/***********************************************************************/
ReadTree::ReadTree() {
try {
- globaldata = GlobalData::getInstance();
m = MothurOut::getInstance();
- globaldata->gTree.clear();
}
catch(exception& e) {
m->errorOut(e, "ReadTree", "ReadTree");
}
}
/***********************************************************************/
+int ReadTree::AssembleTrees() {
+ try {
+ //assemble users trees
+ for (int i = 0; i < Trees.size(); i++) {
+ if (m->control_pressed) { return 0; }
+ Trees[i]->assembleTree();
+ }
+ return 0;
+ }
+ catch(exception& e) {
+ m->errorOut(e, "ReadTree", "AssembleTrees");
+ exit(1);
+ }
+}
+/***********************************************************************/
int ReadTree::readSpecialChar(istream& f, char c, string name) {
try {
- gobble(f);
+ m->gobble(f);
char d = f.get();
if(d == EOF){
exit(1);
}
if(d == ')' && f.peek() == '\n'){
- gobble(f);
+ m->gobble(f);
}
return d;
}
int ReadTree::readNodeChar(istream& f) {
try {
// while(isspace(d=f.get())) {;}
- gobble(f);
+ m->gobble(f);
char d = f.get();
if(d == EOF){
m->mothurOut("Error: Missing branch length in input tree.\n");
exit(1);
}
- gobble(f);
+ m->gobble(f);
return b;
}
catch(exception& e) {
/***********************************************************************/
//This class reads a file in Newick form and stores it in a tree.
-int ReadNewickTree::read() {
+int ReadNewickTree::read(CountTable* ct) {
try {
holder = "";
int c, error;
//if you are not a nexus file
if ((c = filehandle.peek()) != '#') {
- while((c = filehandle.peek()) != EOF) {
+ while((c = filehandle.peek()) != EOF) {
+ if (m->control_pressed) { filehandle.close(); return 0; }
while ((c = filehandle.peek()) != EOF) {
+ if (m->control_pressed) { filehandle.close(); return 0; }
// get past comments
if(c == '[') {
comment = 1;
}
//make new tree
- T = new Tree();
+ T = new Tree(ct);
numNodes = T->getNumNodes();
numLeaves = T->getNumLeaves();
- error = readTreeString();
+ error = readTreeString(ct);
//save trees for later commands
- globaldata->gTree.push_back(T);
- gobble(filehandle);
+ Trees.push_back(T);
+ m->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 right number of seqs from nexus file.
+ Tree* temp = new Tree(ct); delete temp;
+
+ nexusTranslation(ct); //reads file through the translation and updates treemap
+ while((c = filehandle.peek()) != EOF) {
+ if (m->control_pressed) { filehandle.close(); return 0; }
// get past comments
- while ((c = filehandle.peek()) != EOF) {
+ while ((c = filehandle.peek()) != EOF) {
+ if (m->control_pressed) { filehandle.close(); return 0; }
if(holder == "[" || holder == "[!"){
comment = 1;
}
filehandle.putback(c); //put back first ( of tree.
//make new tree
- T = new Tree();
+ T = new Tree(ct);
numNodes = T->getNumNodes();
numLeaves = T->getNumLeaves();
//read tree info
- error = readTreeString();
+ error = readTreeString(ct);
//save trees for later commands
- globaldata->gTree.push_back(T);
+ Trees.push_back(T);
}
}
if (error != 0) { readOk = error; }
filehandle.close();
+
return readOk;
}
catch(exception& e) {
}
/**************************************************************************************************/
//This function read the file through the translation of the sequences names and updates treemap.
-void ReadNewickTree::nexusTranslation() {
+string ReadNewickTree::nexusTranslation(CountTable* ct) {
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 numSeqs = m->Treenames.size(); //must save this some when we clear old names we can still know how many sequences there were
int comment = 0;
// get past comments
comment = 0;
}
filehandle >> holder;
- if(holder == "tree" && comment != 1){return;}
+ if(holder == "tree" && comment != 1){return holder;}
}
-
- //update treemap
- globaldata->gTreemap->namesOfSeqs.clear();
+
+ string number, name;
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);
+ ct->renameSeq(name, toString(number));
}
+
+ return name;
}
catch(exception& e) {
m->errorOut(e, "ReadNewickTree", "nexusTranslation");
}
/**************************************************************************************************/
-int ReadNewickTree::readTreeString() {
+int ReadNewickTree::readTreeString(CountTable* ct) {
try {
int n = 0;
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 (lc == -1) { m->mothurOut("error with lc"); m->mothurOutEndLine(); return -1; } //reports an error in reading
+ lc = readNewickInt(filehandle, n, T, ct);
+ if (lc == -1) { m->mothurOut("error with lc"); m->mothurOutEndLine(); m->control_pressed = true; return -1; } //reports an error in reading
if(filehandle.peek()==','){
readSpecialChar(filehandle,',',"comma");
// ';' means end of tree.
else if((ch=filehandle.peek())==';' || ch=='['){
rooted = 1;
- }
+ }
+
if(rooted != 1){
- rc = readNewickInt(filehandle, n, T);
- if (rc == -1) { m->mothurOut("error with rc"); m->mothurOutEndLine(); return -1; } //reports an error in reading
+ rc = readNewickInt(filehandle, n, T, ct);
+ if (rc == -1) { m->mothurOut("error with rc"); m->mothurOutEndLine(); m->control_pressed = true; return -1; } //reports an error in reading
if(filehandle.peek() == ')'){
readSpecialChar(filehandle,')',"right parenthesis");
}
- }
+ }
}
//note: treeclimber had the code below added - not sure why?
