From c27dc372ad887f33c0cc46c3086c1022f3cfc652 Mon Sep 17 00:00:00 2001
From: Chen Tong <chentong_biology@163.com>
Date: Wed, 27 Sep 2023 10:57:10 +0800
Subject: [PATCH 1/3] add ignore zero anchor parameter for skip no matches in
 batch search (#598)

---
 jcvi/compara/catalog.py | 17 ++++++++++++++++-
 1 file changed, 16 insertions(+), 1 deletion(-)

diff --git a/jcvi/compara/catalog.py b/jcvi/compara/catalog.py
index 80488cc2..eb8e4749 100644
--- a/jcvi/compara/catalog.py
+++ b/jcvi/compara/catalog.py
@@ -666,6 +666,12 @@ def ortholog(args):
     dotplot_group.add_option(
         "--no_dotplot", default=False, action="store_true", help="Do not make dotplot"
     )
+    p.add_option(
+        "--ignore_zero_anchor",
+        default=False,
+        action="store_true",
+        help="Ignore this pair of ortholog identification instead of throwing an error when performing many pairs of cataloging."
+    )
 
     opts, args = p.parse_args(args)
 
@@ -674,6 +680,7 @@ def ortholog(args):
 
     a, b = args
     dbtype = opts.dbtype
+    ignore_zero_anchor = opts.ignore_zero_anchor
     suffix = ".cds" if dbtype == "nucl" else ".pep"
     abed, afasta = a + ".bed", a + suffix
     bbed, bfasta = b + ".bed", b + suffix
@@ -727,7 +734,15 @@ def ortholog(args):
                 dargs += ["--no_strip_names"]
             if opts.liftover_dist:
                 dargs += ["--liftover_dist={}".format(opts.liftover_dist)]
-            scan(dargs)
+            try:
+                scan(dargs)
+            except ValueError as e:
+                if ignore_zero_anchor:
+                    logging.debug(f"{e}")
+                    logging.debug("Ignoring this error and continuing...")
+                    return
+                else:
+                    raise ValueError(e)
         if quota:
             quota_main([lifted_anchors, "--quota={0}".format(quota), "--screen"])
         if need_update(anchors, pdf, warn=True) and not opts.no_dotplot:

From 7e4c5e5c2fd69970de54308788ad3b1b9d50ff3f Mon Sep 17 00:00:00 2001
From: Haibao Tang <tanghaibao@gmail.com>
Date: Thu, 28 Sep 2023 02:59:35 -0700
Subject: [PATCH 2/3] Add a module logger (#600)

* update base.py

* update formats.base

* update compara.base

* update graphics.base

* revert logging changes

* use the correct Console
---
 jcvi/apps/base.py     | 108 +++++++++++++++++++-----------------------
 jcvi/compara/base.py  |  10 ++--
 jcvi/formats/base.py  |  44 +++++++----------
 jcvi/graphics/base.py |  37 ++++++---------
 4 files changed, 85 insertions(+), 114 deletions(-)

diff --git a/jcvi/apps/base.py b/jcvi/apps/base.py
index 719c3cf2..100642c3 100644
--- a/jcvi/apps/base.py
+++ b/jcvi/apps/base.py
@@ -10,8 +10,6 @@
 import sys
 import logging
 import fnmatch
-import functools
-
 
 from collections.abc import Iterable
 from http.client import HTTPSConnection
@@ -29,7 +27,8 @@
 from typing import Any, Collection, List, Optional, Union
 
 from natsort import natsorted
-from rich.logging import Console, RichHandler
+from rich.console import Console
+from rich.logging import RichHandler
 
 from jcvi import __copyright__, __version__
 
@@ -40,28 +39,33 @@
 JCVIHELP = "JCVI utility libraries {} [{}]\n".format(__version__, __copyright__)
 
 
-def patch_debug(func):
-    @functools.wraps(func)
-    def wraps(*args, **kwargs):
-        import inspect
+def debug(level=logging.DEBUG):
+    """
+    Turn on the debugging
+    """
+    logging.basicConfig(
+        level=level,
+        format="%(message)s",
+        datefmt="[%X]",
+        handlers=[RichHandler(console=Console(stderr=True))],
+    )
 
