Segmentation fault when traversing object of type SuffixTreeY

[skuruppu]

I was attempting to construct a compressed suffix tree for this DNA sequence.

The suffix array is constructed successfully (at least I think so) but as soon as you invoke any methods that requires traversing the tree, a segmentation fault occurs. For example, invoking the following method call

cst->Child(node_vl, node_vr, 'c', &next_vl, &next_vr);

where node_vl and node_vr are obtained from cst->Root(), causes a segmentation fault.

The following methods were invoked to construct the suffix tree.

SuffixTree *cst;
SuffixTreeY csty(text, length, DAC, CN_NPR, 32);
cst = &csty;

The stack trace from gdb is as follows: #0 0x00000000004086fa in cds_utils::get_field_64 (A=0xfe2820, len=6, index=65836) at ../includes/libcdsTrees.h:26 #1 0x0000000000411354 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11309, r=1, min_r=0x7fffa70200e0)

at static/suffixtree/NPR_CN.cpp:413

#2 0x00000000004112c9 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=361925, r=0, min_r=0x7fffa7020198)

at static/suffixtree/NPR_CN.cpp:408

#3 0x0000000000411793 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11581659, csa=0xfe2a60, lcp=0xfc76b0)

at static/suffixtree/NPR_CN.cpp:348

#4 0x0000000000402c39 in cds_static::SuffixTreeY::SDepth (this=0xfc7680, vl=0, vr=11581659) at static/suffixtree/SuffixTreeY.cpp:91 #5 0x0000000000404eba in cds_static::SuffixTreeY::Child (this=0xfc7680, vl=0, vr=11581659, a=99 'c', child_l=0x7fffa7020308,

child_r=0x7fffa7020300) at static/suffixtree/SuffixTreeY.cpp:279

#6 0x00000000004022e9 in testSuffixTree (s1=0xfc7680) at testSuffixTree.cpp:52 #7 0x00000000004024fc in main (argc=2, argv=0x7fffa7020698) at testSuffixTree.cpp:92

Hope this is enough information to fix the problem.

[rcanovas]

Hi I don't know what could it be, but I will try to fixed as soon as I can. Anyway if you want to discuss about the implementation, most of the time I'm in room 4.31a in the 4th floor of the ICT building.

Best, Rodrigo

On Mon, Aug 1, 2011 at 9:13 PM, skuruppu < [email protected]>wrote:

I was attempting to construct a compressed suffix tree for this DNA sequence.

The suffix array is constructed successfully (at least I think so) but as soon as you invoke any methods that requires traversing the tree, a segmentation fault occurs. For example, invoking the following method call

cst->Child(node_vl, node_vr, 'c', &next_vl, &next_vr);

where node_vl and node_vr are obtained from cst->Root(), causes a segmentation fault.

The following methods were invoked to construct the suffix tree.

SuffixTree *cst; SuffixTreeY csty(text, length, DAC, CN_NPR, 32); cst = &csty;

The stack trace from gdb is as follows:

#0 0x00000000004086fa in cds_utils::get_field_64 (A=0xfe2820, len=6, index=65836) at ../includes/libcdsTrees.h:26 #1 0x0000000000411354 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11309, r=1, min_r=0x7fffa70200e0) at static/suffixtree/NPR_CN.cpp:413 #2 0x00000000004112c9 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=361925, r=0, min_r=0x7fffa7020198) at static/suffixtree/NPR_CN.cpp:408 #3 0x0000000000411793 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11581659, csa=0xfe2a60, lcp=0xfc76b0) at static/suffixtree/NPR_CN.cpp:348 #4 0x0000000000402c39 in cds_static::SuffixTreeY::SDepth (this=0xfc7680, vl=0, vr=11581659) at static/suffixtree/SuffixTreeY.cpp:91 #5 0x0000000000404eba in cds_static::SuffixTreeY::Child (this=0xfc7680, vl=0, vr=11581659, a=99 'c', child_l=0x7fffa7020308, child_r=0x7fffa7020300) at static/suffixtree/SuffixTreeY.cpp:279 #6 0x00000000004022e9 in testSuffixTree (s1=0xfc7680) at testSuffixTree.cpp:52 #7 0x00000000004024fc in main (argc=2, argv=0x7fffa7020698) at testSuffixTree.cpp:92

Hope this is enough information to fix the problem.

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1

[rcanovas]

Another thing, Where I can get the data set that you use?

On Tue, Aug 2, 2011 at 8:52 AM, Yigo [email protected] wrote:

Hi I don't know what could it be, but I will try to fixed as soon as I can. Anyway if you want to discuss about the implementation, most of the time I'm in room 4.31a in the 4th floor of the ICT building.

Best, Rodrigo

On Mon, Aug 1, 2011 at 9:13 PM, skuruppu < [email protected]>wrote:

I was attempting to construct a compressed suffix tree for this DNA sequence.

The suffix array is constructed successfully (at least I think so) but as soon as you invoke any methods that requires traversing the tree, a segmentation fault occurs. For example, invoking the following method call

cst->Child(node_vl, node_vr, 'c', &next_vl, &next_vr);

where node_vl and node_vr are obtained from cst->Root(), causes a segmentation fault.

The following methods were invoked to construct the suffix tree.

SuffixTree *cst; SuffixTreeY csty(text, length, DAC, CN_NPR, 32); cst = &csty;

The stack trace from gdb is as follows:

#0 0x00000000004086fa in cds_utils::get_field_64 (A=0xfe2820, len=6, index=65836) at ../includes/libcdsTrees.h:26 #1 0x0000000000411354 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11309, r=1, min_r=0x7fffa70200e0) at static/suffixtree/NPR_CN.cpp:413 #2 0x00000000004112c9 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=361925, r=0, min_r=0x7fffa7020198) at static/suffixtree/NPR_CN.cpp:408 #3 0x0000000000411793 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11581659, csa=0xfe2a60, lcp=0xfc76b0) at static/suffixtree/NPR_CN.cpp:348 #4 0x0000000000402c39 in cds_static::SuffixTreeY::SDepth (this=0xfc7680, vl=0, vr=11581659) at static/suffixtree/SuffixTreeY.cpp:91 #5 0x0000000000404eba in cds_static::SuffixTreeY::Child (this=0xfc7680, vl=0, vr=11581659, a=99 'c', child_l=0x7fffa7020308, child_r=0x7fffa7020300) at static/suffixtree/SuffixTreeY.cpp:279 #6 0x00000000004022e9 in testSuffixTree (s1=0xfc7680) at testSuffixTree.cpp:52 #7 0x00000000004024fc in main (argc=2, argv=0x7fffa7020698) at testSuffixTree.cpp:92

Hope this is enough information to fix the problem.

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1

[skuruppu]

Hey Rodrigo,

Sorry I didn't realise it was you. I didn't know your surname. Welcome to Melbourne. Alex told me you were coming.

I don't know much about your suffix tree implementation but I'll try to understand it. I wanted to see if I can use it instead of a suffix array in my RLZ implementation because I need an efficient way to calculate suffix links.

The sequence I used is at http://www.csse.unimelb.edu.au/~kuruppu/testdata/REF.raw.gz

I hope I've given the right permissions for you to access it.

Would be good to see you in any case so let me know if you want to get a coffee sometime.

Enjoy the sunshine :)

Shanika

2011/8/2 rcanovas < [email protected]>

Another thing, Where I can get the data set that you use?

On Tue, Aug 2, 2011 at 8:52 AM, Yigo [email protected] wrote:

Hi I don't know what could it be, but I will try to fixed as soon as I can. Anyway if you want to discuss about the implementation, most of the time I'm in room 4.31a in the 4th floor of the ICT building.

Best, Rodrigo

On Mon, Aug 1, 2011 at 9:13 PM, skuruppu < [email protected]>wrote:

I was attempting to construct a compressed suffix tree for this DNA sequence.

The suffix array is constructed successfully (at least I think so) but as soon as you invoke any methods that requires traversing the tree, a segmentation fault occurs. For example, invoking the following method call

cst->Child(node_vl, node_vr, 'c', &next_vl, &next_vr);

where node_vl and node_vr are obtained from cst->Root(), causes a segmentation fault.

The following methods were invoked to construct the suffix tree.

SuffixTree *cst; SuffixTreeY csty(text, length, DAC, CN_NPR, 32); cst = &csty;

The stack trace from gdb is as follows:

#0 0x00000000004086fa in cds_utils::get_field_64 (A=0xfe2820, len=6, index=65836) at ../includes/libcdsTrees.h:26 #1 0x0000000000411354 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11309, r=1, min_r=0x7fffa70200e0) at static/suffixtree/NPR_CN.cpp:413 #2 0x00000000004112c9 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=361925, r=0, min_r=0x7fffa7020198) at static/suffixtree/NPR_CN.cpp:408 #3 0x0000000000411793 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11581659, csa=0xfe2a60, lcp=0xfc76b0) at static/suffixtree/NPR_CN.cpp:348 #4 0x0000000000402c39 in cds_static::SuffixTreeY::SDepth (this=0xfc7680, vl=0, vr=11581659) at static/suffixtree/SuffixTreeY.cpp:91 #5 0x0000000000404eba in cds_static::SuffixTreeY::Child (this=0xfc7680, vl=0, vr=11581659, a=99 'c', child_l=0x7fffa7020308, child_r=0x7fffa7020300) at static/suffixtree/SuffixTreeY.cpp:279 #6 0x00000000004022e9 in testSuffixTree (s1=0xfc7680) at testSuffixTree.cpp:52 #7 0x00000000004024fc in main (argc=2, argv=0x7fffa7020698) at testSuffixTree.cpp:92

Hope this is enough information to fix the problem.

