Assignment 2: a file synchroniser

version: 1.0.0 last updated: 2024-10-21 12:00:00

Contents

• Aims
• The Task
  • Getting Started
  • Subset 1
  • Subset 2
  • Subset 3
  • Subset 4
  • Subset 5
  • Handling Errors
  • Reference implementation
  • Helper utilities
• Formats of rbuoy indices
  • Type A Buoy Index format
  • Type B Buoy Index format
  • Type C Buoy Index format
• rbuoy vs rsync
• Assumptions and Clarifications
• Subset weighting
• Change Log
• Assessment
  • Testing
  • Submission
  • Due Date
  • Assessment Scheme
  • Intermediate Versions of Work
  • Assignment Conditions

Aims

• building a concrete understanding of file system objects;
• practising C, including bitwise operations and robust error handling;
• understanding file operations, including input-output operations on binary data

The Task

The rsync utility is a useful and popular tool which efficiently transfers files between computers. In this assignment you will be implementing rbuoy, which is a simplified version of rsync.

To copy a file from a sending computer to a receiver, it would theoretically be sufficient to just naïvely send over the entire contents (and possibly metadata) of the file.

However, if the receiver already has an older version of the file which is very close to the sender's version (or even an identical copy!), then a large amount of redundant data is being transmitted. With slow networks or large file sizes this can translate to a unnecessary waiting and cost.

Both the real rsync utility and the rbuoy utility that you'll be implementing in this assignment avoid unnecessary data transfer by only sending the chunks of a file which differ between sender and receiver. The rbuoy algorithm takes place over four stages:

1. Stage 1: the sender constructs a Type A Buoy Index (TABI) file containing a record for each file the sender wants to send. Each record contains metadata about the file, as well as a hash for each block in the file (see the subset 1 description for more information).
2. Stage 2: the receiver uses the TABI file to construct a Type B Buoy Index (TBBI) file containing a record for each TABI record. The TBBI file contains information about which blocks the receiver already has an up-to-date copy of (see the subset 2 description for more information).
3. Stage 3: the sender uses the Type B Buoy Index file to construct a Type C Buoy Index (TCBI) file containing a record for each TBBI record. The TCBI file contains the contents of the blocks which the receiver did not have an up-to-date copy of (see the subset 3 description for more information).
4. Stage 4: the receiver uses the TCBI file to reconstruct an up-to-date copy of the files it is receiving. (see the subset 4 description for more information).

The first four subsets of this assignment correspond to implementing each of these stages for a given list of files. The fifth subset involves adding support for directories.

The real rsync utility is able to transfer files over a network to a remote computer; where the sender would be one computer and the receiver would be a different computer. It can also transfer files locally, where the 'sender' and 'receiver' are two different directories on the same computer. In this assignment, you will only be implementing the local version of rbuoy, where the sender and receiver are two different directories on the same computer.

Getting Started

Create a new directory for this assignment, change to this directory, and fetch the provided code by running

mkdir -m 700 rbuoy
cd rbuoy
1092 fetch rbuoy

If you're not working at CSE, you can download the provided files as a zip file or a tar file.

This will give you the following files:

rbuoy.c

is the only file you need to change: it contains partial definitions of four functions, stage_1, stage_2, stage_3, and stage_4, to which you need to add code to complete the assignment. You can also add your own functions to this file.

rbuoy_main.c

contains a main, which has code to parse the command line arguments, and which then calls one of stage_1, stage_2, stage_3, and stage_4, depending on the command line arguments given to rbuoy. Do not change this file.

rbuoy.h

contains shared function declarations and some useful constant definitions. Do not change this file.

rbuoy_provided.c

contains the hash_block function; you should call this function to calculate hashes for subset 1. Do not change this file.

rbuoy.mk

contains a Makefile fragment for rbuoy.

rbuoy_hash_block.c

contains the source code for the 1092 rbuoy-hash-block helper utility which we have provided you. You may find it useful to look at this code to better understand how the hash_block function can be used. Do not change, attempt to compile with, or submit this file.

You can run make(1) to compile the provided code; and you should be able to run the result.

make
dcc   rbuoy.c rbuoy_main.c rbuoy_provided.c -o rbuoy
./rbuoy
Usage: ./rbuoy [--stage-1|--stage-2|--stage-3|--stage-4]

You may optionally create extra .c or .h files. You can modify the provided Makefile fragment if you choose to do so.

You should run 1092 rbuoy-examples to get a directory called examples/ full of test files and example Buoy Index files to test your program against.

1092 rbuoy-examples
ls examples
aaa  bbb  ccc  tabi  tbbi  tcbi

Subset 1

To complete subset 1, you need to complete the provided stage_1 function.

The stage_1 function should create a TABI file at the specified output path, based on a given array of filenames.

The TABI file should contain the appropriate header, as outlined in the format of the TABI file section below.

It should then produce a TABI record for each file in the given array of in_filenames.

1092 rbuoy-examples
cd examples/aaa
ls
emojis.txt  empty  fizz  fractal_bin  little_endian_shorts  long_path  lyrics.txt  short.txt
../../rbuoy --stage-1 ../out.tabi emojis.txt empty
1092 rbuoy-show ../out.tabi
Field name         Offset        Bytes                    ASCII/Numeric
-----------------------------------------------------------------------
magic              0x00000000    54 41 42 49              chr TABI
num records        0x00000004    02                       dec 2
============================= Record   0 ==============================
pathname len       0x00000005    0a 00                    dec 10
pathname           0x00000007    65 6d 6f 6a 69 73 2e 74  chr emojis.t
                   0x0000000f    78 74                    chr xt
num blocks         0x00000011    03 00 00                 dec 3
hashes[0]          0x00000014    90 30 e3 14 6e e7 0a 90  chr .0..n...
hashes[1]          0x0000001c    91 90 5c 46 fc 07 b3 93  chr ..\F....
hashes[2]          0x00000024    8c ec 01 86 4c dc 63 af  chr ....L.c.
============================= Record   1 ==============================
pathname len       0x0000002c    05 00                    dec 5
pathname           0x0000002e    65 6d 70 74 79           chr empty
num blocks         0x00000033    00 00 00                 dec 0

Subset 2

To complete subset 2, you need to complete the provided stage_2 function.