-        callerframerecord = inspect.stack()[1]
-        frame = callerframerecord[0]
-        info = inspect.getframeinfo(frame)
-        # get caller function name
-        caller = frame.f_code.co_name
-        patch_message = f"{info.filename}:{info.lineno}:{caller}"
-        old_debug = logging.debug
 
-        def my_debug(message: str, *args, **kwargs):
-            old_debug(f"{patch_message} {message}", *args, **kwargs)
+debug()
 
-        logging.debug = my_debug
-        ret = func(*args, **kwargs)
-        logging.debug = old_debug
-        return ret
 
-    return wraps
+def get_logger(name: str):
+    """Return a logger with a default ColoredFormatter."""
+    logger = logging.getLogger(name)
+    if logger.hasHandlers():
+        logger.handlers.clear()
+    logger.addHandler(RichHandler())
+    logger.propagate = False
+    logger.setLevel(logging.INFO)
+    return logger
+
+
+logger = get_logger("jcvi")
 
 
 class ActionDispatcher(object):
@@ -603,12 +607,12 @@ def set_image_options(
         iopts = ImageOptions(opts)
 
         if opts.notex:
-            logging.info("--notex={}. latex use is disabled.".format(opts.notex))
+            logger.info("--notex=%s. latex use is disabled.", opts.notex)
         elif not is_tex_available():
             if not bool(which("latex")):
-                logging.info("`latex` not found. latex use is disabled.")
+                logger.info("`latex` not found. latex use is disabled.")
             if not bool(which("lp")):
-                logging.info("`lp` not found. latex use is disabled.")
+                logger.info("`lp` not found. latex use is disabled.")
 
         setup_theme(style=opts.style, font=opts.font, usetex=iopts.usetex)
 
@@ -1048,9 +1052,9 @@ def ConfigSectionMap(Config, section):
         try:
             cfg[option] = Config.get(section, option)
             if cfg[option] == -1:
-                logging.debug("skip: {0}".format(option))
+                logger.debug("skip: %s", option)
         except:
-            logging.debug("exception on {0}!".format(option))
+            logger.debug("exception on %s!", option)
             cfg[option] = None
     return cfg
 
@@ -1131,7 +1135,7 @@ def dmain(mainfile, type="action"):
 def backup(filename):
     bakname = filename + ".bak"
     if op.exists(filename):
-        logging.debug("Backup `{0}` to `{1}`".format(filename, bakname))
+        logger.debug("Backup `%s` to `%s`", filename, bakname)
         sh("mv {0} {1}".format(filename, bakname))
     return bakname
 
@@ -1205,7 +1209,7 @@ def sh(
             cmd += " &"
 
         if log:
-            logging.debug(cmd)
+            logger.debug(cmd)
 
         call_func = check_output if check else call
         return call_func(cmd, shell=True, executable=shell, stderr=redirect_error)
@@ -1218,7 +1222,7 @@ def Popen(cmd, stdin=None, stdout=PIPE, debug=False, shell="/bin/bash"):
     from subprocess import Popen as P
 
     if debug:
-        logging.debug(cmd)
+        logger.debug(cmd)
     # See: <https://blog.nelhage.com/2010/02/a-very-subtle-bug/>
     proc = P(cmd, bufsize=1, stdin=stdin, stdout=stdout, shell=True, executable=shell)
     return proc
@@ -1307,7 +1311,7 @@ def mkdir(dirname, overwrite=False):
         if overwrite:
             cleanup(dirname)
             os.mkdir(dirname)
-            logging.debug("Overwrite folder `{0}`.".format(dirname))
+            logger.debug("Overwrite folder `%s`", dirname)
         else:
             return False  # Nothing is changed
     else:
@@ -1315,7 +1319,7 @@ def mkdir(dirname, overwrite=False):
             os.mkdir(dirname)
         except:
             os.makedirs(dirname)
-        logging.debug("`{0}` not found. Creating new.".format(dirname))
+        logger.debug("`%s` not found. Creating new.", dirname)
 
     return True
 
@@ -1377,7 +1381,7 @@ def need_update(a: str, b: str, warn: bool = False) -> bool:
         or any(is_newer_file(x, y) for x in a for y in b)
     )
     if (not should_update) and warn:
-        logging.debug("File `{}` found. Computation skipped.".format(", ".join(b)))
+        logger.debug("File `%s` found. Computation skipped.", ", ".join(b))
     return should_update
 
