badly formed tree input causes out of memory crash

[chinchliff]

If the last tree in the input treefile is not followed by a semicolon, the program will apparently keep reading until it completely fills the memory of the computer. I have crashed my 48GB RAM desktop several times doing this.

[josephwb]

Seems to occur in file tree_reader.cpp, line 65:

else if (nextChar == ';') {
keepGoing = false;
}

This is the only exit I can see. Should have a condition:

if (nextChar == null) {
Yell at user.
}

Better yet, as we discussed, perhaps deal with the length of the tree string.

int counter = 0;
int treeLength = tree.size();
bool semicolonEncountered = false;

while (counter < (treeLength - 1)) { // iterate through the characters in the tree string
Do stuff...
counter ++;
}
if (!semicolonEncountered) {
cout << "You messed up, yo." << endl;
exit(0);
}

[josephwb]

Noticed the exact same code is in treemachine, despite being in java. Java doesn't do the infinite loop thing when a semicolon is missing, but crashes with informative error.

[josephwb]

For what it is worth, I patched this in treemachine:

// pb is the newick tree string
if (pb.charAt(pb.length() - 1) != ';') {
    System.out.println("Tree is invalid: missing concluding semicolon. Exiting.");
    System.exit(1);
}

[blackrim]

look good. I will get this in treepl this weekend

On Wed, Mar 13, 2013 at 4:11 PM, Joseph W. Brown
notifications@github.comwrote:

I patched this in treemachinehttps://github.com/OpenTreeOfLife/treemachine/blob/master/src/main/java/jade/tree/TreeReader.java:

// pb is the newick tree string
if (pb.charAt(pb.length() - 1) != ';') {
System.out.println("Tree is invalid: missing concluding semicolon.
Exiting.");
System.exit(1);
}

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/3#issuecomment-14865561
.

[josephwb]

I've fixed this in a pull request I just sent.

[josephwb]

Okay, in pull request #9

[blackrim]

thanks

On Tue, Apr 15, 2014 at 8:07 AM, Joseph W. Brown
notifications@github.comwrote:

Okay, in pull request #9 #9

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/3#issuecomment-40473851
.