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705138

[rcanovas]

Every Wednesday I have time (I'm not completely sure if this Wednesday) if you want to meet and discuss or if u want me to explain the CST to u.

On Tue, Aug 2, 2011 at 9:59 AM, skuruppu < [email protected]>wrote:

Hey Rodrigo,

Sorry I didn't realise it was you. I didn't know your surname. Welcome to Melbourne. Alex told me you were coming.

I don't know much about your suffix tree implementation but I'll try to understand it. I wanted to see if I can use it instead of a suffix array in my RLZ implementation because I need an efficient way to calculate suffix links.

The sequence I used is at http://www.csse.unimelb.edu.au/~kuruppu/testdata/REF.raw.gz

I hope I've given the right permissions for you to access it.

Would be good to see you in any case so let me know if you want to get a coffee sometime.

Enjoy the sunshine :)

Shanika

2011/8/2 rcanovas < [email protected]>

Another thing, Where I can get the data set that you use?

On Tue, Aug 2, 2011 at 8:52 AM, Yigo [email protected] wrote:

Hi I don't know what could it be, but I will try to fixed as soon as I can. Anyway if you want to discuss about the implementation, most of the time I'm in room 4.31a in the 4th floor of the ICT building.

Best, Rodrigo

On Mon, Aug 1, 2011 at 9:13 PM, skuruppu < [email protected]>wrote:

I was attempting to construct a compressed suffix tree for this DNA sequence.

The suffix array is constructed successfully (at least I think so) but as soon as you invoke any methods that requires traversing the tree, a segmentation fault occurs. For example, invoking the following method call

cst->Child(node_vl, node_vr, 'c', &next_vl, &next_vr);

where node_vl and node_vr are obtained from cst->Root(), causes a segmentation fault.

The following methods were invoked to construct the suffix tree.

SuffixTree *cst; SuffixTreeY csty(text, length, DAC, CN_NPR, 32); cst = &csty;

The stack trace from gdb is as follows:

#0 0x00000000004086fa in cds_utils::get_field_64 (A=0xfe2820, len=6, index=65836) at ../includes/libcdsTrees.h:26 #1 0x0000000000411354 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11309, r=1, min_r=0x7fffa70200e0) at static/suffixtree/NPR_CN.cpp:413 #2 0x00000000004112c9 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=361925, r=0, min_r=0x7fffa7020198) at static/suffixtree/NPR_CN.cpp:408 #3 0x0000000000411793 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11581659, csa=0xfe2a60, lcp=0xfc76b0) at static/suffixtree/NPR_CN.cpp:348 #4 0x0000000000402c39 in cds_static::SuffixTreeY::SDepth (this=0xfc7680, vl=0, vr=11581659) at static/suffixtree/SuffixTreeY.cpp:91 #5 0x0000000000404eba in cds_static::SuffixTreeY::Child (this=0xfc7680, vl=0, vr=11581659, a=99 'c', child_l=0x7fffa7020308, child_r=0x7fffa7020300) at static/suffixtree/SuffixTreeY.cpp:279 #6 0x00000000004022e9 in testSuffixTree (s1=0xfc7680) at testSuffixTree.cpp:52 #7 0x00000000004024fc in main (argc=2, argv=0x7fffa7020698) at testSuffixTree.cpp:92

Hope this is enough information to fix the problem.

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705138

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705399

[skuruppu]

Sure, it would be good to discuss the CST. But first I'd like to read the paper so I'm not too lost. Is this the paper relating to this implementation?

Rodrigo Cnovas and Gonzalo Navarro. Practical Compressed Suffix Trees. Proc. SEA'10

Wed next week would be good if you have time.

2011/8/2 rcanovas < [email protected]>

Every Wednesday I have time (I'm not completely sure if this Wednesday) if you want to meet and discuss or if u want me to explain the CST to u.

On Tue, Aug 2, 2011 at 9:59 AM, skuruppu < [email protected]>wrote:

Hey Rodrigo,

Sorry I didn't realise it was you. I didn't know your surname. Welcome to Melbourne. Alex told me you were coming.

I don't know much about your suffix tree implementation but I'll try to understand it. I wanted to see if I can use it instead of a suffix array in my RLZ implementation because I need an efficient way to calculate suffix links.

The sequence I used is at http://www.csse.unimelb.edu.au/~kuruppu/testdata/REF.raw.gz

I hope I've given the right permissions for you to access it.

Would be good to see you in any case so let me know if you want to get a coffee sometime.

Enjoy the sunshine :)

Shanika

2011/8/2 rcanovas < [email protected]>

Another thing, Where I can get the data set that you use?

On Tue, Aug 2, 2011 at 8:52 AM, Yigo [email protected] wrote:

Hi I don't know what could it be, but I will try to fixed as soon as I can. Anyway if you want to discuss about the implementation, most of the time I'm in room 4.31a in the 4th floor of the ICT building.

Best, Rodrigo

On Mon, Aug 1, 2011 at 9:13 PM, skuruppu < [email protected]>wrote:

I was attempting to construct a compressed suffix tree for this DNA sequence.

The suffix array is constructed successfully (at least I think so) but as soon as you invoke any methods that requires traversing the tree, a segmentation fault occurs. For example, invoking the following method call

cst->Child(node_vl, node_vr, 'c', &next_vl, &next_vr);

where node_vl and node_vr are obtained from cst->Root(), causes a segmentation fault.

The following methods were invoked to construct the suffix tree.

SuffixTree *cst; SuffixTreeY csty(text, length, DAC, CN_NPR, 32); cst = &csty;

The stack trace from gdb is as follows:

#0 0x00000000004086fa in cds_utils::get_field_64 (A=0xfe2820, len=6, index=65836) at ../includes/libcdsTrees.h:26 #1 0x0000000000411354 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11309, r=1, min_r=0x7fffa70200e0) at static/suffixtree/NPR_CN.cpp:413 #2 0x00000000004112c9 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=361925, r=0, min_r=0x7fffa7020198) at static/suffixtree/NPR_CN.cpp:408 #3 0x0000000000411793 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11581659, csa=0xfe2a60, lcp=0xfc76b0) at static/suffixtree/NPR_CN.cpp:348 #4 0x0000000000402c39 in cds_static::SuffixTreeY::SDepth (this=0xfc7680, vl=0, vr=11581659) at static/suffixtree/SuffixTreeY.cpp:91 #5 0x0000000000404eba in cds_static::SuffixTreeY::Child (this=0xfc7680, vl=0, vr=11581659, a=99 'c', child_l=0x7fffa7020308, child_r=0x7fffa7020300) at static/suffixtree/SuffixTreeY.cpp:279 #6 0x00000000004022e9 in testSuffixTree (s1=0xfc7680) at testSuffixTree.cpp:52 #7 0x00000000004024fc in main (argc=2, argv=0x7fffa7020698) at testSuffixTree.cpp:92

Hope this is enough information to fix the problem.

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705138

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705399

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705478

[rcanovas]

Yes is suppose to be related. Just in case I attach the presentation that I did at SEA... and I also attach my thesis if you want to read something more specific. Any way I have no problem if you prefer to review the thing starting from zero... It could be easy for you and for me it could be good just to review the topic =)... Just let me know.

Best, Yigo.

On Tue, Aug 2, 2011 at 10:30 AM, skuruppu < [email protected]>wrote:

Sure, it would be good to discuss the CST. But first I'd like to read the paper so I'm not too lost. Is this the paper relating to this implementation?

Rodrigo Cnovas and Gonzalo Navarro. Practical Compressed Suffix Trees. Proc. SEA'10

Wed next week would be good if you have time.

2011/8/2 rcanovas < [email protected]>

Every Wednesday I have time (I'm not completely sure if this Wednesday) if you want to meet and discuss or if u want me to explain the CST to u.

On Tue, Aug 2, 2011 at 9:59 AM, skuruppu < [email protected]>wrote:

Hey Rodrigo,

Sorry I didn't realise it was you. I didn't know your surname. Welcome to Melbourne. Alex told me you were coming.

I don't know much about your suffix tree implementation but I'll try to understand it. I wanted to see if I can use it instead of a suffix array in my RLZ implementation because I need an efficient way to calculate suffix links.

The sequence I used is at http://www.csse.unimelb.edu.au/~kuruppu/testdata/REF.raw.gz

I hope I've given the right permissions for you to access it.

Would be good to see you in any case so let me know if you want to get a coffee sometime.

Enjoy the sunshine :)

Shanika

2011/8/2 rcanovas < [email protected]>

Another thing, Where I can get the data set that you use?

On Tue, Aug 2, 2011 at 8:52 AM, Yigo [email protected] wrote:

Hi I don't know what could it be, but I will try to fixed as soon as I can. Anyway if you want to discuss about the implementation, most of the time I'm in room 4.31a in the 4th floor of the ICT building.

Best, Rodrigo

On Mon, Aug 1, 2011 at 9:13 PM, skuruppu < [email protected]>wrote:

I was attempting to construct a compressed suffix tree for this DNA sequence.

The suffix array is constructed successfully (at least I think so) but as soon as you invoke any methods that requires traversing the tree, a segmentation fault occurs. For example, invoking the following method call

cst->Child(node_vl, node_vr, 'c', &next_vl, &next_vr);

where node_vl and node_vr are obtained from cst->Root(), causes a segmentation fault.

The following methods were invoked to construct the suffix tree.

