-// NewickFile.java
-// Tree I/O
-// http://evolution.genetics.washington.edu/phylip/newick_doc.html
-package jalview.io;
-
-import java.io.*;
-import java.util.*;
-import jalview.datamodel.*;
-
-public class NewickFile extends FileParse
-{
- SequenceNode root;
-
- private boolean HasBootstrap = false;
- private boolean HasDistances = false;
- private boolean RootHasDistance = false;
-
- private String ErrorStringrange(String Error, String Er, int r, int p, String s) {
- return ((Error==null) ? "" : Error)
- + Er +
- " at position "+p+" ( "
- + s.substring((p-r)<0 ? 0 : (p-r),
- (p+r)>s.length() ? s.length() : (p+r))
- + " )\n";
- }
-
- // @tree annotations
- // These are set automatically by the reader
- public boolean HasBootstrap() {
- return HasBootstrap;
- }
-
- public boolean HasDistances() {
- return HasDistances;
- }
- public NewickFile(String inStr) throws IOException
- {
- super(inStr, "Paste");
- }
-
- public NewickFile(String inFile, String type)
- throws IOException
- {
-
- super(inFile, type);
- }
- // File IO Flags
- boolean ReplaceUnderscores = false;
-
- public void parse() throws IOException
- {
- String nf;
-
- { // fill nf with complete tree file
- StringBuffer file = new StringBuffer();
- while ( (nf = nextLine()) != null)
- {
- file.append(nf);
- }
- nf = file.toString();
- }
-
- root = new SequenceNode();
- SequenceNode realroot = null;
- SequenceNode c = root;
- int d = -1;
- int cp = 0;
- int flen = nf.length();
- String Error = null;
- String nodename = null;
- float distance=-99999;
- int bootstrap=-99999;
- boolean ascending = false; // flag indicating that we are leaving the current node
- com.stevesoft.pat.Regex majorsyms = new com.stevesoft.pat.Regex("[(\\['),;]");
- while (majorsyms.searchFrom(nf, cp) && Error==null) {
- int fcp = majorsyms.matchedFrom();
- switch (nf.charAt(fcp)) {
-
- case '[': // Comment or structured/extended NH format info
- com.stevesoft.pat.Regex comment = new com.stevesoft.pat.Regex("]");
- if (comment.searchFrom(nf, fcp))
- {
- // Skip the comment field
- cp = 1 + comment.matchedFrom();
- }
- else
- {
- Error = ErrorStringrange(Error, "Unterminated comment", 3, fcp, nf);
- }
- ;
- break;
-
- case '(':
-
- // ascending should not be set
- // New Internal node
- if (ascending)
- {
- Error = ErrorStringrange(Error, "Unexpected '('", 7, fcp, nf);
- continue;
- }
- ;
- d++;
- if (c.right() == null)
- {
- c.setRight(new SequenceNode(null, c, 0, null));
- c = (SequenceNode) c.right();
- }
- else
- {
- if (c.left() != null)
- {
- // Dummy node for polytomy - keeps c.left free for new node
- SequenceNode tmpn = new SequenceNode(null, c, null, 0, 0, true);
- tmpn.SetChildren(c.left(), c.right());
- c.setRight(tmpn);
- }
- c.setLeft(new SequenceNode(null, c, 0, null));
- c = (SequenceNode) c.left();
- }
- if (realroot==null) {
- realroot = c;
- }
- nodename = null;
- distance = -99999;
- bootstrap = -99999;
- cp = fcp + 1;
- break;
-
- // Deal with quoted fields
- case '\'':
- com.stevesoft.pat.Regex qnodename = new com.stevesoft.pat.Regex(
- "([^']|'')+'");
- if (qnodename.searchFrom(nf, fcp))
- {
- int nl = qnodename.stringMatched().length();
- nodename = new String(qnodename.stringMatched().substring(0, nl - 1));
- cp = fcp + nl + 1;
- }
- else
- {
- Error = ErrorStringrange(Error, "Unterminated quotes for nodename",
- 7, fcp, nf);
- }
- break;
-
- case ';':
- if (d != -1)
- {
- Error = ErrorStringrange(Error,
- "Wayward semicolon (depth=" + d + ")", 7,
- fcp, nf);
- }
- // cp advanced at the end of default
-
- default:
-
- // Parse simpler field strings
- String fstring = nf.substring(cp, fcp);
- com.stevesoft.pat.Regex uqnodename = new com.stevesoft.pat.Regex(
- "\\b([^' :;\\](),]+)");
- com.stevesoft.pat.Regex nbootstrap = new com.stevesoft.pat.Regex(
- "\\S+([0-9+]+)\\S*:");
- com.stevesoft.pat.Regex ndist = new com.stevesoft.pat.Regex(
- ":([-0-9.+]+)");
- if (uqnodename.search(fstring)
- && (uqnodename.matchedFrom(1) == 0
- || fstring.charAt(uqnodename.matchedFrom(1) - 1) != ':')) // JBPNote HACK!