The stage 2 function receives a path to an input TABI file and a path to an output TBBI file.

The TBBI file should contain the appropriate header, as outlined in the format of the TBBI file section below.

It should then produce a TBBI record for each file in the given TABI file.

[continued from subset 1 example]
cd ../bbb
../../rbuoy --stage-2 ../out.tbbi ../out.tabi
1092 rbuoy-show ../out.tbbi
Field name         Offset        Bytes                    ASCII/Numeric
-----------------------------------------------------------------------
magic              0x00000000    54 42 42 49              chr TBBI
num records        0x00000004    02                       dec 2
============================= Record   0 ==============================
pathname len       0x00000005    0a 00                    dec 10
pathname           0x00000007    65 6d 6f 6a 69 73 2e 74  chr emojis.t
                   0x0000000f    78 74                    chr xt
num blocks         0x00000011    03 00 00                 dec 3
matches[0]         0x00000014    a0                       bin 10100000
============================= Record   1 ==============================
pathname len       0x00000015    05 00                    dec 5
pathname           0x00000017    65 6d 70 74 79           chr empty
num blocks         0x0000001c    00 00 00                 dec 0

Subset 3

In subset 3, you will need to complete the provided stage_3 function, you will need to produce a TCBI file given a TBBI file as input.

The TCBI file should contain the appropriate header, as outlined in the format of the TCBI file section below. It should also contain a TCBI record for each file in the given TBBI file, containing the data for the blocks the receiver didn't already have an up-to-date copy of.

[continued from subset 2 example]
cd ../aaa
../../rbuoy --stage-3 ../out.tcbi ../out.tbbi
1092 rbuoy-show ../out.tcbi
Field name         Offset        Bytes                    ASCII/Numeric
-----------------------------------------------------------------------
magic              0x00000000    54 43 42 49              chr TCBI
num records        0x00000004    02                       dec 2
============================= Record   0 ==============================
pathname len       0x00000005    0a 00                    dec 10
pathname           0x00000007    65 6d 6f 6a 69 73 2e 74  chr emojis.t
                   0x0000000f    78 74                    chr xt
file type          0x00000011    2d                       chr -
owner perms        0x00000012    72 77 2d                 chr rw-
group perms        0x00000015    72 2d 2d                 chr r--
other perms        0x00000018    2d 2d 2d                 chr ---
file size          0x0000001b    01 02 00 00              dec 513
num updates        0x0000001f    01 00 00                 dec 1
(0) block num      0x00000022    01 00 00                 dec 1
(0) update len     0x00000025    00 01                    dec 256
(0) update data    0x00000027    54 68 65 20 73 65 63 6f  chr The seco
                   0x0000002f    6e 64 20 62 6c 6f 63 6b  chr nd block
                [... omitted for brevity ...]
                   0x00000117    73 20 61 73 74 65 72 69  chr s asteri
                   0x0000011f    73 6b 20 2d 2d 3e 20 2a  chr sk --> *
============================= Record   1 ==============================
pathname len       0x00000127    05 00                    dec 5
pathname           0x00000129    65 6d 70 74 79           chr empty
file type          0x0000012e    2d                       chr -
owner perms        0x0000012f    72 77 2d                 chr rw-
group perms        0x00000132    72 2d 2d                 chr r--
other perms        0x00000135    2d 2d 2d                 chr ---
file size          0x00000138    00 00 00 00              dec 0
num updates        0x0000013c    00 00 00                 dec 0

Subset 4

So far, we've created several types of rbuoy indices files in order to communicate the current state of the receiver's files to the sender, and to communicate updated blocks from the sender to the receiver. In this subset, you will need to complete the provided stage_4 function, which will be invoked with a TCBI file as input. You will then need to apply the changes described in the TCBI file to the receiver's files. This includes updating the contents of the receiver's files, and creating any new files that are required. You will also need to update the mode of the receiver's files such that the permissions match those described in the TCBI file.

[continued from subset 3 example]
cd ../bbb
../../rbuoy --stage-4 ../out.tcbi
diff ../aaa/empty ../bbb/empty identical
diff ../aaa/emojis.txt ../bbb/emojis.txt identical
we have now synchronised `empty` and `emojis.txt` from aaa/ to bbb/

Subset 5

Subset 5 requires you to add support for directories. You will need to update your stage_1, stage_2, stage_3 and stage_4 implementations to complete subset 5:

• In stage_1, if the value of num_in_filenames is zero, then you should create a TABI file containing the contents of the entire current working directory. When num_in_filenames is non-zero you can still make the assumption that all paths in in_filenames are either regular files or don't exist.

  When creating a TABI file for the current directory, you should include a record for every directory, as well as every file. Records for directories should have their number of blocks as zero. The record for a parent directory should be placed in the TABI file before any records for files or sub-directories in that parent directory. Apart from that restriction, you may choose any order for records in the generated TABI file.

• In stage_2, a record with a path which is a directory for the receiver should result in all match bits being set to zero.

• In stage_3, a record with a path which is a directory for the sender should be treated as an empty file. That is, the number of blocks should be checked to be zero, and a record with no updates should be generated. Note that the file type of the mode should be a d rather than a -.

• In stage_4, you should create directories for directory records if they do not exist. You should also set the correct permissions for directories. If a record for a file has the path of an existing directory, or a record for a directory has the path of an existing file, then you should output an appropriate error message and exit with status 1.