SuffixTree *cst; SuffixTreeY csty(text, length, DAC, CN_NPR, 32); cst = &csty;

The stack trace from gdb is as follows:

#0 0x00000000004086fa in cds_utils::get_field_64 (A=0xfe2820, len=6, index=65836) at ../includes/libcdsTrees.h:26 #1 0x0000000000411354 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11309, r=1, min_r=0x7fffa70200e0) at static/suffixtree/NPR_CN.cpp:413 #2 0x00000000004112c9 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=361925, r=0, min_r=0x7fffa7020198) at static/suffixtree/NPR_CN.cpp:408 #3 0x0000000000411793 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11581659, csa=0xfe2a60, lcp=0xfc76b0) at static/suffixtree/NPR_CN.cpp:348 #4 0x0000000000402c39 in cds_static::SuffixTreeY::SDepth (this=0xfc7680, vl=0, vr=11581659) at static/suffixtree/SuffixTreeY.cpp:91 #5 0x0000000000404eba in cds_static::SuffixTreeY::Child (this=0xfc7680, vl=0, vr=11581659, a=99 'c', child_l=0x7fffa7020308, child_r=0x7fffa7020300) at static/suffixtree/SuffixTreeY.cpp:279 #6 0x00000000004022e9 in testSuffixTree (s1=0xfc7680) at testSuffixTree.cpp:52 #7 0x00000000004024fc in main (argc=2, argv=0x7fffa7020698) at testSuffixTree.cpp:92

Hope this is enough information to fix the problem.

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705138

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705399

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705478

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705537

[rcanovas]

Are u sure that you have the last libcds version?? because I just try your problem and its work well

On Tue, Aug 2, 2011 at 10:37 AM, Yigo [email protected] wrote:

Yes is suppose to be related. Just in case I attach the presentation that I did at SEA... and I also attach my thesis if you want to read something more specific. Any way I have no problem if you prefer to review the thing starting from zero... It could be easy for you and for me it could be good just to review the topic =)... Just let me know.

Best, Yigo.

On Tue, Aug 2, 2011 at 10:30 AM, skuruppu < [email protected]>wrote:

Sure, it would be good to discuss the CST. But first I'd like to read the paper so I'm not too lost. Is this the paper relating to this implementation?

Rodrigo Cnovas and Gonzalo Navarro. Practical Compressed Suffix Trees. Proc. SEA'10

Wed next week would be good if you have time.

2011/8/2 rcanovas < [email protected]>

Every Wednesday I have time (I'm not completely sure if this Wednesday) if you want to meet and discuss or if u want me to explain the CST to u.

On Tue, Aug 2, 2011 at 9:59 AM, skuruppu < [email protected]>wrote:

Hey Rodrigo,

Sorry I didn't realise it was you. I didn't know your surname. Welcome to Melbourne. Alex told me you were coming.

I don't know much about your suffix tree implementation but I'll try to understand it. I wanted to see if I can use it instead of a suffix array in my RLZ implementation because I need an efficient way to calculate suffix links.

The sequence I used is at http://www.csse.unimelb.edu.au/~kuruppu/testdata/REF.raw.gz

I hope I've given the right permissions for you to access it.

Would be good to see you in any case so let me know if you want to get a coffee sometime.

Enjoy the sunshine :)

Shanika

2011/8/2 rcanovas < [email protected]>

Another thing, Where I can get the data set that you use?

On Tue, Aug 2, 2011 at 8:52 AM, Yigo [email protected] wrote:

Hi I don't know what could it be, but I will try to fixed as soon as I can. Anyway if you want to discuss about the implementation, most of the time I'm in room 4.31a in the 4th floor of the ICT building.

Best, Rodrigo

On Mon, Aug 1, 2011 at 9:13 PM, skuruppu < [email protected]>wrote:

I was attempting to construct a compressed suffix tree for this DNA sequence.

The suffix array is constructed successfully (at least I think so) but as soon as you invoke any methods that requires traversing the tree, a segmentation fault occurs. For example, invoking the following method call

cst->Child(node_vl, node_vr, 'c', &next_vl, &next_vr);

where node_vl and node_vr are obtained from cst->Root(), causes a segmentation fault.

The following methods were invoked to construct the suffix tree.

SuffixTree *cst; SuffixTreeY csty(text, length, DAC, CN_NPR, 32); cst = &csty;

The stack trace from gdb is as follows:

#0 0x00000000004086fa in cds_utils::get_field_64 (A=0xfe2820, len=6, index=65836) at ../includes/libcdsTrees.h:26 #1 0x0000000000411354 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11309, r=1, min_r=0x7fffa70200e0) at static/suffixtree/NPR_CN.cpp:413 #2 0x00000000004112c9 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=361925, r=0, min_r=0x7fffa7020198) at static/suffixtree/NPR_CN.cpp:408 #3 0x0000000000411793 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11581659, csa=0xfe2a60, lcp=0xfc76b0) at static/suffixtree/NPR_CN.cpp:348 #4 0x0000000000402c39 in cds_static::SuffixTreeY::SDepth (this=0xfc7680, vl=0, vr=11581659) at static/suffixtree/SuffixTreeY.cpp:91 #5 0x0000000000404eba in cds_static::SuffixTreeY::Child (this=0xfc7680, vl=0, vr=11581659, a=99 'c', child_l=0x7fffa7020308, child_r=0x7fffa7020300) at static/suffixtree/SuffixTreeY.cpp:279 #6 0x00000000004022e9 in testSuffixTree (s1=0xfc7680) at testSuffixTree.cpp:52 #7 0x00000000004024fc in main (argc=2, argv=0x7fffa7020698) at testSuffixTree.cpp:92

Hope this is enough information to fix the problem.

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705138

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705399

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705478

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705537

[skuruppu]

I installed version 1.0.8 from http://libcds.recoded.cl/system/files/libcds-v1.0.8.tgz

This was the only version available from the libcds website.

Are you using it on a 64bit machine? Because I am, maybe that's the cause?

2011/8/2 rcanovas < [email protected]>

Are u sure that you have the last libcds version?? because I just try your problem and its work well

On Tue, Aug 2, 2011 at 10:37 AM, Yigo [email protected] wrote:

Yes is suppose to be related. Just in case I attach the presentation that I did at SEA... and I also attach my thesis if you want to read something more specific. Any way I have no problem if you prefer to review the thing starting from zero... It could be easy for you and for me it could be good just to review the topic =)... Just let me know.

Best, Yigo.

On Tue, Aug 2, 2011 at 10:30 AM, skuruppu < [email protected]>wrote:

Sure, it would be good to discuss the CST. But first I'd like to read the paper so I'm not too lost. Is this the paper relating to this implementation?

Rodrigo Cnovas and Gonzalo Navarro. Practical Compressed Suffix Trees. Proc. SEA'10

Wed next week would be good if you have time.

2011/8/2 rcanovas < [email protected]>

Every Wednesday I have time (I'm not completely sure if this Wednesday) if you want to meet and discuss or if u want me to explain the CST to u.

On Tue, Aug 2, 2011 at 9:59 AM, skuruppu < [email protected]>wrote:

Hey Rodrigo,

Sorry I didn't realise it was you. I didn't know your surname. Welcome to Melbourne. Alex told me you were coming.

I don't know much about your suffix tree implementation but I'll try to understand it. I wanted to see if I can use it instead of a suffix array in my RLZ implementation because I need an efficient way to calculate suffix links.

The sequence I used is at http://www.csse.unimelb.edu.au/~kuruppu/testdata/REF.raw.gz

I hope I've given the right permissions for you to access it.

Would be good to see you in any case so let me know if you want to get a coffee sometime.

Enjoy the sunshine :)

Shanika

2011/8/2 rcanovas < [email protected]>

Another thing, Where I can get the data set that you use?

On Tue, Aug 2, 2011 at 8:52 AM, Yigo [email protected] wrote:

Hi I don't know what could it be, but I will try to fixed as soon as I can. Anyway if you want to discuss about the implementation, most of the time I'm in room 4.31a in the 4th floor of the ICT building.

Best, Rodrigo

On Mon, Aug 1, 2011 at 9:13 PM, skuruppu < [email protected]>wrote:

I was attempting to construct a compressed suffix tree for this DNA sequence.

The suffix array is constructed successfully (at least I think so) but as soon as you invoke any methods that requires traversing the tree, a segmentation fault occurs. For example, invoking the following method call

cst->Child(node_vl, node_vr, 'c', &next_vl, &next_vr);

where node_vl and node_vr are obtained from cst->Root(), causes a segmentation fault.

The following methods were invoked to construct the suffix tree.

SuffixTree *cst; SuffixTreeY csty(text, length, DAC, CN_NPR, 32); cst = &csty;

The stack trace from gdb is as follows:

#0 0x00000000004086fa in cds_utils::get_field_64 (A=0xfe2820, len=6, index=65836) at ../includes/libcdsTrees.h:26 #1 0x0000000000411354 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11309, r=1, min_r=0x7fffa70200e0) at static/suffixtree/NPR_CN.cpp:413 #2 0x00000000004112c9 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=361925, r=0, min_r=0x7fffa7020198) at static/suffixtree/NPR_CN.cpp:408 #3 0x0000000000411793 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11581659, csa=0xfe2a60, lcp=0xfc76b0) at static/suffixtree/NPR_CN.cpp:348 #4 0x0000000000402c39 in cds_static::SuffixTreeY::SDepth (this=0xfc7680, vl=0, vr=11581659) at static/suffixtree/SuffixTreeY.cpp:91 #5 0x0000000000404eba in cds_static::SuffixTreeY::Child (this=0xfc7680, vl=0, vr=11581659, a=99 'c', child_l=0x7fffa7020308, child_r=0x7fffa7020300) at static/suffixtree/SuffixTreeY.cpp:279 #6 0x00000000004022e9 in testSuffixTree (s1=0xfc7680) at testSuffixTree.cpp:52 #7 0x00000000004024fc in main (argc=2, argv=0x7fffa7020698) at testSuffixTree.cpp:92

Hope this is enough information to fix the problem.

