1 // ***************************************************************************
2 // SamFormatParser.cpp (c) 2010 Derek Barnett
3 // Marth Lab, Department of Biology, Boston College
4 // All rights reserved.
5 // ---------------------------------------------------------------------------
6 // Last modified: 19 April 2011 (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);
44 void SamFormatParser::ParseSamLine(const string& line) {
46 // skip if line is not long enough to contain true values
47 if (line.length() < 5 ) return;
49 // determine token at beginning of line
50 const string firstToken = line.substr(0,3);
51 string restOfLine = line.substr(4);
52 if ( firstToken == Constants::SAM_HD_BEGIN_TOKEN) ParseHDLine(restOfLine);
53 else if ( firstToken == Constants::SAM_SQ_BEGIN_TOKEN) ParseSQLine(restOfLine);
54 else if ( firstToken == Constants::SAM_RG_BEGIN_TOKEN) ParseRGLine(restOfLine);
55 else if ( firstToken == Constants::SAM_PG_BEGIN_TOKEN) ParsePGLine(restOfLine);
56 else if ( firstToken == Constants::SAM_CO_BEGIN_TOKEN) ParseCOLine(restOfLine);
58 cerr << "SamFormatParser ERROR: unknown token: " << firstToken << endl;
61 void SamFormatParser::ParseHDLine(const string& line) {
63 // split HD lines into tokens
64 vector<string> tokens = Split(line, Constants::SAM_TAB);
66 // iterate over tokens
67 vector<string>::const_iterator tokenIter = tokens.begin();
68 vector<string>::const_iterator tokenEnd = tokens.end();
69 for ( ; tokenIter != tokenEnd; ++tokenIter ) {
72 const string tokenTag = (*tokenIter).substr(0,2);
73 const string tokenValue = (*tokenIter).substr(3);
75 // set header contents
76 if ( tokenTag == Constants::SAM_HD_VERSION_TAG ) m_header.Version = tokenValue;
77 else if ( tokenTag == Constants::SAM_HD_SORTORDER_TAG ) m_header.SortOrder = tokenValue;
78 else if ( tokenTag == Constants::SAM_HD_GROUPORDER_TAG ) m_header.GroupOrder = tokenValue;
80 cerr << "SamFormatParser ERROR: unknown HD tag: " << tokenTag << endl;
83 // if @HD line exists, VN must be provided
84 if ( !m_header.HasVersion() )
85 cerr << "SamFormatParser ERROR: @HD line is missing VN tag" << endl;
88 void SamFormatParser::ParseSQLine(const string& line) {
92 // split SQ line into tokens
93 vector<string> tokens = Split(line, Constants::SAM_TAB);
95 // iterate over tokens
96 vector<string>::const_iterator tokenIter = tokens.begin();
97 vector<string>::const_iterator tokenEnd = tokens.end();
98 for ( ; tokenIter != tokenEnd; ++tokenIter ) {
101 const string tokenTag = (*tokenIter).substr(0,2);
102 const string tokenValue = (*tokenIter).substr(3);
104 // set sequence contents
105 if ( tokenTag == Constants::SAM_SQ_NAME_TAG ) seq.Name = tokenValue;
106 else if ( tokenTag == Constants::SAM_SQ_LENGTH_TAG ) seq.Length = tokenValue;
107 else if ( tokenTag == Constants::SAM_SQ_ASSEMBLYID_TAG ) seq.AssemblyID = tokenValue;
108 else if ( tokenTag == Constants::SAM_SQ_CHECKSUM_TAG ) seq.Checksum = tokenValue;
109 else if ( tokenTag == Constants::SAM_SQ_SPECIES_TAG ) seq.Species = tokenValue;
110 else if ( tokenTag == Constants::SAM_SQ_URI_TAG ) seq.URI = tokenValue;
112 cerr << "SamFormatParser ERROR: unknown SQ tag: " << tokenTag << endl;
115 bool isMissingRequiredFields = false;
117 // if @SQ line exists, SN must be provided
118 if ( !seq.HasName() ) {
119 isMissingRequiredFields = true;
