COMP2521 Assignment 1

Textbuffer
Submission
Specification
Jump to FAQ

Your task is to implement an abstract textbuffer data type that meets the given interface.
You will submit the C code implementing the textbuffer ADT (textbuffer.c).
This page describes the interface of the textbuffer ADT that you are to implement. For your implementation, download textbuffer.c below and implement the type struct textbuffer as well as all of the functions whose prototypes are given in the header file textbuffer.h. All your code should go in textbuffer.c, which you have to submit.
Changelog
1st October
Proofreading
Line numbers now always start at 1
Changed charIndex in struct _matchNode to columnNumber
Changed int showLineNumbers in dumpTB to bool showLineNumbers
Added a stub test file testTextbuffer.c for students to write tests
Improved formatting/style of textbuffer.c and textbuffer.h
Added some clarifications to the spec
2nd October
Added clarifications for searchTB to the FAQ
Added a line to testTextbuffer.c that calls linesTB
Updated the comment for addPrefixTB in textbuffer.c to be consistent with textbuffer.h
3rd October
Added a clarification for pasteTB to the FAQ
Added a clarification about bonus marks
Added a clarification for formRichText Note that the # character must be the first character in a line and there must be more characters on that line for it to be treated as a special character otherwise, it does nothing.
Moved testing for memory leaks from Style Marks to Autotesting Marks
4th October
Added a question to the FAQ under newTB on the input text being NULL.
Fixed the answer to one of the newTB questions: Unless text is the empty string, it will always have a newline at the end.
8th October
Added details to the FAQ on how diffTB will be tested
Added an example for diffTB to the FAQ
9th October
Added details to Late Submission section
Bonus marks can now only make up for lost marks in the labs and the second assignment not the midterm exam. Sorry!
11th October
Removed confusing question about abort from the FAQ
13th October
Submission instructions are now available
14th October
Added a clarification for dumpTB to the FAQ the returned string should always be allocated such that that the user can free it.
Added a clarification for searchTB to the FAQ about returning an empty list
15th October
Removed the -O flag from the compilation line and replaced it with -g, which supports debugging. (-O is just an optimisation flag, which is unnecessary for this assignment.)
Clarified and removed vagueness from the Compactness requirement for diffTB. There is now no model solution. Instead, youll pass each test if the number of commands in your edit solution is smaller than some threshold, determined by the sizes of the optimal and brute-force solutions (see the section for diffTB for details).
Added a link to download all dryrun files
Added clarifications on undoing cutTB and pasteTB to the FAQ.
19th October
Added some clarifications/examples based on questions asked on WebCMS.
Marks
The assignment is worth 10 marks. The mark breakdown is as follows:
Component
Mark
Autotesting of functionality
8 (+2 bonus)
Subjective evaluation of style
2
Due to the bonus challenges, you could get up to 12 marks for the assignment. Any extra marks obtained during this assignment (in excess of the usual 10 marks) can be used to make up for lost marks in the labs and the second assignment.
Automarking 8 (+2) Marks
We will run a number of tests against your textbuffer implementation. These will be much more comprehensive than the tests we run during submission. You get marks for each test you pass.
We will also test your program for memory leaks (memory you have allocated and have responsibility to free but never freed) and memory errors. Your program will be tested for memory leaks/errors via valgrind.
Style 2 Marks
Style marks will include comments, indentation, variable names, etc., and will also include marks for choosing an appropriate representation for your ADT and for the efficiency of the functions you implement. For example, you will lose marks if your implementation of a function has a time complexity of O(n^2) when there is a solution with a time complexity of O(n) or O(n * log n).
Submission
Deadline: 9am, Saturday 26 October 2019.
You need to submit one file: textbuffer.c
You can submit from the assignment page on WebCMS via the give interface or by running the command below:
give cs2521 assign1 textbuffer.c
The submission system runs a few simple dryrun tests. All files used by the dryrun are available here (click here to download the whole lot). After the deadline, all functions will be more thoroughly tested by the automarking system.
You can submit multiple times only your last submission will count.
Late Submission
A late penalty of 15% per day will be applied. The latest you can submit the assignment is 9am, 31 October 2019, of course with late penalty.
Files
.
.textbuffer.h
.
.textbuffer.c
.
.testTextbuffer.c
Note: When we test your assignment, it with be compiled with gcc and the following flags:

gcc -Wall -Werror -std=c11 -g -lm -o testTextbuffer testTextbuffer.c textbuffer.c

ADT Specification

The following is a description of the components of the interface.

