5 * Created by Sarah Westcott on 1/22/09.
6 * Copyright 2009 Schloss Lab UMASS Amherst. All rights reserved.
12 /***********************************************************************/
13 ReadTree::ReadTree() {
15 m = MothurOut::getInstance();
18 m->errorOut(e, "ReadTree", "ReadTree");
22 /***********************************************************************/
23 int ReadTree::AssembleTrees() {
25 //assemble users trees
26 for (int i = 0; i < Trees.size(); i++) {
27 if (m->control_pressed) { return 0; }
28 Trees[i]->assembleTree();
33 m->errorOut(e, "ReadTree", "AssembleTrees");
37 /***********************************************************************/
38 int ReadTree::readSpecialChar(istream& f, char c, string name) {
45 m->mothurOut("Error: Input file ends prematurely, expecting a " + name + "\n");
49 m->mothurOut("Error: Expected " + name + " in input file. Found " + toString(d) + ".\n");
52 if(d == ')' && f.peek() == '\n'){
58 m->errorOut(e, "ReadTree", "readSpecialChar");
62 /**************************************************************************************************/
64 int ReadTree::readNodeChar(istream& f) {
66 // while(isspace(d=f.get())) {;}
71 m->mothurOut("Error: Input file ends prematurely, expecting a left parenthesis\n");
77 m->errorOut(e, "ReadTree", "readNodeChar");
82 /**************************************************************************************************/
84 float ReadTree::readBranchLength(istream& f) {
89 m->mothurOut("Error: Missing branch length in input tree.\n");
96 m->errorOut(e, "ReadTree", "readBranchLength");
101 /***********************************************************************/
102 /***********************************************************************/
104 //Child Classes Below
106 /***********************************************************************/
107 /***********************************************************************/
108 //This class reads a file in Newick form and stores it in a tree.
110 int ReadNewickTree::read(CountTable* ct) {
116 //if you are not a nexus file
117 if ((c = filehandle.peek()) != '#') {
118 while((c = filehandle.peek()) != EOF) {
119 if (m->control_pressed) { filehandle.close(); return 0; }
120 while ((c = filehandle.peek()) != EOF) {
121 if (m->control_pressed) { filehandle.close(); return 0; }
129 if((c == '(') && (comment != 1)){ break; }
136 numNodes = T->getNumNodes();
137 numLeaves = T->getNumLeaves();
139 error = readTreeString(ct);
141 //save trees for later commands
143 m->gobble(filehandle);
145 //if you are a nexus file
146 }else if ((c = filehandle.peek()) == '#') {
147 //get right number of seqs from nexus file.
148 Tree* temp = new Tree(ct); delete temp;
150 nexusTranslation(ct); //reads file through the translation and updates treemap
151 while((c = filehandle.peek()) != EOF) {
152 if (m->control_pressed) { filehandle.close(); return 0; }
154 while ((c = filehandle.peek()) != EOF) {
155 if (m->control_pressed) { filehandle.close(); return 0; }
156 if(holder == "[" || holder == "[!"){
162 if((holder == "tree" || holder == "end;") && comment != 1){ holder = ""; comment = 0; break;}
163 filehandle >> holder;
166 //pass over the "tree rep.6878900 = "
167 while (((c = filehandle.get()) != '(') && ((c = filehandle.peek()) != EOF) ) {;}
169 if (c == EOF ) { break; }
170 filehandle.putback(c); //put back first ( of tree.
174 numNodes = T->getNumNodes();
175 numLeaves = T->getNumLeaves();
178 error = readTreeString(ct);
180 //save trees for later commands
185 if (error != 0) { readOk = error; }
191 catch(exception& e) {
192 m->errorOut(e, "ReadNewickTree", "read");
196 /**************************************************************************************************/
197 //This function read the file through the translation of the sequences names and updates treemap.
198 string ReadNewickTree::nexusTranslation(CountTable* ct) {
202 int numSeqs = m->Treenames.size(); //must save this some when we clear old names we can still know how many sequences there were
206 while(holder != "translate" && holder != "Translate"){
207 if(holder == "[" || holder == "[!"){
213 filehandle >> holder;
214 if(holder == "tree" && comment != 1){return holder;}
218 for(int i=0;i<numSeqs;i++){
220 filehandle >> number;
222 name.erase(name.end()-1); //erase the comma
223 ct->renameSeq(name, toString(number));
228 catch(exception& e) {
229 m->errorOut(e, "ReadNewickTree", "nexusTranslation");
234 /**************************************************************************************************/
235 int ReadNewickTree::readTreeString(CountTable* ct) {
243 int ch = filehandle.peek();
246 n = numLeaves; //number of leaves / sequences, we want node 1 to start where the leaves left off
248 lc = readNewickInt(filehandle, n, T, ct);
249 if (lc == -1) { m->mothurOut("error with lc"); m->mothurOutEndLine(); m->control_pressed = true; return -1; } //reports an error in reading
251 if(filehandle.peek()==','){
252 readSpecialChar(filehandle,',',"comma");
254 // ';' means end of tree.