-
- {
- if (nodename == null)
- {
- if (ReplaceUnderscores) {
- nodename = uqnodename.stringMatched(1).replace('_', ' ');
- } else {
- nodename = uqnodename.stringMatched(1);
- }
- }
- else
- {
- Error = ErrorStringrange(Error,
- "File has broken algorithm - overwritten nodename",
- 10, fcp,
- nf);
- }
- }
- if (nbootstrap.search(fstring)
- &&
- nbootstrap.matchedFrom(1)
- > uqnodename.matchedFrom(1) + uqnodename.stringMatched().length())
- {
- try
- {
- bootstrap = (new Integer(nbootstrap.stringMatched(1))).intValue();
- HasBootstrap = true;
- }
- catch (Exception e)
- {
- Error = ErrorStringrange(Error, "Can't parse bootstrap value", 4,
- cp + nbootstrap.matchedFrom(), nf);
- }
- }
- boolean nodehasdistance=false;
- if (ndist.search(fstring))
- {
- try
- {
- distance = (new Float(ndist.stringMatched(1))).floatValue();
- HasDistances = true;
- nodehasdistance = true;
- }
- catch (Exception e)
- {
- Error = ErrorStringrange(Error, "Can't parse node distance value",
- 7, cp + ndist.matchedFrom(), nf);
- }
- }
-
- if (ascending)
- {
- // Write node info here
- c.setName(nodename);
- c.dist = (HasDistances) ? distance : 0;
- c.setBootstrap((HasBootstrap) ? bootstrap : 0);
- if (c==realroot) {
- RootHasDistance = nodehasdistance; // JBPNote This is really UGLY!!!
- }
- }
- else
- {
- // Find a place to put the leaf
- SequenceNode newnode = new SequenceNode(null, c,
- (HasDistances) ? distance : 0,
- nodename);
-
- newnode.setBootstrap(HasBootstrap ? bootstrap : 0);
-
- if (c.right() == null)
- {
- c.setRight(newnode);
- }
- else
- {
- if (c.left() == null)
- {
- c.setLeft(newnode);
- }
- else
- {
- // Insert a dummy node for polytomy
- SequenceNode newdummy = new SequenceNode(null, c, null, 0, 0, true);
- newdummy.SetChildren(c.left(), newnode);
- c.setLeft(newdummy);
- }
- }
- }
- if (ascending)
- {
- // move back up the tree from preceding closure
- c = c.AscendTree();
- if (d > -1 && c == null)
- {
- Error = ErrorStringrange(Error,
- "File broke algorithm: Lost place in tree (is there an extra ')' ?)",
- 7, fcp, nf);
- }
- }
-
- if (nf.charAt(fcp) == ')') {
- d--;
- ascending = true;
- } else {
- if (nf.charAt(fcp) == ',') {
- if (ascending) {
- ascending = false;
- } else {
- // Just advance focus, if we need to
- if (c.left()!=null && (!c.left().isLeaf())) {
- c = (SequenceNode) c.left();
- }
- }
- } // else : We do nothing if ';' is encountered.
- }
- // Reset new node properties to obvious fakes
- nodename = null;
- distance = -99999;
- bootstrap = -99999;
-
- cp=fcp+1;
- }
- }
-
- if (Error!=null) {
- throw(new IOException("NewickFile: "+Error+"\n"));
- }
-
- root = (SequenceNode) root.right().detach(); // remove the imaginary root.
- if (!RootHasDistance) {
- root.dist = 0;
- }
- }
-
- public NewickFile(SequenceNode newtree) {
- root = newtree;
- }
-
- public NewickFile(SequenceNode newtree, boolean bootstrap) {
- HasBootstrap = bootstrap;
- root=newtree;
- }
- public NewickFile(SequenceNode newtree, boolean bootstrap, boolean distances) {
- root = newtree;
- HasBootstrap = bootstrap;
- HasDistances = distances;
- }
-
- public NewickFile(SequenceNode newtree, boolean bootstrap, boolean distances, boolean rootdistance) {
- root = newtree;
- HasBootstrap = bootstrap;
- HasDistances = distances;
- RootHasDistance = rootdistance;
- }
-
- public SequenceNode getTree() {
- return root;
- }
-
- public String print() {
- synchronized (this) {
- StringBuffer tf = new StringBuffer();
- print(tf, root);
- return (tf.append(";").toString());
- }
- }
-
- public String print(boolean withbootstraps) {
- synchronized(this) {
- boolean boots = this.HasBootstrap;
- this.HasBootstrap = withbootstraps;
- String rv = print();
- this.HasBootstrap = boots;
- return rv;
- }
- }
-
- public String print(boolean withbootstraps, boolean withdists) {