else{
lc = rc = -1;
}
- while(((ch=filehandle.get())!=';') && (filehandle.eof() != true)){;}
+ while(((ch=filehandle.get())!=';') && (filehandle.eof() != true)){;}
+
if(rooted != 1){
T->tree[n].setChildren(lc,rc);
T->tree[n].setBranchLength(0);
if(lc!=-1){ T->tree[lc].setParent(n); }
if(rc!=-1){ T->tree[rc].setParent(n); }
}
+
+ //T->printTree(); cout << endl;
return 0;
}
}
/**************************************************************************************************/
-int ReadNewickTree::readNewickInt(istream& f, int& n, Tree* T) {
+int ReadNewickTree::readNewickInt(istream& f, int& n, Tree* T, CountTable* ct) {
try {
if (m->control_pressed) { return -1; }
int c = readNodeChar(f);
-
+
if(c == '('){
- int lc = readNewickInt(f, n, T);
- if (lc == -1) { return -1; } //reports an error in reading
+
+ //to account for multifurcating trees generated by fasttree, we are forcing them to be bifurcating
+ //read all children
+ vector<int> childrenNodes;
+ while(f.peek() != ')'){
+ int child = readNewickInt(f, n, T, ct);
+ if (child == -1) { return -1; } //reports an error in reading
+ //cout << "child = " << child << endl;
+ childrenNodes.push_back(child);
+
+ //after a child you either have , or ), check for both
+ if(f.peek()==')'){ break; }
+ else if (f.peek()==',') { readSpecialChar(f,',',"comma"); }
+ else {;}
+ }
+ //cout << childrenNodes.size() << endl;
+ if (childrenNodes.size() < 2) { m->mothurOut("Error in tree, please correct."); m->mothurOutEndLine(); return -1; }
+
+ //then force into 2 node structure
+ for (int i = 1; i < childrenNodes.size(); i++) {
+
+ int lc, rc;
+ if (i == 1) { lc = childrenNodes[i-1]; rc = childrenNodes[i]; }
+ else { lc = n-1; rc = childrenNodes[i]; }
+ //cout << i << '\t' << lc << '\t' << rc << endl;
+ T->tree[n].setChildren(lc,rc);
+ T->tree[lc].setParent(n);
+ T->tree[rc].setParent(n);
+
+ //T->printTree(); cout << endl;
+ n++;
+ }
+
+ //to account for extra ++ in looping
+ n--;
- readSpecialChar(f,',',"comma");
-
- int rc = readNewickInt(f, n, T);
- if (rc == -1) { return -1; } //reports an error in reading
if(f.peek()==')'){
readSpecialChar(f,')',"right parenthesis");
//to pass over labels in trees
c=filehandle.get();
- while((c!=',') && (c != -1) && (c!= ':') && (c!=';')){ c=filehandle.get(); }
+ while((c!=',') && (c != -1) && (c!= ':') && (c!=';')&& (c!=')')){ c=filehandle.get(); }
filehandle.putback(c);
-
}
if(f.peek() == ':'){
readSpecialChar(f,':',"colon");
- if(n >= numNodes){ m->mothurOut("Error: Too many nodes in input tree\n"); readOk = -1; return -1; }
+ if(n >= numNodes){ m->mothurOut("Error: Too many nodes in input tree\n"); readOk = -1; return -1; }
T->tree[n].setBranchLength(readBranchLength(f));
}else{
T->tree[n].setBranchLength(0.0);
}
-
- T->tree[n].setChildren(lc,rc);
- T->tree[lc].setParent(n);
- T->tree[rc].setParent(n);
-
+
return n++;
+
}else{
f.putback(c);
string name = "";
name += d;
d=f.get();
}
-
+//cout << name << endl;
int blen = 0;
if(d == ':') { blen = 1; }
f.putback(d);
//set group info
- string group = globaldata->gTreemap->getGroup(name);
-
+ vector<string> group = ct->getGroups(name);
+ //cout << name << endl;
//find index in tree of name
int n1 = T->getIndex(name);
//adds sequence names that are not in group file to the "xxx" group
- if(group == "not found") {
+ if(group.size() == 0) {
m->mothurOut("Name: " + name + " is not in your groupfile, and will be disregarded. \n"); //readOk = -1; return n1;
- globaldata->gTreemap->namesOfSeqs.push_back(name);
- globaldata->gTreemap->treemap[name].groupname = "xxx";
-
- map<string, int>::iterator it;
- it = globaldata->gTreemap->seqsPerGroup.find("xxx");
- if (it == globaldata->gTreemap->seqsPerGroup.end()) { //its a new group
- globaldata->gTreemap->namesOfGroups.push_back("xxx");
- globaldata->gTreemap->seqsPerGroup["xxx"] = 1;
- }else {
- globaldata->gTreemap->seqsPerGroup["xxx"]++;
- }
-
- group = "xxx";
- }
-
- vector<string> tempGroup; tempGroup.push_back(group);
-
- T->tree[n1].setGroup(tempGroup);
+ vector<string> currentGroups = ct->getNamesOfGroups();
+ if (!m->inUsersGroups("xxx", currentGroups)) { ct->addGroup("xxx"); }
+ currentGroups = ct->getNamesOfGroups();
+ vector<int> thisCounts; thisCounts.resize(currentGroups.size(), 0);
+ for (int h = 0; h < currentGroups.size(); h++) {
+ if (currentGroups[h] == "xxx") { thisCounts[h] = 1; break; }
+ }
+ ct->push_back(name, thisCounts);
+
+ group.push_back("xxx");
+ }
+ T->tree[n1].setGroup(group);
T->tree[n1].setChildren(-1,-1);
if(blen == 1){
}
while((c=f.get())!=0 && (c != ':' && c != ',' && c!=')') ) {;}
+
f.putback(c);
return n1;
projects / mothur.git / blobdiff