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705138

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705399

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705478

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705537

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706154

[skuruppu]

Yeah I just ran the same code on my laptop which is running 32bit Linux and the code works fine. The segmentation fault only happens when I run it on the 64bit Linux machine.

2011/8/2 Shanika Kuruppu [email protected]

I installed version 1.0.8 from http://libcds.recoded.cl/system/files/libcds-v1.0.8.tgz

This was the only version available from the libcds website.

Are you using it on a 64bit machine? Because I am, maybe that's the cause?

2011/8/2 rcanovas < [email protected]>

Are u sure that you have the last libcds version?? because I just try your problem and its work well

On Tue, Aug 2, 2011 at 10:37 AM, Yigo [email protected] wrote:

Yes is suppose to be related. Just in case I attach the presentation that I did at SEA... and I also attach my thesis if you want to read something more specific. Any way I have no problem if you prefer to review the thing starting from zero... It could be easy for you and for me it could be good just to review the topic =)... Just let me know.

Best, Yigo.

On Tue, Aug 2, 2011 at 10:30 AM, skuruppu < [email protected]>wrote:

Sure, it would be good to discuss the CST. But first I'd like to read the paper so I'm not too lost. Is this the paper relating to this implementation?

Rodrigo Cnovas and Gonzalo Navarro. Practical Compressed Suffix Trees. Proc. SEA'10

Wed next week would be good if you have time.

2011/8/2 rcanovas < [email protected]>

Every Wednesday I have time (I'm not completely sure if this Wednesday) if you want to meet and discuss or if u want me to explain the CST to u.

On Tue, Aug 2, 2011 at 9:59 AM, skuruppu < [email protected]>wrote:

Hey Rodrigo,

Sorry I didn't realise it was you. I didn't know your surname. Welcome to Melbourne. Alex told me you were coming.

I don't know much about your suffix tree implementation but I'll try to understand it. I wanted to see if I can use it instead of a suffix array in my RLZ implementation because I need an efficient way to calculate suffix links.

The sequence I used is at http://www.csse.unimelb.edu.au/~kuruppu/testdata/REF.raw.gz

I hope I've given the right permissions for you to access it.

Would be good to see you in any case so let me know if you want to get a coffee sometime.

Enjoy the sunshine :)

Shanika

2011/8/2 rcanovas < [email protected]>

Another thing, Where I can get the data set that you use?

On Tue, Aug 2, 2011 at 8:52 AM, Yigo [email protected] wrote:

Hi I don't know what could it be, but I will try to fixed as soon as I can. Anyway if you want to discuss about the implementation, most of the time I'm in room 4.31a in the 4th floor of the ICT building.

Best, Rodrigo

On Mon, Aug 1, 2011 at 9:13 PM, skuruppu < [email protected]>wrote:

I was attempting to construct a compressed suffix tree for this DNA sequence.

The suffix array is constructed successfully (at least I think so) but as soon as you invoke any methods that requires traversing the tree, a segmentation fault occurs. For example, invoking the following method call

cst->Child(node_vl, node_vr, 'c', &next_vl, &next_vr);

where node_vl and node_vr are obtained from cst->Root(), causes a segmentation fault.

The following methods were invoked to construct the suffix tree.

SuffixTree *cst; SuffixTreeY csty(text, length, DAC, CN_NPR, 32); cst = &csty;

The stack trace from gdb is as follows:

#0 0x00000000004086fa in cds_utils::get_field_64 (A=0xfe2820, len=6, index=65836) at ../includes/libcdsTrees.h:26 #1 0x0000000000411354 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11309, r=1, min_r=0x7fffa70200e0) at static/suffixtree/NPR_CN.cpp:413 #2 0x00000000004112c9 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=361925, r=0, min_r=0x7fffa7020198) at static/suffixtree/NPR_CN.cpp:408 #3 0x0000000000411793 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11581659, csa=0xfe2a60, lcp=0xfc76b0) at static/suffixtree/NPR_CN.cpp:348 #4 0x0000000000402c39 in cds_static::SuffixTreeY::SDepth (this=0xfc7680, vl=0, vr=11581659) at static/suffixtree/SuffixTreeY.cpp:91 #5 0x0000000000404eba in cds_static::SuffixTreeY::Child (this=0xfc7680, vl=0, vr=11581659, a=99 'c', child_l=0x7fffa7020308, child_r=0x7fffa7020300) at static/suffixtree/SuffixTreeY.cpp:279 #6 0x00000000004022e9 in testSuffixTree (s1=0xfc7680) at testSuffixTree.cpp:52 #7 0x00000000004024fc in main (argc=2, argv=0x7fffa7020698) at testSuffixTree.cpp:92

Hope this is enough information to fix the problem.

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705138

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705399

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705478

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705537

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706154

[rcanovas]

Did you try running the experiment in 64-bits replacing the suffixtree files with the ones that I send you?

On Tue, Aug 2, 2011 at 1:04 PM, skuruppu < [email protected]>wrote:

Yeah I just ran the same code on my laptop which is running 32bit Linux and the code works fine. The segmentation fault only happens when I run it on the 64bit Linux machine.

2011/8/2 Shanika Kuruppu [email protected]

I installed version 1.0.8 from http://libcds.recoded.cl/system/files/libcds-v1.0.8.tgz

This was the only version available from the libcds website.

Are you using it on a 64bit machine? Because I am, maybe that's the cause?

2011/8/2 rcanovas < [email protected]>

Are u sure that you have the last libcds version?? because I just try your problem and its work well

On Tue, Aug 2, 2011 at 10:37 AM, Yigo [email protected] wrote:

Yes is suppose to be related. Just in case I attach the presentation that I did at SEA... and I also attach my thesis if you want to read something more specific. Any way I have no problem if you prefer to review the thing starting from zero... It could be easy for you and for me it could be good just to review the topic =)... Just let me know.

Best, Yigo.

On Tue, Aug 2, 2011 at 10:30 AM, skuruppu < [email protected]>wrote:

Sure, it would be good to discuss the CST. But first I'd like to read the paper so I'm not too lost. Is this the paper relating to this implementation?

Rodrigo Cnovas and Gonzalo Navarro. Practical Compressed Suffix Trees. Proc. SEA'10

Wed next week would be good if you have time.

2011/8/2 rcanovas < [email protected]>

Every Wednesday I have time (I'm not completely sure if this Wednesday) if you want to meet and discuss or if u want me to explain the CST to u.

On Tue, Aug 2, 2011 at 9:59 AM, skuruppu < [email protected]>wrote:

Hey Rodrigo,

Sorry I didn't realise it was you. I didn't know your surname. Welcome to Melbourne. Alex told me you were coming.

I don't know much about your suffix tree implementation but I'll try to understand it. I wanted to see if I can use it instead of a suffix array in my RLZ implementation because I need an efficient way to calculate suffix links.

The sequence I used is at http://www.csse.unimelb.edu.au/~kuruppu/testdata/REF.raw.gz

I hope I've given the right permissions for you to access it.

Would be good to see you in any case so let me know if you want to get a coffee sometime.

Enjoy the sunshine :)

Shanika

2011/8/2 rcanovas < [email protected]>

Another thing, Where I can get the data set that you use?

On Tue, Aug 2, 2011 at 8:52 AM, Yigo [email protected] wrote:

Hi I don't know what could it be, but I will try to fixed as soon as I can. Anyway if you want to discuss about the implementation, most of the time I'm in room 4.31a in the 4th floor of the ICT building.

Best, Rodrigo

On Mon, Aug 1, 2011 at 9:13 PM, skuruppu < [email protected]>wrote:

I was attempting to construct a compressed suffix tree for this DNA sequence.

The suffix array is constructed successfully (at least I think so) but as soon as you invoke any methods that requires traversing the tree, a segmentation fault occurs. For example, invoking the following method call

cst->Child(node_vl, node_vr, 'c', &next_vl, &next_vr);

where node_vl and node_vr are obtained from cst->Root(), causes a segmentation fault.

The following methods were invoked to construct the suffix tree.

SuffixTree *cst; SuffixTreeY csty(text, length, DAC, CN_NPR, 32); cst = &csty;

The stack trace from gdb is as follows:

#0 0x00000000004086fa in cds_utils::get_field_64 (A=0xfe2820, len=6, index=65836) at ../includes/libcdsTrees.h:26 #1 0x0000000000411354 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11309, r=1, min_r=0x7fffa70200e0) at static/suffixtree/NPR_CN.cpp:413 #2 0x00000000004112c9 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=361925, r=0, min_r=0x7fffa7020198) at static/suffixtree/NPR_CN.cpp:408 #3 0x0000000000411793 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11581659, csa=0xfe2a60, lcp=0xfc76b0) at static/suffixtree/NPR_CN.cpp:348 #4 0x0000000000402c39 in cds_static::SuffixTreeY::SDepth (this=0xfc7680, vl=0, vr=11581659) at static/suffixtree/SuffixTreeY.cpp:91 #5 0x0000000000404eba in cds_static::SuffixTreeY::Child (this=0xfc7680, vl=0, vr=11581659, a=99 'c', child_l=0x7fffa7020308, child_r=0x7fffa7020300) at static/suffixtree/SuffixTreeY.cpp:279 #6 0x00000000004022e9 in testSuffixTree (s1=0xfc7680) at testSuffixTree.cpp:52 #7 0x00000000004024fc in main (argc=2, argv=0x7fffa7020698) at testSuffixTree.cpp:92

Hope this is enough information to fix the problem.

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705138

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705399

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705478

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705537

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706154

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706219

[skuruppu]

I haven't been getting any attachments in the emails that you sent. Maybe you try sending them again? Or let me know where I can download them from.

2011/8/2 rcanovas < [email protected]>

Did you try running the experiment in 64-bits replacing the suffixtree files with the ones that I send you?