- synchronized(this) {
- boolean dists = this.HasDistances;
- this.HasDistances = withdists;
- String rv = print(withbootstraps);
- this.HasDistances = dists;
- return rv;
- }
- }
- boolean printRootInfo = false;
-
-
- public String print(boolean withbootstraps, boolean withdists, boolean printRootInfo) {
- synchronized(this) {
- boolean rootinfo = printRootInfo;
- this.printRootInfo = printRootInfo;
- String rv = print(withbootstraps, withdists);
- this.printRootInfo = rootinfo;
- return rv;
- }
- }
- private com.stevesoft.pat.Regex[] NodeSafeName =
- new com.stevesoft.pat.Regex[]
- { new com.stevesoft.pat.Regex().perlCode("m/[\\[,:'()]/"), // test for requiring quotes
- new com.stevesoft.pat.Regex().perlCode("s/'/''/"), // escaping quote characters
- new com.stevesoft.pat.Regex().perlCode("s/\\/w/_/") // unqoted whitespace transformation
- };
- char QuoteChar='\'';
- char getQuoteChar() {
- return QuoteChar;
- }
-
- char setQuoteChar(char c) {
- char old = QuoteChar;
- QuoteChar = c;
- return old;
- }
-
- private String nodeName(String name) {
- if (NodeSafeName[0].search(name)) {
- return QuoteChar + NodeSafeName[1].replaceAll(name) + QuoteChar;
- } else {
- return NodeSafeName[2].replaceAll(name);
- }
- }
- private String printNodeField(SequenceNode c) {
-
- return ( (c.getName() == null) ? ""
- : nodeName(c.getName()))
- + ( (HasBootstrap)
- ? ( (c.getBootstrap() > -1)
- ? " " + c.getBootstrap()
- : "")
- : "")
- + ( (HasDistances)
- ? ":" + c.dist : "");
- }
- private String printRootField(SequenceNode root) {
-
- return (printRootInfo)
- ? (( (root.getName() == null) ? ""
- : nodeName(root.getName()))
- + ( (HasBootstrap)
- ? ( (root.getBootstrap() > -1)
- ? " " + root.getBootstrap()
- : "")
- : "")
- + ( (RootHasDistance)
- ? ":" + root.dist : ""))
- : "";
- }
-
-// Non recursive call deals with root node properties
- public void print(StringBuffer tf, SequenceNode root) {
- if (root!=null) {
- if (root.isLeaf() && printRootInfo) {
- tf.append(printRootField(root));
- } else {
- if (root.isDummy()) {
- _print(tf, (SequenceNode) root.right());
- _print(tf, (SequenceNode) root.left());
- } else {
- tf.append("(");
- _print(tf, (SequenceNode) root.right());
- if (root.left() != null) {
- tf.append(",");
- }
- _print(tf, (SequenceNode) root.left());
- tf.append(")" + printRootField(root));
- }
- }
- }
- }
-
- // Recursive call for non-root nodes
- public void _print(StringBuffer tf, SequenceNode c) {
- if (c!=null) {
- if (c.isLeaf()) {
- tf.append(printNodeField(c));
- } else {
- if (c.isDummy()) {
- _print(tf, (SequenceNode) c.right());
- _print(tf, (SequenceNode) c.left());
- } else {
- tf.append("(");
- _print(tf, (SequenceNode) c.right());
- if (c.left() != null) {
- tf.append(",");
- }
- _print(tf, (SequenceNode) c.left());
- tf.append(")" + printNodeField(c));
- }
- }
- }
- }
-
-// Test
- public static void main(String[] args)
-{
- try
- {
- File fn = new File(args[0]);
-
- StringBuffer newickfile = new StringBuffer();
- BufferedReader treefile =
- new BufferedReader(new FileReader(fn));
- String l;
- while ((l = treefile.readLine())!=null) {
- newickfile.append(l);
- }
- treefile.close();
- System.out.println("Read file :\n");
- NewickFile trf = new NewickFile(args[0], "File");
- trf.parse();
- System.out.println("Original file :\n");
- com.stevesoft.pat.Regex nonl = new com.stevesoft.pat.Regex("\n+", "");
- System.out.println(nonl.replaceAll(newickfile.toString())+"\n");
-
- System.out.println("Parsed file.\n");
- System.out.println("Default output type for original input.\n");
- System.out.println(trf.print());
- System.out.println("Without bootstraps.\n");
- System.out.println(trf.print(false));
- System.out.println("Without distances.\n");
- System.out.println(trf.print(true,false));
