[physfs] PhysicsFS 3.0.1 released!

Ryan C. Gordon icculus at icculus.org
Thu Oct 26 16:48:16 EDT 2017


( This work was funded by my Patreon! If you like me working on things 
like PhysicsFS and other useful pieces of code, consider throwing some 
money in the pot: https://patreon.com/icculus )


PhysicsFS 3.0.1 is now available.

This is a collection of emergency fixes to 3.0.0. All 3.0 users should 
definitely upgrade right away.


You can grab it here...

       https://icculus.org/physfs/downloads/physfs-3.0.1.tar.bz2

...or from Mercurial with this command...

       hg clone -u release-3.0.1 https://hg.icculus.org/icculus/physfs

...or just stare at the webview here...

       https://hg.icculus.org/icculus/physfs/file/release-3.0.1


No changes to the 1.0 or 2.0 branches are planned, even if they might 
apply. Upgrade to PhysicsFS 3.


Notable changes in this release:

- Trying to use the new mount functions (PHYSFS_mountIo, 
PHYSFS_mountMemory, PHYSFS_mountHandle) with a NULL filename would cause 
several problems. This now reports an error without mounting. Docs have 
been corrected to reflect this reality.
- Trying to mounting an archive inside a .zip file with 
PHYSFS_mountHandle() would crash with a NULL pointer dereference. Fixed.
- Trying to access a file in the search path named "." or ".." no longer 
succeeds. Previously, this bug meant you could enumerate (but not 
access) files in the parent of a mounted native directory. Trying to 
open "." or ".." might have crashed or hung PhysicsFS. This only applies 
to the literal string "." or ".."; if there was a '/' char in the path, 
PhysicsFS would correctly reject a path with an element named as such.
- Trying to mount a symlink to a directory would (incorrectly) fail on 
Unix and Apple platforms instead of mounting the directory the symlink 
points to. Fixed.
- Fixed several crashes/hangs that a maliciously-crafted .iso file could 
trigger.

Thanks to those that reported these issues in 3.0.0, and those that 
funded this work on Patreon! Enjoy!

--ryan.



More information about the physfs mailing list