 
@@ -1476,8 +1480,7 @@ def download(
     final_filename = filename or target
     if op.exists(final_filename):
         if debug:
-            msg = "File `{}` exists. Download skipped.".format(final_filename)
-            logging.info(msg)
+            logger.info("File `%s` exists. Download skipped.", final_filename)
         success = True
     else:
         from jcvi.utils.ez_setup import get_best_downloader
@@ -1527,26 +1530,11 @@ def getfilesize(filename, ratio=None):
     while size < heuristicsize:
         size += 2**32
     if size > 2**32:
-        logging.warning("Gzip file estimated uncompressed size: {0}.".format(size))
+        logger.warning("Gzip file estimated uncompressed size: %d", size)
 
     return size
 
 
-def debug(level=logging.DEBUG):
-    """
-    Turn on the debugging
-    """
-    logging.basicConfig(
-        level=level,
-        format="%(message)s",
-        datefmt="[%X]",
-        handlers=[RichHandler(console=Console(stderr=True))],
-    )
-
-
-debug()
-
-
 def main():
     actions = (
         ("less", "enhance the unix `less` command"),
@@ -1603,7 +1591,7 @@ def expand(args):
     for a in args:
         oa = a.replace("/", "_")
         if oa in seen:
-            logging.debug("Name collision `{0}`, ignored.".format(oa))
+            logger.debug("Name collision `%s`, ignored", oa)
             continue
 
         cmd = "cp -s" if opts.symlink else "mv"
@@ -1995,7 +1983,7 @@ def notify(args):
     opts, args = p.parse_args(args)
 
     if len(args) == 0:
-        logging.error("Please provide a brief message to be sent")
+        logger.error("Please provide a brief message to be sent")
         sys.exit(not p.print_help())
 
     subject = opts.subject
@@ -2005,7 +1993,7 @@ def notify(args):
         toaddr = opts.email.split(",")  # TO address should be in a list
         for addr in toaddr:
             if not is_valid_email(addr):
-                logging.debug("Email address `{0}` is not valid!".format(addr))
+                logger.debug("Email address `%s` is not valid!", addr)
                 sys.exit()
         send_email(fromaddr, toaddr, subject, message)
     else:
@@ -2165,7 +2153,7 @@ def waitpid(args):
             msg = check_output(shlex.split(get_origcmd)).strip()
         _waitpid(pid, interval=opts.interval)
     else:
-        logging.debug("Process with PID {0} does not exist".format(pid))
+        logger.debug("Process with PID %d does not exist", pid)
         sys.exit()
 
     if opts.notify:
@@ -2188,9 +2176,9 @@ def get_config(path):
         config.read(path)
     except ParsingError:
         e = sys.exc_info()[1]
-        logging.error(
-            "There was a problem reading or parsing "
-            "your credentials file: %s" % (e.args[0],),
+        logger.error(
+            "There was a problem reading or parsing your credentials file: %s",
+            e.args[0],
         )
     return config
 
@@ -2241,16 +2229,16 @@ def getpath(
     else:
         err_msg = f"Cannot execute binary `{path}`. Please verify and rerun."
         if warn == "exit":
-            logging.fatal(err_msg)
+            logger.fatal(err_msg)
         else:
-            logging.warning(err_msg)
+            logger.warning(err_msg)
         return None
 
     if changed:
         configfile = open(cfg, "w")
         config.write(configfile)
         configfile.close()
-        logging.debug("Configuration written to `{0}`.".format(cfg))
+        logger.debug("Configuration written to `%s`", cfg)
 
     return path
 
diff --git a/jcvi/compara/base.py b/jcvi/compara/base.py
index 9919159e..eae6ec6b 100644
--- a/jcvi/compara/base.py
+++ b/jcvi/compara/base.py
@@ -1,6 +1,6 @@
-import logging
 from collections import defaultdict
 
