2 * splitgroupscommand.cpp
5 * Created by westcott on 9/20/10.
6 * Copyright 2010 Schloss Lab. All rights reserved.
10 #include "splitgroupscommand.h"
11 #include "sharedutilities.h"
13 //**********************************************************************************************************************
14 vector<string> SplitGroupCommand::setParameters(){
16 CommandParameter pfasta("fasta", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(pfasta);
17 CommandParameter pname("name", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pname);
18 CommandParameter pgroup("group", "InputTypes", "", "", "none", "none", "none",false,true); parameters.push_back(pgroup);
19 CommandParameter pgroups("groups", "String", "", "", "", "", "",false,false); parameters.push_back(pgroups);
20 CommandParameter pinputdir("inputdir", "String", "", "", "", "", "",false,false); parameters.push_back(pinputdir);
21 CommandParameter poutputdir("outputdir", "String", "", "", "", "", "",false,false); parameters.push_back(poutputdir);
23 vector<string> myArray;
24 for (int i = 0; i < parameters.size(); i++) { myArray.push_back(parameters[i].name); }
28 m->errorOut(e, "SplitGroupCommand", "setParameters");
32 //**********************************************************************************************************************
33 string SplitGroupCommand::getHelpString(){
35 string helpString = "";
36 helpString += "The split.group command reads a group file, and parses your fasta and names files by groups. \n";
37 helpString += "The split.group command parameters are fasta, name, group and groups.\n";
38 helpString += "The fasta and group parameters are required.\n";
39 helpString += "The groups parameter allows you to select groups to create files for. \n";
40 helpString += "For example if you set groups=A-B-C, you will get a .A.fasta, .A.names, .B.fasta, .B.names, .C.fasta, .C.names files. \n";
41 helpString += "If you want .fasta and .names files for all groups, set groups=all. \n";
42 helpString += "The split.group command should be used in the following format: split.group(fasta=yourFasta, group=yourGroupFile).\n";
43 helpString += "Example: split.group(fasta=abrecovery.fasta, group=abrecovery.groups).\n";
44 helpString += "Note: No spaces between parameter labels (i.e. fasta), '=' and parameters (i.e.yourFasta).\n";
48 m->errorOut(e, "SplitGroupCommand", "getHelpString");
52 //**********************************************************************************************************************
53 SplitGroupCommand::SplitGroupCommand(){
55 abort = true; calledHelp = true;
57 vector<string> tempOutNames;
58 outputTypes["fasta"] = tempOutNames;
59 outputTypes["name"] = tempOutNames;
62 m->errorOut(e, "SplitGroupCommand", "SplitGroupCommand");
66 //**********************************************************************************************************************
67 SplitGroupCommand::SplitGroupCommand(string option) {
69 abort = false; calledHelp = false;
71 //allow user to run help
72 if(option == "help") { help(); abort = true; calledHelp = true; }
73 else if(option == "citation") { citation(); abort = true; calledHelp = true;}
76 vector<string> myArray = setParameters();
78 OptionParser parser(option);
79 map<string, string> parameters = parser.getParameters();
81 ValidParameters validParameter;
82 map<string, string>::iterator it;
84 //check to make sure all parameters are valid for command
85 for (it = parameters.begin(); it != parameters.end(); it++) {
86 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
89 //initialize outputTypes
90 vector<string> tempOutNames;
91 outputTypes["fasta"] = tempOutNames;
92 outputTypes["name"] = tempOutNames;
94 //if the user changes the input directory command factory will send this info to us in the output parameter
95 string inputDir = validParameter.validFile(parameters, "inputdir", false);
96 if (inputDir == "not found"){ inputDir = ""; }
99 it = parameters.find("group");
100 //user has given a template file
101 if(it != parameters.end()){
102 path = m->hasPath(it->second);
103 //if the user has not given a path then, add inputdir. else leave path alone.
104 if (path == "") { parameters["group"] = inputDir + it->second; }
107 it = parameters.find("fasta");
108 //user has given a template file
109 if(it != parameters.end()){
110 path = m->hasPath(it->second);
111 //if the user has not given a path then, add inputdir. else leave path alone.
112 if (path == "") { parameters["fasta"] = inputDir + it->second; }
115 it = parameters.find("name");
116 //user has given a template file
117 if(it != parameters.end()){
118 path = m->hasPath(it->second);
119 //if the user has not given a path then, add inputdir. else leave path alone.