On Tue, Aug 2, 2011 at 1:04 PM, skuruppu < [email protected]>wrote:

Yeah I just ran the same code on my laptop which is running 32bit Linux and the code works fine. The segmentation fault only happens when I run it on the 64bit Linux machine.

2011/8/2 Shanika Kuruppu [email protected]

I installed version 1.0.8 from http://libcds.recoded.cl/system/files/libcds-v1.0.8.tgz

This was the only version available from the libcds website.

Are you using it on a 64bit machine? Because I am, maybe that's the cause?

2011/8/2 rcanovas < [email protected]>

Are u sure that you have the last libcds version?? because I just try your problem and its work well

On Tue, Aug 2, 2011 at 10:37 AM, Yigo [email protected] wrote:

Yes is suppose to be related. Just in case I attach the presentation that I did at SEA... and I also attach my thesis if you want to read something more specific. Any way I have no problem if you prefer to review the thing starting from zero... It could be easy for you and for me it could be good just to review the topic =)... Just let me know.

Best, Yigo.

On Tue, Aug 2, 2011 at 10:30 AM, skuruppu < [email protected]>wrote:

Sure, it would be good to discuss the CST. But first I'd like to read the paper so I'm not too lost. Is this the paper relating to this implementation?

Rodrigo Cnovas and Gonzalo Navarro. Practical Compressed Suffix Trees. Proc. SEA'10

Wed next week would be good if you have time.

2011/8/2 rcanovas < [email protected]>

Every Wednesday I have time (I'm not completely sure if this Wednesday) if you want to meet and discuss or if u want me to explain the CST to u.

On Tue, Aug 2, 2011 at 9:59 AM, skuruppu < [email protected]>wrote:

Hey Rodrigo,

Sorry I didn't realise it was you. I didn't know your surname. Welcome to Melbourne. Alex told me you were coming.

I don't know much about your suffix tree implementation but I'll try to understand it. I wanted to see if I can use it instead of a suffix array in my RLZ implementation because I need an efficient way to calculate suffix links.

The sequence I used is at http://www.csse.unimelb.edu.au/~kuruppu/testdata/REF.raw.gz

I hope I've given the right permissions for you to access it.

Would be good to see you in any case so let me know if you want to get a coffee sometime.

Enjoy the sunshine :)

Shanika

2011/8/2 rcanovas < [email protected]>

Another thing, Where I can get the data set that you use?

On Tue, Aug 2, 2011 at 8:52 AM, Yigo [email protected] wrote:

Hi I don't know what could it be, but I will try to fixed as soon as I can. Anyway if you want to discuss about the implementation, most of the time I'm in room 4.31a in the 4th floor of the ICT building.

Best, Rodrigo

On Mon, Aug 1, 2011 at 9:13 PM, skuruppu < [email protected]>wrote:

I was attempting to construct a compressed suffix tree for this DNA sequence.

The suffix array is constructed successfully (at least I think so) but as soon as you invoke any methods that requires traversing the tree, a segmentation fault occurs. For example, invoking the following method call

cst->Child(node_vl, node_vr, 'c', &next_vl, &next_vr);

where node_vl and node_vr are obtained from cst->Root(), causes a segmentation fault.

The following methods were invoked to construct the suffix tree.

SuffixTree *cst; SuffixTreeY csty(text, length, DAC, CN_NPR, 32); cst = &csty;

The stack trace from gdb is as follows:

#0 0x00000000004086fa in cds_utils::get_field_64 (A=0xfe2820, len=6, index=65836) at ../includes/libcdsTrees.h:26 #1 0x0000000000411354 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11309, r=1, min_r=0x7fffa70200e0) at static/suffixtree/NPR_CN.cpp:413 #2 0x00000000004112c9 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=361925, r=0, min_r=0x7fffa7020198) at static/suffixtree/NPR_CN.cpp:408 #3 0x0000000000411793 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11581659, csa=0xfe2a60, lcp=0xfc76b0) at static/suffixtree/NPR_CN.cpp:348 #4 0x0000000000402c39 in cds_static::SuffixTreeY::SDepth (this=0xfc7680, vl=0, vr=11581659) at static/suffixtree/SuffixTreeY.cpp:91 #5 0x0000000000404eba in cds_static::SuffixTreeY::Child (this=0xfc7680, vl=0, vr=11581659, a=99 'c', child_l=0x7fffa7020308, child_r=0x7fffa7020300) at static/suffixtree/SuffixTreeY.cpp:279 #6 0x00000000004022e9 in testSuffixTree (s1=0xfc7680) at testSuffixTree.cpp:52 #7 0x00000000004024fc in main (argc=2, argv=0x7fffa7020698) at testSuffixTree.cpp:92

Hope this is enough information to fix the problem.

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705138

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705399

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705478

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705537

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706154

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706219

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706246

[rcanovas]

http://www.dcc.uchile.cl/~rcanovas/downloads/new_cst.zip

On Tue, Aug 2, 2011 at 1:17 PM, skuruppu < [email protected]>wrote:

I haven't been getting any attachments in the emails that you sent. Maybe you try sending them again? Or let me know where I can download them from.

2011/8/2 rcanovas < [email protected]>

Did you try running the experiment in 64-bits replacing the suffixtree files with the ones that I send you?

On Tue, Aug 2, 2011 at 1:04 PM, skuruppu < [email protected]>wrote:

Yeah I just ran the same code on my laptop which is running 32bit Linux and the code works fine. The segmentation fault only happens when I run it on the 64bit Linux machine.

2011/8/2 Shanika Kuruppu [email protected]

I installed version 1.0.8 from http://libcds.recoded.cl/system/files/libcds-v1.0.8.tgz

This was the only version available from the libcds website.

Are you using it on a 64bit machine? Because I am, maybe that's the cause?

2011/8/2 rcanovas < [email protected]>

Are u sure that you have the last libcds version?? because I just try your problem and its work well

On Tue, Aug 2, 2011 at 10:37 AM, Yigo [email protected] wrote:

Yes is suppose to be related. Just in case I attach the presentation that I did at SEA... and I also attach my thesis if you want to read something more specific. Any way I have no problem if you prefer to review the thing starting from zero... It could be easy for you and for me it could be good just to review the topic =)... Just let me know.

Best, Yigo.

On Tue, Aug 2, 2011 at 10:30 AM, skuruppu < [email protected]>wrote:

Sure, it would be good to discuss the CST. But first I'd like to read the paper so I'm not too lost. Is this the paper relating to this implementation?

Rodrigo Cnovas and Gonzalo Navarro. Practical Compressed Suffix Trees. Proc. SEA'10

Wed next week would be good if you have time.

2011/8/2 rcanovas < [email protected]>

Every Wednesday I have time (I'm not completely sure if this Wednesday) if you want to meet and discuss or if u want me to explain the CST to u.

On Tue, Aug 2, 2011 at 9:59 AM, skuruppu < [email protected]>wrote:

Hey Rodrigo,

Sorry I didn't realise it was you. I didn't know your surname. Welcome to Melbourne. Alex told me you were coming.

I don't know much about your suffix tree implementation but I'll try to understand it. I wanted to see if I can use it instead of a suffix array in my RLZ implementation because I need an efficient way to calculate suffix links.

The sequence I used is at http://www.csse.unimelb.edu.au/~kuruppu/testdata/REF.raw.gz

I hope I've given the right permissions for you to access it.

Would be good to see you in any case so let me know if you want to get a coffee sometime.

Enjoy the sunshine :)

Shanika

2011/8/2 rcanovas < [email protected]>

Another thing, Where I can get the data set that you use?

On Tue, Aug 2, 2011 at 8:52 AM, Yigo [email protected] wrote:

Hi I don't know what could it be, but I will try to fixed as soon as I can. Anyway if you want to discuss about the implementation, most of the time I'm in room 4.31a in the 4th floor of the ICT building.

Best, Rodrigo

On Mon, Aug 1, 2011 at 9:13 PM, skuruppu < [email protected]>wrote:

I was attempting to construct a compressed suffix tree for this DNA sequence.

The suffix array is constructed successfully (at least I think so) but as soon as you invoke any methods that requires traversing the tree, a segmentation fault occurs. For example, invoking the following method call

cst->Child(node_vl, node_vr, 'c', &next_vl, &next_vr);

where node_vl and node_vr are obtained from cst->Root(), causes a segmentation fault.

The following methods were invoked to construct the suffix tree.

SuffixTree *cst; SuffixTreeY csty(text, length, DAC, CN_NPR, 32); cst = &csty;

The stack trace from gdb is as follows:

#0 0x00000000004086fa in cds_utils::get_field_64 (A=0xfe2820, len=6, index=65836) at ../includes/libcdsTrees.h:26 #1 0x0000000000411354 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11309, r=1, min_r=0x7fffa70200e0) at static/suffixtree/NPR_CN.cpp:413 #2 0x00000000004112c9 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=361925, r=0, min_r=0x7fffa7020198) at static/suffixtree/NPR_CN.cpp:408 #3 0x0000000000411793 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11581659, csa=0xfe2a60, lcp=0xfc76b0) at static/suffixtree/NPR_CN.cpp:348 #4 0x0000000000402c39 in cds_static::SuffixTreeY::SDepth (this=0xfc7680, vl=0, vr=11581659) at static/suffixtree/SuffixTreeY.cpp:91 #5 0x0000000000404eba in cds_static::SuffixTreeY::Child (this=0xfc7680, vl=0, vr=11581659, a=99 'c', child_l=0x7fffa7020308, child_r=0x7fffa7020300) at static/suffixtree/SuffixTreeY.cpp:279 #6 0x00000000004022e9 in testSuffixTree (s1=0xfc7680) at testSuffixTree.cpp:52 #7 0x00000000004024fc in main (argc=2, argv=0x7fffa7020698) at testSuffixTree.cpp:92

Hope this is enough information to fix the problem.