As marks are awarded by an automated marking program, you must follow this specification precisely. Otherwise, you risk getting few or no marks! You must NOT modify the textbuffer.h file.
The ADT type
We represent the ADT by way of a handle of type TB. The handle type is declared in the header file, but you will have to provide an implementation of the handle representation i.e. of struct textbuffer as part of your implementation:
typedef struct textbuffer *TB;
Refer to the lecture about ADTs for examples of this construction.
Required properties of the implementation
A textbuffer is an ordered collection of strings, where each string represents one line of a text file. Your implementation must keep the lines of a textbuffer in a linked data structure (such as a linked list or a variant of that). Each line must be represented as a (dynamically allocated) string. Adding, deleting, or moving lines requires manipulation of the linked structure. Such a data structure may, for example, be used as part of a text editor.
Constructor and destructor
newTB
TB newTB (char *text);
newTB allocates a new textbuffer and initialises its contents with the text in the given string. Each fragment of the string that ends with a newline character (
) indicates a separate line in the textbuffer.
releaseTB
void releaseTB (TB tb);
releaseTB frees the memory occupied by the given textbuffer. It is an error to access a textbuffer after freeing it.
Query functions
The following functions do not alter their textbuffer argument.
dumpTB
char *dumpTB (TB tb, bool showLineNumbers);
dumpTB allocates and returns a string containing the text in the given textbuffer. The returned string should contain newline characters (
) to indicate the end of each line in the textbuffer. It is the callers responsibility to free the memory occupied by the returned string. If there are no lines in the textbuffer, return an empty string (the string should still be allocated). If showLineNumbers is true, prepend a line number (along with a dot and space) to each line of the output.
For example, if dumpTB was called on a textbuffer containing the lines hello world and amazing, and showLineNumbers was true, it should return 1. hello world
2. amazing
. If showLineNumbers was false, it should instead return hello world
amazing
.
linesTB
int linesTB (TB tb);
linesTB returns the number of lines in the given textbuffer.
Textbuffer editing
For all editing functions, if any of the arguments indicating a line number is out of range (i.e., smaller than 1 or bigger than the number of lines in the textbuffer), the function must print a suitable error message and terminate the program with the standard function abort().
The first line of a textbuffer is at position/index 1.
addPrefixTB
void addPrefixTB (TB tb, int from, int to, char *prefix);
addPrefixTB adds the supplied prefix to all lines between from and to (inclusive). If to is less than from, abort.
For example, consider calling addPrefixTB (tb, 1, 3, goodnight ):
+ – ++ +
| room || goodnight room |
| moon |>| goodnight moon |
| cow jumping over the moon|| goodnight cow jumping over the moon|
| light|| light|
+ – ++ +
deleteTB
void deleteTB (TB tb, int from, int to);
deleteTB deletes the lines between from and to (inclusive) from the textbuffer tb. It should free the memory of the deleted lines. If to is less than from, abort.
Combining textbuffers
For all combining functions, if any of the arguments indicating a line number is out of range, the function must print a suitable error message and terminate the program with the standard function abort().
Note that for these functions, if the number of lines in tb1 is n, then n + 1 is a valid argument for pos (the lines in tb2 are added to the end of tb1).
mergeTB
void mergeTB (TB tb1, int pos, TB tb2);
mergeTB merges tb2 into tb1 at line pos. Afterwards, what was at line 1 of tb2 will now be at line pos of tb1. Line pos of tb1 will be moved to line pos + linesTB (tb2), after the merged-in lines from tb2. After this operation, tb2 cannot be used anymore (as if we had used releaseTB on it).
pasteTB
void pasteTB (TB tb1, int pos, TB tb2);
pasteTB copies (i.e., pastes) all lines from tb2 into tb1 at line pos. It is like mergeTB, but tb2 remains unmodified and is still usable independent of tb1.
Extracting textbuffers
For all extracting functions, if any of the arguments indicating a line number is out of range, the function must print a suitable error message and terminate the program with the standard function abort().
The textbuffers returned by the extracting functions are as if they were newly created with newTB().
cutTB
TB cutTB (TB tb, int from, int to);
cutTB cuts the lines between from and to (inclusive) out of the textbuffer tb into a new textbuffer, which is then returned. If to is less than from, return NULL.
Searching textbuffers
Match searchTB (TB tb, char *search);
searchTB returns a linked list of all non-overlapping matches in tb of a certain string. The search is case sensitive and the textbuffer tb must remain unmodified. The matches must be returned in order of their appearance in the textbuffer. It is the callers responsibility to free the returned list.
Consider calling searchTB (tb, love) on the following TB:
1 Hello World My
2 name is jarred lovegood
3 and i love carley ray jepson
This should return a list:
+====================++====================+
| lineNumber: 2|| lineNumber: 3|
| columnNumber: 16 || columnNumber: 7|
| next: >| next: > NULL
+====================++====================+
Note that the line number and column number are both 1-indexed (i.e., start at 1). The column number refers to a position within the line where there is a match.
Note that Match is a pointer to the first node in the list. If there are no matches, then return NULL.
Rich text
formRichText
void formRichText (TB tb);
formRichText searches every line of tb and performs the following substitutions:
String
Replacement
Example
*some string*
some string
*hello* -> hello
_some string_
some string
_hello_ -> hello
#some string

