]> git.donarmstrong.com Git - mothur.git/blobdiff - readtree.cpp
fixed bug in read.tree
[mothur.git] / readtree.cpp
index 9b96861a661a6a794120f2d47899222f3664e33b..1910c4a5a1761e4cf73cd95ce8a1df9b97ee4b33 100644 (file)
 #include "readtree.h"
 
 /***********************************************************************/
-//Parent Class
-// The following functions are used by all reading formats.
-/***********************************************************************/
-ReadTree::ReadTree() { 
+ReadTree::ReadTree() {
        try {
-               globaldata = GlobalData::getInstance(); 
-               T = new Tree(); 
-               numNodes = T->getNumNodes();
-               numLeaves = T->getNumLeaves();
+               globaldata = GlobalData::getInstance();
+               globaldata->gTree.clear();
        }
        catch(exception& e) {
                cout << "Standard Error: " << e.what() << " has occurred in the ReadTree class Function ReadTree. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
@@ -32,9 +27,10 @@ ReadTree::ReadTree() {
 /***********************************************************************/
 int ReadTree::readSpecialChar(istream& f, char c, string name) {
     try {
-               char d;
        
-               while(isspace(d=f.get()))               {;}
+               gobble(f);
+               char d = f.get();
+       
                if(d == EOF){
                        cerr << "Error: Input file ends prematurely, expecting a " << name << "\n";
                        exit(1);
@@ -44,10 +40,8 @@ int ReadTree::readSpecialChar(istream& f, char c, string name) {
                        exit(1);
                }
                if(d == ')' && f.peek() == '\n'){
-                       while(isspace(d=f.get()))               {;}
-                       f.putback(d);
+                       gobble(f);
                }       
-       
                return d;
        }
        catch(exception& e) {
@@ -63,8 +57,10 @@ int ReadTree::readSpecialChar(istream& f, char c, string name) {
 
 int ReadTree::readNodeChar(istream& f) {
        try {
-               char d;
-               while(isspace(d=f.get()))               {;}
+//             while(isspace(d=f.get()))               {;}
+               gobble(f);
+               char d = f.get();
+
                if(d == EOF){
                        cerr << "Error: Input file ends prematurely, expecting a left parenthesis\n";
                        exit(1);
@@ -91,7 +87,7 @@ float ReadTree::readBranchLength(istream& f) {
                        cerr << "Error: Missing branch length in input tree.\n";
                        exit(1);
                }
-    
+               gobble(f);
                return b;
        }
        catch(exception& e) {
@@ -117,6 +113,116 @@ float ReadTree::readBranchLength(istream& f) {
 
 void ReadNewickTree::read() {
        try {
+               int c;
+               int comment = 0;
+               
+               //if you are not a nexus file 
+               if ((c = filehandle.peek()) != '#') {  
+                       while((c = filehandle.peek()) != EOF) { 
+                               //make new tree
+                               T = new Tree(); 
+                               numNodes = T->getNumNodes();
+                               numLeaves = T->getNumLeaves();
+                               
+                               readTreeString();  
+                               //save trees for later commands
+                               globaldata->gTree.push_back(T); 
+                               gobble(filehandle);
+                       }
+               //if you are a nexus file
+               }else if ((c = filehandle.peek()) == '#') {
+                       nexusTranslation();  //reads file through the translation and updates treemap
+                       while((c = filehandle.peek()) != EOF) { 
+                               // get past comments
+                               while ((c = filehandle.peek()) != EOF) {        
+                                       if(holder == "[" || holder == "[!"){
+                                               comment = 1;
+                                       }
+                                       if(holder == "]"){
+                                               comment = 0;
+                                       }
+                                       if((holder == "tree" || holder == "end;") && comment != 1){ holder = ""; comment = 0; break;}
+                                       filehandle >> holder;
+                               }
+                       
+                               //pass over the "tree rep.6878900 = "
+                               while (((c = filehandle.get()) != '(') && ((c = filehandle.peek()) != EOF) ) {;}
+                                       
+                               if (c == EOF ) { break; }
+                               filehandle.putback(c);  //put back first ( of tree.
+                               
+                               //make new tree
+                               T = new Tree(); 
+                               numNodes = T->getNumNodes();
+                               numLeaves = T->getNumLeaves();
+                               
+                               //read tree info
+                               readTreeString();  
+                               //save trees for later commands
+                               globaldata->gTree.push_back(T); 
+                       }
+               }
+               
+       }
+       catch(exception& e) {
+               cout << "Standard Error: " << e.what() << " has occurred in the ReadNewickTree class Function read. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+               exit(1);
+       }
+       catch(...) {
+               cout << "An unknown error has occurred in the ReadNewickTree class function read. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+               exit(1);
+       }               
+}
+/**************************************************************************************************/
+//This function read the file through the translation of the sequences names and updates treemap.
+void ReadNewickTree::nexusTranslation() {
+       try {
+               
+               holder = "";
+               int numSeqs = globaldata->gTreemap->getNumSeqs(); //must save this some when we clear old names we can still know how many sequences there were
+               int comment = 0;
+               
+               // get past comments
+               while(holder != "translate" && holder != "Translate"){  
+                       if(holder == "[" || holder == "[!"){
+                               comment = 1;
+                       }
+                       if(holder == "]"){
+                               comment = 0;
+                       }
+                       filehandle >> holder; 
+                       if(holder == "tree" && comment != 1){return;}
+               }
+               
+               //update treemap
+               globaldata->gTreemap->namesOfSeqs.clear();
+               for(int i=0;i<numSeqs;i++){
+                       string number, name;
+                       filehandle >> number;
+                       filehandle >> name;
+                       name.erase(name.end()-1);  //erase the comma
+                       //insert new one with new name
+                       globaldata->gTreemap->treemap[toString(number)].groupname = globaldata->gTreemap->treemap[name].groupname;
+                       globaldata->gTreemap->treemap[toString(number)].vectorIndex = globaldata->gTreemap->treemap[name].vectorIndex;
+                       //erase old one.  so treemap[sarah].groupnumber is now treemap[1].groupnumber. if number is 1 and name is sarah.
+                       globaldata->gTreemap->treemap.erase(name);
+                       globaldata->gTreemap->namesOfSeqs.push_back(number);
+               }
+       }
+       catch(exception& e) {
+               cout << "Standard Error: " << e.what() << " has occurred in the ReadNewickTree class Function nexus. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+               exit(1);
+       }
+       catch(...) {
+               cout << "An unknown error has occurred in the ReadNewickTree class function nexus. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+               exit(1);
+       }               
+}
+
+/**************************************************************************************************/
+void ReadNewickTree::readTreeString() {
+       try {
+               
                int n = 0;
                int lc, rc; 
                