Additionally, you must add checks in stage_2, stage_3 and stage_4 to detect if any paths referenced in the input rbuoy indices reference files outside the current working directory. When that occurs, you should output an appropriate error message and exit with status 1. In real code, it is important that untrusted user input such as paths cannot be used to do damage to the wider system. You may assume that if any initial segment of the path exits the current working directory then the whole path will exit the current working directory.

You are encouraged to use the reference implementation to check that your understanding of the above subset 5 requirements are correct.

Error handling

Error checking is an important part of this assignment. Automarking will test error handling.

Error messages should be one line (only) and be written to stderr (not stdout).

rbuoy should exit with status 1 after an error.

You do not have to free memory or close files before exiting in the event of an error.

rbuoy should check all file operations for errors.

As much as possible match the reference implementation error messages exactly.

The reference implementation uses perror(3) to report errors from file operations and other system calls.

It is not necessary to remove files and directories already created or partially created when an error occurs.

You may leave any created rbuoy indices in an indeterminate state.

Where multiple error messages could be produced, rbuoy may produce any one of the error messages.

In stages 2, 3, and 4 you cannot assume that the input rbuoy indices are in a valid format. If your program is given an invalid Buoy Index file, you must output an appropriate error message to stderr and exit with status 1.

During automarking to be awarded marks for the error handling tests you'll need to also have passed a sufficient proportion of the non-error tests for that subset.

Reference implementation

A reference implementation is a common, efficient, and effective method to provide or define an operational specification; and it's something you will likely work with after you leave UNSW.

We've provided a reference implementation, 1092 rbuoy, which you can use to find the correct outputs and behaviours for any input:

1092 rbuoy-examples
cd examples
cd aaa
1092 rbuoy --stage-1 ../out.tabi short.txt
1092 rbuoy-show ../out.tabi 
Field name         Offset        Bytes                    ASCII/Numeric
-----------------------------------------------------------------------
magic              0x00000000    54 41 42 49              chr TABI
num records        0x00000004    01                       dec 1
============================= Record   0 ==============================
pathname len       0x00000005    09 00                    dec 9
pathname           0x00000007    73 68 6f 72 74 2e 74 78  chr short.tx
                   0x0000000f    74                       chr t
num blocks         0x00000010    01 00 00                 dec 1
hashes[0]          0x00000013    15 b8 4c 98 fe c3 b7 d6  chr ..L.....

Every concrete example shown below is runnable using the helper utilities; run 1092 rbuoy instead of ./rbuoy.

The command 1092 rbuoy-show <name of index> display the contents of TABI, TBBI and TCBI files in a human readable format. It is useful for understanding the output of both the reference implementation and your own implementation.

Where any aspect of this assignment is undefined in this specification, you should match the behaviour exhibited by the reference implementation. Discovering and matching the reference implementation's behaviour is deliberately a part of this assignment.

If you discover what you believe to be a bug in the reference implementation, please report it in the class forum. If it is a bug, we may fix the bug; or otherwise indicate that you do not need to match the reference implementation's behaviour in that specific case.

Helper utilities

Alongside 1092 rbuoy-show, which was used above, we have also provided you two additional utilities - 1092 rbuoy-dump-blocks and 1092 rbuoy-hash-block. These utilities have been provided to assist you in understanding the requirements of the assignment, and to help you debug your program.

1092 rbuoy-dump-blocks takes a file as input and splits it into 256 (BLOCK_SIZE) byte blocks, and outputs it to stdout either in hex format or raw bytes. This is useful for ensuring that your program is correctly splitting files into blocks.

1092 rbuoy-dump-blocks ---raw examples/aaa/emojis.txt
=== block 0 ===
This file should be broken up by your program into three blocks: the
first 256 bytes spans lines one to four (and includes the newline on line
four), the second 256 bytes is from line 5 to the asterisk (inclusive), and
the final block is only 1 byte long!

=== block 1 ===
The second block started on this line. Now for an assortment of emoji:
✨ ✨ ✨ 1️⃣ 5️⃣ 2️⃣ 1️⃣  ✨ ✨ ✨
📚 🎓 📈 📈 💾 💽 💿 🖥️ 💻 🚀 🌌 🤯 🎉 🥳
The last character of this block is this asterisk --> *
=== block 2 ===
a
[... no newline after output ...]
1092 rbuoy-dump-blocks --hex examples/aaa/emojis.txt
=== block 0 ===
    54 68 69 73 20 66 69 6c 65 20 73 68 6f 75 6c 64
    20 62 65 20 62 72 6f 6b 65 6e 20 75 70 20 62 79
    20 79 6f 75 72 20 70 72 6f 67 72 61 6d 20 69 6e
    74 6f 20 74 68 72 65 65 20 62 6c 6f 63 6b 73 3a
    20 74 68 65 0a 66 69 72 73 74 20 32 35 36 20 62
    79 74 65 73 20 73 70 61 6e 73 20 6c 69 6e 65 73
    20 6f 6e 65 20 74 6f 20 66 6f 75 72 20 28 61 6e
    64 20 69 6e 63 6c 75 64 65 73 20 74 68 65 20 6e
    65 77 6c 69 6e 65 20 6f 6e 20 6c 69 6e 65 0a 66
    6f 75 72 29 2c 20 74 68 65 20 73 65 63 6f 6e 64
    20 32 35 36 20 62 79 74 65 73 20 69 73 20 66 72
    6f 6d 20 6c 69 6e 65 20 35 20 74 6f 20 74 68 65
    20 61 73 74 65 72 69 73 6b 20 28 69 6e 63 6c 75
    73 69 76 65 29 2c 20 61 6e 64 0a 74 68 65 20 66
    69 6e 61 6c 20 62 6c 6f 63 6b 20 69 73 20 6f 6e
    6c 79 20 31 20 62 79 74 65 20 6c 6f 6e 67 21 0a