some string

#hello ->

hello

The matching is simplistic in that you would begin scanning at the first special character and continue to consume characters (ignoring any further special characters) until a matching special character. If there is no matching special character, nothing is done and the next special character (if there is one) is processed.
Note that the # character must be the first character in a line and there must be more characters on that line for it to be treated as a special character otherwise, it does nothing. Furthermore, it matches until the end of the line and not until a matching #. See example below.
Example
Result
*some string
*some string
*some string*lol*
some stringlol*
* *

*some_string*again_
some_stringagain_
*some* _string_
some string
some *string_again_
some *string again
some#string*once_again*
some#string once_again
#string_stuff_

string_stuff_

#
#
###

##

In the case of nested special characters, for example:
*some_string_*
#some _string_
Take the outermost element and ignore any nesting.
Example
Result
*some_string_*
some_string_
#some _string_

some _string_

If there are no characters between a pair of consecutive special characters, for example, hello ** world, ignore it and continue to the next pair of special characters (if there is one). For example:
hello ** world> hello ** world
hello **world*> hello * world
hello **world** > hello * world*
**hello***world** > * hello* world*
***hello* > ** hello
Note that in the last case, the first * does nothing because there are no characters between it and the next *. In that case the first * is ignored and the next one is processed as normal.
Assignment 1 Bonus Challenges
diffTB (1 bonus mark)
char *diffTB (TB tb1, TB tb2);
Given two text files, we sometimes want to know what changes are made from one file to another file.
The function diffTB works out which lines of texts are added or deleted from tb1 to get tb2. The string returned from the function is an edit solution consisting of a series of add and delete commands. Applying such commands on tb1 in sequence should result in tb2.
An edit solution should have one command per line to either add or delete a line of text at a specific line number. An example is given below. The first command adds a line of text add this line please at line 2 of the current textbuffer (counting from 1). The existing line 2 is moved to line 3, and so on. The second command deletes line 3 of the textbuffer. The last command adds the specified text at line 12 of the textbuffer.
+,2,add this line please
-,3
+,12,add this line as well please
A mark is given if your solution satisfies two criteria given below:
Correctness applying your edit solution on tb1 results in tb2.
Compactness the size of your edit solution (i.e., number of commands/lines) is smaller than or equal to the average of the sizes of the optimal solution and the brute-force solution (which is delete all lines of tb1 and add all lines of tb2). This is to avoid brute-force solutions, such as the one just described.
undoTB and redoTB (1 bonus mark)
void undoTB (TB tb);
undoTB allows the user to reverse up to 10 of the most-recently called operations on tb. Applicable operations are: deleteTB, mergeTB, pasteTB, and cutTB. Each time undoTB is called, one operation is reversed on tb. When the maximum number of allowable undo operations is reached, further calls to undoTB should do nothing (until one of the applicable operations is performed again or redoTB is called).
void redoTB (TB tb);
The function redoTB allows the user to redo operations that have been reversed by undoTB. Similar to undoTB, this function should redo one operation on tb per function call. However, when a new operation is called on tb, any reversed operations cannot be executed again with redoTB.
Note: When testing your undoTB and redoTB functions, we will not call inapplicable operations such as addPrefixTB and formRichText, so you do not have to worry about undoing such operations.
COMP2521 Assignment 1 FAQ

