NCL: nxsmultiformat.h Source File

00001 //  Copyright (C) 1999-2003 Paul O. Lewis and Mark T. Holder
00002 //
00003 //  This file is part of NCL (Nexus Class Library) version 2.0.
00004 //
00005 //  NCL is free software; you can redistribute it and/or modify
00006 //  it under the terms of the GNU General Public License as published by
00007 //  the Free Software Foundation; either version 2 of the License, or
00008 //  (at your option) any later version.
00009 //
00010 //  NCL is distributed in the hope that it will be useful,
00011 //  but WITHOUT ANY WARRANTY; without even the implied warranty of
00012 //  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
00013 //  GNU General Public License for more details.
00014 //
00015 //  You should have received a copy of the GNU General Public License
00016 //  along with NCL; if not, write to the Free Software Foundation, Inc.,
00017 //  59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
00018 //
00019 
00020 #ifndef NCL_NXSMULTIFORMAT_H
00021 #define NCL_NXSMULTIFORMAT_H
00022 #include <iostream>
00023 
00024 #include "ncl/nxsdefs.h"
00025 #include "ncl/nxspublicblocks.h"
00026 class FileToCharBuffer;
00038 class MultiFormatReader: public PublicNexusReader
00039 {
00040     public:
00046         enum DataFormatType
00047             {
00048                 NEXUS_FORMAT,
00049                 FASTA_DNA_FORMAT,
00050                 FASTA_AA_FORMAT,
00051                 FASTA_RNA_FORMAT,
00052                 PHYLIP_DNA_FORMAT,
00053                 PHYLIP_RNA_FORMAT,
00054                 PHYLIP_AA_FORMAT,
00055                 PHYLIP_DISC_FORMAT,
00056                 INTERLEAVED_PHYLIP_DNA_FORMAT,
00057                 INTERLEAVED_PHYLIP_RNA_FORMAT,
00058                 INTERLEAVED_PHYLIP_AA_FORMAT,
00059                 INTERLEAVED_PHYLIP_DISC_FORMAT,
00060                 RELAXED_PHYLIP_DNA_FORMAT,
00061                 RELAXED_PHYLIP_RNA_FORMAT,
00062                 RELAXED_PHYLIP_AA_FORMAT,
00063                 RELAXED_PHYLIP_DISC_FORMAT,
00064                 INTERLEAVED_RELAXED_PHYLIP_DNA_FORMAT,
00065                 INTERLEAVED_RELAXED_PHYLIP_RNA_FORMAT,
00066                 INTERLEAVED_RELAXED_PHYLIP_AA_FORMAT,
00067                 INTERLEAVED_RELAXED_PHYLIP_DISC_FORMAT,
00068                 ALN_DNA_FORMAT,
00069                 ALN_RNA_FORMAT,
00070                 ALN_AA_FORMAT,
00071                 PHYLIP_TREE_FORMAT,
00072                 RELAXED_PHYLIP_TREE_FORMAT,
00073                 NEXML_FORMAT,
00074                 UNSUPPORTED_FORMAT // keep this last
00075             };
00076 
00082         static std::vector<std::string> getFormatNames();
00087         static DataFormatType formatNameToCode(const std::string &);
00088 
00089 
00097         MultiFormatReader(const int blocksToRead = -1, NxsReader::WarningHandlingMode mode=NxsReader::WARNINGS_TO_STDERR)
00098             :PublicNexusReader(blocksToRead, mode)
00099             {}
00100         virtual ~MultiFormatReader(){}
00105         void ReadStream(std::istream & inp, const char * formatName);
00110         void ReadStream(std::istream & inp, DataFormatType format, const char * filepath=0L);
00111 
00116         void ReadFilepath(const char * filepath, const char * formatName);
00121         void ReadFilepath(const char * filepath, DataFormatType format);
00122 
00127         void readFastaFile(std::istream & inf, NxsCharactersBlock::DataTypesEnum dt);
00128 
00129     private:
00130         void addTaxaNames(const std::list<std::string> & taxaName, NxsTaxaBlockAPI * taxa);
00131         void moveDataToDataBlock(const std::list<std::string> & taxaNames, std::list<NxsDiscreteStateRow> & matList, const unsigned nchar, NxsDataBlock * dataB);
00132         void moveDataToMatrix(std::list<NxsDiscreteStateRow> & matList,  NxsDiscreteStateMatrix &mat);
00133         void moveDataToUnalignedBlock(const std::list<std::string> & taxaNames, std::list<NxsDiscreteStateRow> & matList, NxsUnalignedBlock * uB);
00134         bool readFastaSequences(FileToCharBuffer & ftcb, const NxsDiscreteDatatypeMapper &dm, std::list<std::string> & taxaNames, std::list<NxsDiscreteStateRow> & matList, size_t & longest);
00135         void readPhylipFile(std::istream & inf, NxsCharactersBlock::DataTypesEnum dt, bool relaxedNames, bool interleaved);
00136         void readPhylipTreeFile(std::istream & inf, bool relaxedNames);
00137         void readAlnFile(std::istream & inf, NxsCharactersBlock::DataTypesEnum dt);
00138         bool readAlnData(FileToCharBuffer & ftcb, const NxsDiscreteDatatypeMapper &dm, std::list<std::string> & taxaNames, std::list<NxsDiscreteStateRow> & matList);
00139 
00140         unsigned readPhylipHeader(std::istream & inf, unsigned & ntax, unsigned & nchar);
00141         void readPhylipData(FileToCharBuffer & ftcb, const NxsDiscreteDatatypeMapper &dm, std::list<std::string> & taxaNames, std::list<NxsDiscreteStateRow> & matList, const unsigned n_taxa, const unsigned n_char, bool relaxedNames);
00142         void readInterleavedPhylipData(FileToCharBuffer & ftcb, const NxsDiscreteDatatypeMapper &dm, std::list<std::string> & taxaNames, std::list<NxsDiscreteStateRow> & matList, const unsigned n_taxa, const unsigned n_char, bool relaxedNames);
00143         std::string readPhylipName(FileToCharBuffer & ftcb, unsigned i, bool relaxedNames);
00144 
00145 
00146 };
00147 
00236 #endif
00237