 === block 1 ===
    54 68 65 20 73 65 63 6f 6e 64 20 62 6c 6f 63 6b
    20 73 74 61 72 74 65 64 20 6f 6e 20 74 68 69 73
    20 6c 69 6e 65 2e 20 4e 6f 77 20 66 6f 72 20 61
    6e 20 61 73 73 6f 72 74 6d 65 6e 74 20 6f 66 20
    65 6d 6f 6a 69 3a 0a e2 9c a8 20 e2 9c a8 20 e2
    9c a8 20 31 ef b8 8f e2 83 a3 20 35 ef b8 8f e2
    83 a3 20 32 ef b8 8f e2 83 a3 20 31 ef b8 8f e2
    83 a3 20 20 e2 9c a8 20 e2 9c a8 20 e2 9c a8 0a
    f0 9f 93 9a 20 f0 9f 8e 93 20 f0 9f 93 88 20 f0
    9f 93 88 20 f0 9f 92 be 20 f0 9f 92 bd 20 f0 9f
    92 bf 20 f0 9f 96 a5 ef b8 8f 20 f0 9f 92 bb 20
    f0 9f 9a 80 20 f0 9f 8c 8c 20 f0 9f a4 af 20 f0
    9f 8e 89 20 f0 9f a5 b3 0a 54 68 65 20 6c 61 73
    74 20 63 68 61 72 61 63 74 65 72 20 6f 66 20 74
    68 69 73 20 62 6c 6f 63 6b 20 69 73 20 74 68 69
    73 20 61 73 74 65 72 69 73 6b 20 2d 2d 3e 20 2a

 === block 2 ===
    61

Additionally, 1092 rbuoy-dump-blocks is able to only output a single block, specified by the --index option. For example, to only output the first block of the file examples/aaa/emojis.txt as hex, you would run:

1092 rbuoy-dump-blocks --index 0 --hex rbuoy/examples/aaa/emojis.txt
=== block 0 ===
    54 68 69 73 20 66 69 6c 65 20 73 68 6f 75 6c 64
    20 62 65 20 62 72 6f 6b 65 6e 20 75 70 20 62 79
    20 79 6f 75 72 20 70 72 6f 67 72 61 6d 20 69 6e
    74 6f 20 74 68 72 65 65 20 62 6c 6f 63 6b 73 3a
    20 74 68 65 0a 66 69 72 73 74 20 32 35 36 20 62
    79 74 65 73 20 73 70 61 6e 73 20 6c 69 6e 65 73
    20 6f 6e 65 20 74 6f 20 66 6f 75 72 20 28 61 6e
    64 20 69 6e 63 6c 75 64 65 73 20 74 68 65 20 6e
    65 77 6c 69 6e 65 20 6f 6e 20 6c 69 6e 65 0a 66
    6f 75 72 29 2c 20 74 68 65 20 73 65 63 6f 6e 64
    20 32 35 36 20 62 79 74 65 73 20 69 73 20 66 72
    6f 6d 20 6c 69 6e 65 20 35 20 74 6f 20 74 68 65
    20 61 73 74 65 72 69 73 6b 20 28 69 6e 63 6c 75
    73 69 76 65 29 2c 20 61 6e 64 0a 74 68 65 20 66
    69 6e 61 6c 20 62 6c 6f 63 6b 20 69 73 20 6f 6e
    6c 79 20 31 20 62 79 74 65 20 6c 6f 6e 67 21 0a

We have also provided a 1092 rbuoy-hash-block command that reads up to 256 (BLOCK_SIZE) bytes from standard input and outputs the 64-bit hash of the data as a hex string, using the same hash_block function as provided for the assignment. We've also provided you the source code for this command in rbuoy_hash_block.c for your reference.

You can combine these commands to check the hash of any given block of an input file, for example:

1092 rbuoy-dump-blocks --index 0 --raw examples/aaa/emojis.txt | 1092 rbuoy-hash-block
900ae76e14e33090

It is important to use the --raw option and specify a block index in order to produce the expected hash for that block.

Formats of rbuoy indices

The rbuoy indices emitted by your implementation must follow the exact format produced by the reference implementation.

Type A Buoy Index format

When a sender wants to send files, it first creates a TABI file. This file contains a record for each file that is going to be sent. In each record is the pathname of the file, the number of blocks in the file (computed by number_of_blocks_in_file), and the hash of each block in the file.

A TABI file consists of a header, followed by 0 or more records. The format of the header is:

name	length	type	description
magic number	4 B	characters sequence	The magic number for TABI files, which is the sequence of bytes 0x54, 0x41, 0x42, 0x49 (ASCII TABI).
number of records	1 B	unsigned, 8-bit	The number of records in this TABI file.

The TABI header is followed by the specified number of records. Each TABI record has the following format:

name	length	type	description
pathname length	2 B	unsigned, 16-bit, little-endian	The length of the pathname of this record.
pathname	pathname-length	character sequence	The pathname of the file of this record. It is not nul-terminated.
number of blocks	3 B	unsigned, 24-bit, little-endian	The number of 256-byte blocks in the sender's version of the file (the final block may be shorter than 256 bytes).
hashes	8 B × num-blocks	sequence of unsigned, 64-bit, little-endian integers	The hashes the sender has computed for their version of the file (using the hash_block function), with one 64-bit hash for each block.