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705138

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705399

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705478

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1705537

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706154

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706219

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706246

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706261

[skuruppu]

Thanks for the link to the new source files but sorry the new code doesn't compile on a 64 bit machine. I get the following errors:

static/suffixtree/LCP_naive.cpp: In constructor cds_static::LCP_naive::LCP_naive(cds_static::TextIndex_, char_, size_t): static/suffixtree/LCP_naive.cpp:44: error: cannot convert unsigned int* to size_t* for argument 1 to void cds_utils::set_field_64(size_t_, size_t, size_t, size_t) static/suffixtree/LCP_naive.cpp:50: error: cannot convert unsigned int_ to size_t* for argument 1 to void cds_utils::set_field_64(size_t_, size_t, size_t, size_t) static/suffixtree/LCP_naive.cpp: In member function virtual size_t cds_static::LCP_naive::get_LCP(size_t, cds_static::TextIndex_) const: static/suffixtree/LCP_naive.cpp:57: error: cannot convert unsigned int* const to size_t* for argument 1 to size_t cds_utils::get_field_64(size_t*, size_t, size_t)

I noticed that you've changed some of the size_t type variables into unsigned int types. Unfortunately, g++ and gcc on a 64 bit machine doesn't allow you to cast anything of type unsigned int into size_t because size_t is 8 bytes.

It'll probably take a lot of work to make this work in a 64 bit machine so don't worry about it. I'll just use this library on my laptop for the moment and keep a look out for a 64 bit version. If I get it to work, I'll send you a patch. Thanks very much for helping.

2011/8/2 rcanovas < [email protected]>

http://www.dcc.uchile.cl/~rcanovas/downloads/new_cst.zip

On Tue, Aug 2, 2011 at 1:17 PM, skuruppu < [email protected]>wrote:

I haven't been getting any attachments in the emails that you sent. Maybe you try sending them again? Or let me know where I can download them from.

2011/8/2 rcanovas < [email protected]>

Did you try running the experiment in 64-bits replacing the suffixtree files with the ones that I send you?

On Tue, Aug 2, 2011 at 1:04 PM, skuruppu < [email protected]>wrote:

Yeah I just ran the same code on my laptop which is running 32bit Linux and the code works fine. The segmentation fault only happens when I run it on the 64bit Linux machine.

2011/8/2 Shanika Kuruppu [email protected]

I installed version 1.0.8 from http://libcds.recoded.cl/system/files/libcds-v1.0.8.tgz

This was the only version available from the libcds website.

Are you using it on a 64bit machine? Because I am, maybe that's the cause?

2011/8/2 rcanovas < [email protected]>

Are u sure that you have the last libcds version?? because I just try your problem and its work well

On Tue, Aug 2, 2011 at 10:37 AM, Yigo [email protected] wrote:

Yes is suppose to be related. Just in case I attach the presentation that I did at SEA... and I also attach my thesis if you want to read something more specific. Any way I have no problem if you prefer to review the thing starting from zero... It could be easy for you and for me it could be good just to review the topic =)... Just let me know.

Best, Yigo.

On Tue, Aug 2, 2011 at 10:30 AM, skuruppu < [email protected]>wrote:

Sure, it would be good to discuss the CST. But first I'd like to read the paper so I'm not too lost. Is this the paper relating to this implementation?

Rodrigo Cnovas and Gonzalo Navarro. Practical Compressed Suffix Trees. Proc. SEA'10

Wed next week would be good if you have time.

2011/8/2 rcanovas < [email protected]>

Every Wednesday I have time (I'm not completely sure if this Wednesday) if you want to meet and discuss or if u want me to explain the CST to u.

On Tue, Aug 2, 2011 at 9:59 AM, skuruppu < [email protected]>wrote:

Hey Rodrigo,

Sorry I didn't realise it was you. I didn't know your surname. Welcome to Melbourne. Alex told me you were coming.

I don't know much about your suffix tree implementation but I'll try to understand it. I wanted to see if I can use it instead of a suffix array in my RLZ implementation because I need an efficient way to calculate suffix links.

The sequence I used is at

http://www.csse.unimelb.edu.au/~kuruppu/testdata/REF.raw.gz

I hope I've given the right permissions for you to access it.

Would be good to see you in any case so let me know if you want to get a coffee sometime.

Enjoy the sunshine :)

Shanika

2011/8/2 rcanovas < [email protected]>

Another thing, Where I can get the data set that you use?

On Tue, Aug 2, 2011 at 8:52 AM, Yigo [email protected] wrote:

Hi I don't know what could it be, but I will try to fixed as soon as I can. Anyway if you want to discuss about the implementation, most of the time I'm in room 4.31a in the 4th floor of the ICT building.

Best, Rodrigo

On Mon, Aug 1, 2011 at 9:13 PM, skuruppu < [email protected]>wrote:

I was attempting to construct a compressed suffix tree for this DNA sequence.

The suffix array is constructed successfully (at least I think so) but as soon as you invoke any methods that requires traversing the tree, a segmentation fault occurs. For example, invoking the following method call

cst->Child(node_vl, node_vr, 'c', &next_vl, &next_vr);

where node_vl and node_vr are obtained from cst->Root(), causes a segmentation fault.

The following methods were invoked to construct the suffix tree.

SuffixTree *cst; SuffixTreeY csty(text, length, DAC, CN_NPR, 32); cst = &csty;

The stack trace from gdb is as follows:

#0 0x00000000004086fa in cds_utils::get_field_64 (A=0xfe2820, len=6, index=65836) at ../includes/libcdsTrees.h:26 #1 0x0000000000411354 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11309, r=1, min_r=0x7fffa70200e0) at static/suffixtree/NPR_CN.cpp:413 #2 0x00000000004112c9 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=361925, r=0, min_r=0x7fffa7020198) at static/suffixtree/NPR_CN.cpp:408 #3 0x0000000000411793 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11581659, csa=0xfe2a60, lcp=0xfc76b0) at static/suffixtree/NPR_CN.cpp:348 #4 0x0000000000402c39 in cds_static::SuffixTreeY::SDepth (this=0xfc7680, vl=0, vr=11581659) at static/suffixtree/SuffixTreeY.cpp:91 #5 0x0000000000404eba in cds_static::SuffixTreeY::Child (this=0xfc7680, vl=0, vr=11581659, a=99 'c', child_l=0x7fffa7020308, child_r=0x7fffa7020300) at static/suffixtree/SuffixTreeY.cpp:279 #6 0x00000000004022e9 in testSuffixTree (s1=0xfc7680) at testSuffixTree.cpp:52 #7 0x00000000004024fc in main (argc=2, argv=0x7fffa7020698) at testSuffixTree.cpp:92

Hope this is enough information to fix the problem.

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705138

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705399

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705478

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705537

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706154

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706219

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706246

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706261

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706833

[rcanovas]

Sorry for all the problems, but in this link: www.dcc.uchile.cl/~rcanovas/downloads/licds.tar is the last version of the library and at least in my computer (64bits) I still don't have any problem =). Please let me know how if it's work.

On Tue, Aug 2, 2011 at 5:47 PM, skuruppu < [email protected]>wrote:

Thanks for the link to the new source files but sorry the new code doesn't compile on a 64 bit machine. I get the following errors:

static/suffixtree/LCP_naive.cpp: In constructor cds_static::LCP_naive::LCP_naive(cds_static::TextIndex_, char_, size_t): static/suffixtree/LCP_naive.cpp:44: error: cannot convert unsigned int* to size_t* for argument 1 to void cds_utils::set_field_64(size_t_, size_t, size_t, size_t) static/suffixtree/LCP_naive.cpp:50: error: cannot convert unsigned int_ to size_t* for argument 1 to void cds_utils::set_field_64(size_t_, size_t, size_t, size_t) static/suffixtree/LCP_naive.cpp: In member function virtual size_t cds_static::LCP_naive::get_LCP(size_t, cds_static::TextIndex_) const: static/suffixtree/LCP_naive.cpp:57: error: cannot convert unsigned int* const to size_t* for argument 1 to size_t cds_utils::get_field_64(size_t*, size_t, size_t)

I noticed that you've changed some of the size_t type variables into unsigned int types. Unfortunately, g++ and gcc on a 64 bit machine doesn't allow you to cast anything of type unsigned int into size_t because size_t is 8 bytes.

It'll probably take a lot of work to make this work in a 64 bit machine so don't worry about it. I'll just use this library on my laptop for the moment and keep a look out for a 64 bit version. If I get it to work, I'll send you a patch. Thanks very much for helping.

2011/8/2 rcanovas < [email protected]>

http://www.dcc.uchile.cl/~rcanovas/downloads/new_cst.zip

On Tue, Aug 2, 2011 at 1:17 PM, skuruppu < [email protected]>wrote:

I haven't been getting any attachments in the emails that you sent. Maybe you try sending them again? Or let me know where I can download them from.

2011/8/2 rcanovas < [email protected]>

Did you try running the experiment in 64-bits replacing the suffixtree files with the ones that I send you?

On Tue, Aug 2, 2011 at 1:04 PM, skuruppu < [email protected]>wrote:

Yeah I just ran the same code on my laptop which is running 32bit Linux and the code works fine. The segmentation fault only happens when I run it on the 64bit Linux machine.

2011/8/2 Shanika Kuruppu [email protected]

I installed version 1.0.8 from http://libcds.recoded.cl/system/files/libcds-v1.0.8.tgz

This was the only version available from the libcds website.

Are you using it on a 64bit machine? Because I am, maybe that's the cause?