120 cerr << "SamFormatParser ERROR: @SQ line is missing SN tag" << endl;
123 // if @SQ line exists, LN must be provided
124 if ( !seq.HasLength() ) {
125 isMissingRequiredFields = true;
126 cerr << "SamFormatParser ERROR: @SQ line is missing LN tag" << endl;
129 // store SAM sequence entry
130 if ( !isMissingRequiredFields )
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_DESCRIPTION_TAG ) rg.Description = tokenValue;
153 else if ( tokenTag == Constants::SAM_RG_FLOWORDER_TAG ) rg.FlowOrder = tokenValue;
154 else if ( tokenTag == Constants::SAM_RG_KEYSEQUENCE_TAG ) rg.KeySequence = tokenValue;
155 else if ( tokenTag == Constants::SAM_RG_LIBRARY_TAG ) rg.Library = tokenValue;
156 else if ( tokenTag == Constants::SAM_RG_PLATFORMUNIT_TAG ) rg.PlatformUnit = tokenValue;
157 else if ( tokenTag == Constants::SAM_RG_PREDICTEDINSERTSIZE_TAG ) rg.PredictedInsertSize = tokenValue;
158 else if ( tokenTag == Constants::SAM_RG_PRODUCTIONDATE_TAG ) rg.ProductionDate = tokenValue;
159 else if ( tokenTag == Constants::SAM_RG_PROGRAM_TAG ) rg.Program = tokenValue;
160 else if ( tokenTag == Constants::SAM_RG_SAMPLE_TAG ) rg.Sample = tokenValue;
161 else if ( tokenTag == Constants::SAM_RG_SEQCENTER_TAG ) rg.SequencingCenter = tokenValue;
162 else if ( tokenTag == Constants::SAM_RG_SEQTECHNOLOGY_TAG ) rg.SequencingTechnology = tokenValue;
164 cerr << "SamFormatParser ERROR: unknown RG tag: " << tokenTag << endl;
167 bool isMissingRequiredFields = false;
169 // if @RG line exists, ID must be provided
171 isMissingRequiredFields = true;
172 cerr << "SamFormatParser ERROR: @RG line is missing ID tag" << endl;
175 // store SAM read group entry
176 if ( !isMissingRequiredFields )
177 m_header.ReadGroups.Add(rg);
180 void SamFormatParser::ParsePGLine(const string& line) {
184 // split string into tokens
185 vector<string> tokens = Split(line, Constants::SAM_TAB);
187 // iterate over tokens
188 vector<string>::const_iterator tokenIter = tokens.begin();
189 vector<string>::const_iterator tokenEnd = tokens.end();
190 for ( ; tokenIter != tokenEnd; ++tokenIter ) {
192 // get token tag/value
193 const string tokenTag = (*tokenIter).substr(0,2);
194 const string tokenValue = (*tokenIter).substr(3);
196 // set program record contents
197 if ( tokenTag == Constants::SAM_PG_ID_TAG ) pg.ID = tokenValue;
198 else if ( tokenTag == Constants::SAM_PG_NAME_TAG ) pg.Name = tokenValue;
199 else if ( tokenTag == Constants::SAM_PG_COMMANDLINE_TAG ) pg.CommandLine = tokenValue;
200 else if ( tokenTag == Constants::SAM_PG_PREVIOUSPROGRAM_TAG ) pg.PreviousProgramID = tokenValue;
201 else if ( tokenTag == Constants::SAM_PG_VERSION_TAG ) pg.Version = tokenValue;
203 cerr << "SamFormatParser ERROR: unknown PG tag: " << tokenTag << endl;
206 bool isMissingRequiredFields = false;
208 // if @PG line exists, ID must be provided
210 isMissingRequiredFields = true;
211 cerr << "SamFormatParser ERROR: @PG line is missing ID tag" << endl;
214 // store SAM program record
215 if ( !isMissingRequiredFields )
216 m_header.Programs.Add(pg);
219 void SamFormatParser::ParseCOLine(const string& line) {
220 // simply add line to comments list
221 m_header.Comments.push_back(line);
224 const vector<string> SamFormatParser::Split(const string& line, const char delim) {
225 vector<string> tokens;
226 stringstream lineStream(line);
228 while ( getline(lineStream, token, delim) )
229 tokens.push_back(token);