An example TABI file, displayed using 1092 rbuoy-show:

1092 rbuoy-examples
cd examples
1092 rbuoy-show tabi/my_text_files.tabi
Field name         Offset        Bytes                    ASCII/Numeric
-----------------------------------------------------------------------
magic              0x00000000    54 41 42 49              chr TABI
num records        0x00000004    03                       dec 3
============================= Record   0 ==============================
filename len       0x00000005    09 00                    dec 9
filename           0x00000007    73 68 6f 72 74 2e 74 78  chr short.tx
                   0x0000000f    74                       chr t
num blocks         0x00000010    01 00 00                 dec 1
hashes[0]          0x00000013    15 b8 4c 98 fe c3 b7 d6  chr ..L.....
============================= Record   1 ==============================
filename len       0x0000001b    0a 00                    dec 10
filename           0x0000001d    65 6d 6f 6a 69 73 2e 74  chr emojis.t
                   0x00000025    78 74                    chr xt
num blocks         0x00000027    03 00 00                 dec 3
hashes[0]          0x0000002a    90 30 e3 14 6e e7 0a 90  chr .0..n...
hashes[1]          0x00000032    91 90 5c 46 fc 07 b3 93  chr ..\F....
hashes[2]          0x0000003a    8c ec 01 86 4c dc 63 af  chr ....L.c.
============================= Record   2 ==============================
filename len       0x00000042    05 00                    dec 5
filename           0x00000044    65 6d 70 74 79           chr empty
num blocks         0x00000049    00 00 00                 dec 0

The above example shows that the sender is sending three files: short.txt, emojis.txt, and empty. The file short.txt has one block of data (so its length must be between 1 and 256), and that block has a hash 0xd6b7c3fe984cb815.

The second file emojis.txt has 3 blocks, so its length must be between 513 and 768. The first block (bytes at indices 0..255) hashes to 0x900ae76e14e33090, the second block (bytes and indices 256..511) has a hash of 0x93b307fc465c9091 and the final block (bytes from index 512 to the end of the file) has a hash of 0xaf63dc4c8601ec8c.

The final record is for the file named empty. Since it has zero blocks it must be, as its name suggests, empty.

Type B Buoy Index format

After a receiver receives a TABI file, it responds with a TBBI file, containing information about which blocks the receiver already has a copy of. A TBBI file contains a header, followed by zero or more records. The format for the header is:

name	length	type	description
magic number	4 B	characters sequence	The magic number for TBBI files, which is the sequence of bytes 0x54, 0x42, 0x42, 0x49 (ASCII TBBI).
number of records	1 B	unsigned, 8-bit	The number of records in this TBBI file.

Following the TBBI header are the records. The receiver creates one record for each record in the TABI file. Each TBBI record has the following format:

name	length	type	description
pathname length	2 B	unsigned, 16-bit, little-endian	The length of the pathname of this record.
pathname	pathname-length	character sequence	The pathname of the file of this record. It is not nul-terminated.
number of blocks	3 B	unsigned, 24-bit, little-endian	The number of blocks in the sender's version of the file. This is the same value as the number of blocks in the TABI file.
matches	ceil(num-blocks ÷ 8) (num_tbbi_match_bytes)	bit sequence	A sequence of bits, with a single bit for each hash in the TABI file. For each hash in the TABI file, the receiver computes the hash for the corresponding block in their own copy of the file. If the two hashes match, then the corresponding match bit is a 1. Otherwise (if the hashes don't match, there is no corresponding block because the receiver's file is too small, or the file doesn't exist) the corresponding bit is 0. This means that if the file doesn't exist for the receiver, all the bits in the matches field will be 0. The first bit is the most significant bit of the first byte. In the case where the number of blocks is not a multiple of 8, the last byte of the matches field is right-padded with 0 bits.

An example TBBI file, displayed using 1092 rbuoy-show:

1092 rbuoy-examples
cd examples
1092 rbuoy-show tbbi/bbb_text_files.tbbi
Field name         Offset        Bytes                    ASCII/Numeric
-----------------------------------------------------------------------
magic              0x00000000    54 42 42 49              chr TBBI
num records        0x00000004    03                       dec 3
============================= Record   0 ==============================
pathname len       0x00000005    09 00                    dec 9
pathname           0x00000007    73 68 6f 72 74 2e 74 78  chr short.tx
                   0x0000000f    74                       chr t
num blocks         0x00000010    01 00 00                 dec 1
matches[0]         0x00000013    00                       bin 00000000
============================= Record   1 ==============================
pathname len       0x00000014    0a 00                    dec 10
pathname           0x00000016    65 6d 6f 6a 69 73 2e 74  chr emojis.t
                   0x0000001e    78 74                    chr xt
num blocks         0x00000020    03 00 00                 dec 3
matches[0]         0x00000023    a0                       bin 10100000
============================= Record   2 ==============================
pathname len       0x00000024    05 00                    dec 5
pathname           0x00000026    65 6d 70 74 79           chr empty
num blocks         0x0000002b    00 00 00                 dec 0

The above example file shows a response to the tabi/my_text_files.tabi TABI file. The first file, short.txt has only one block. Since the first bit of matches is a 0, this means that either the first block of the receiver's version of short.txt didn't have a hash of 0xd6b7c3fe984cb815, or the receiver didn't have the file short.txt at all. The remaining 7 bits are padding bits, and so are all zero.

The second file, emojis.txt has 3 blocks. The first of the 3 match bits is a 1. This means that the the first block of the receiver's emojis.txt had a hash which matched the hash of the first block in the TABI record (0x900ae76e14e33090). The second bit is a 0, meaning that the receiver's second block didn't match the second hash in the TABI record. The third bit is a 1, so the third block did match. The remaining 5 bits are padding bits, and so are all zero.

The final file, empty, has zero blocks. Since num_tbbi_match_bytes(0) == 0, this means that there are no match bytes included in the TBBI file.

