken();
errormsg += " was specified";
throw NxsException(errormsg, token.GetFilePosition(), token.GetFileLine(), token.GetFileColumn());
}
else if (token.IsPunctuationToken() && !token.IsPlusMinusToken())
{
errormsg = "MISSING symbol specified cannot be a punctuation token (";
errormsg += token.GetToken();
errormsg += " was specified)";
throw NxsException(errormsg, token.GetFilePosition(), token.GetFileLine(), token.GetFileColumn());
}
else if (token.IsWhitespaceToken())
{
errormsg = "MISSING symbol specified cannot be a whitespace character (";
errormsg += token.GetToken();
errormsg += " was specified)";
throw NxsException(errormsg, token.GetFilePosition(), token.GetFileLine(), token.GetFileColumn());
}
missing = token.GetToken()[0];
ignoreCaseAssumed = true;
standardDataTypeAssumed = true;
}
else if (token.Equals("SYMBOLS") || token.Equals("SYMBOL"))
{
NxsDiscreteStateCell numDefStates;
unsigned maxNewStates;
switch(datatype)
{
case NxsCharactersBlock::dna:
case NxsCharactersBlock::rna:
case NxsCharactersBlock::nucleotide:
numDefStates = 4;
maxNewStates = NCL_MAX_STATES-4;
break;
case NxsCharactersBlock::protein:
numDefStates = 21;
maxNewStates = NCL_MAX_STATES-21;
break;
default:
numDefStates = 0; // replace symbols list for standard datatype
symbols[0] = '\0';
maxNewStates = NCL_MAX_STATES;
}
DemandEquals(token, "after keyword SYMBOLS");
// This should be the symbols list
token.SetLabileFlagBit(NxsToken::doubleQuotedToken);
token.GetNextToken();
token.StripWhitespace();
unsigned numNewSymbols = token.GetTokenLength();
if (numNewSymbols > maxNewStates)
{
errormsg = "SYMBOLS defines ";
errormsg += numNewSymbols;
errormsg += " new states but only ";
errormsg += maxNewStates;
errormsg += " new states allowed for this DATATYPE";
throw NxsException(errormsg, token.GetFilePosition(), token.GetFileLine(), token.GetFileColumn());
}
NxsString to = token.GetToken();
unsigned tlen = (unsigned)to.size();
NxsString processedS;
// Check to make sure user has not used any symbols already in the
// default symbols list for this data type
for (unsigned i = 0; i < tlen; i++)
{
if (IsInSymbols(to[i]))
{
errormsg = "The character ";
errormsg << to[i] << " defined in SYMBOLS is predefined for this DATATYPE and shoud not occur in a SYMBOLS subcommand of a FORMAT command.";
if (nexusReader)
{
nexusReader->NexusWarnToken(errormsg, NxsReader::SKIPPING_CONTENT_WARNING, token);
errormsg.clear();
}
}
else
processedS += to[i];
}
// If we've made it this far, go ahead and add the user-defined
// symbols to the end of the list of predefined symbols
symbols.append(processedS);
ignoreCaseAssumed = true;
standardDataTypeAssumed = true;
}
else if (token.Equals("EQUATE"))
{
DemandEquals(token, "after keyword EQUATE");
// This should be a double-quote character
token.GetNextToken();
if (!token.Equals("\""))
{
errormsg = "Expecting '\"' after keyword EQUATE but found ";
errormsg += token.GetToken();
errormsg += " instead";
throw NxsException(errormsg, token.GetFilePosition(), token.GetFileLine(), token.GetFileColumn());
}
// Loop until second double-quote character is encountered
for (;;)
{
token.GetNextToken();
if (token.Equals("\""))
break;
// If token is not a double-quote character, then it must be the equate symbol (i.e., the
// character to be replaced in the data matrix)
if (token.GetTokenLength() != 1)
{
errormsg = "Expecting single-character EQUATE symbol but found ";
errormsg += token.GetToken();
errormsg += " instead";
throw NxsException(errormsg, token.GetFilePosition(), token.GetFileLine(), token.GetFileColumn());
}
// Check for bad choice of equate symbol
NxsString t = token.GetToken();
const char ch = t[0];
bool badEquateSymbol = false;
// The character '^' cannot be an equate symbol
if (ch == '^')
badEquateSymbol = true;
// Equate symbols cannot be punctuation (except for + and -)
if (token.IsPunctuationToken() && !token.IsPlusMinusToken())
badEquateSymbol = true;
// Equate symbols cannot be same as matchchar, missing, or gap
if (ch == missing || ch == gap)
badEquateSymbol = true;
// Equate symbols cannot be one of the state symbols currently defined
if (IsInSymbols(ch))
badEquateSymbol = true;
if (badEquateSymbol)
{
errormsg = "EQUATE symbol specified (";
errormsg += token.GetToken();
errormsg += ") is not valid; must not be same as missing, \nmatchchar, gap, state symbols, or any of the following: ()[]{}/\\,;:=*'\"`<>^";
throw NxsException(errormsg, token.GetFilePosition(), token.GetFileLine(), token.GetFileColumn());
}
NxsString k = token.GetToken();
DemandEquals(token, "in EQUATE definition");
// This should be the token to be substituted in for the equate symbol
token.SetLabileFlagBit(NxsToken::parentheticalToken);
token.SetLabileFlagBit(NxsToken::curlyBracketedToken);
token.GetNextToken();
NxsString v = token.GetToken();
// Add the new equate association to the equates list
equates[ch] = v;
}
standardDataTypeAssumed = true;
}
else if (token.Equals("LABELS"))
{
labels = true;
standardDataTypeAssumed = true;
}
else if (token.Equals("NOLABELS"))
{
labels = false;
standardDataTypeAssumed = true;
}
else if (token.Equals(";"))
{
break;
}
}
ResetDatatypeMapper();
}
void NxsTaxaAssociationBlock::HandleAssociatesCommand(
NxsToken &token)
{
if (this->firstTaxaBlock == 0L || this->secondTaxaBlock == 0L)
{
errormsg << "Expecting TAXA command to precede an ASSOCIATES command.";
throw NxsException(errormsg, token);
}
token.GetNextToken();
for (;; )
{
std::set<unsigned> fSet;
while (!token.IsPunctuationToken() || !(token.Equals(";") || token.Equals(",") || token.Equals("/")))
{
try {
this->firstTaxaBlock->GetIndicesForLabel(token.GetTokenReference(), &fSet);
}
catch(...)
