target branch for v0.9.1

.github/workflows/main.yml

@@ -1,11 +1,19 @@
 name: main
-on: [push]
+on:
+  push:
+    branches:
+      - master
+  pull_request:
+    types:
+      - opened
+      - reopened
+      - synchronize
 
 jobs:
   build-and-test:
     strategy:
       matrix:
-        python-version: [3.6, 3.7, 3.8, 3.9]
+        python-version: ["3.8", "3.9", "3.10", "3.11"]
     runs-on: ubuntu-latest
     steps:
     - uses: actions/checkout@v2
@@ -30,7 +38,7 @@ jobs:
       # This only requires Cython, no other dependencies.
       run: |
         eval "$(conda shell.bash hook)"
-        conda create -p ./cython-env -y cython
+        conda create -p ./cython-env -y cython python=${{ matrix.python-version }} numpy
         conda activate ./cython-env
         python setup.py clean cythonize sdist
         (cd dist && pip install pybedtools-*.tar.gz && cd $TMPDIR && python -c 'import pybedtools; print(pybedtools.__file__)')
@@ -53,31 +61,53 @@ jobs:
       # Tests below will operate in this newly-installed directory.
       run: |
         eval "$(conda shell.bash hook)"
-        conda create -y -p ./test-env \
-          --channel conda-forge \
-          --channel bioconda python=${{ matrix.python-version }} \
-          --file requirements.txt \
-          --file test-requirements.txt \
-          --file optional-requirements.txt
+        conda install mamba python=${{ matrix.python-version }} -y --channel conda-forge
+
+        if [ ${{ matrix.python-version }} != "3.11" ]; then
+          mamba create -y -p ./test-env \
+            --channel conda-forge \
+            --channel bioconda python=${{ matrix.python-version }} \
+            --file requirements.txt \
+            --file test-requirements.txt \
+            --file optional-requirements.txt
+          conda activate ./test-env
+        else
+          # Only install bedtools; let pip take care of the rest for 3.11 until
+          # bioconda catches up.
+          #
+          # We still install the test requirements though, and the optional
+          # requirements except for genomepy which is in bioconda.
+          grep -v "genomepy" optional-requirements.txt > optional-requirements-3.11.txt
+          mamba create -y -p ./test-env \
+            --channel conda-forge \
+            --channel bioconda \
+            bedtools \
+            python=${{ matrix.python-version }} \
+            --file test-requirements.txt \
+            --file optional-requirements-3.11.txt
+          conda activate ./test-env
+          pip install genomepy
+
+        fi
         conda activate ./test-env
 
         mkdir -p /tmp/pybedtools-uncompressed
         cd /tmp/pybedtools-uncompressed
         tar -xf $WORKDIR/dist/pybedtools-*.tar.gz
-        cd pybedtools-*
-        pip install -e .
-        python -c 'import pybedtools; print(pybedtools.__file__)'
-        ls *
+        pip install -e /tmp/pybedtools-uncompressed/pybedtools-*
 
+        # Trying import in the same directory will complain that cbedtools
+        # can't be imported
+        (cd / && python -c 'import pybedtools; print(pybedtools.__file__)')
 
     - name: tests
       # Run pytest and sphinx doctests
       run: |
-        eval "$(conda shell.bash hook)"
         cd $WORKDIR
+        eval "$(conda shell.bash hook)"
         conda activate ./test-env
 