- System.out.println("Without bootstraps but with distanecs.\n");
- System.out.println(trf.print(false, true));
- System.out.println("Without bootstraps or distanecs.\n");
- System.out.println(trf.print(false, false));
- System.out.println("With bootstraps and with distances.\n");
- System.out.println(trf.print(true, true));
- }
- catch (java.io.IOException e)
- {
- System.out.println("Exception\n" + e);
- e.printStackTrace();
- }
-}
-
-}
+/*\r
+ * Jalview - A Sequence Alignment Editor and Viewer\r
+ * Copyright (C) 2007 AM Waterhouse, J Procter, G Barton, M Clamp, S Searle\r
+ *\r
+ * This program is free software; you can redistribute it and/or\r
+ * modify it under the terms of the GNU General Public License\r
+ * as published by the Free Software Foundation; either version 2\r
+ * of the License, or (at your option) any later version.\r
+ *\r
+ * This program is distributed in the hope that it will be useful,\r
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of\r
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the\r
+ * GNU General Public License for more details.\r
+ *\r
+ * You should have received a copy of the GNU General Public License\r
+ * along with this program; if not, write to the Free Software\r
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA\r
+ */\r
+\r
+// NewickFile.java\r
+// Tree I/O\r
+// http://evolution.genetics.washington.edu/phylip/newick_doc.html\r
+// TODO: Implement Basic NHX tag parsing and preservation\r
+// TODO: http://evolution.genetics.wustl.edu/eddy/forester/NHX.html\r
+// TODO: Extended SequenceNodeI to hold parsed NHX strings\r
+package jalview.io;\r
+\r
+import java.io.*;\r
+\r
+import jalview.datamodel.*;\r
+\r
+/**\r
+ * DOCUMENT ME!\r
+ *\r
+ * @author $author$\r
+ * @version $Revision$\r
+ */\r
+public class NewickFile\r
+ extends FileParse\r
+{\r
+ SequenceNode root;\r
+ private boolean HasBootstrap = false;\r
+ private boolean HasDistances = false;\r
+ private boolean RootHasDistance = false;\r
+\r
+ // File IO Flags\r
+ boolean ReplaceUnderscores = false;\r
+ boolean printRootInfo = false;\r
+ private com.stevesoft.pat.Regex[] NodeSafeName = new com.stevesoft.pat.Regex[]\r
+ {\r
+ new com.stevesoft.pat.Regex().perlCode("m/[\\[,:'()]/"), // test for requiring quotes\r
+ new com.stevesoft.pat.Regex().perlCode("s/'/''/"), // escaping quote characters\r
+ new com.stevesoft.pat.Regex().perlCode("s/\\/w/_/") // unqoted whitespace transformation\r
+ };\r
+ char QuoteChar = '\'';\r
+\r
+ /**\r
+ * Creates a new NewickFile object.\r
+ *\r
+ * @param inStr DOCUMENT ME!\r
+ *\r
+ * @throws IOException DOCUMENT ME!\r
+ */\r
+ public NewickFile(String inStr)\r
+ throws IOException\r
+ {\r
+ super(inStr, "Paste");\r
+ }\r
+\r
+ /**\r
+ * Creates a new NewickFile object.\r
+ *\r
+ * @param inFile DOCUMENT ME!\r
+ * @param type DOCUMENT ME!\r
+ *\r
+ * @throws IOException DOCUMENT ME!\r
+ */\r
+ public NewickFile(String inFile, String type)\r
+ throws IOException\r
+ {\r
+ super(inFile, type);\r
+ }\r
+\r
+ /**\r
+ * Creates a new NewickFile object.\r
+ *\r
+ * @param newtree DOCUMENT ME!\r
+ */\r
+ public NewickFile(SequenceNode newtree)\r
+ {\r
+ root = newtree;\r
+ }\r
+\r
+ /**\r
+ * Creates a new NewickFile object.\r
+ *\r
+ * @param newtree DOCUMENT ME!\r
+ * @param bootstrap DOCUMENT ME!\r
+ */\r
+ public NewickFile(SequenceNode newtree, boolean bootstrap)\r
+ {\r
+ HasBootstrap = bootstrap;\r
+ root = newtree;\r
+ }\r
+\r
+ /**\r
+ * Creates a new NewickFile object.\r
+ *\r
+ * @param newtree DOCUMENT ME!\r
+ * @param bootstrap DOCUMENT ME!\r
+ * @param distances DOCUMENT ME!\r
+ */\r
+ public NewickFile(SequenceNode newtree, boolean bootstrap, boolean distances)\r
+ {\r
+ root = newtree;\r
+ HasBootstrap = bootstrap;\r
+ HasDistances = distances;\r
+ }\r
+\r
+ /**\r
+ * Creates a new NewickFile object.\r