Type C Buoy Index format

After the sender receives the TBBI file, it responds with a TCBI file, containing the data for the blocks which the receiver didn't already have a copy of. A TCBI file contains a header, followed by zero or more records. The format for the header is:

name	length	type	description
magic number	4 B	characters sequence	The magic number for TCBI files, which is the sequence of bytes 0x54, 0x43, 0x42, 0x49 (ASCII TCBI).
number of records	1 B	unsigned, 8-bit	The number of records in this TCBI file.

Following the TCBI header are the records. The sender creates one record for each record in the TBBI file. Each TCBI record has two sections. The first section has the following format:

name	length	type	description
pathname length	2 B	unsigned, 16-bit, little-endian	The length of the pathname of this record.
pathname	pathname-length	character sequence	The pathname of the file of this record. It is not nul-terminated.
mode	10 B	characters	The type and permissions as a ls(1)-like character array; e.g., "-rwxr-xr-x". It is not nul-terminated.
file size	4 B	unsigned, 32-bit, little-endian	The size of the sender's version of the file, in bytes.
number of updates	3 B	unsigned, 24-bit, little-endian	The number of updates in this record.

The second section of the record contains the updates for that file. An update contains a block of data which the receiver needs. The number of updates for a record is equal to the number of non-padding 0 bits in the TBBI record. Each non-padding 0 bit creates an update. An update has the following format:

name	length	type	description
block index	3 B	unsigned, 24-bit, little-endian	The index of the block that this update is for. This is zero-indexed - the first block in a file has an index of 0, the second block has an index of 1, and so on.
update length	2 B	unsigned, 16-bit, little-endian	The number of bytes in the block that is being updated. For any block apart from the trailing block, this is equal to 256. But the final block in a file might be shorter than that.
update data	update-length	bytes	The block at block-index from the sender's version of the file.

An example TCBI file, displayed using 1092 rbuoy-show:

1092 rbuoy-examples
cd examples
1092 rbuoy-show tcbi/bbb_text_files.tcbi
Field name         Offset        Bytes                    ASCII/Numeric
-----------------------------------------------------------------------
magic              0x00000000    54 43 42 49              chr TCBI
num records        0x00000004    03                       dec 3
============================= Record   0 ==============================
pathname len       0x00000005    09 00                    dec 9
pathname           0x00000007    73 68 6f 72 74 2e 74 78  chr short.tx
                   0x0000000f    74                       chr t
file type          0x00000010    2d                       chr -
owner perms        0x00000011    72 77 2d                 chr rw-
group perms        0x00000014    72 2d 2d                 chr r--
other perms        0x00000017    2d 2d 2d                 chr ---
file size          0x0000001a    40 00 00 00              dec 64
num updates        0x0000001e    01 00 00                 dec 1
(0) block num      0x00000021    00 00 00                 dec 0
(0) update len     0x00000024    40 00                    dec 64
(0) update data    0x00000026    54 68 69 73 20 74 65 78  chr This tex
                   0x0000002e    74 20 66 69 6c 65 20 68  chr t file h
                   0x00000036    61 73 20 73 69 78 74 79  chr as sixty
                   0x0000003e    20 66 6f 75 72 20 62 79  chr  four by
                   0x00000046    74 65 73 2c 20 74 77 65  chr tes, twe
                   0x0000004e    6c 76 65 20 77 6f 72 64  chr lve word
                   0x00000056    73 20 61 6e 64 20 6f 6e  chr s and on
                   0x0000005e    65 20 6c 69 6e 65 2e 0a  chr e line..
============================= Record   1 ==============================
pathname len       0x00000066    0a 00                    dec 10
pathname           0x00000068    65 6d 6f 6a 69 73 2e 74  chr emojis.t
                   0x00000070    78 74                    chr xt
file type          0x00000072    2d                       chr -
owner perms        0x00000073    72 77 2d                 chr rw-
group perms        0x00000076    72 2d 2d                 chr r--
other perms        0x00000079    2d 2d 2d                 chr ---
file size          0x0000007c    01 02 00 00              dec 513
num updates        0x00000080    01 00 00                 dec 1
(0) block num      0x00000083    01 00 00                 dec 1
(0) update len     0x00000086    00 01                    dec 256
(0) update data    0x00000088    54 68 65 20 73 65 63 6f  chr The seco
                   0x00000090    6e 64 20 62 6c 6f 63 6b  chr nd block
                   0x00000098    20 73 74 61 72 74 65 64  chr  started
               [... omitted for brevity ...]
                   0x00000170    6b 20 69 73 20 74 68 69  chr k is thi
                   0x00000178    73 20 61 73 74 65 72 69  chr s asteri
                   0x00000180    73 6b 20 2d 2d 3e 20 2a  chr sk --> *
============================= Record   2 ==============================
pathname len       0x00000188    05 00                    dec 5
pathname           0x0000018a    65 6d 70 74 79           chr empty
file type          0x0000018f    2d                       chr -
owner perms        0x00000190    72 77 2d                 chr rw-
group perms        0x00000193    72 2d 2d                 chr r--
other perms        0x00000196    2d 2d 2d                 chr ---
file size          0x00000199    00 00 00 00              dec 0
num updates        0x0000019d    00 00 00                 dec 0

The above example file shows a response to the tbbi/bbb_text_files.tbbi TBBI file.

Since the receiver indicated that it didn't have the first block of shorts.txt, a single update is sent containing the contents of that first block. Since the first block (which is also the last block) has a length of 64 bytes, the update length is also 64 bytes.

The second file, emojis.txt has 3 blocks. But since the receiver indicated it had the first and third block, only the second block needs to be sent across, so there is only one update. The second block (since it's not the final block) has a length of 256 bytes.

The final file, empty, does not require any updates (there are no non-padding zero bits in the TBBI file), so the number of updates is zero.

The file type (- for file and d for directory), as well as the permissions, are also included in every record, even if there are no updates.

