aboutsummaryrefslogtreecommitdiff
path: root/lib
AgeCommit message (Collapse)Author
2021-03-06Cleanup: replace ad-hoc dynamic array in sqfs_frag_table_tDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-03-06Add a generic implementation of a dynamic array to libutilDavid Oberhollenzer
The intention is to get rid of all the ad-hoc array implementations in the other components and cut down code size. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-03-06Store xattr writer block description in a red-black treeDavid Oberhollenzer
By storing the blocks in a tree, the de-duplication can lookup existing blocks in logartihmic instead of linear time. The linked list is still maintained, because we need to iterate over the blocks in creation order during serialization. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-03-06Add a context pointer to the rbtree key comparisonDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-03-06Add a copy function to the rb-tree implementationDavid Oberhollenzer
If we use the rb-tree in libsquashfs objects, we need to be able top copy an entire tree as part of the object. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-03-06Cleanup: replace the void-ptr with an inode-ptr in the file tree nodeDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-02-28added shared read access when opening sqfs image with read-only flags (win32)Thomas Lang
2021-02-19Fix: libfstree: add an assert the canonicalize_name return valueDavid Oberhollenzer
Since the canonicalize_name function only fails if the path contains ".." and the one we are constructing from the scanned fstree (built using canonicalized names), it should NEVER fail. However, coverity does get concerned, because we are checking the return value elesewhere. So do what we do at other, similar locations and add an assert(). Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-02-19Fix: canonicalize path names in glob pattern matchingDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-02-19libfstree: reject unknown glob options to allow future expansionsDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-02-19fstree_from_file: Add fnmatch() pattern matching to file globbingDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-02-19fstree_from_file: Implement basic file globbingDavid Oberhollenzer
The base path is passed to the fstree_from_file function and in turn to the individual callbacks. The line parsing function is modified to allow '*' as mode, uid and gid for specifically marked callbacks. A glob callback is added that internally uses the fstree_from_dir scanning functions in combination with a filter callback. Directory scanning flags are parsed from the extra arguments before interpreting it as a path fragment. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-02-19libfstree: Add a filter callback to the directory scanning functionDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-02-18libfstree: add a subdirectory scanning functionDavid Oberhollenzer
So we can scan a sub-directory within a the base directory without having to do string operations first. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-02-18fstree_from_dir: add filtering flags to skip certain inode typesDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-02-10cleanup: fstree_from_file: split & simplify line parsing functionDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-02-10Always use the correct data type for realloc return valueDavid Oberhollenzer
This commit mainly serves the static analysis tooling. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-01-19libsqfs: Implement exact matching of fragmentsDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-01-19Add a user pointer to the hash table implementationDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-01-19libcommon: Use the newer API for the block processorDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-01-19libsqfs: Add a sqfs_block_processor_create_ex functionDavid Oberhollenzer
This function creates a block processor from a structure describing it. A stub implementation for the old sqfs_block_processor_create is added that simply sets up such a struct and forwards the call. The current version of the description struct only contains the exact same parameters and a size field at the beginning. This approach is supposed to make extending the range of parameters easier without breaking ABI compatibillity. Currently already planned are: - Adding a sqfs_file_t pointer to double-check when deduplicating fragments. - When the scanning code reaches a usable state, add the abillity to pass scanned fragment data, so the block processor can be used for appending to an existing image. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-01-19libsqfs: block processor: removed unused chunk next pointerDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-01-19Fix: Move fragment consolidation back to block processor serial partDavid Oberhollenzer
Keeping a list of fragments stored away in the current fragment block and consolidating them in the thread pool takes them out of circulation. If we have a lot of tiny fragments, this can lead to a situation where all the limit is reached, but we cannot do anything, because we are waiting for a block to complete, but they are all attached to the current fragment block and the queue is empty. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2021-01-15Fix more normalization of slashes in filenames.Scott Moser
It looks like the last commit missed a couple more occurences where '\' was treated incorrectly. Fixes were still needed in sqfs_dir_reader_find_by_path and sqfs_dir_reader_get_full_hierarchy. This path is used in extras/browse.c.
