pyranges · endrebak · Oct 7, 2019 · Oct 7, 2019 · Oct 7, 2019 · Oct 7, 2019
diff --git a/.travis.yml b/.travis.yml
@@ -2,8 +2,8 @@ language: python
 python:
   - "3.6"
 install:
-  - pip install cython
+  - pip install cython pytest
   - python setup.py install
 
 # functionality is tested in the pyranges package
-script: cd ~/; python -c 'import ncls; print(ncls.__version__)'
+script: py.test tests/test_ncls.py && cd ~/ && python -c 'import ncls; print(ncls.__version__)'
diff --git a/CHANGELOG b/CHANGELOG
@@ -1,6 +1,10 @@
-# 0.0.44 (unreleased)
+# 0.0.45 (07.10.19)
+- try to fix another windows error
+
+# 0.0.44 (10.09.19)
 - remove sys.time dep. perhaps it now works on windows?
 - add k_find_both
+- fix bug when large ints were used
 
 # 0.0.43 (23.07.19)
 - add coverage-method which finds bp overlap of intervals

diff --git a/README.md b/README.md
@@ -1,6 +1,6 @@
 # Nested containment list
 
-[![Build Status](https://travis-ci.org/hunt-genes/ncls.svg?branch=master)](https://travis-ci.org/hunt-genes/ncls) [![PyPI version](https://badge.fury.io/py/ncls.svg)](https://badge.fury.io/py/ncls)
+[![Build Status](https://travis-ci.org/biocore-ntnu/ncls.svg?branch=master)](https://travis-ci.org/hunt-genes/ncls) [![PyPI version](https://badge.fury.io/py/ncls.svg)](https://badge.fury.io/py/ncls)
 
 The Nested Containment List is a datastructure for interval overlap queries,
 like the interval tree. It is usually an order of magnitude faster than the
@@ -14,7 +14,8 @@ gains.
 It was implemented to be the cornerstone of the PyRanges project, but I have made
 it available to the Python community as a stand-alone library. Enjoy.
 
-Paper: https://academic.oup.com/bioinformatics/article/23/11/1386/199545
+Original Paper: https://academic.oup.com/bioinformatics/article/23/11/1386/199545
+Cite: http://dx.doi.org/10.1093/bioinformatics/btz615
 
