FGETS from text to array

Hey guys,

I need to access a text file and store the data into a 3 dimensional array.

I'm having issues getting the info into an array, from there I can manipulate the text as needed. But the first step is stumping me.

Code:

#include <stdio.h> int main(void) { int count = 0; char text[50][50][21]; FILE *ptr_file; char buf[1000]; ptr_file =fopen("proj4.txt","r"); if (!ptr_file) return 1; while (fgets(buf,1000, ptr_file)!=NULL){ text[count] = fgets(buf,1000, ptr_file); count++; } fclose(ptr_file); getch(); return 0; }

Look up the strcpy() function ...

A couple of things:

1. How is the data stored in your file and what is the framework behind the need for a 3d array. Usually if you are just reading in lines of text an array would be 2d, e.g. [NUMLINES][LINESOFTEXT]. If the data is structured in a certain way in your file you could use a fgets-sscanf combination, or fscanf.

2. You actually wind up reading in two lines from your file before you try to save it(in red). In addition I am not sure what you are trying to do with your array(in green), did this code compile?

Code:

while (fgets(buf,1000, ptr_file)!=NULL){ text[count]=fgets(buf,1000, ptr_file); count++;

I'm actually goign with a different approach:

Code:

void get_words(){ int ch, i = 0, j = 0, k = 0; char ch_holder; while ((ch = getc(fp)) != EOF) { ch_holder = ch; if(!isalpha(ch)){ if(iscntrl(ch)) { j++; } if(isspace(ch)) { j++; k = 0; } if(isdigit(ch)){ j++; k = 0; } if(ispunct(ch)){ k++; } } if(isalpha(ch) && !isspace(ch) && !ispunct(ch) && !iscntrl(ch)){ ch_holder = tolower(ch); textArray[i][j][k] = ch_holder; if(textArray[i][j][k] == '\0'){ k--; } k++; } } }

but still getting some weird errors in my array, can't seem to figure out how to avoid it storing certain characters in an array. Still going through the logic and understanding how to approach this. ... tips would be greatly appreciated. As the output is looking like this:
1: information => 11
2: technology => 10
3: => 0 <-- this part here is what i'm trying to avoid and just store regular characters instead of these null spaces.
4: => 0
5: => 0
....

You don't actually need ch_holder. Just use ch. Also:

Code:

if(isalpha(ch) && !isspace(ch) && !ispunct(ch) && !iscntrl(ch)){

If it is alpha, then obviously it is not punct, space, cntrl. Did you mean: NOT isalpha?

Quzah.

no it doesn't really do anything, had that in there before i added the other code. I wanted it to store the alpha's only. Thanks for the catch. but it doesn't resolve my issue.

"weird errors" doesn't tell us anything. What is textArray defned as, and what do you think you are doing here:

Code:

textArray[i][j][k] = ch_holder;

Input string: "::11aa"

Code:

ispunct so k = 1 you didn't actually do anything with k = 0 second if does nothing, because it is punct ispunct so k = 2 you didn't actually do anything with k = 1 same as before, but with k = 2 and k = 3 for the numbers isalpha = true, so skip first iff textarray[ 0 ][ 0 ][ 3 ] = ch k = 4 isalpha = true, so skip first iff textarray[ 0 ][ 0 ][ 4 ] = ch k = 5

When in doubt, walk through it on paper.

Quzah.

Yeah i've been trying. Will give it another go around :) you cleared a few things up! thanks quzah!

Code:

void get_words(){ int ch, i = 0, j = 0, k = 0, q = 0; while ((ch = getc(fp)) != EOF) { if(isspace(ch) || iscntrl(ch) || isdigit(ch) || ispunct(ch)){ q = 1; } if(isalpha(ch)){ if(q == 1) { j++; k=0;} ch = tolower(ch); textArray[i][j][k] = ch; k++; q = 0; } } }

You still aren't doing anything with most of your array's cells, which leads me to question why you are using a 3D array for this. If your file has "Hello World!" in it, and nothing else, how do you expect your 3D array to look at the end?

Quzah.

The way he broke it down was as follows:

[i] = lines
[j] = words
[k] = characters

but for the project i don't see the necessity in that. Do you?

We are just storing the text into arrays, sorting through it doing different things, like finding unique words => put those into a different array, and calculate the mean, min, max of the words from the text file.

[ line 0 ][ word 0 ][ characters of "Hello" ]
[ line 0 ][ word 1 ][ characters of "World" ]

So what about spaces, and punctuation, where do those go? Where would the ' ' and the '!' go?

Quzah.

We are simply doing a word counter, ignoring all punctuation and digits

Quote:

Originally Posted by Oonej

[i] = lines
[j] = words
[k] = characters

but for the project i don't see the necessity in that. Do you?

Ahhh! I must admit I've been looking at your code for about 10 minutes trying to figure out what on earth you want a 3 dimensional array for.