255 else if((ch=filehandle.peek())==';' || ch=='['){
260 rc = readNewickInt(filehandle, n, T, ct);
261 if (rc == -1) { m->mothurOut("error with rc"); m->mothurOutEndLine(); m->control_pressed = true; return -1; } //reports an error in reading
262 if(filehandle.peek() == ')'){
263 readSpecialChar(filehandle,')',"right parenthesis");
267 //note: treeclimber had the code below added - not sure why?
269 filehandle.putback(ch);
271 filehandle.get(name, MAX_LINE,'\n');
272 SKIPLINE(filehandle, ch);
274 n = T->getIndex(name);
277 m->mothurOut("Internal error: The only taxon is not taxon 0.\n");
279 readOk = -1; return -1;
284 while(((ch=filehandle.get())!=';') && (filehandle.eof() != true)){;}
287 T->tree[n].setChildren(lc,rc);
288 T->tree[n].setBranchLength(0);
289 T->tree[n].setParent(-1);
290 if(lc!=-1){ T->tree[lc].setParent(n); }
291 if(rc!=-1){ T->tree[rc].setParent(n); }
294 //T->printTree(); cout << endl;
298 catch(exception& e) {
299 m->errorOut(e, "ReadNewickTree", "readTreeString");
303 /**************************************************************************************************/
305 int ReadNewickTree::readNewickInt(istream& f, int& n, Tree* T, CountTable* ct) {
308 if (m->control_pressed) { return -1; }
310 int c = readNodeChar(f);
314 //to account for multifurcating trees generated by fasttree, we are forcing them to be bifurcating
316 vector<int> childrenNodes;
317 while(f.peek() != ')'){
318 int child = readNewickInt(f, n, T, ct);
319 if (child == -1) { return -1; } //reports an error in reading
320 //cout << "child = " << child << endl;
321 childrenNodes.push_back(child);
323 //after a child you either have , or ), check for both
324 if(f.peek()==')'){ break; }
325 else if (f.peek()==',') { readSpecialChar(f,',',"comma"); }
328 //cout << childrenNodes.size() << endl;
329 if (childrenNodes.size() < 2) { m->mothurOut("Error in tree, please correct."); m->mothurOutEndLine(); return -1; }
331 //then force into 2 node structure
332 for (int i = 1; i < childrenNodes.size(); i++) {
335 if (i == 1) { lc = childrenNodes[i-1]; rc = childrenNodes[i]; }
336 else { lc = n-1; rc = childrenNodes[i]; }
337 //cout << i << '\t' << lc << '\t' << rc << endl;
338 T->tree[n].setChildren(lc,rc);
339 T->tree[lc].setParent(n);
340 T->tree[rc].setParent(n);
342 //T->printTree(); cout << endl;
346 //to account for extra ++ in looping
350 readSpecialChar(f,')',"right parenthesis");
351 //to pass over labels in trees
353 while((c!=',') && (c != -1) && (c!= ':') && (c!=';')&& (c!=')')){ c=filehandle.get(); }
354 filehandle.putback(c);
358 readSpecialChar(f,':',"colon");
360 if(n >= numNodes){ m->mothurOut("Error: Too many nodes in input tree\n"); readOk = -1; return -1; }
362 T->tree[n].setBranchLength(readBranchLength(f));
364 T->tree[n].setBranchLength(0.0);
373 while(d != ':' && d != ',' && d!=')' && d!='\n'){
377 //cout << name << endl;
379 if(d == ':') { blen = 1; }
384 vector<string> group = ct->getGroups(name);
385 //cout << name << endl;
386 //find index in tree of name
387 int n1 = T->getIndex(name);
389 //adds sequence names that are not in group file to the "xxx" group
390 if(group.size() == 0) {
391 m->mothurOut("Name: " + name + " is not in your groupfile, and will be disregarded. \n"); //readOk = -1; return n1;
393 vector<string> currentGroups = ct->getNamesOfGroups();
394 if (!m->inUsersGroups("xxx", currentGroups)) { ct->addGroup("xxx"); }
395 currentGroups = ct->getNamesOfGroups();
396 vector<int> thisCounts; thisCounts.resize(currentGroups.size(), 0);
397 for (int h = 0; h < currentGroups.size(); h++) {
398 if (currentGroups[h] == "xxx") { thisCounts[h] = 1; break; }
400 ct->push_back(name, thisCounts);
402 group.push_back("xxx");
404 T->tree[n1].setGroup(group);
405 T->tree[n1].setChildren(-1,-1);
409 T->tree[n1].setBranchLength(readBranchLength(f));
411 T->tree[n1].setBranchLength(0.0);
414 while((c=f.get())!=0 && (c != ':' && c != ',' && c!=')') ) {;}
421 catch(exception& e) {
422 m->errorOut(e, "ReadNewickTree", "readNewickInt");
426 /**************************************************************************************************/
427 /**************************************************************************************************/