+ *\r
+ * @param newtree DOCUMENT ME!\r
+ * @param bootstrap DOCUMENT ME!\r
+ * @param distances DOCUMENT ME!\r
+ * @param rootdistance DOCUMENT ME!\r
+ */\r
+ public NewickFile(SequenceNode newtree, boolean bootstrap,\r
+ boolean distances, boolean rootdistance)\r
+ {\r
+ root = newtree;\r
+ HasBootstrap = bootstrap;\r
+ HasDistances = distances;\r
+ RootHasDistance = rootdistance;\r
+ }\r
+\r
+ /**\r
+ * DOCUMENT ME!\r
+ *\r
+ * @param Error DOCUMENT ME!\r
+ * @param Er DOCUMENT ME!\r
+ * @param r DOCUMENT ME!\r
+ * @param p DOCUMENT ME!\r
+ * @param s DOCUMENT ME!\r
+ *\r
+ * @return DOCUMENT ME!\r
+ */\r
+ private String ErrorStringrange(String Error, String Er, int r, int p,\r
+ String s)\r
+ {\r
+ return ( (Error == null) ? "" : Error) + Er + " at position " + p +\r
+ " ( " +\r
+ s.substring( ( (p - r) < 0) ? 0 : (p - r),\r
+ ( (p + r) > s.length()) ? s.length() : (p + r)) + " )\n";\r
+ }\r
+\r
+ // @tree annotations\r
+ // These are set automatically by the reader\r
+ public boolean HasBootstrap()\r
+ {\r
+ return HasBootstrap;\r
+ }\r
+\r
+ /**\r
+ * DOCUMENT ME!\r
+ *\r
+ * @return DOCUMENT ME!\r
+ */\r
+ public boolean HasDistances()\r
+ {\r
+ return HasDistances;\r
+ }\r
+\r
+ public boolean HasRootDistance()\r
+ {\r
+ return RootHasDistance;\r
+ }\r
+\r
+ /**\r
+ * DOCUMENT ME!\r
+ *\r
+ * @throws IOException DOCUMENT ME!\r
+ */\r
+ public void parse()\r
+ throws IOException\r
+ {\r
+ String nf;\r
+\r
+ { // fill nf with complete tree file\r
+\r
+ StringBuffer file = new StringBuffer();\r
+\r
+ while ( (nf = nextLine()) != null)\r
+ {\r
+ file.append(nf);\r
+ }\r
+\r
+ nf = file.toString();\r
+ }\r
+\r
+ root = new SequenceNode();\r
+\r
+ SequenceNode realroot = null;\r
+ SequenceNode c = root;\r
+\r
+ int d = -1;\r
+ int cp = 0;\r
+ //int flen = nf.length();\r
+\r
+ String Error = null;\r
+ String nodename = null;\r
+\r
+ float DefDistance = (float) 0.001; // @param Default distance for a node - very very small\r
+ int DefBootstrap = 0; // @param Default bootstrap for a node\r
+\r
+ float distance = DefDistance;\r
+ int bootstrap = DefBootstrap;\r
+\r
+ boolean ascending = false; // flag indicating that we are leaving the current node\r
+\r
+ com.stevesoft.pat.Regex majorsyms = new com.stevesoft.pat.Regex(\r
+ "[(\\['),;]");\r
+\r
+ while (majorsyms.searchFrom(nf, cp) && (Error == null))\r
+ {\r
+ int fcp = majorsyms.matchedFrom();\r
+\r
+ switch (nf.charAt(fcp))\r
+ {\r
+ case '[': // Comment or structured/extended NH format info\r
+\r
+ com.stevesoft.pat.Regex comment = new com.stevesoft.pat.Regex(\r
+ "]");\r
+\r
+ if (comment.searchFrom(nf, fcp))\r
+ {\r
+ // Skip the comment field\r
+ cp = 1 + comment.matchedFrom();\r
+ }\r
+ else\r
+ {\r
+ Error = ErrorStringrange(Error, "Unterminated comment", 3,\r
+ fcp, nf);\r
+ }\r
+\r
+ ;\r
+\r
+ break;\r
+\r
+ case '(':\r
+\r
+ // ascending should not be set\r
+ // New Internal node\r
+ if (ascending)\r
+ {\r
+ Error = ErrorStringrange(Error, "Unexpected '('", 7, fcp, nf);\r
+\r
+ continue;\r
+ }\r
+\r
+ ;\r
+ d++;\r
+\r
+ if (c.right() == null)\r
+ {\r
+ c.setRight(new SequenceNode(null, c, null, DefDistance,\r
+ DefBootstrap, false));\r
+ c = (SequenceNode) c.right();\r
+ }\r
+ else\r
+ {\r
+ if (c.left() != null)\r
+ {\r
+ // Dummy node for polytomy - keeps c.left free for new node\r
+ SequenceNode tmpn = new SequenceNode(null, c, null, 0,\r
+ 0, true);\r
+ tmpn.SetChildren(c.left(), c.right());\r
+ c.setRight(tmpn);\r
+ }\r
+\r
+ c.setLeft(new SequenceNode(null, c, null, DefDistance,\r
+ DefBootstrap, false));\r
+ c = (SequenceNode) c.left();\r
+ }\r
+\r
+ if (realroot == null)\r
+ {\r
+ realroot = c;\r
+ }\r
+\r
+ nodename = null;\r
+ distance = DefDistance;\r
+ bootstrap = DefBootstrap;\r
+ cp = fcp + 1;\r
+\r
+ break;\r
+\r
+ // Deal with quoted fields\r
