summaryrefslogtreecommitdiff
path: root/32_kvs/README
blob: 35b0a181ab403bfd49e245e110e8ae2889b291e8 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
For the next 4 problems, you will be doing one slightly larger program, which
we will cut up into 4 discretely testable pieces. 

The big picture is that your program will read two types of input files.
The first type of input file (which your program will read one of) will have the format:

key1=value1
key2=value2
....
keyN=valueN

That is, it might say

Jean Luc Picard=Captain
Will Riker=Commander
Beverly Crusher=Commander
Data=Lt. Commander
Geordi LaForge=Lt. Commander
Worf=Lt. Commander
Deanna Troi=Commander

Note that the division between the key and the value is the first equals sign (the
values could have = in them, but the keys cannot.  So a=b=c, would have a key of a, and
a value of b=c).

The second type of input file will contain a list of lines (which will typically match
the keys from the first input file).  Your program will read one or more of this
type of file.  For example, it might contain:

Jean Luc Picard
Will Riker
Worf
Deanna Troi
Q

For each such input file that your program reads, it will print out the counts
of the values for the corresponding keys (or <unknown> for anything that did not
match any known key from the first input file).   E.g. given the above to input files,
it would print to the corresponding output file (which will be named the same
as the input file, but with ".count" appended to the name).

Captain: 1
Commander: 2
Lt. Commander: 1
<unknown> : 1




In thinking about our program, we might come up with the following 
generalized high-level algorithm:

 //read the key/value pairs from the file named by argv[1] (call the result kv)
 //count from 2 to argc (call the number you count i)
    //count the values that appear in the file named by argv[i], using kv as the key/value pair
    //   (call this result c)
    //compute the output file name from argv[i] (call this outName)
    //open the file named by outName (call that f)
    //print the counts from c into the FILE f
    //close f
    //free the memory for outName and c
 //free the memory for kv

This high-level algorithm suggests many functions which we can split our task into.
We will cut them up into 4 groups to make the 4 problems of this assignment:

The first step (this problem):
  - read the key/value pairs from a file
  - free the memory for the key/value pairs

The second step (next problem):
 - compute the output file name

The third step (problem after that):
 - print the counts to a file
 - free the memory for the counts

The fourth step (the problem after that):
 - compute the counts of  values that appear in a particular input file

Each subsequent problem will have more details about it.

For this particular problem, you will need to:

  - read the key/value pairs from a file
  - free the memory for the key/value pairs

To start with this problem, you are going to want to define two structs, in the
file kv.h.  The first one (struct _kvpair_t) should define the structure for
one key/value pair.   The second should define the structure for an 
array of key/value pairs (hint: you will want to include the length of the
array in this structure).

Now you will want to write the four functions in kv.c.  

In readKVs, you will want to open the file, read the lines of text, split them into
key/value pairs, add the resulting pairs to an array (hint: realloc it to make it larger
each time), close the file, and return the kvarray_t * that has your array.

Remember that you will want to abstract out complex steps into functions (you should
see at least 2 pieces to naturally pull out into their own functions).

Next, you will write freeKVs, which should free all the memory allocated by readKVs.
That is,  freeKVs(readKVs(filename)) should not leak any memory.

Third, write printKVs which should take a kvarray_t *, and print out

 "key = '%s' value = '%s'\n"
for each key/value pair in the array, with the first %s being whatever the key is,
and the second being whatever value (e.g., key = 'Jean Luc Picard' value = 'Captain').

Finally, write the lookupValue function, which takes a kvarray_t * and a char * (string).
The char * that is passed in represents a key.  This function searches the kvarray_t 
for a matching key, and if found, returns the corresponding value.  If no match is found,
this function returns NULL.

Once you complete these functions, test them using the main in kv_test.c before
proceeding to the next problem.