I don't see the benefit. It sounds like you just want an array of words (so a 2D array). Then you can just loop through them figuring stuff out. Much easier to understand, too.

The fact that you've been asked to use a 3d array implies that next you might be asked to figure out the same stats per-line. So then you have a set of arrays of words... meh. It's a bit wasteful.

Anyway, to your code:
What is q meant to represent?

I think you're trying to do too much with too little code.
You have 3 dimensions: line, word, character.
A newline means increment i.
A new word (a space) means increment j
Punctuation means continue on without doing anything
Alpha characters need to be stored, and k incremented.

So I think you need at least that many "if" statements in your loop.

You should also mark the end of each word with a '\0' terminator if you want to use any string functions or printing functions. Otherwise the functions will access the garbage beyond the end of the word.
And I guess you'll need to mark the end of line somehow too, maybe just a 0 value.

Will you be able to do all the statistics at once? If yes, then there isn't really a reason to use 3D, since you won't need to remember all this:

Quote:

[ line 0 ][ word 0 ][ characters of "Hello" ]
[ line 0 ][ word 1 ][ characters of "World" ]

That would only be helpful if for some reason the user got to control when certain statistics where calculated.

We wont be doing anything past this little bit dealing with text i/o from files, we already have our last project which is dealing with a whole different ball park.

He just had a multi-dimensional array in his "guide" to help students work through the project, but as you said, a third array seems meaningless and i agree with adding the \0 .. forgot to put that in :)

Definitely drop the third dimension then. I had a bit of a play with this and knocked together a working program with a 3d array, but I found the third dimension quite awkward. I made some mistakes that lead to NULL words being in the array, as you described earlier. So I bet you were just incrementing the word number (j) when you shouldn't have been, or not resetting k.

I imagine it'll be easier with 2 dimensions. Good luck :)

Yeah I dropped the third dimensionally , made it SOO much easier to work with... but ran into a snag :(

trying to find unique words but my logic is flawed again, any tips?

textArray carries all the words in the text file and i need to search through text array and put the words into uArray, if uarray already has the word, then move on :(

any help would be great. below is my code:

Code:

int i, j, count = 1; strcpy(uArray[0], textArray[0]); for(i = 0; i < 500; i++){ for(j = 0; j < count ; j++){ if(strcmp(uArray[j], textArray[i]) != 0){ strcpy(uArray[count], textArray[i]); count++; break; } } }

If you sort the words, any duplicates will be adjacent to each other. So finding unique word(s) is a matter of stepping through the list, comparing each word with every other word. You should end up with a list of unique words.

Sorting is a good idea! I think for a data set of this size you probably won't lose or gain a lot by doing it one way over the other, but for llarger amounts of input data you probably should be thinking of alternatives to re examining all the elements in nArray.

So... what you're trying to do should work, and would have been my first thought too.

Code:

int i, j, count = 1; strcpy(uArray[0], textArray[0]);

So far so good.

Code:

for(i = 0; i < 500; i++){

Where did this 500 come from? I remember your arrays being quite a lot smaller. Style suggestion: #define MAX_WORD_LENGTH and maybe MAX_WORDS to avoid 'magic' constants appearing in your code.

Code:

for(j = 0; j < count ; j++){ if(strcmp(uArray[j], textArray[i]) != 0){ strcpy(uArray[count], textArray[i]); count++; break; } }

This is where your logical error is. You're definitely on the right track - The idea is right but the implementation is not quite there. For example if you had

"cat dog cat" in textArray
cat goes into uArray
dog considered ... dog != cat so dog added to uArray
cat considered. Cat == cat for uArray[0]. But then j is incrememnted and the compare is tried again: Is uArray[1] - "dog" a duplicate of cat" -- no, add it again. Whoops. Got too many cats.

You've got a "break" in there for when the word is added. That's good, but also, finding a duplicate is also time for "break" -- there's nothing else we can do with the word. I suggest something like this

Code:

bool seen = false; for(j = 0; j < count ; j++) { seen = false; // reset "seen" for new word // if strings equal... if(strcmp(uArray[j], textArray[i]) == 0) { seen = true; // break - we don't want to add this word to uarray break; } // this is outside of the 'j' loop so can only be executed once for each word now if (!seen) { strcpy(uArray[count], textArray[i]); count++; } }

First glance I noticed a problem with the code, it doesn't increment i at all..

I think I get it. Is this the final code you are suggesting?

Code:

void get_Uwords(){ int i, j, count = 1; strcpy(uArray[0], textArray[0]); for(i = 0; i < 500; i++){ bool seen = false; for(j = 0; j < count ; j++) { seen = false; // reset "seen" for new word // if strings equal... if(strcmp(uArray[j], textArray[i]) == 0) { seen = true; // break - we don't want to add this word to uarray break; } // this is outside of the 'j' loop so can only be executed once for each word now if (!seen) { strcpy(uArray[count], textArray[i]); count++; } } } }

if thats the code, my program is crashing with this code implemented :(

not sure why

Oh dear. No, that was meant more to just be a suggestion as to how you could do it. I haven't put any effort into making sure it's safe....