+ case '\'':\r
+\r
+ com.stevesoft.pat.Regex qnodename = new com.stevesoft.pat.Regex(\r
+ "([^']|'')+'");\r
+\r
+ if (qnodename.searchFrom(nf, fcp))\r
+ {\r
+ int nl = qnodename.stringMatched().length();\r
+ nodename = new String(qnodename.stringMatched().substring(0,\r
+ nl - 1));\r
+ cp = fcp + nl + 1;\r
+ }\r
+ else\r
+ {\r
+ Error = ErrorStringrange(Error,\r
+ "Unterminated quotes for nodename", 7, fcp,\r
+ nf);\r
+ }\r
+\r
+ break;\r
+\r
+ case ';':\r
+\r
+ if (d != -1)\r
+ {\r
+ Error = ErrorStringrange(Error,\r
+ "Wayward semicolon (depth=" + d + ")", 7,\r
+ fcp, nf);\r
+ }\r
+\r
+ // cp advanced at the end of default\r
+ default:\r
+\r
+ // Parse simpler field strings\r
+ String fstring = nf.substring(cp, fcp);\r
+ com.stevesoft.pat.Regex uqnodename = new com.stevesoft.pat.Regex(\r
+ "\\b([^' :;\\](),]+)");\r
+ com.stevesoft.pat.Regex nbootstrap = new com.stevesoft.pat.Regex(\r
+ "\\S+([0-9+]+)\\S*:");\r
+ com.stevesoft.pat.Regex ndist = new com.stevesoft.pat.Regex(\r
+ ":([-0-9Ee.+]+)");\r
+\r
+ if (uqnodename.search(fstring) &&\r
+ ( (uqnodename.matchedFrom(1) == 0) ||\r
+ (fstring.charAt(uqnodename.matchedFrom(1) - 1) != ':'))) // JBPNote HACK!\r
+ {\r
+ if (nodename == null)\r
+ {\r
+ if (ReplaceUnderscores)\r
+ {\r
+ nodename = uqnodename.stringMatched(1).replace('_',\r
+ ' ');\r
+ }\r
+ else\r
+ {\r
+ nodename = uqnodename.stringMatched(1);\r
+ }\r
+ }\r
+ else\r
+ {\r
+ Error = ErrorStringrange(Error,\r
+ "File has broken algorithm - overwritten nodename",\r
+ 10, fcp, nf);\r
+ }\r
+ }\r
+\r
+ if (nbootstrap.search(fstring) &&\r
+ (nbootstrap.matchedFrom(1) > (uqnodename.matchedFrom(1) +\r
+ uqnodename.stringMatched().length())))\r
+ {\r
+ try\r
+ {\r
+ bootstrap = (new Integer(nbootstrap.stringMatched(1))).intValue();\r
+ HasBootstrap = true;\r
+ }\r
+ catch (Exception e)\r
+ {\r
+ Error = ErrorStringrange(Error,\r
+ "Can't parse bootstrap value", 4,\r
+ cp + nbootstrap.matchedFrom(), nf);\r
+ }\r
+ }\r
+\r
+ boolean nodehasdistance = false;\r
+\r
+ if (ndist.search(fstring))\r
+ {\r
+ try\r
+ {\r
+ distance = (new Float(ndist.stringMatched(1))).floatValue();\r
+ HasDistances = true;\r
+ nodehasdistance = true;\r
+ }\r
+ catch (Exception e)\r
+ {\r
+ Error = ErrorStringrange(Error,\r
+ "Can't parse node distance value", 7,\r
+ cp + ndist.matchedFrom(), nf);\r
+ }\r
+ }\r
+\r
+ if (ascending)\r
+ {\r
+ // Write node info here\r
+ c.setName(nodename);\r
+ // Trees without distances still need a render distance\r
+ c.dist = (HasDistances) ? distance : DefDistance;\r
+ // be consistent for internal bootstrap defaults too\r
+ c.setBootstrap( (HasBootstrap) ? bootstrap : DefBootstrap);\r
+ if (c == realroot)\r
+ {\r
+ RootHasDistance = nodehasdistance; // JBPNote This is really UGLY!!! Ensure root node gets its given distance\r
+ }\r
+ }\r
+ else\r
+ {\r
+ // Find a place to put the leaf\r
+ SequenceNode newnode = new SequenceNode(null, c, nodename,\r
+ (HasDistances) ? distance : DefDistance,\r
+ (HasBootstrap) ? bootstrap : DefBootstrap, false);\r
+\r
+ if (c.right() == null)\r
+ {\r
+ c.setRight(newnode);\r
+ }\r
+ else\r
+ {\r
+ if (c.left() == null)\r
+ {\r
+ c.setLeft(newnode);\r
+ }\r
+ else\r
+ {\r
+ // Insert a dummy node for polytomy\r
+ // dummy nodes have distances\r
+ SequenceNode newdummy = new SequenceNode(null, c,\r
+ null, (HasDistances ? 0 : DefDistance), 0, true);\r
+ newdummy.SetChildren(c.left(), newnode);\r
+ c.setLeft(newdummy);\r
+ }\r
+ }\r
+ }\r
+\r
+ if (ascending)\r
+ {\r
+ // move back up the tree from preceding closure\r
+ c = c.AscendTree();\r
+\r
+ if ( (d > -1) && (c == null))\r
+ {\r
+ Error = ErrorStringrange(Error,\r
+ "File broke algorithm: Lost place in tree (is there an extra ')' ?)",\r
+ 7, fcp, nf);\r
+ }\r
+ }\r
+\r
+ if (nf.charAt(fcp) == ')')\r
+ {\r
+ d--;\r