-        # Move to extracted tarball dir, see above notes
+        # Extract the package tarball built above, and use that for running the tests.
         cd /tmp/pybedtools-uncompressed/pybedtools-*
         pytest -v --doctest-modules
         pytest -v pybedtools/test/genomepy_integration.py
@@ -89,29 +119,31 @@ jobs:
       # Build docs and commit to gh-pages branch. Note that no push happens
       # unless we're on the master branch
       run: |
-        eval "$(conda shell.bash hook)"
-        conda activate ./test-env
-
-        # Move to extracted tarball dir, see above notes
-        cd /tmp/pybedtools-uncompressed/pybedtools-*
-        (cd docs && make html)
-
-        git clone \
-          --single-branch \
-          --branch gh-pages "https://x-access-token:${{ secrets.GITHUB_TOKEN }}@github.com/$GITHUB_REPOSITORY" \
-          /tmp/docs
-
-        rm -rf /tmp/docs/*
-        cp -r docs/build/html/* /tmp/docs
-        touch /tmp/docs/.nojekyll
-        cd /tmp/docs
-        git add .
-        if git diff --cached --quiet; then
-          echo "no changes, nothing to commit"
-        else
-          git commit -m 'update docs'
+        if [ ${{ matrix.python-version }} != "3.11" ]; then
+          eval "$(conda shell.bash hook)"
+          conda activate ./test-env
+
+          # Move to extracted tarball dir, see above notes
+          cd /tmp/pybedtools-uncompressed/pybedtools-*
+          (cd docs && make html)
+
+          git clone \
+            --single-branch \
+            --branch gh-pages "https://x-access-token:${{ secrets.GITHUB_TOKEN }}@github.com/$GITHUB_REPOSITORY" \
+            /tmp/docs
+
+          rm -rf /tmp/docs/*
+          cp -r docs/build/html/* /tmp/docs
+          touch /tmp/docs/.nojekyll
+          cd /tmp/docs
+          git add .
+          if git diff --cached --quiet; then
+            echo "no changes, nothing to commit"
+          else
+            git commit -m 'update docs'
+          fi
+          cd $WORKDIR
         fi
-        cd $WORKDIR
 
 
     - name: docs artifact

README.rst

@@ -1,3 +1,4 @@
+
 Overview
 --------
 

docs/source/changes.rst

@@ -3,6 +3,26 @@
 Changelog
 =========
 
+Changes in v0.9.1
+-----------------
+
+2023-07-23
+
+* Dropping support for Python 3.6 and 3.7
+* Respect sorting of chromsize files (thanks @mgperry)
+* Updated setup.py to correctly reflect the MIT license change elsewhere (`#374
+  <https://github.com/daler/pybedtools/issues/374>`, thanks @hyandell)
+*  Support plotting lengths of intervals and custom DPI (`#367
+   <https://github.com/daler/pybedtools/issues/367>`, `#366
+   <https://github.com/daler/pybedtools/issues/366>`), thanks @yunfeiguo)
+* Remove outdated hard-coded check for 510 files in ``intersect`` and instead
+  defer to local machine's ``ulimit``
+* Enabling building/installing on Python 3.11 (thanks @daz10000)
+* Allow np.int64 start/stop positions to be used when creating Interval objects (`#390 <https://github.com/daler/pybedtools/issues/390>`)
+* properly close filehandles in .save_seq (thanks @PeterRobots)
+* include minimal pyproject.toml file (thanks @afg1)
+
+
 Changes in v0.9
 ---------------
 

docs/source/topical-genome.rst

@@ -98,20 +98,18 @@ will create a file from a dictionary or string:
     'dm3.genome'
     >>> print(open('dm3.genome').read())
     chr2L	23011544
-    chr2LHet	368872
     chr2R	21146708
-    chr2RHet	3288761
     chr3L	24543557
-    chr3LHet	2555491
     chr3R	27905053
-    chr3RHet	2517507
     chr4	1351857
+    chrX	22422827
+    chr2LHet	368872
+    chr2RHet	3288761
+    chr3LHet	2555491
+    chr3RHet	2517507
     chrM	19517
     chrU	10049037
     chrUextra	29004656
-    chrX	22422827
     chrXHet	204112
     chrYHet	347038
     <BLANKLINE>
-
-

pybedtools/_Window.pyx

@@ -1,4 +1,5 @@
 # cython: profile=True
+# cython: language_level=2
 
 import os
 from collections import deque

pybedtools/bedtool.py

@@ -338,20 +338,6 @@ def wrapped(self, *args, **kwargs):
             if check_for_genome:
                 kwargs = self.check_genome(**kwargs)
 