General questions
Can I modify textbuffer.h?
No.
Can I add my own structs and functions to textbuffer.c?
Yes! Make sure your helper functions are declared static, and that you document what the functions and structures you add are for.
Can I use functions from ?
Yes. Its much, much harder if you dont.
Will I need to check my code with valgrind?
Well certainly be checking your submission with valgrind for memory leaks.
Can TB ever be NULL?
It can be in the case something goes wrong but in any case you have a TB which is NULL the correct course of action would be to print a suitable error message and abort().
Can I use the math.h library?
Yes.
Will the time complexity of bonus functions affect the bonus marks?
No.
newTB
How does newTB work?
If the input text is, for example, Hi there,
how
are
things
, the textbuffer should contain the following lines: { Hi there,, how, are, things }. You will have to process the input text, extract all the substrings separated by newlines, and copy them into your textbuffer structure.
Should I leave the
characters in?
Depending on your approach to splitting text, they may already be gone. The only other place you need the
characters is in dumpTB, so you could probably get away without storing them. But it is up to you.
Is it safe to assume that the input text will always have a newline at the end?
Unless text is the empty string, it will always have a newline at the end.
What should happen with multiple consecutive newlines?
Every newline marks a new line in the textbuffer, so a newline that immediately follows another newline (or a newline at the beginning of the input text) would represent an empty line. You need to track empty lines.
Can I assume a maximum length for lines?
No. Your program should be able to dynamically allocate any memory needed for your strings depending on the input text.
What if the input text is a empty string?
Create an empty TB.
What if the input text consists of just a single newline character?
Create a TB with one empty line.
What if the input text is NULL?
We wont be testing this (as NULL is not a valid string), but in this case a sensible thing to do would be to print a suitable error message and abort().
releaseTB
How can I test releaseTB?
You cant. You cant write a black-box test for a destructor.
When you free() memory, what youre saying is that you no longer need the block of memory you had a pointer to; it should be irrelevant to you whether that memorys value changes or becomes invalid in some way, because you are absolutely forbidden from accessing the memory once freed. Use-after-free is an illegal and undefined operation.
A good test that your releaseTB worked is that your program is still running after you do so.
Do note though that valgrind may be useful to help diagnose memory leaks which can indirectly signal a error with your releaseTB.
dumpTB
My textbuffer has no lines; what should dumpTB return?
It should return an empty string, regardless of whether showLineNumbers is true or false. Note that this string should still be allocated so that the user can free it.
addPrefixTB
Can the prefix string have newlines in it?
No. We will not test these cases.
Can the prefix string be the empty string?
Yes. In this case, do nothing.
Can the prefix string be NULL?
No. In this case, abort().
mergeTB
What should happen if I mergeTB (tb1, 1, tb1)?
Attempts to merge a textbuffer with itself should be ignored.
Should I call releaseTB as well?
No! This will probably destroy both the source and destination textbuffers. However, youve moved the contents of the source textbuffer, so you can just free() as you would in releaseTB. You must not subsequently dereference it; thats a use-after-free and (say it with me, folks!) use-after-free is illegal.
Can I concatenate text buffers with mergeTB?
The correct behaviour should be as follows, for mergeTB (dest, pos, src):
pos == 1: Insert src before the start of dest.
pos == linesTB (dest): Insert src before the last line of dest.
pos == linesTB (dest) + 1: Append src to the end of dest.
What should happen if tb1 or tb2 are empty?
Both may be empty. If dest is empty then the only valid value for pos is 1, which would cause src to be appended to the end of the empty TB.
pasteTB
Can a textbuffer be pasted onto itself?
Yes! For example, suppose tb was:
1 Never gonna give you up
2 Never gonna let you down
Then after calling pasteTB (tb, 2, tb), tb would look like:
1 Never gonna give you up
2 Never gonna give you up
3 Never gonna let you down
4 Never gonna let you down
searchTB
Can the search string have newlines in it?
No. We will not test these cases.
Can the search string be the empty string?
Yes. In this case, return an empty list.
How should I return an empty list?
In general, this depends on the representation of the list. If the list is represented by a structure containing metadata about the list (such as its size, the pointer to the first node, etc.), like in the Week 1 and Week 2 labs, then an empty list is represented by a (pointer to a) metadata structure where the size field is set to 0, and the pointers to the first/last nodes are set to NULL. If the list is represented by a pointer to the first node, then an empty list is represented by a NULL pointer, as there are no nodes in the list. In this assignment, because a list of matches is merely represented by a pointer to the first match node, an empty list is represented by NULL.
Can the search string be the NULL?
No. In this case, abort().
Can the search string occur multiple times on the same line?
Yes. In this case, the returned list of matches should have a node for each of the occurrences on that line. For example, if searchTB (tb, bird) is called, and tb is:
1 A well a everybodys heard about the bird
2 B-b-b bird, bird, bird, b-birds the word
3 A well a bird, bird, bird, the bird is the word
4 A well a bird, bird, bird, well the bird is the word
5 A well a bird, bird, bird, b-birds the word
The returned list should be:
(1, 38) > (2, 7) > (2, 13) > (2, 19) > (2, 27) > (3, 10) > (3, 16) >
How do you handle the case where the search string is a repeated pattern (e.g., looking for abab in ababab)?
The matches you return should not overlap. After you find a match on a line, the search should resume from after the part of the line that was matched. For example, if we searched for abracadabra in the string abracadabracadabracadabracadabra, the matches are abracadabracadabracadabracadabra. So if we searched for abracadabra in this textbuffer:
1 abracadabra alacazam
2 abracadabracadabracadabracadabra
The returned list should be:
(1, 1) > (2, 1) > (2, 15) > X
formRichText
How should I handle cases where there are no characters between a pair of special characters (such as **)?
In this case, nothing should happen. Only add the tags if there is at least 1 character being acted on.
Can substitutions occur across lines?
No.
diffTB
Does diffTB change either of its textbuffer arguments?
No. diffTB is non-destructive.
How will diffTB be tested?
There are many possible valid sequences of commands that could be returned from diffTB, so comparing the output against an expected output would not work. Instead, we will parse your edit solution and apply the commands one by one to tb1. After applying the commands, we will call dumpTB on tb1 and tb2. If they return the same string, and your edit solution consists of fewer commands than the threshold, then you pass the test.