2011/8/2 rcanovas < [email protected]>

Are u sure that you have the last libcds version?? because I just try your problem and its work well

On Tue, Aug 2, 2011 at 10:37 AM, Yigo [email protected] wrote:

Yes is suppose to be related. Just in case I attach the presentation that I did at SEA... and I also attach my thesis if you want to read something more specific. Any way I have no problem if you prefer to review the thing starting from zero... It could be easy for you and for me it could be good just to review the topic =)... Just let me know.

Best, Yigo.

On Tue, Aug 2, 2011 at 10:30 AM, skuruppu < [email protected]>wrote:

Sure, it would be good to discuss the CST. But first I'd like to read the paper so I'm not too lost. Is this the paper relating to this implementation?

Rodrigo Cnovas and Gonzalo Navarro. Practical Compressed Suffix Trees. Proc. SEA'10

Wed next week would be good if you have time.

2011/8/2 rcanovas < [email protected]>

Every Wednesday I have time (I'm not completely sure if this Wednesday) if you want to meet and discuss or if u want me to explain the CST to u.

On Tue, Aug 2, 2011 at 9:59 AM, skuruppu < [email protected]>wrote:

Hey Rodrigo,

Sorry I didn't realise it was you. I didn't know your surname. Welcome to Melbourne. Alex told me you were coming.

I don't know much about your suffix tree implementation but I'll try to understand it. I wanted to see if I can use it instead of a suffix array in my RLZ implementation because I need an efficient way to calculate suffix links.

The sequence I used is at

http://www.csse.unimelb.edu.au/~kuruppu/testdata/REF.raw.gz

I hope I've given the right permissions for you to access it.

Would be good to see you in any case so let me know if you want to get a coffee sometime.

Enjoy the sunshine :)

Shanika

2011/8/2 rcanovas < [email protected]>

Another thing, Where I can get the data set that you use?

On Tue, Aug 2, 2011 at 8:52 AM, Yigo <[email protected]

wrote:

Hi I don't know what could it be, but I will try to fixed as soon as I can. Anyway if you want to discuss about the implementation, most of the time I'm in room 4.31a in the 4th floor of the ICT building.

Best, Rodrigo

On Mon, Aug 1, 2011 at 9:13 PM, skuruppu < [email protected]>wrote:

I was attempting to construct a compressed suffix tree for this [DNA sequence](

http://www.csse.unimelb.edu.au/~kuruppu/testdata/REF.raw.gz

).

The suffix array is constructed successfully (at least I think so) but as soon as you invoke any methods that requires traversing the tree, a segmentation fault occurs. For example, invoking the following method call

cst->Child(node_vl, node_vr, 'c', &next_vl, &next_vr);

where node_vl and node_vr are obtained from cst->Root(), causes a segmentation fault.

The following methods were invoked to construct the suffix tree.

SuffixTree *cst; SuffixTreeY csty(text, length, DAC, CN_NPR, 32); cst = &csty;

The stack trace from gdb is as follows:

#0 0x00000000004086fa in cds_utils::get_field_64 (A=0xfe2820, len=6, index=65836) at ../includes/libcdsTrees.h:26 #1 0x0000000000411354 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11309, r=1, min_r=0x7fffa70200e0) at static/suffixtree/NPR_CN.cpp:413 #2 0x00000000004112c9 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=361925, r=0, min_r=0x7fffa7020198) at static/suffixtree/NPR_CN.cpp:408 #3 0x0000000000411793 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11581659, csa=0xfe2a60, lcp=0xfc76b0) at static/suffixtree/NPR_CN.cpp:348 #4 0x0000000000402c39 in cds_static::SuffixTreeY::SDepth (this=0xfc7680, vl=0, vr=11581659) at static/suffixtree/SuffixTreeY.cpp:91 #5 0x0000000000404eba in cds_static::SuffixTreeY::Child (this=0xfc7680, vl=0, vr=11581659, a=99 'c', child_l=0x7fffa7020308, child_r=0x7fffa7020300) at static/suffixtree/SuffixTreeY.cpp:279 #6 0x00000000004022e9 in testSuffixTree (s1=0xfc7680) at testSuffixTree.cpp:52 #7 0x00000000004024fc in main (argc=2, argv=0x7fffa7020698) at testSuffixTree.cpp:92

Hope this is enough information to fix the problem.

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705138

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705399

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705478

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705537

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706154

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706219

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706246

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706261

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706833

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1707170

[skuruppu]

Perfect, it works :) Thank you very much for fixing the problem.

PS: I think the link is meant to be www.dcc.uchile.cl/~rcanovas/downloads/libcds.tar

Shanika

2011/8/3 rcanovas < [email protected]>

Sorry for all the problems, but in this link: www.dcc.uchile.cl/~rcanovas/downloads/licds.tar is the last version of the library and at least in my computer (64bits) I still don't have any problem =). Please let me know how if it's work.

On Tue, Aug 2, 2011 at 5:47 PM, skuruppu < [email protected]>wrote:

Thanks for the link to the new source files but sorry the new code doesn't compile on a 64 bit machine. I get the following errors:

static/suffixtree/LCP_naive.cpp: In constructor cds_static::LCP_naive::LCP_naive(cds_static::TextIndex_, char_, size_t): static/suffixtree/LCP_naive.cpp:44: error: cannot convert unsigned int* to size_t* for argument 1 to void cds_utils::set_field_64(size_t_, size_t, size_t, size_t) static/suffixtree/LCP_naive.cpp:50: error: cannot convert unsigned int_ to size_t* for argument 1 to void cds_utils::set_field_64(size_t_, size_t, size_t, size_t) static/suffixtree/LCP_naive.cpp: In member function virtual size_t cds_static::LCP_naive::get_LCP(size_t, cds_static::TextIndex_) const: static/suffixtree/LCP_naive.cpp:57: error: cannot convert unsigned int* const to size_t* for argument 1 to size_t cds_utils::get_field_64(size_t*, size_t, size_t)

I noticed that you've changed some of the size_t type variables into unsigned int types. Unfortunately, g++ and gcc on a 64 bit machine doesn't allow you to cast anything of type unsigned int into size_t because size_t is 8 bytes.

It'll probably take a lot of work to make this work in a 64 bit machine so don't worry about it. I'll just use this library on my laptop for the moment and keep a look out for a 64 bit version. If I get it to work, I'll send you a patch. Thanks very much for helping.

2011/8/2 rcanovas < [email protected]>

http://www.dcc.uchile.cl/~rcanovas/downloads/new_cst.zip

On Tue, Aug 2, 2011 at 1:17 PM, skuruppu < [email protected]>wrote:

I haven't been getting any attachments in the emails that you sent. Maybe you try sending them again? Or let me know where I can download them from.

2011/8/2 rcanovas < [email protected]>

Did you try running the experiment in 64-bits replacing the suffixtree files with the ones that I send you?

On Tue, Aug 2, 2011 at 1:04 PM, skuruppu < [email protected]>wrote:

Yeah I just ran the same code on my laptop which is running 32bit Linux and the code works fine. The segmentation fault only happens when I run it on the 64bit Linux machine.

2011/8/2 Shanika Kuruppu [email protected]

I installed version 1.0.8 from http://libcds.recoded.cl/system/files/libcds-v1.0.8.tgz

This was the only version available from the libcds website.

Are you using it on a 64bit machine? Because I am, maybe that's the cause?

2011/8/2 rcanovas < [email protected]>

Are u sure that you have the last libcds version?? because I just try your problem and its work well

On Tue, Aug 2, 2011 at 10:37 AM, Yigo [email protected] wrote:

Yes is suppose to be related. Just in case I attach the presentation that I did at SEA... and I also attach my thesis if you want to read something more specific. Any way I have no problem if you prefer to review the thing starting from zero... It could be easy for you and for me it could be good just to review the topic =)... Just let me know.

Best, Yigo.

On Tue, Aug 2, 2011 at 10:30 AM, skuruppu < [email protected]>wrote:

Sure, it would be good to discuss the CST. But first I'd like to read the paper so I'm not too lost. Is this the paper relating to this implementation?

Rodrigo Cnovas and Gonzalo Navarro. Practical Compressed Suffix Trees. Proc. SEA'10

Wed next week would be good if you have time.

2011/8/2 rcanovas < [email protected]>

Every Wednesday I have time (I'm not completely sure if this Wednesday) if you want to meet and discuss or if u want me to explain the CST to u.

On Tue, Aug 2, 2011 at 9:59 AM, skuruppu < [email protected]>wrote:

Hey Rodrigo,

Sorry I didn't realise it was you. I didn't know your surname. Welcome to Melbourne. Alex told me you were coming.

I don't know much about your suffix tree implementation but I'll try to understand it. I wanted to see if I can use it instead of a suffix array in my RLZ implementation because I need an efficient way to calculate suffix links.

The sequence I used is at

http://www.csse.unimelb.edu.au/~kuruppu/testdata/REF.raw.gz

I hope I've given the right permissions for you to access it.

Would be good to see you in any case so let me know if you want to get a coffee sometime.

Enjoy the sunshine :)

Shanika

2011/8/2 rcanovas < [email protected]>

Another thing, Where I can get the data set that you use?

On Tue, Aug 2, 2011 at 8:52 AM, Yigo < [email protected]

wrote:

Hi I don't know what could it be, but I will try to fixed as soon as I can. Anyway if you want to discuss about the implementation, most of the time I'm in room 4.31a in the 4th floor of the ICT building.

Best, Rodrigo

On Mon, Aug 1, 2011 at 9:13 PM, skuruppu < [email protected]>wrote:

I was attempting to construct a compressed suffix tree for this [DNA sequence](

http://www.csse.unimelb.edu.au/~kuruppu/testdata/REF.raw.gz

).