-            # TODO: should this be implemented as a generic function that can
-            # be passed in for a each tool to check kwargs? Currently this is
-            # the only check I can think of.
-            if prog in ("intersect", "intersectBed"):
-                if (
-                    isinstance(kwargs["b"], list)
-                    and len(kwargs["b"]) > 510
-                    and all([isinstance(i, str) for i in kwargs["b"]])
-                ):
-                    raise pybedtoolsError(
-                        "BEDTools intersect does not support > 510 filenames for -b "
-                        "argument. Consider passing these as BedTool objects instead"
-                    )
-
             # For sequence methods, we may need to make a tempfile that will
             # hold the resulting sequence.  For example, fastaFromBed needs to
             # make a tempfile for 'fo' if no 'fo' was explicitly specified by
@@ -2130,7 +2116,7 @@ def shuffle(self):
         """
 
     @_log_to_history
-    @_wraps(prog="sortBed", implicit="i")
+    @_wraps(prog="sortBed", implicit="i", uses_genome=True, genome_if=["g", "genome"])
     def sort(self):
         """
         Wraps `bedtools sort`.
@@ -2320,8 +2306,8 @@ def complement(self):
         chr1	0	1
         chr1	500	900
         chr1	950	249250621
-        chr10	0	135534747
-        chr11	0	135006516
+        chr2	0	243199373
+        chr3	0	198022430
         """
 
     @_log_to_history
@@ -2726,9 +2712,11 @@ def save_seqs(self, fn):
 
         if not hasattr(self, "seqfn"):
             raise ValueError("Use .sequence(fasta) to get the sequence first")
-        fout = open(fn, "w")
-        fout.write(open(self.seqfn).read())
-        fout.close()
+
+        with open(fn, "w") as fout:
+            with open(self.seqfn) as seqfile:
+                fout.write(seqfile.read())
+
         new_bedtool = BedTool(self.fn)
         new_bedtool.seqfn = fn
         return new_bedtool

pybedtools/cbedtools.pyx

@@ -1,4 +1,5 @@
 # distutils: language = c++
+# cython: language_level=2
 
 # String notes:
 #
@@ -15,6 +16,7 @@
 
 from cpython.version cimport PY_MAJOR_VERSION
 from libcpp.string cimport string
+import numpy as np
 
 # Python byte strings automatically coerce to/from C++ strings.
 
@@ -23,7 +25,7 @@ cdef _cppstr(s):
     #
     # C++ uses bytestrings. PY2 strings need no conversion; bare PY3 strings
     # are unicode and so must be encoded to bytestring.
-    if isinstance(s, int):
+    if isinstance(s, integer_types):
         s = str(s)
     if isinstance(s, unicode):
         s = s.encode('UTF-8')
@@ -36,9 +38,9 @@ cdef _pystr(string s):
     return s.decode('UTF-8', 'strict')
 
 if PY_MAJOR_VERSION < 3:
-    integer_types = (int, long)
+    integer_types = (int, long, np.int64)
 else:
-    integer_types = (int,)
+    integer_types = (int, np.int64)
 