+ ascending = true;\r
+ }\r
+ else\r
+ {\r
+ if (nf.charAt(fcp) == ',')\r
+ {\r
+ if (ascending)\r
+ {\r
+ ascending = false;\r
+ }\r
+ else\r
+ {\r
+ // Just advance focus, if we need to\r
+ if ( (c.left() != null) && (!c.left().isLeaf()))\r
+ {\r
+ c = (SequenceNode) c.left();\r
+ }\r
+ }\r
+ }\r
+\r
+ // else : We do nothing if ';' is encountered.\r
+ }\r
+\r
+ // Reset new node properties to obvious fakes\r
+ nodename = null;\r
+ distance = DefDistance;\r
+ bootstrap = DefBootstrap;\r
+\r
+ cp = fcp + 1;\r
+ }\r
+ }\r
+\r
+ if (Error != null)\r
+ {\r
+ throw (new IOException("NewickFile: " + Error + "\n"));\r
+ }\r
+\r
+ root = (SequenceNode) root.right().detach(); // remove the imaginary root.\r
+\r
+ if (!RootHasDistance)\r
+ {\r
+ root.dist = (HasDistances) ? 0 : DefDistance;\r
+ }\r
+ }\r
+\r
+ /**\r
+ * DOCUMENT ME!\r
+ *\r
+ * @return DOCUMENT ME!\r
+ */\r
+ public SequenceNode getTree()\r
+ {\r
+ return root;\r
+ }\r
+\r
+ /**\r
+ * Generate a newick format tree according to internal flags\r
+ * for bootstraps, distances and root distances.\r
+ *\r
+ * @return new hampshire tree in a single line\r
+ */\r
+ public String print()\r
+ {\r
+ synchronized (this)\r
+ {\r
+ StringBuffer tf = new StringBuffer();\r
+ print(tf, root);\r
+\r
+ return (tf.append(";").toString());\r
+ }\r
+ }\r
+\r
+ /**\r
+ *\r
+ *\r
+ * Generate a newick format tree according to internal flags\r
+ * for distances and root distances and user specificied writing of\r
+ * bootstraps.\r
+ * @param withbootstraps controls if bootstrap values are explicitly written.\r
+ *\r
+ * @return new hampshire tree in a single line\r
+ */\r
+ public String print(boolean withbootstraps)\r
+ {\r
+ synchronized (this)\r
+ {\r
+ boolean boots = this.HasBootstrap;\r
+ this.HasBootstrap = withbootstraps;\r
+\r
+ String rv = print();\r
+ this.HasBootstrap = boots;\r
+\r
+ return rv;\r
+ }\r
+ }\r
+\r
+ /**\r
+ *\r
+ * Generate newick format tree according to internal flags\r
+ * for writing root node distances.\r
+ *\r
+ * @param withbootstraps explicitly write bootstrap values\r
+ * @param withdists explicitly write distances\r
+ *\r
+ * @return new hampshire tree in a single line\r
+ */\r
+ public String print(boolean withbootstraps, boolean withdists)\r
+ {\r
+ synchronized (this)\r
+ {\r
+ boolean dists = this.HasDistances;\r
+ this.HasDistances = withdists;\r
+\r
+ String rv = print(withbootstraps);\r
+ this.HasDistances = dists;\r
+\r
+ return rv;\r
+ }\r
+ }\r
+\r
+ /**\r
+ * Generate newick format tree according to user specified flags\r
+ *\r
+ * @param withbootstraps explicitly write bootstrap values\r
+ * @param withdists explicitly write distances\r
+ * @param printRootInfo explicitly write root distance\r
+ *\r
+ * @return new hampshire tree in a single line\r
+ */\r
+ public String print(boolean withbootstraps, boolean withdists,\r
+ boolean printRootInfo)\r
+ {\r
+ synchronized (this)\r
+ {\r
+ boolean rootinfo = printRootInfo;\r
+ this.printRootInfo = printRootInfo;\r
+\r
+ String rv = print(withbootstraps, withdists);\r
+ this.printRootInfo = rootinfo;\r
+\r
+ return rv;\r
+ }\r
+ }\r
+\r
+ /**\r
+ * DOCUMENT ME!\r
+ *\r
+ * @return DOCUMENT ME!\r
+ */\r
+ char getQuoteChar()\r
+ {\r
+ return QuoteChar;\r
+ }\r
+\r
+ /**\r
+ * DOCUMENT ME!\r
+ *\r
+ * @param c DOCUMENT ME!\r
+ *\r
+ * @return DOCUMENT ME!\r
+ */\r
+ char setQuoteChar(char c)\r
+ {\r
+ char old = QuoteChar;\r
+ QuoteChar = c;\r
+\r
+ return old;\r
+ }\r
+\r
+ /**\r
+ * DOCUMENT ME!\r
+ *\r
+ * @param name DOCUMENT ME!\r
+ *\r
+ * @return DOCUMENT ME!\r
+ */\r
+ private String nodeName(String name)\r
+ {\r
+ if (NodeSafeName[0].search(name))\r
+ {\r
+ return QuoteChar + NodeSafeName[1].replaceAll(name) + QuoteChar;\r
+ }\r
+ else\r
+ {\r
+ return NodeSafeName[2].replaceAll(name);\r