120 if (path == "") { parameters["name"] = inputDir + it->second; }
126 namefile = validParameter.validFile(parameters, "name", true);
127 if (namefile == "not open") { abort = true; }
128 else if (namefile == "not found") { namefile = ""; }
129 else { m->setNameFile(namefile); }
131 fastafile = validParameter.validFile(parameters, "fasta", true);
132 if (fastafile == "not open") { abort = true; }
133 else if (fastafile == "not found") {
134 fastafile = m->getFastaFile();
135 if (fastafile != "") { m->mothurOut("Using " + fastafile + " as input file for the fasta parameter."); m->mothurOutEndLine(); }
136 else { m->mothurOut("You have no current fastafile and the fasta parameter is required."); m->mothurOutEndLine(); abort = true; }
137 }else { m->setFastaFile(fastafile); }
139 groupfile = validParameter.validFile(parameters, "group", true);
140 if (groupfile == "not open") { groupfile = ""; abort = true; }
141 else if (groupfile == "not found") {
142 groupfile = m->getGroupFile();
143 if (groupfile != "") { m->mothurOut("Using " + groupfile + " as input file for the group parameter."); m->mothurOutEndLine(); }
144 else { m->mothurOut("You have no current groupfile and the group parameter is required."); m->mothurOutEndLine(); abort = true; }
145 }else { m->setGroupFile(groupfile); }
147 groups = validParameter.validFile(parameters, "groups", false);
148 if (groups == "not found") { groups = ""; }
149 else { m->splitAtDash(groups, Groups); }
151 //if the user changes the output directory command factory will send this info to us in the output parameter
152 outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){ outputDir = m->hasPath(groupfile); }
156 catch(exception& e) {
157 m->errorOut(e, "SplitGroupCommand", "SplitAbundCommand");
161 //**********************************************************************************************************************
162 int SplitGroupCommand::execute(){
165 if (abort == true) { if (calledHelp) { return 0; } return 2; }
167 groupMap = new GroupMap(groupfile);
170 vector<string> namesGroups = groupMap->getNamesOfGroups();
171 SharedUtil util; util.setGroups(Groups, namesGroups);
173 if (namefile != "") { readNames(); }
178 if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; }
181 itTypes = outputTypes.find("fasta");
182 if (itTypes != outputTypes.end()) {
183 if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setFastaFile(current); }
186 itTypes = outputTypes.find("name");
187 if (itTypes != outputTypes.end()) {
188 if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setNameFile(current); }
191 m->mothurOutEndLine();
192 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
193 for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
194 m->mothurOutEndLine();
198 catch(exception& e) {
199 m->errorOut(e, "SplitGroupCommand", "execute");
203 /**********************************************************************************************************************/
204 int SplitGroupCommand::readNames() {
208 m->openInputFile(namefile, in);
211 if (m->control_pressed) { break; }
213 string firstCol, secondCol;
214 in >> firstCol >> secondCol; m->gobble(in);
216 vector<string> names;
217 m->splitAtComma(secondCol, names);
219 nameMap[firstCol] = names;
226 catch(exception& e) {
227 m->errorOut(e, "SplitGroupCommand", "readNames");
232 /**********************************************************************************************************************/
233 int SplitGroupCommand::splitFasta() {
236 string filerootFasta = outputDir + m->getRootName(m->getSimpleName(fastafile));
237 string filerootName = outputDir + m->getRootName(m->getSimpleName(namefile));
240 map<string, ofstream*> filehandles;
241 map<string, ofstream*>::iterator it3;
243 for (int i=0; i<Groups.size(); i++) {
245 filehandles[Groups[i]+"fasta"] = temp;
246 m->openOutputFile(filerootFasta + Groups[i] + ".fasta", *(filehandles[Groups[i]+"fasta"]));
247 outputNames.push_back(filerootFasta + Groups[i] + ".fasta"); outputTypes["fasta"].push_back(filerootFasta + Groups[i] + ".fasta");
249 if (namefile != "") {
250 temp2 = new ofstream;
251 filehandles[Groups[i]+"name"] = temp2;
252 m->openOutputFile(filerootName + Groups[i] + ".names", *(filehandles[Groups[i]+"name"]));
253 outputNames.push_back(filerootName + Groups[i] + ".names"); outputTypes["name"].push_back(filerootFasta + Groups[i] + ".names");
259 m->openInputFile(fastafile, in);
262 if (m->control_pressed) { break; }
264 Sequence seq(in); m->gobble(in);
266 if (seq.getName() != "") {
268 itNames = nameMap.find(seq.getName());
270 if (itNames == nameMap.end()) {
271 if (namefile != "") { m->mothurOut(seq.getName() + " is not in your namesfile, ignoring."); m->mothurOutEndLine(); }
272 else { //no names file given
273 string group = groupMap->getGroup(seq.getName());
275 if (m->inUsersGroups(group, Groups)) { //only add if this is in a group we want
276 seq.printSequence(*(filehandles[group+"fasta"]));
277 }else if(group == "not found") {
278 m->mothurOut(seq.getName() + " is not in your groupfile. Ignoring."); m->mothurOutEndLine();
282 vector<string> names = itNames->second;
283 map<string, string> group2Names;
284 map<string, string>::iterator it;
286 for (int i = 0; i < names.size(); i++) { //build strings for new group names file, will select rep below
287 string group = groupMap->getGroup(names[i]);
289 if (m->inUsersGroups(group, Groups)) { //only add if this is in a group we want
290 it = group2Names.find(group);
291 if (it == group2Names.end()) {
292 group2Names[group] = names[i] + ",";
294 group2Names[group] += names[i] + ",";
296 }else if(group == "not found") {
297 m->mothurOut(names[i] + " is not in your groupfile. Ignoring."); m->mothurOutEndLine();
301 for (map<string, string>::iterator itGroups = group2Names.begin(); itGroups != group2Names.end(); itGroups++) {
302 //edit names string, by grabbing the first guy as the rep and removing the last comma
303 string newNames = itGroups->second;
304 newNames = newNames.substr(0, newNames.length()-1);
307 int pos = newNames.find_first_of(',');
308 if (pos == newNames.npos) { //only one sequence so it represents itself
311 repName = newNames.substr(0, pos);
314 *(filehandles[itGroups->first+"name"]) << repName << '\t' << newNames << endl;
315 *(filehandles[itGroups->first+"fasta"]) << ">" << repName << endl << seq.getAligned() << endl;
323 for (it3 = filehandles.begin(); it3 != filehandles.end(); it3++) {
324 (*(filehandles[it3->first])).close();
328 vector<string> newOutputNames;
330 for (int i = 0; i < outputNames.size(); i++) {
331 if (m->isBlank(outputNames[i])) {
332 m->mothurRemove(outputNames[i]);
333 }else { newOutputNames.push_back(outputNames[i]); }
335 outputNames = newOutputNames;
340 catch(exception& e) {
341 m->errorOut(e, "SplitGroupCommand", "splitFasta");
345 /**********************************************************************************************************************/