Could we get an example?
Certainly!
Suppose that these are tb1 (left) and tb2 (right):
1 first line1 first line
2 second line 2 2nd line
3 third line3 third line
4 fourth line 4 quatre
Here are some examples of correct command strings (there are others):
+,2,2nd line
-,3
+,4,quatre
-,5

-,2
+,2,2nd line
-,4
+,4,quatre

-,2
-,3
+,2,2nd line
+,4,quatre

-,4
-,2
+,3,quatre
+,2,2nd line

undoTB and redoTB
My implementation of undoTB and redoTB requires me to modify existing functions (e.g., mergeTB), which affects their time complexities. Will I be penalised for having a slow time complexity for these functions?
Your implementation of the applicable functions will probably be split into two parts: (1) one part that allows undoTB and redoTB to work, and (2) a second part that actually does the main work of the function. In checking your functions time complexities we will only consider the second part of the code.
Should I record an operation that has no effect on the TB, such as merging an empty textbuffer?
Its up to you we wont be testing these cases.
What should happen if I undo a merge? Is tb2 alive again?
If you called mergeTB (dest, pos, src), src no longer exists, so calling undoTB (src) is invalid. Calling undoTB (dest) should simply remove the merged lines from dest (of course, they may reappear again if redoTB (dest) is called).
What should happen if I undo a cut?
If you called cutTB (tb1, from, to), the returned TB (lets call it tb2) is completely independent of tb1. Calling undoTB (tb1) should restore the lines that were cut from tb1, but have no effect on tb2.
What should happen if I undo a paste?
If you called pasteTB (tb1, pos, tb2), the paste operation is performed on tb1, not tb2, so tb2 has no record of the operation taking place. Thus, calling undoTB (tb1) should remove the pasted lines from tb1 (they may reappear if redoTB (tb1) is called), while calling undoTB (tb2) should do nothing unless some operation was performed on tb2 before the paste.

Plagiarism
This is an individual assignment. Each student will have to develop their own solution without help from other people. In particular, it is not permitted to exchange code or pseudocode. You are not allowed to use code developed by persons other than yourself. If you have questions about the assignment, ask your tutor.
Plagiarism is defined as using the words or ideas of others and presenting them as your own. UNSW and CSE treat plagiarism as academic misconduct, which means that it carries penalties as severe as being excluded from further study at UNSW. There are several on-line sources to help you understand what plagiarism is and how it is dealt with at UNSW:
Plagiarism and Academic Integrity
UNSW Plagiarism Procedure
Make sure that you read and understand these. Ignorance is not accepted as an excuse for plagiarism. In particular, you are also responsible that your assignment files are not accessible by anyone but you by setting the correct permissions in your CSE directory and code repository, if using. Note also that plagiarism includes paying or asking another person to do a piece of work for you and then submitting it as your own work.
UNSW has an ongoing commitment to fostering a culture of learning informed by academic integrity. All UNSW staff and students have a responsibility to adhere to this principle of academic integrity. Plagiarism undermines academic integrity and is not tolerated at UNSW. Plagiarism at UNSW is defined as using the words or ideas of others and passing them off as your own.
If you havent done so yet, please take the time to read the full text of
UNSWs policy regarding academic honesty and plagiarism
The pages below describe the policies and procedures in more detail:
Student Code Policy
Student Misconduct Procedure
Plagiarism Policy Statement
Plagiarism Procedure
You should also read the following page which describes your rights and responsibilities in the CSE context:
Essential Advice for CSE Students