 """
     bedtools.pyx: A Cython wrapper for the BEDTools BedFile class

pybedtools/featurefuncs.pyx

@@ -1,3 +1,4 @@
+# cython: language_level=2
 # distutils: language = c++
 from cbedtools cimport Interval
 from cbedtools import create_interval_from_list

pybedtools/helpers.py

@@ -815,7 +815,7 @@ def chromsizes_to_file(chrom_sizes, fn=None):
     if isinstance(chrom_sizes, str):
         chrom_sizes = chromsizes(chrom_sizes)
     fout = open(fn, "wt")
-    for chrom, bounds in sorted(chrom_sizes.items()):
+    for chrom, bounds in chrom_sizes.items():
         line = chrom + "\t" + str(bounds[1]) + "\n"
         fout.write(line)
     fout.close()

pybedtools/scripts/venn_mpl.py

@@ -17,8 +17,7 @@
 import os
 import pybedtools
 
-
-def venn_mpl(a, b, c, colors=None, outfn="out.png", labels=None, dpi=300):
+def venn_mpl(a, b, c, colors=None, outfn="out.png", labels=None, by_length=False, dpi=300):
     """
     *a*, *b*, and *c* are filenames to BED-like files.
 
@@ -30,7 +29,10 @@ def venn_mpl(a, b, c, colors=None, outfn="out.png", labels=None, dpi=300):
 
     *labels* is a list of labels to use for each of the files; by default the
     labels are ['a','b','c']
-
+
+    *by_length* if True, then instead of plotting number of intervals, plot combined
+    lengths of intervals
+
     *dpi* is the dpi setting passed to matplotlib savefig
     """
     try:
@@ -46,6 +48,9 @@ def venn_mpl(a, b, c, colors=None, outfn="out.png", labels=None, dpi=300):
     a = pybedtools.BedTool(a)
     b = pybedtools.BedTool(b)
     c = pybedtools.BedTool(c)
+    count_features = lambda x:x.count()
+    if by_length:
+        count_features = lambda x:x.total_coverage()
 
     if colors is None:
         colors = ["r", "b", "g"]
@@ -91,31 +96,31 @@ def venn_mpl(a, b, c, colors=None, outfn="out.png", labels=None, dpi=300):
     kwargs = dict(horizontalalignment="center")
 
     # Unique to A
-    ax.text(center - 2 * offset, center + offset, str((a - b - c).count()), **kwargs)
+    ax.text(center - 2 * offset, center + offset, str(count_features(a - b - c)), **kwargs)
 
     # Unique to B
-    ax.text(center + 2 * offset, center + offset, str((b - a - c).count()), **kwargs)
+    ax.text(center + 2 * offset, center + offset, str(count_features(b - a - c)), **kwargs)
 
     # Unique to C
-    ax.text(center, center - 2 * offset, str((c - a - b).count()), **kwargs)
+    ax.text(center, center - 2 * offset, str(count_features(c - a - b)), **kwargs)
 
     # A and B not C
     ax.text(
-        center, center + 2 * offset - 0.5 * offset, str((a + b - c).count()), **kwargs
+        center, center + 2 * offset - 0.5 * offset, str(count_features(a + b - c)), **kwargs
     )
 
     # A and C not B
     ax.text(
-        center - 1.2 * offset, center - 0.5 * offset, str((a + c - b).count()), **kwargs
+        center - 1.2 * offset, center - 0.5 * offset, str(count_features(a + c - b)), **kwargs
     )
 
     # B and C not A
     ax.text(
-        center + 1.2 * offset, center - 0.5 * offset, str((b + c - a).count()), **kwargs
+        center + 1.2 * offset, center - 0.5 * offset, str(count_features(b + c - a)), **kwargs
     )
 
     # all
-    ax.text(center, center, str((a + b + c).count()), **kwargs)
+    ax.text(center, center, str(count_features(a + b + c)), **kwargs)
 
     ax.legend(loc="best")
 

pybedtools/test/data/example.narrowPeak

@@ -0,0 +1,5 @@
+track type=narrowPeak visibility=3 db=hg19 name="nPk" description="ENCODE narrowPeak Example"
+browser position chr1:9356000-9365000
+chr1	9356548	9356648	.	0	.	182	5.0945	-1	50
+chr1	9358722	9358822	.	0	.	91	4.6052	-1	40
+chr1	9361082	9361182	.	0	.	182	9.2103	-1	75