2020-12-29Fix normalization of slashes in filenamesDavid Oberhollenzer
All paths were canonicalized internally, which includes filtering sequences of slashes and converting backslashes to slashes. Furthermore, when unpacking files, filenames are sanity checked and rejected if they contain forward OR backward slashes. This is a problem on Unix-like systems, where files containing backslashes are a legitimate use case (*cough* SystemD *cough*). This patch removes the backslash conversion from the canonicalization and modifies the sanity check to reject backslashes only on Windows. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-12-15libfstree: make the directory scanning code a little more genericDavid Oberhollenzer
- Instead of using the fstree root, let the caller specify it. - Add a flag to prevent recursion into sub directories. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-12-10Move fstree dirscan code back to libfstreeDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-12-06libsqfs: implement exact matching in the default block writer.David Oberhollenzer
Instead of comparing (compresed, disk-size, checksum) tuples to find block matches, do an exact, byte-for-byte comparison of the data stored on disk to avoid the possibility of a spurious colision. Since this is the desired behaviour, make it the default, optionally overrideable through a flag. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-11-13Fix: include MacOS compat byteswab definitions in LZO compressorDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-11-07Fix: libsquashfs: xattr_writer: return NULL if calloc failsDavid Oberhollenzer
Instead of dereferencing the NULL pointer and crashing. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-11-07Cleanup: libcommon: try to split up some of the larger filesDavid Oberhollenzer
The sqsf_writer is moved to a separate header, as well as the compressor related stuff. The statistics function is moved into the writer code, as this is the only place that actually uses it. The writer code itself is split up into a hand full of file in their own subdirectory. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-09-29Update tar2sqfs/sqfs2tar documentationDavid Oberhollenzer
Modify the man pages and help texts to mention compression support, make the help option display a list of supported compressors. Also clarify the squashfs compressor list in the help text to state that the list shows block compressors supported for SquashFS compression. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-09-29Replace file/getline usage with istreamDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-09-29Add a more usefull getline-like function to libfstreamDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-09-29libfstream: Rework unix sparse file writingDavid Oberhollenzer
Instead of immediately seeking forward, record the sparse bytes "written" and then do the seek the next time we write data, or do an ftruncate if we do a sync. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-09-17Add a check whether libzstd supports stream compressionDavid Oberhollenzer
Streaming compression was added fairly recently to zstd (and then the API was changed a few times). Rather than rely on libzstd versioning macros, this commit adds an m4 script to test at configure time if the enums/functions we need are available by trying to compile a small sample. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-09-17Fix: add missing compiler/linker flagsDavid Oberhollenzer
If tar2sqfs and sqfs2tar want to use comperssion via libfstream, they need to link against the compressor libraries. Also, libfstream may need the apropriate cflags to access the headers on some platforms. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-09-16Add bzip2 stream compression supportDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-09-16Add zstd stream compressor implementation to libfstreamDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-09-16Re-import crc32 code to builtin zlib copyDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-09-16Implement istream decompression supportDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-09-16Remodel libtar/tar2sqfs to read data from an istream_tDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-09-16Implement ostream compressor wrapperDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-09-16Remodel file extraction tools to use libfstreamDavid Oberhollenzer
This commit rewrites the libtar write paths to use libfstream insead of a FILE pointer. Also, the libcommon file extraction function is remodeled to use libfstream. In accordance, rdsquashfs, sqfs2tar and sqfsdiff have some minor adjustments made to work with the ported libtar and libcommon. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-09-16Implement a data writer based output streamDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-09-16Add stream I/O abstraction libraryDavid Oberhollenzer
Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-09-03Fix integer bounds checking in GNU tar sparse format 1.0 parserDavid Oberhollenzer
- Make sure the file actually has that many records before trying to read one and fail if not. - Use the helper macros for size_t overflow checking instead of assuming size_t == uint64_t. - Impose a "reasonable" upper bound on the number of data segments and insist that there is at least one entry. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-09-02Fix nonexistant gnu tar sparse format 1.0 supportDavid Oberhollenzer
Contrary to previous claims, support for the GNU tar sparse format 1.0 was missing entirely (the newest of their 3 different sparse mapping formats). This oversight wasn't caught, because the unit test was compiling the wrong source file and tar2sqfs had no problem processing the test file because it is still a valid POSIX-ish tar archive (but the sparse part was missing and the mapping embedded in the file). Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-08-26Fix tree node path generation for detached sub treesDavid Oberhollenzer
The function sqfs_tree_node_get_path is used in several places within rdsquashfs to produce a path for a tree node, either when describing the file system, or when unpacking it. Unpacking can be done on sub-trees as well as the entire tree, in which case the root of the sub-tree has its parent pointer removed, so the full path terminates at the new root. This works with directories, since they receive special case handling anyway, but fails if the sub-tree to unpack is only a single file because the sqfs_tree_node_get_path function assumes that we are at the tree root and returns "/" as a path, which gets normalized to "". This commit adds a workaround to the function to simply use the nodes name (if available) in that case instead. The describe case in rdsquashfs is unaffacted, since it always starts at the root. Likewise, the sqfs2tar case should also be unaffacted, since it already employs special case handling for the [sub] tree root node. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>
2020-08-16Fix libtar treatment of link targets that fill the header fieldDavid Oberhollenzer
The tar header has a 100 byte field for symlink and hard link targets. If the target is longer than 100 bytes, an extension header has to be used. However, it is perfectly valid to fill all 100 bytes to the brim without adding a null terminator. In case of a symlink, this can result in garbage link targets, while for hard links it results in an immediate error since the target cannot be resolved later on. This commit attempts to fix the problem by replacing the strdup of the link target with an strndup that copies at most the size of the target header field. Signed-off-by: David Oberhollenzer <david.oberhollenzer@sigma-star.at>