Hashing and the hash_block function

A hash function is a function which takes a sequence of bytes and returns a fixed-length value called a hash. The hash is usually much smaller than the input, and is often used to verify that the input has not been modified without having to store the entire input. For example, if you download a file from the internet, you can verify that the file hasn't been corrupted by comparing the hash of the file you downloaded to the hash of the file published by the author. If the hashes are the same, then the file is almost certainly the same as well, as hash functions are designed to produce different hashes for even slightly different inputs, and be very unlikely to produce the same hash for two given inputs.

The supplied hash_block function takes a sequence of bytes and returns a hash. The hash produced is a 64-bit integer regardless of the size of the input. You are not required to understand how the hash_block function works, but you are required to use it in your implementation of rbuoy to compute the hashes of blocks.

rbuoy vs rsync (optional extra information)

Click here to view more

There are a few interesting differences between how the real rsync algorithm works and what rbuoy does. This section describes what they are, but it's not necessary to know this to complete this assignment, nor is it in scope for this course. If you just want to work on the assignment, you can safely skip this section.

One serious issue the rbuoy algorithm has is that while it can efficiently deal with bytes being modified in‑place in a file, it fails to efficiently work when bytes have been inserted or deleted. This is because any insertions or deletions not only alter the block where the change occurred, but by shifting all the following bytes over, it also causes all subsequent blocks to be considered as modified. This means that, for example, inserting a single character at the start of a text file forces rbuoy to transmit the whole file.

For certain workloads this is not such a big deal (for example most databases, which avoid needing to shift bytes in files where it's not necessary). But for other workloads, such as editing source code, this is far from ideal.

The solution is to use 'rolling hashes'. These are hashes where if you've already computed the hash for some block, it's very easy to compute the hash for a block beginning 1 byte after the start of the original block. Then, it's possible to search for shared data blocks not just when the two blocks are at exactly the same position across the two files, but also when the two blocks are at any offset from each other. This means that the real rsync can operate effectively even when there is insertion or deletion of bytes.

Another issue is the choice of hash function. In rbuoy, the hash_block function implements the 64-bit variant of the Fowler–Noll–Vo hash function. We then operate under the assumption that if the hashes of two blocks are the same, that means that the two blocks must also be identical. This assumption is mathematically false, due to the pigeon-hole principle, and the phenomenon where two different inputs hash to the same value is called hash collision.

But worse yet, the design of the FNV hash means that it is computationally feasible for an adversary to find FNV hash collisions (that is, FNV is not collisions resistant). This problem of collisions can cause rbuoy to run successfully, but still not have correctly synchronised files (which defeats the whole point of running rbuoy)!

The rolling hash used in rsync suffers a similar problem. The solution rsync uses is to use a secondary hash to confirm that two blocks are indeed the same, if the rolling hashes match. And although this still leaves open the mathematical possibility of a collision, because this secondary hash is designed to be cryptographically secure (and thus collision resistant), in practice it's not just unlikely for a collision to happen randomly, but it's also computationally very difficult to deliberately cause a collision to occur.

Assumptions and Clarifications

Like all good programmers, you should make as few assumptions as possible. If in doubt, match the output of the reference implementation.

• Your submitted code must be a single C program only. You may not submit code in other languages.

• You can call functions from the C standard library available by default on CSE Linux systems: including, e.g., stdio.h, stdlib.h, string.h, math.h, assert.h, as well as any C POSIX libraries used in lectures or lecture slides such as unistd.h, sys/types.h, sys/stat.h, fcntl.h, dirent.h.

• We will compile your code with dcc when marking. Run-time errors from illegal or invalid C will cause your code to fail automarking (and will likely result in you losing marks).

• Your program must not require extra compile options. It must compile successfully with:

  dcc *.c -o rbuoy
  

• You may not use functions from other libraries. In other words, you cannot use the dcc -l flag.

• If your program prints debugging output, Make sure you disable any debugging output before submission. it will fail automarking tests.

• You may not create or use temporary files.

• You may not create subprocesses: you may not use posix_spawn(3), posix_spawnp(3), system(3), popen(3), fork(2), vfork(2), clone(2), or any of the exec* family of functions, like execve(2).

• rbuoy only has to handle ordinary files and directories.

  rbuoy does not have to handle symbolic links, devices or other special files.

  rbuoy will not be run in directories containing symbolic links, devices or other special files.

  rbuoy does not have to handle hard links.

• Outside of the cases of errors or early termination, rbuoy must make a reasonable attempt to free all memory it has allocated and close any open files.
• rbuoy will never need to delete any files.
• You may not make any assumptions based off file extensions.
• You must not assume that your program is being run on a system using little-endian byte ordering - you will be assessed on portability with respect to byte ordering.

If you need clarification on what you can and cannot use or do for this assignment, ask in the class forum.

You are required to submit intermediate versions of your assignment. See below for details.

Subset weighting

The weighting of each subset in the performance mark is as follows:

• Subset 1: 40%
• Subset 2: 25%
• Subset 3: 15%
• Subset 4: 12%
• Subset 5: 8%

Change Log

Version 1.0.0
(2024-10-21 12:00:00)

• Initial release.

Assessment

Testing

When you think your program is working, you can use autotest to run some simple automated tests:

1092 autotest rbuoy [optionally: any extra .c or .h files]

You can also run autotests for a specific subset. For example, to run all tests from subset 1:

1092 autotest rbuoy S1 [optionally: any extra .c or .h files]

Some tests are more complex than others. If you are failing more than one test, you are encouraged to focus on solving the first of those failing tests. To do so, you can run a specific test by giving its name to the autotest command:

1092 autotest rbuoy S1_0 [optionally: any extra .c or .h files]

1092 autotest will not test everything.
Always do your own testing.

Automarking will be run by the lecturer after the submission deadline, using a superset of tests to those autotest runs for you.

