1 // ***************************************************************************
2 // SamFormatParser.cpp (c) 2010 Derek Barnett
3 // Marth Lab, Department of Biology, Boston College
4 // All rights reserved.
5 // ---------------------------------------------------------------------------
6 // Last modified: 23 December 2010 (DB)
7 // ---------------------------------------------------------------------------
8 // Provides functionality for parsing SAM header text into SamHeader object
9 // ***************************************************************************
11 #include <api/SamConstants.h>
12 #include <api/SamHeader.h>
13 #include <api/internal/SamFormatParser_p.h>
14 using namespace BamTools;
15 using namespace BamTools::Internal;
22 SamFormatParser::SamFormatParser(SamHeader& header)
26 SamFormatParser::~SamFormatParser(void) { }
28 void SamFormatParser::Parse(const string& headerText) {
30 // clear header's prior contents
33 // empty header is OK, but skip processing
34 if ( headerText.empty() )
37 // other wise parse SAM lines
38 istringstream headerStream(headerText);
39 string headerLine = "";
40 while ( getline(headerStream, headerLine) )
41 ParseSamLine(headerLine);
45 void SamFormatParser::ParseSamLine(const string& line) {
47 // skip if line is not long enough to contain true values
48 if (line.length() < 5 ) return;
50 // determine token at beginning of line
51 const string firstToken = line.substr(0,3);
52 string restOfLine = line.substr(4);
53 if ( firstToken == Constants::SAM_HD_BEGIN_TOKEN) ParseHDLine(restOfLine);
54 else if ( firstToken == Constants::SAM_SQ_BEGIN_TOKEN) ParseSQLine(restOfLine);
55 else if ( firstToken == Constants::SAM_RG_BEGIN_TOKEN) ParseRGLine(restOfLine);
56 else if ( firstToken == Constants::SAM_PG_BEGIN_TOKEN) ParsePGLine(restOfLine);
57 else if ( firstToken == Constants::SAM_CO_BEGIN_TOKEN) ParseCOLine(restOfLine);
58 else cerr << "SAM Format Error - unknown token: " << firstToken << endl;
62 void SamFormatParser::ParseHDLine(const string& line) {
64 // split HD lines into tokens
65 vector<string> tokens = Split(line, Constants::SAM_TAB);
67 // iterate over tokens
68 vector<string>::const_iterator tokenIter = tokens.begin();
69 vector<string>::const_iterator tokenEnd = tokens.end();
70 for ( ; tokenIter != tokenEnd; ++tokenIter ) {
73 const string tokenTag = (*tokenIter).substr(0,2);
74 const string tokenValue = (*tokenIter).substr(3);
76 // set header contents
77 if ( tokenTag == Constants::SAM_HD_VERSION_TAG ) m_header.Version = tokenValue;
78 else if ( tokenTag == Constants::SAM_HD_GROUPORDER_TAG ) m_header.GroupOrder = tokenValue;
79 else if ( tokenTag == Constants::SAM_HD_SORTORDER_TAG ) m_header.SortOrder = tokenValue;
81 cerr << "SAM Format Error - unknown HD tag: " << tokenTag << endl;
84 // if @HD line exists, VN must be provided
85 if ( !m_header.HasVersion() ) {
86 cerr << "SAM Format Error - @HD line is missing VN tag!" << endl;
91 void SamFormatParser::ParseSQLine(const string& line) {
95 // split SQ line into tokens
96 vector<string> tokens = Split(line, Constants::SAM_TAB);
98 // iterate over tokens
99 vector<string>::const_iterator tokenIter = tokens.begin();
100 vector<string>::const_iterator tokenEnd = tokens.end();
101 for ( ; tokenIter != tokenEnd; ++tokenIter ) {
104 const string tokenTag = (*tokenIter).substr(0,2);
105 const string tokenValue = (*tokenIter).substr(3);
107 // set sequence contents
108 if ( tokenTag == Constants::SAM_SQ_NAME_TAG ) seq.Name = tokenValue;
109 else if ( tokenTag == Constants::SAM_SQ_LENGTH_TAG ) seq.Length = tokenValue;
110 else if ( tokenTag == Constants::SAM_SQ_ASSEMBLYID_TAG ) seq.AssemblyID = tokenValue;
111 else if ( tokenTag == Constants::SAM_SQ_URI_TAG ) seq.URI = tokenValue;
