why ? reversing postion results in different result

source code1:

Code:

---

    char qseq[] = {'G'};
    char sseq[] = {'A','C','G','T','G','T','A','T'};
    printf("%s\n",qseq);

---

result:G?"
source code2:

Code:

---

    char sseq[] = {'A','C','G','T','G','T','A','T'};
    char qseq[] = {'G'};
    printf("%s\n",qseq);

---

result:
GACGTGTAT?"

>> why ? reversing postion results in different result

printf traverses the string until it encounters a null terminator. In this case, you haven't supplied any, so it just keeps going until it happens to encounter one on the stack (or crashes). You could have used:

Code:

---

char qseq[] = {'G', 0};
// or perhaps
char qseq[] = "G";

---

Note that in the second example the null terminator is implied.

Quote:

---

Originally Posted by Sebastiani

>> why ? reversing postion results in different result

printf traverses the string until it encounters a null terminator. In this case, you haven't supplied any, so it just keeps going until it happens to encounter one on the stack (or crashes). You could have used:

Code:

---

char qseq[] = {'G', 0};
// or perhaps
char qseq[] = "G";

---

Note that in the second example the null terminator is implied.

---

Thanks.

On the 64bit server the array maxium length? Is the length of the stack?

maxium length?

Quote:

---

Originally Posted by zcrself

Thanks.

On the 64bit server the array maxium length? Is the length of the stack?

maxium length?

---

There is no definite maximum size for an automatic (stack) variable, but it does depend on the stack size. Since the stack is used for both local variables and to keep track of function calls, it's not a good idea to use up all your stack space for variable storage by declaring very large automatic variables. If you are using a data structure (array, struct) of significant size, allocate it dynamically with malloc().

Quote:

---

Originally Posted by brewbuck

There is no definite maximum size for an automatic (stack) variable, but it does depend on the stack size. Since the stack is used for both local variables and to keep track of function calls, it's not a good idea to use up all your stack space for variable storage by declaring very large automatic variables. If you are using a data structure (array, struct) of significant size, allocate it dynamically with malloc().

---

Thanks your reply!

My server has 64G memory,and now I have a 50G file,now I want 50G putting into memory!
which data struct is best?
How to read into memory?

Quote:

---

Originally Posted by zcrself

Thanks your reply!

My server has 64G memory,and now I have a 50G file,now I want 50G putting into memory!
which data struct is best?
How to read into memory?

---

Loading a file that size into memory is almost certainly not the right thing to do. Unless you are going to be doing processing that demands that all the file data be immediately available, it seems like an almost crazy waste of resources.

Even if you could read 100 MB/s, it would take about 8.5 minutes to load 50 GB into memory.

EDIT: Upon glancing at your code again, are you doing some kind of bioinformatics processing? Then it MIGHT be justifiable to have the whole thing in RAM, but even then maybe not. I don't know a whole lot about bioinformatics algorithms.

Quote:

---

Originally Posted by brewbuck

Loading a file that size into memory is almost certainly not the right thing to do. Unless you are going to be doing processing that demands that all the file data be immediately available, it seems like an almost crazy waste of resources.

Even if you could read 100 MB/s, it would take about 8.5 minutes to load 50 GB into memory.

EDIT: Upon glancing at your code again, are you doing some kind of bioinformatics processing? Then it MIGHT be justifiable to have the whole thing in RAM, but even then maybe not. I don't know a whole lot about bioinformatics algorithms.

---

Yes I am doing bioinformatics processing!
Now there are two files:
file1:
cigar::50 HWI-EAS-249_35:6:1:6:1154#0/2 76 32 - chromosome06 8365857 8365901 +
....
....
....

file2:
@HWI-EAS-249_35:6:1:6:1154#0/2
GGGGGGCTGAGAAGGTTGAGACAAGTAAGGTATTTCTACGTGATACTAGT GTTATTTCTCCTTACTCGCTCCTTCT
+
BBBBBC@CC@B?@C@;A>:<58@><4;>>@*8@AB;B;6>7>3=66?315 <8@8@@?/>47?;88@%%%%%%%%%%
....
....
....
....

these two files have more than 3,000,000 lines ,I want to produce the structure(as follows)

HWI-EAS-249_35:6:1:6:1154#0/2 76 32 - chromosome06 8365857 8365901 + GGGGGGCTGAGAAGGTTGAGACAAGTAAGGTATTTCTACGTGATACTAGT GTTATTTCTCCTTACTCGCTCCTTCT
.....
......
.....

according to keyword search the sequences,

As I think , I want to create hash table to store the data so that giving me a keyword I can find its sequences quickly!

Have best method to complete this job?

the keyword is the second column every line of the file1

Quote:

---

Originally Posted by brewbuck

Upon glancing at your code again, are you doing some kind of bioinformatics processing? Then it MIGHT be justifiable to have the whole thing in RAM, but even then maybe not. I don't know a whole lot about bioinformatics algorithms.

---

If they're DNA sequences it might even be useful to store them in a more compressed structure - since there are only four possible values, you really only need two bits to encode each sequence. Percentage-wise, that may not sound like a lot of savings, but when you're talking about a 50GB file that works out to like 37.5GB less storage!