Submission

When you are finished working on the assignment, you must submit your work by running give:

give dp1092 ass2_rbuoy rbuoy.c [optionally: any extra .c or .h files]

You must run give before Week 12 Friday 09:00am to obtain the marks for this assignment. Note that this is an individual exercise, the work you submit with give must be entirely your own.

You can run give multiple times.
Only your last submission will be marked.

If you are working at home, you may find it more convenient to upload your work via give's web interface.

You cannot obtain marks by emailing your code to tutors or lecturers.

You can check your latest submission on CSE servers with:

1092 classrun check ass2_rbuoy

You can check the files you have submitted here.

Manual marking will be done by your tutor, who will mark for style and readability, as described in the Assessment section below. After your tutor has assessed your work, you can view your results here; The resulting mark will also be available via give's web interface.

Due Date

This assignment is due Week 12 Friday 09:00am (2024-11-15 09:00:00).

The UNSW standard late penalty for assessment is 5% per day for 5 days - this is implemented hourly for this assignment.

Your assignment mark will be reduced by 0.2% for each hour (or part thereof) late past the submission deadline.

For example, if an assignment worth 60% was submitted half an hour late, it would be awarded 59.8%, whereas if it was submitted past 10 hours late, it would be awarded 57.8%.

Beware - submissions 5 or more days late will receive zero marks. This again is the UNSW standard assessment policy.

Assessment Scheme

This assignment will contribute 15 marks to your final DPST1092 mark.

80% of the marks for assignment 2 will come from the performance of your code on a large series of tests.

20% of the marks for assignment 2 will come from hand marking. These marks will be awarded on the basis of clarity, commenting, elegance and style. In other words, you will be assessed on how easy it is for a human to read and understand your program.

An indicative assessment scheme for style follows. The lecturer may vary the assessment scheme after inspecting the assignment submissions, but it is likely to be broadly similar to the following:

100% for style	perfect style
90% for style	great style, almost all style characteristics perfect.
80% for style	good style, one or two style characteristics not well done.
70% for style	good style, a few style characteristics not well done.
60% for style	ok style, an attempt at most style characteristics.
≤ 50% for style	an attempt at style.

An indicative style rubric follows:

• Formatting (6/20):
  • Whitespace (e.g. 1 + 2 instead of 1+2)
  • Indentation (consistent, tabs or spaces are okay)
  • Line length (below 80 characters unless very exceptional)
  • Line breaks (using vertical whitespace to improve readability)
• Documentation (8/20):
  • Header comment (with name and zID)
  • Function comments (above each function with a good description)
  • Descriptive variable names (e.g. char *home_directory instead of char *h)
  • Descriptive function names (e.g. get_home_directory instead of get_hd)
  • Sensible commenting throughout the code (don't comment every single line; leave comments when necessary)
• Elegance (5/20):
  • Does this code avoid redundancy? (e.g. Don't repeat yourself!)
  • Are helper functions used to reduce complexity? (functions should be small and simple where possible)
  • Are constants appropriately created and used? (magic numbers should be avoided)
• Portability (1/20):
  • Would this code be able to compile and behave as expected on other POSIX-compliant machines? (using standard libraries without platform-specific code)
  • Does this code make any assumptions about the endianness of the machine it is running on?

Note that the following penalties apply to your total mark for plagiarism:

0 for asst2	knowingly providing your work to anyone and it is subsequently submitted (by anyone).
0 FL for DPST1092	submitting any other person's work; this includes joint work.
academic misconduct	submitting another person's work without their consent; paying another person to do work for you.

Intermediate Versions of Work

You are required to submit intermediate versions of your assignment.

Every time you work on the assignment and make some progress you should copy your work to your CSE account and submit it using the give command above. It is fine if intermediate versions do not compile or otherwise fail submission tests. Only the final submitted version of your assignment will be marked.

Assignment Conditions

• Joint work is not permitted on this assignment.

  This is an individual assignment. The work you submit must be entirely your own work: submission of work even partly written by any other person is not permitted.

  Do not request help from anyone other than the teaching staff of DPST1092 — for example, in the course forum, or in help sessions.

  Do not post your assignment code to the course forum. The teaching staff can view code you have recently submitted with give, or recently autotested.

  Assignment submissions are routinely examined both automatically and manually for work written by others.

  Rationale: this assignment is designed to develop the individual skills needed to produce an entire working program. Using code written by, or taken from, other people will stop you learning these skills. Other courses focus on skills needed for working in a team.

• The use of generative tools such as Github Copilot, ChatGPT, Google Bard is not permitted on this assignment.

  Rationale: this assignment is designed to develop your understanding of basic concepts. Using synthesis tools will stop you learning these fundamental concepts, which will significantly impact your ability to complete future courses.

• Sharing, publishing, or distributing your assignment work is not permitted.

  Do not provide or show your assignment work to any other person, other than the teaching staff of DPST1092. For example, do not message your work to friends.

  Do not publish your assignment code via the Internet. For example, do not place your assignment in a public GitHub repository.

  Rationale: by publishing or sharing your work, you are facilitating other students using your work. If other students find your assignment work and submit part or all of it as their own work, you may become involved in an academic integrity investigation.

• Sharing, publishing, or distributing your assignment work after the completion of DPST1092 is not permitted.

  For example, do not place your assignment in a public GitHub repository after this offering of DPST1092 is over.

  Rationale: DPST1092 may reuse assignment themes covering similar concepts and content. If students in future terms find your assignment work and submit part or all of it as their own work, you may become involved in an academic integrity investigation.

Violation of any of the above conditions may result in an academic integrity investigation, with possible penalties up to and including a mark of 0 in DPST1092, and exclusion from future studies at UNSW. For more information, read the UNSW Student Code, or contact your lecturer.