PyTables User's Guide: Hierarchical datasets in Python - Release 1.3.2 | ||
---|---|---|
Prev | Chapter 6. FileNode - simulating a filesystem with PyTables | Next |
FileNode is still a young piece of software, so it lacks some functionality. This is a list of known current limitations:
Node files can only be opened for read-only or read and append mode. This will be enhanced in the future.
There is no universal newline support yet. This is likely to be implemented in a near future.
Sparse files (files with lots of zeros) are not treated specially; if you want them to take less space, you should be better off using compression.
Node file and attribute names are constrained to valid PyTables node names, which happen to be more or less those accepted by Unix filesystems (see 4.4).
These limitations still make FileNode entirely adequate to work with most binary and text files. Of course, suggestions and patches are welcome.