112 else if ( tokenTag == Constants::SAM_SQ_CHECKSUM_TAG ) seq.Checksum = tokenValue;
113 else if ( tokenTag == Constants::SAM_SQ_SPECIES_TAG ) seq.Species = tokenValue;
115 cerr << "SAM Format Error - unknown SQ tag: " << tokenTag << endl;
118 // if @SQ line exists, SN must be provided
119 if ( !seq.HasName() ) {
120 cerr << "SAM Format Error - @SQ line is missing SN tag!" << endl;
124 // if @SQ line exists, LN must be provided
125 if ( !seq.HasLength() ) {
126 cerr << "SAM Format Error - @SQ line is missing LN tag!" << endl;
130 // store SAM sequence entry
131 m_header.Sequences.Add(seq);
134 void SamFormatParser::ParseRGLine(const string& line) {
138 // split string into tokens
139 vector<string> tokens = Split(line, Constants::SAM_TAB);
141 // iterate over tokens
142 vector<string>::const_iterator tokenIter = tokens.begin();
143 vector<string>::const_iterator tokenEnd = tokens.end();
144 for ( ; tokenIter != tokenEnd; ++tokenIter ) {
146 // get token tag/value
147 const string tokenTag = (*tokenIter).substr(0,2);
148 const string tokenValue = (*tokenIter).substr(3);
150 // set read group contents
151 if ( tokenTag == Constants::SAM_RG_ID_TAG ) rg.ID = tokenValue;
152 else if ( tokenTag == Constants::SAM_RG_SAMPLE_TAG ) rg.Sample = tokenValue;
153 else if ( tokenTag == Constants::SAM_RG_LIBRARY_TAG ) rg.Library = tokenValue;
154 else if ( tokenTag == Constants::SAM_RG_DESCRIPTION_TAG ) rg.Description = tokenValue;
155 else if ( tokenTag == Constants::SAM_RG_PLATFORMUNIT_TAG ) rg.PlatformUnit = tokenValue;
156 else if ( tokenTag == Constants::SAM_RG_PREDICTEDINSERTSIZE_TAG ) rg.PredictedInsertSize = tokenValue;
157 else if ( tokenTag == Constants::SAM_RG_SEQCENTER_TAG ) rg.SequencingCenter = tokenValue;
158 else if ( tokenTag == Constants::SAM_RG_PRODUCTIONDATE_TAG ) rg.ProductionDate = tokenValue;
159 else if ( tokenTag == Constants::SAM_RG_SEQTECHNOLOGY_TAG ) rg.SequencingTechnology = tokenValue;
161 cerr << "SAM Format Error - unknown RG tag: " << tokenTag << endl;
164 // if @RG line exists, ID must be provided
166 cerr << "SAM Format Error - @RG line is missing ID tag!" << endl;
170 // if @RG line exists, SM must be provided
171 if ( !rg.HasSample() ) {
172 cerr << "SAM Format Error - @RG line is missing SM tag!" << endl;
176 // store SAM read group entry
177 m_header.ReadGroups.Add(rg);
180 void SamFormatParser::ParsePGLine(const string& line) {
182 // split string into tokens
183 vector<string> tokens = Split(line, Constants::SAM_TAB);
185 // iterate over tokens
186 vector<string>::const_iterator tokenIter = tokens.begin();
187 vector<string>::const_iterator tokenEnd = tokens.end();
188 for ( ; tokenIter != tokenEnd; ++tokenIter ) {
190 // get token tag/value
191 const string tokenTag = (*tokenIter).substr(0,2);
192 const string tokenValue = (*tokenIter).substr(3);
194 // set header contents
195 if ( tokenTag == Constants::SAM_PG_NAME_TAG ) m_header.ProgramName = tokenValue;
196 else if ( tokenTag == Constants::SAM_PG_VERSION_TAG ) m_header.ProgramVersion = tokenValue;
197 else if ( tokenTag == Constants::SAM_PG_COMMANDLINE_TAG ) m_header.ProgramCommandLine = tokenValue;
199 cerr << "SAM Format Error - unknown PG tag: " << tokenTag << endl;
202 // if @PG line exists, ID must be provided
203 if ( !m_header.HasProgramName() ) {
204 cerr << "SAM Format Error - @PG line is missing ID tag!" << endl;
209 void SamFormatParser::ParseCOLine(const string& line) {
210 // simply add line to comments list
211 m_header.Comments.push_back(line);
214 const vector<string> SamFormatParser::Split(const string& line, const char delim) {
215 vector<string> tokens;
216 stringstream lineStream(line);
218 while ( getline(lineStream, token, delim) )
219 tokens.push_back(token);