 ## Install
 
@@ -25,7 +26,6 @@ pip install ncls
 ## Usage
 
 ```python
-# see the examples/ folder for more examples
 from ncls import NCLS
 
 import pandas as pd
@@ -34,6 +34,16 @@ starts = pd.Series(range(0, 5))
 ends = starts + 100
 ids = starts
 
+subject_df = pd.DataFrame({"Start": starts, "End": ends}, index=ids)
+
+print(subject_df)
+#    Start  End
+# 0      0  100
+# 1      1  101
+# 2      2  102
+# 3      3  103
+# 4      4  104
+
 ncls = NCLS(starts.values, ends.values, ids.values)
 
 # python API, slower
@@ -47,11 +57,45 @@ starts_query = pd.Series([1, 3])
 ends_query = pd.Series([52, 14])
 indexes_query = pd.Series([10000, 100])
 
+query_df = pd.DataFrame({"Start": starts_query.values, "End": ends_query.values}, index=indexes_query.values)
+
+query_df
+#        Start  End
+# 10000      1   52
+# 100        3   14
+
+
 # everything done in C/Cython; faster
-ncls.all_overlaps_both(starts_query.values, ends_query.values, indexes_query.values)
+l_idxs, r_idxs = ncls.all_overlaps_both(starts_query.values, ends_query.values, indexes_query.values)
+l_idxs, r_idxs
 # (array([10000, 10000, 10000, 10000, 10000,   100,   100,   100,   100,
 #          100]), array([0, 1, 2, 3, 4, 0, 1, 2, 3, 4]))
 
+print(query_df.loc[l_idxs])
+#        Start  End
+# 10000      1   52
+# 10000      1   52
+# 10000      1   52
+# 10000      1   52
+# 10000      1   52
+# 100        3   14
+# 100        3   14
+# 100        3   14
+# 100        3   14
+# 100        3   14
+print(subject_df.loc[r_idxs])
+#    Start  End
+# 0      0  100
+# 1      1  101
+# 2      2  102
+# 3      3  103
+# 4      4  104
+# 0      0  100
+# 1      1  101
+# 2      2  102
+# 3      3  103
+# 4      4  104
+
 # return intervals in python (slow/mem-consuming)
 intervals = ncls.intervals()
 intervals
@@ -74,7 +118,7 @@ usage is one fifth and one ninth.
 
 ## Cite
 
-https://www.biorxiv.org/content/10.1101/609396v1
+http://dx.doi.org/10.1093/bioinformatics/btz615
 
 ## Original paper
 

diff --git a/ncls/src/cncls.pxd b/ncls/src/cncls.pxd
@@ -1,3 +1,4 @@
+from libc.stdint cimport int64_t
 
 # cdef extern from "string.h":
 #   ctypedef int size_t
@@ -39,8 +40,8 @@ cdef extern from "stdio.h":
 
 cdef extern from "ncls/src/intervaldb.h":
     ctypedef struct IntervalMap:
-        int start
-        int end
+        int64_t start
+        int64_t end
         int target_id
         int sublist
 
@@ -51,7 +52,7 @@ cdef extern from "ncls/src/intervaldb.h":
         int start
         int len
 
-    int find_overlap_start(int start, int end, IntervalMap im[], int n)
+    int find_overlap_start(int64_t start, int64_t end, IntervalMap im[], int n)
     int imstart_qsort_cmp(void *void_a,void *void_b)
     # int target_qsort_cmp(void *void_a,void *void_b)
     IntervalMap *read_intervals(int n,FILE *ifile)
@@ -68,8 +69,8 @@ cdef extern from "ncls/src/intervaldb.h":
                              int *nfound)
 
     int find_intervals(IntervalIterator *it0,
-                       int start,
-                       int end,
+                       int64_t start,
+                       int64_t end,
                        IntervalMap im[],
                        int n,
                        SublistHeader subheader[],

diff --git a/ncls/src/cncls32.pxd b/ncls/src/cncls32.pxd
@@ -1,5 +1,5 @@
 
-from libc.stdint cimport int32_t
+from libc.stdint cimport int32_t, int64_t
 
 cdef extern from "stdlib.h":
   void free(void *)

diff --git a/ncls/src/intervaldb.c b/ncls/src/intervaldb.c
@@ -1,5 +1,6 @@
 
 
+#include <stdint.h>
 #include "intervaldb.h"
 
 int C_int_max=INT_MAX; /* KLUDGE TO LET PYREX CODE ACCESS VALUE OF INT_MAX MACRO */
@@ -331,9 +332,9 @@ IntervalMap *interval_map_alloc(int n)
 
 
 
-inline int find_overlap_start(int start,int end,IntervalMap im[],int n)
+inline int64_t find_overlap_start(int64_t start,int64_t end,IntervalMap im[],int n)
 {
-  int l=0,mid,r;
+  int64_t l=0,mid,r;
 
   r=n-1;
   while (l<r) {
@@ -426,14 +427,14 @@ void reorient_intervals(int n,IntervalMap im[],int ori_sign)
 }
 
 
-int find_intervals(IntervalIterator *it0, int start, int end,
+int64_t find_intervals(IntervalIterator *it0, int64_t start, int64_t end,
                    IntervalMap im[],int n,
                    SublistHeader subheader[], int nlists,
                    IntervalMap buf[], int nbuf,
                    int *p_nreturn, IntervalIterator **it_return)
 {
   IntervalIterator *it=NULL,*it2=NULL;
-  int ibuf=0,j,k,ori_sign=1;
+  int64_t ibuf=0,j,k,ori_sign=1;
   if (!it0) { /* ALLOCATE AN ITERATOR IF NOT SUPPLIED*/
     CALLOC(it,1,IntervalIterator);
   }

diff --git a/ncls/src/intervaldb.h b/ncls/src/intervaldb.h
@@ -4,21 +4,23 @@
 #include "default.h"
 #include <limits.h>
 
+#include <stdint.h>
+
 extern int C_int_max;
 
 typedef struct {
-  int start;
-  int end;
-  int target_id;
+  int64_t start;
+  int64_t end;
+  int64_t target_id;
   /* int target_start; */
   /* int target_end; */
-  int sublist;
+  int64_t sublist;
 } IntervalMap;
 
 
 typedef struct {
-  int start;
-  int end;
+  int64_t start;
+  int64_t end;
 } IntervalIndex;
 
 typedef struct {
@@ -75,7 +77,7 @@ typedef struct {
 
 extern int *alloc_array(int n);
 
-extern int find_overlap_start(int start,int end,IntervalMap im[],int n);
+extern int64_t find_overlap_start(int64_t start,int64_t end,IntervalMap im[],int n);
 extern int imstart_qsort_cmp(const void *void_a,const void *void_b);
 extern int target_qsort_cmp(const void *void_a,const void *void_b);
 extern IntervalMap *read_intervals(int n,FILE *ifile);
@@ -88,7 +90,7 @@ extern IntervalDB *build_interval_db(IntervalMap im[],int n);
 extern IntervalIterator *interval_iterator_alloc(void);
 extern int free_interval_iterator(IntervalIterator *it);
 extern IntervalIterator *reset_interval_iterator(IntervalIterator *it);
-extern int find_intervals(IntervalIterator *it0,int start,int end,IntervalMap im[],int n,SublistHeader subheader[],int nlists,IntervalMap buf[],int nbuf,int *p_nreturn,IntervalIterator **it_return);
+extern int64_t find_intervals(IntervalIterator *it0,int64_t start,int64_t end,IntervalMap im[],int n,SublistHeader subheader[],int nlists,IntervalMap buf[],int nbuf,int *p_nreturn,IntervalIterator **it_return);
 extern int read_imdiv(FILE *ifile,IntervalMap imdiv[],int div,int i_div,int ntop);
 extern IntervalMap *read_sublist(FILE *ifile,SublistHeader *subheader,IntervalMap *im);
 extern int find_file_intervals(IntervalIterator *it0,int start,int end,
@@ -110,10 +112,6 @@ extern int save_text_file(char filestem[],char err_msg[],
 extern int text_file_to_binaries(FILE *infile,char buildpath[],char err_msg[]);
 extern void reorient_intervals(int n,IntervalMap im[],int ori_sign);
 
-extern int find_intervals_stack(int start_stack[], int end_stack[], int sp, int start,
-                         int end, IntervalMap im[], int n,
-                         SublistHeader subheader[], IntervalMap buf[],
-                                int *nfound);
 /* extern int find_k_next(int start, int end, */
 /*                 IntervalMap im[], int n, */
 /*                 SublistHeader subheader[], int nlists, */