@@ -126,6 +232,7 @@ void ReadNewickTree::read() {
                
                if(ch == '('){
                        n = numLeaves;  //number of leaves / sequences, we want node 1 to start where the leaves left off
+
                        lc = readNewickInt(filehandle, n, T);
                
                        if(filehandle.peek()==','){                                                     
@@ -143,14 +250,14 @@ void ReadNewickTree::read() {
                        }                                                                                               
                }
                //note: treeclimber had the code below added - not sure why?
-                else{
+               else{
                        filehandle.putback(ch);
                        char name[MAX_LINE];
                        filehandle.get(name, MAX_LINE,'\n');
                        SKIPLINE(filehandle, ch);
                
-               
                        n = T->getIndex(name);
+
                        if(n!=0){
                                cerr << "Internal error: The only taxon is not taxon 0.\n";
                                exit(1);
@@ -159,25 +266,24 @@ void ReadNewickTree::read() {
                } 
                
                while((ch=filehandle.get())!=';'){;}                                            
-                       if(rooted != 1){                                                                        
+               if(rooted != 1){                                                                        
                        T->tree[n].setChildren(lc,rc);
                        T->tree[n].setBranchLength(0);
                        T->tree[n].setParent(-1);
                        if(lc!=-1){             T->tree[lc].setParent(n);               }
                        if(rc!=-1){             T->tree[rc].setParent(n);               }
                }
-               
-               //save tree for later commands
-               globaldata->gTree = T;
+       
        }
        catch(exception& e) {
-               cout << "Standard Error: " << e.what() << " has occurred in the ReadNewickTree class Function read. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+               cout << "Standard Error: " << e.what() << " has occurred in the ReadNewickTree class Function readTreeString. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
                exit(1);
        }
        catch(...) {
-               cout << "An unknown error has occurred in the ReadNewickTree class function read. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
+               cout << "An unknown error has occurred in the ReadNewickTree class function readTreeString. Please contact Pat Schloss at pschloss@microbio.umass.edu." << "\n";
                exit(1);
        }               
+
 }
 /**************************************************************************************************/
 
@@ -188,7 +294,7 @@ int ReadNewickTree::readNewickInt(istream& f, int& n, Tree* T) {
                if(c == '('){
                        int lc = readNewickInt(f, n, T);
                        readSpecialChar(f,',',"comma");
-               
+
                        int rc = readNewickInt(f, n, T);                
                        if(f.peek()==')'){      
                                readSpecialChar(f,')',"right parenthesis");                                     
@@ -255,4 +361,4 @@ int ReadNewickTree::readNewickInt(istream& f, int& n, Tree* T) {
 }
 /**************************************************************************************************/
 /**************************************************************************************************/
-    
\ No newline at end of file
+