{
errormsg << "Unrecognized taxon \"" << token.GetTokenReference() << "\" in ASSOCIATES command";
throw NxsException(errormsg, token);
}
token.GetNextToken();
}
if (!token.Equals("/"))
{
errormsg << "Expecting / in ASSOCIATES command, found \"" << token.GetTokenReference() << "\"";
throw NxsException(errormsg, token);
}
if (fSet.empty())
{
errormsg << "Expecting taxon labels from the first TAXA block before the / in ASSOCIATES command.";
throw NxsException(errormsg, token);
}
token.GetNextToken();
std::set<unsigned> sSet;
while (!token.IsPunctuationToken() || !(token.Equals(";") || token.Equals(",") || token.Equals("/")))
{
try {
this->secondTaxaBlock->GetIndicesForLabel(token.GetTokenReference(), &sSet);
}
catch(...)
{
errormsg << "Unrecognized taxon \"" << token.GetTokenReference() << "\" in ASSOCIATES command";
throw NxsException(errormsg, token);
}
token.GetNextToken();
}
if (!(token.Equals(";") || token.Equals(",")))
{
errormsg << "Expecting , or ; in ASSOCIATES command, found \"" << token.GetTokenReference() << "\"";
throw NxsException(errormsg, token);
}
if (sSet.empty())
{
errormsg << "Expecting taxon labels from the second TAXA block after the / in ASSOCIATES command.";
throw NxsException(errormsg, token);
}
for (std::set<unsigned>::const_iterator fIt = fSet.begin(); fIt != fSet.end(); ++fIt)
{
this->AddAssociation(*fIt, sSet);
}
if (token.Equals(";"))
{
break;
}
token.GetNextToken();
}
}
/*----------------------------------------------------------------------------------------------------------------------
| This function provides the ability to read everything following the block name (which is read by the NxsReader
| object) to the END or ENDBLOCK statement. Characters are read from the input stream `in'. Overrides the virtual
| function in the base class.
*/
void GarliBlock::Read(
NxsToken &token) /* the token used to read from `in' */
{
isEmpty = false;
//if we already read a garli block with a model string, clear it
modelString.clear();
// This should be the semicolon after the block name
//
token.GetNextToken();
if (!token.Equals(";"))
{
errormsg = "Expecting ';' after ";
errormsg += id;
errormsg += " block name, but found ";
errormsg += token.GetToken();
errormsg += " instead";
throw NxsException(errormsg, token.GetFilePosition(), token.GetFileLine(), token.GetFileColumn());
}
for (;;)
{//only allowing three things to happen here
//1. endblock is reached, sucessfully exiting the garli block
//2. something besides an endblock is read. This is interpreted as part of the model string, with minimal error checking
//3. eof is hit before an endblock
token.GetNextToken();
if (token.Abbreviation("ENdblock"))
{
HandleEndblock(token);
break;
}
else if(token.AtEOF() == false){
NxsString s = token.GetToken();
if(s.size() > 1 && (s.IsADouble() == false && s.IsALong() == false)){
errormsg = "Unexpected character(s) in Garli block.\n See manual for model parameter format.";
throw NxsException(errormsg, token.GetFilePosition(), token.GetFileLine(), token.GetFileColumn());
}
if(token.IsPunctuationToken() == false){//toss semicolons and such
modelString += token.GetToken();
modelString += ' ';
}
}
else
{
errormsg = "Unexpected end of file encountered before \"end;\" or\n \"endblock;\" command in Garli block";
throw NxsException(errormsg, token.GetFilePosition(), token.GetFileLine(), token.GetFileColumn());
}
}
/*
else if (token.Abbreviation("GarliReader"))
{
HandleGarliReader(token);
}
else if (token.Abbreviation("Help"))
{
HandleHelp(token);
}
else if (token.Abbreviation("Log"))
{
HandleLog(token);
}
else if (token.Abbreviation("EXecute"))
{
HandleExecute(token);
}
else if (token.Abbreviation("Show"))
{
HandleShow(token);
}
else if (token.Abbreviation("Quit"))
{
quit_now = true;
message = "\nGarliReader says goodbye\n";
PrintMessage();
break;
}
else
{
SkippingCommand(token.GetToken());
do
{
token.GetNextToken();
}
while (!token.AtEOF() && !token.Equals(";"));
if (token.AtEOF())
{
errormsg = "Unexpected end of file encountered";
throw NxsException(errormsg, token.GetFilePosition(), token.GetFileLine(), token.GetFileColumn());
}
}
*/
}