It might be crashing because of the < 500 loop condition, obviously that should be adjusted appropriately if your array is smaller

I did make that change, i made a counter when storing the words to be used in this function.. so if it only stores 137 words it will only loop 137 times in that function. So it works well that way.

Its actually hanging up on the 6th word or something, thats why its crashing.

by hanging up i mean, it doesn't change words and keeps checking the same words over and over again

You should initialize count to 0, since at first go, there are 0 words in uArray (remember, 0 is the first element of an array in C). Also:

Code:

// this is outside of the 'j' loop so can only be executed once for each word now

That statement is not true. That code is inside the 'j' loop (look at your braces and indentation). It should be moved outside.

Lastly, how are you passing any arguments to get_Uwords? Can we get a full code listing to see how you call this function? Please don't tell me you're using globals for this. Read why you shouldn't here: Global Variables Are Bad.

i'm not using global variables, i'm simply taking the function that gets the words and sending back an int value from that function to a variable in the main function. From there i'm sending the size to the other functions.

also sending the arrays as pointers to the functions.

as for the count to 0, remember we are intializing the first array element uArray, copied from textArray.

so 1 is fine. since we know that is already in the array.

Quote:

Originally Posted by Oonej

i'm not using global variables, i'm simply taking the function that gets the words and sending back an int value from that function to a variable in the main function. From there i'm sending the size to the other functions.

also sending the arrays as pointers to the functions.

Glad you're not using globals. But I don't see you actually sending the size or uArray or textArray to get_Uwords, and those variables are not declared locally in that function either, so that function shouldn't even compile as shown. Please show me the full code that you're using and that causes errors on the 6th word. It's quite possible an error elsewhere is manifesting itself in this function (are you sure the program locks up in this function? How do you know?). Also, you didn't mention whether you actually moved that if (!seen) block outside of the j loop where it belongs.

Quote:

Originally Posted by Oonej

as for the count to 0, remember we are intializing the first array element uArray, copied from textArray.

so 1 is fine. since we know that is already in the array.

Ahh, yes, I missed that. Sorry.

I can't post my full code, since i can't delete it and had people from my class taking my work on the last several projects.. not looking to get into trouble for posting full code, but i can post this segment.

Code:

int main(void){ int total_words = 1; char textArray[400][21]; char uArray[400][21]; //missing code here.. not relevant to the part im working on. total_words = get_words(textArray); get_Uwords(total_words, textArray, uArray); // more missing code }

Code:

void get_Uwords(int total_words, char (*textArray)[21], char (*uArray)[21]){ int i, j, count = 1, g = 0; printf("%d", total_words); strcpy(uArray[0], textArray[0]); for(i = 0; i < total_words; i++){ g = 0; for(j = 0; j < count ; j++) { g = 0; // reset "seen" for new word // if strings equal... if(strcmp(uArray[j], textArray[i]) == 0) { g = 1; // break - we don't want to add this word to uarray break; } // this is outside of the 'j' loop so can only be executed once for each word now if (g == 0) { strcpy(uArray[count], textArray[i]); count++; } } } }

I didn't want to do bool's so I simply made a similar type int validator.

I hacked together my own get_words and your program crashed on larger input files (more than 50 or 100words or so). get_Uwords is not descriptive, since U could be uppercase, or unique or any other U word. Try calling it get_unique_words. Never sacrifice clarity in the name of saving a few keystrokes. It looks like you're trying to only get unique words from textArray, but you still have a problem. For the third and final time, move the following code outside your 'j' loop:

Code:

// this is outside of the 'j' loop so can only be executed once for each word now if (g == 0) { strcpy(uArray[count], textArray[i]); count++; }

That should fix your crash.

Quote:

Originally Posted by smokeyangel

Sorting is a good idea! I think for a data set of this size you probably won't lose or gain a lot by doing it one way over the other, but for llarger amounts of input data you probably should be thinking of alternatives to re examining all the elements in nArray.

I think the ideas are very different. My advice was:

Quote:

Originally Posted by whiteflags

If you sort the words, any duplicates will be adjacent to each other. So finding unique word(s) is a matter of stepping through the list, comparing each word with every other word. You should end up with a list of unique words.

Because he said:

Quote:

Originally Posted by Oonej

trying to find unique words but my logic is flawed again, any tips?

Now I took that to mean that out of a list like, ("cat", "planet", "factory", "factory", "factory") the list of unique words should be ("cat", "planet"). But you could certainly make a different list of unique words by removing repeated words, and that might be what is meant. But if you compare every word with every other word, you can find the list I just gave. It's simple logic -- if the first occurrence of the word is the only occurrence, then it is unique. (Sorting is intended to be a bit of a speedup, but I have to admit that it could end up being more trouble than it is worth.)

ETA: cat planet