The suffix array is constructed successfully (at least I think so) but as soon as you invoke any methods that requires traversing the tree, a segmentation fault occurs. For example, invoking the following method call

cst->Child(node_vl, node_vr, 'c', &next_vl, &next_vr);

where node_vl and node_vr are obtained from cst->Root(), causes a segmentation fault.

The following methods were invoked to construct the suffix tree.

SuffixTree *cst; SuffixTreeY csty(text, length, DAC, CN_NPR, 32); cst = &csty;

The stack trace from gdb is as follows:

#0 0x00000000004086fa in cds_utils::get_field_64 (A=0xfe2820, len=6, index=65836) at ../includes/libcdsTrees.h:26 #1 0x0000000000411354 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11309, r=1, min_r=0x7fffa70200e0) at static/suffixtree/NPR_CN.cpp:413 #2 0x00000000004112c9 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=361925, r=0, min_r=0x7fffa7020198) at static/suffixtree/NPR_CN.cpp:408 #3 0x0000000000411793 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11581659, csa=0xfe2a60, lcp=0xfc76b0) at static/suffixtree/NPR_CN.cpp:348 #4 0x0000000000402c39 in cds_static::SuffixTreeY::SDepth (this=0xfc7680, vl=0, vr=11581659) at static/suffixtree/SuffixTreeY.cpp:91 #5 0x0000000000404eba in cds_static::SuffixTreeY::Child (this=0xfc7680, vl=0, vr=11581659, a=99 'c', child_l=0x7fffa7020308, child_r=0x7fffa7020300) at static/suffixtree/SuffixTreeY.cpp:279 #6 0x00000000004022e9 in testSuffixTree (s1=0xfc7680) at testSuffixTree.cpp:52 #7 0x00000000004024fc in main (argc=2, argv=0x7fffa7020698) at testSuffixTree.cpp:92

Hope this is enough information to fix the problem.

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705138

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705399

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705478

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705537

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1706154

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706219

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706246

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706261

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706833

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1707170

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1713949

[rcanovas]

To say the true I fix that problem long time ago, but it seems that with Francisco Claude we forgot to upload in the new library

On Wed, Aug 3, 2011 at 11:09 AM, skuruppu < [email protected]>wrote:

Perfect, it works :) Thank you very much for fixing the problem.

PS: I think the link is meant to be www.dcc.uchile.cl/~rcanovas/downloads/libcds.tar

Shanika

2011/8/3 rcanovas < [email protected]>

Sorry for all the problems, but in this link: www.dcc.uchile.cl/~rcanovas/downloads/licds.tar is the last version of the library and at least in my computer (64bits) I still don't have any problem =). Please let me know how if it's work.

On Tue, Aug 2, 2011 at 5:47 PM, skuruppu < [email protected]>wrote:

Thanks for the link to the new source files but sorry the new code doesn't compile on a 64 bit machine. I get the following errors:

static/suffixtree/LCP_naive.cpp: In constructor cds_static::LCP_naive::LCP_naive(cds_static::TextIndex_, char_, size_t): static/suffixtree/LCP_naive.cpp:44: error: cannot convert unsigned int* to size_t* for argument 1 to void cds_utils::set_field_64(size_t_, size_t, size_t, size_t) static/suffixtree/LCP_naive.cpp:50: error: cannot convert unsigned int_ to size_t* for argument 1 to void cds_utils::set_field_64(size_t_, size_t, size_t, size_t) static/suffixtree/LCP_naive.cpp: In member function virtual size_t cds_static::LCP_naive::get_LCP(size_t, cds_static::TextIndex_) const: static/suffixtree/LCP_naive.cpp:57: error: cannot convert unsigned int* const to size_t* for argument 1 to size_t cds_utils::get_field_64(size_t*, size_t, size_t)

I noticed that you've changed some of the size_t type variables into unsigned int types. Unfortunately, g++ and gcc on a 64 bit machine doesn't allow you to cast anything of type unsigned int into size_t because size_t is 8 bytes.

It'll probably take a lot of work to make this work in a 64 bit machine so don't worry about it. I'll just use this library on my laptop for the moment and keep a look out for a 64 bit version. If I get it to work, I'll send you a patch. Thanks very much for helping.

2011/8/2 rcanovas < [email protected]>

http://www.dcc.uchile.cl/~rcanovas/downloads/new_cst.zip

On Tue, Aug 2, 2011 at 1:17 PM, skuruppu < [email protected]>wrote:

I haven't been getting any attachments in the emails that you sent. Maybe you try sending them again? Or let me know where I can download them from.

2011/8/2 rcanovas < [email protected]>

Did you try running the experiment in 64-bits replacing the suffixtree files with the ones that I send you?

On Tue, Aug 2, 2011 at 1:04 PM, skuruppu < [email protected]>wrote:

Yeah I just ran the same code on my laptop which is running 32bit Linux and the code works fine. The segmentation fault only happens when I run it on the 64bit Linux machine.

2011/8/2 Shanika Kuruppu [email protected]

I installed version 1.0.8 from http://libcds.recoded.cl/system/files/libcds-v1.0.8.tgz

This was the only version available from the libcds website.

Are you using it on a 64bit machine? Because I am, maybe that's the cause?

2011/8/2 rcanovas < [email protected]>

Are u sure that you have the last libcds version?? because I just try your problem and its work well

On Tue, Aug 2, 2011 at 10:37 AM, Yigo [email protected] wrote:

Yes is suppose to be related. Just in case I attach the presentation that I did at SEA... and I also attach my thesis if you want to read something more specific. Any way I have no problem if you prefer to review the thing starting from zero... It could be easy for you and for me it could be good just to review the topic =)... Just let me know.

Best, Yigo.

On Tue, Aug 2, 2011 at 10:30 AM, skuruppu < [email protected]>wrote:

Sure, it would be good to discuss the CST. But first I'd like to read the paper so I'm not too lost. Is this the paper relating to this implementation?

Rodrigo Cnovas and Gonzalo Navarro. Practical Compressed Suffix Trees. Proc. SEA'10

Wed next week would be good if you have time.

2011/8/2 rcanovas < [email protected]>

Every Wednesday I have time (I'm not completely sure if this Wednesday) if you want to meet and discuss or if u want me to explain the CST to u.

On Tue, Aug 2, 2011 at 9:59 AM, skuruppu < [email protected]>wrote:

Hey Rodrigo,

Sorry I didn't realise it was you. I didn't know your surname. Welcome to Melbourne. Alex told me you were coming.

I don't know much about your suffix tree implementation but I'll try to understand it. I wanted to see if I can use it instead of a suffix array in my RLZ implementation because I need an efficient way to calculate suffix links.

The sequence I used is at

http://www.csse.unimelb.edu.au/~kuruppu/testdata/REF.raw.gz

I hope I've given the right permissions for you to access it.

Would be good to see you in any case so let me know if you want to get a coffee sometime.

Enjoy the sunshine :)

Shanika

2011/8/2 rcanovas < [email protected]>

Another thing, Where I can get the data set that you use?

On Tue, Aug 2, 2011 at 8:52 AM, Yigo < [email protected]

wrote:

Hi I don't know what could it be, but I will try to fixed as soon as I can. Anyway if you want to discuss about the implementation, most of the time I'm in room 4.31a in the 4th floor of the ICT building.

Best, Rodrigo

On Mon, Aug 1, 2011 at 9:13 PM, skuruppu < [email protected]>wrote:

I was attempting to construct a compressed suffix tree for this [DNA sequence](

http://www.csse.unimelb.edu.au/~kuruppu/testdata/REF.raw.gz

).

The suffix array is constructed successfully (at least I think so) but as soon as you invoke any methods that requires traversing the tree, a segmentation fault occurs. For example, invoking the following method call

cst->Child(node_vl, node_vr, 'c', &next_vl, &next_vr);

where node_vl and node_vr are obtained from cst->Root(), causes a segmentation fault.

The following methods were invoked to construct the suffix tree.

SuffixTree *cst; SuffixTreeY csty(text, length, DAC, CN_NPR, 32); cst = &csty;

The stack trace from gdb is as follows:

#0 0x00000000004086fa in cds_utils::get_field_64 (A=0xfe2820, len=6, index=65836) at ../includes/libcdsTrees.h:26 #1 0x0000000000411354 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11309, r=1, min_r=0x7fffa70200e0) at static/suffixtree/NPR_CN.cpp:413 #2 0x00000000004112c9 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=361925, r=0, min_r=0x7fffa7020198) at static/suffixtree/NPR_CN.cpp:408 #3 0x0000000000411793 in cds_static::NPR_CN::find_RMQ (this=0xfc77e0, x=1, y=11581659, csa=0xfe2a60, lcp=0xfc76b0) at static/suffixtree/NPR_CN.cpp:348 #4 0x0000000000402c39 in cds_static::SuffixTreeY::SDepth (this=0xfc7680, vl=0, vr=11581659) at static/suffixtree/SuffixTreeY.cpp:91 #5 0x0000000000404eba in cds_static::SuffixTreeY::Child (this=0xfc7680, vl=0, vr=11581659, a=99 'c', child_l=0x7fffa7020308, child_r=0x7fffa7020300) at static/suffixtree/SuffixTreeY.cpp:279 #6 0x00000000004022e9 in testSuffixTree (s1=0xfc7680) at testSuffixTree.cpp:52 #7 0x00000000004024fc in main (argc=2, argv=0x7fffa7020698) at testSuffixTree.cpp:92

Hope this is enough information to fix the problem.

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705138

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705399

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705478

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1705537

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1706154

Reply to this email directly or view it on GitHub:

https://github.com/fclaude/libcds/issues/1#issuecomment-1706219

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706246

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706261

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1706833

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1707170

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1713949

Reply to this email directly or view it on GitHub: https://github.com/fclaude/libcds/issues/1#issuecomment-1714717