+ }\r
+ }\r
+\r
+ /**\r
+ * DOCUMENT ME!\r
+ *\r
+ * @param c DOCUMENT ME!\r
+ *\r
+ * @return DOCUMENT ME!\r
+ */\r
+ private String printNodeField(SequenceNode c)\r
+ {\r
+ return ( (c.getName() == null) ? "" : nodeName(c.getName())) +\r
+ ( (HasBootstrap)\r
+ ? ( (c.getBootstrap() > -1) ? (" " + c.getBootstrap()) : "") : "") +\r
+ ( (HasDistances) ? (":" + c.dist) : "");\r
+ }\r
+\r
+ /**\r
+ * DOCUMENT ME!\r
+ *\r
+ * @param root DOCUMENT ME!\r
+ *\r
+ * @return DOCUMENT ME!\r
+ */\r
+ private String printRootField(SequenceNode root)\r
+ {\r
+ return (printRootInfo)\r
+ ? ( ( (root.getName() == null) ? "" : nodeName(root.getName())) +\r
+ ( (HasBootstrap)\r
+ ? ( (root.getBootstrap() > -1) ? (" " + root.getBootstrap()) : "") :\r
+ "") +\r
+ ( (RootHasDistance) ? (":" + root.dist) : "")) : "";\r
+ }\r
+\r
+ // Non recursive call deals with root node properties\r
+ public void print(StringBuffer tf, SequenceNode root)\r
+ {\r
+ if (root != null)\r
+ {\r
+ if (root.isLeaf() && printRootInfo)\r
+ {\r
+ tf.append(printRootField(root));\r
+ }\r
+ else\r
+ {\r
+ if (root.isDummy())\r
+ {\r
+ _print(tf, (SequenceNode) root.right());\r
+ _print(tf, (SequenceNode) root.left());\r
+ }\r
+ else\r
+ {\r
+ tf.append("(");\r
+ _print(tf, (SequenceNode) root.right());\r
+\r
+ if (root.left() != null)\r
+ {\r
+ tf.append(",");\r
+ }\r
+\r
+ _print(tf, (SequenceNode) root.left());\r
+ tf.append(")" + printRootField(root));\r
+ }\r
+ }\r
+ }\r
+ }\r
+\r
+ // Recursive call for non-root nodes\r
+ public void _print(StringBuffer tf, SequenceNode c)\r
+ {\r
+ if (c != null)\r
+ {\r
+ if (c.isLeaf())\r
+ {\r
+ tf.append(printNodeField(c));\r
+ }\r
+ else\r
+ {\r
+ if (c.isDummy())\r
+ {\r
+ _print(tf, (SequenceNode) c.left());\r
+ if (c.left() != null)\r
+ {\r
+ tf.append(",");\r
+ }\r
+ _print(tf, (SequenceNode) c.right());\r
+ }\r
+ else\r
+ {\r
+ tf.append("(");\r
+ _print(tf, (SequenceNode) c.right());\r
+\r
+ if (c.left() != null)\r
+ {\r
+ tf.append(",");\r
+ }\r
+\r
+ _print(tf, (SequenceNode) c.left());\r
+ tf.append(")" + printNodeField(c));\r
+ }\r
+ }\r
+ }\r
+ }\r
+\r
+ // Test\r
+ public static void main(String[] args)\r
+ {\r
+ try\r
+ {\r
+ if (args == null || args.length != 1)\r
+ {\r
+ System.err.println(\r
+ "Takes one argument - file name of a newick tree file.");\r
+ System.exit(0);\r
+ }\r
+\r
+ File fn = new File(args[0]);\r
+\r
+ StringBuffer newickfile = new StringBuffer();\r
+ BufferedReader treefile = new BufferedReader(new FileReader(fn));\r
+ String l;\r
+\r
+ while ( (l = treefile.readLine()) != null)\r
+ {\r
+ newickfile.append(l);\r
+ }\r
+\r
+ treefile.close();\r
+ System.out.println("Read file :\n");\r
+\r
+ NewickFile trf = new NewickFile(args[0], "File");\r
+ trf.parse();\r
+ System.out.println("Original file :\n");\r
+\r
+ com.stevesoft.pat.Regex nonl = new com.stevesoft.pat.Regex("\n+", "");\r
+ System.out.println(nonl.replaceAll(newickfile.toString()) + "\n");\r
+\r
+ System.out.println("Parsed file.\n");\r
+ System.out.println("Default output type for original input.\n");\r
+ System.out.println(trf.print());\r
+ System.out.println("Without bootstraps.\n");\r
+ System.out.println(trf.print(false));\r
+ System.out.println("Without distances.\n");\r
+ System.out.println(trf.print(true, false));\r
+ System.out.println("Without bootstraps but with distanecs.\n");\r
+ System.out.println(trf.print(false, true));\r
+ System.out.println("Without bootstraps or distanecs.\n");\r
+ System.out.println(trf.print(false, false));\r
+ System.out.println("With bootstraps and with distances.\n");\r
+ System.out.println(trf.print(true, true));\r
+ }\r
+ catch (java.io.IOException e)\r
+ {\r
+ System.err.println("Exception\n" + e);\r
+ e.printStackTrace();\r
+ }\r
+ }\r
+}\r
git://source.jalview.org / jalview.git / blobdiff