+from ..apps.base import logger
 from ..formats.base import BaseFile, read_block, must_open
 from ..utils.range import Range
 
@@ -69,9 +69,9 @@ def print_to_file(self, filename="stdout", accepted=None):
                 print("\t".join((a, b, score)), file=fw)
         fw.close()
 
-        logging.debug("Removed %d existing anchors.", nremoved)
-        logging.debug("Corrected scores for %d anchors.", ncorrected)
-        logging.debug("Anchors written to `%s`.", filename)
+        logger.debug("Removed %d existing anchors", nremoved)
+        logger.debug("Corrected scores for %d anchors", ncorrected)
+        logger.debug("Anchors written to `%s`", filename)
 
     def blast(self, blastfile=None, outfile=None):
         """
@@ -102,7 +102,7 @@ def blast(self, blastfile=None, outfile=None):
             nlines += 1
         fw.close()
 
-        logging.debug("A total of %d BLAST lines written to `%s`.", nlines, outfile)
+        logger.debug("A total of %d BLAST lines written to `%s`", nlines, outfile)
 
         return outfile
 
diff --git a/jcvi/formats/base.py b/jcvi/formats/base.py
index 79aa1ac0..8617a145 100644
--- a/jcvi/formats/base.py
+++ b/jcvi/formats/base.py
@@ -5,20 +5,20 @@
 import os.path as op
 import math
 import sys
-import logging
 
 from collections import OrderedDict
 from itertools import groupby, islice, cycle
 
 from Bio import SeqIO
-from jcvi.apps.base import (
+from ..apps.base import (
     OptionParser,
     ActionDispatcher,
     cleanup,
-    sh,
-    need_update,
+    logger,
     mkdir,
+    need_update,
     popen,
+    sh,
 )
 
 
@@ -28,10 +28,9 @@
 
 class BaseFile(object):
     def __init__(self, filename):
-
         self.filename = filename
         if filename:
-            logging.debug("Load file `{0}`".format(filename))
+            logger.debug("Load file `%s`", filename)
 
 
 class LineFile(BaseFile, list):
@@ -40,15 +39,12 @@ class LineFile(BaseFile, list):
     """
 
     def __init__(self, filename, comment=None, load=False):
-
         super(LineFile, self).__init__(filename)
 
         if load:
             fp = must_open(filename)
             self.lines = [l.strip() for l in fp if l[0] != comment]
-            logging.debug(
-                "Load {0} lines from `{1}`.".format(len(self.lines), filename)
-            )
+            logger.debug("Load %d lines from `%s`", len(self.lines), filename)
 
 
 class DictFile(BaseFile, OrderedDict):
@@ -66,7 +62,6 @@ def __init__(
         keycast=None,
         cast=None,
     ):
-
         BaseFile.__init__(self, filename)
         OrderedDict.__init__(self)
         self.keypos = keypos
@@ -84,7 +79,7 @@ def __init__(
 
                 msg = "Must contain >= {0} columns.  {1}.\n".format(ncols, action)
                 msg += "  --> Line {0}: {1}".format(lineno + 1, row)
-                logging.error(msg)
+                logger.error(msg)
                 if strict:
                     sys.exit(1)
                 else:
@@ -100,7 +95,7 @@ def __init__(
 
         assert thiscols, "File empty"
         self.ncols = thiscols
-        logging.debug("Imported {0} records from `{1}`.".format(len(self), filename))
+        logger.debug("Imported %d records from `%s`", len(self), filename)
 
     @classmethod
     def num_columns(cls, filename, delimiter=None):
@@ -136,7 +131,6 @@ class FileMerger(object):
     """
 
     def __init__(self, filelist, outfile):
-
         self.filelist = filelist
         self.outfile = outfile
         self.ingz = filelist[0].endswith(".gz")
@@ -166,7 +160,7 @@ def __init__(self, filename, outputdir=None, format="fasta", mode="cycle"):
         self.mode = mode
 
         format = format or self._guess_format(filename)
-        logging.debug("format is %s" % format)
+        logger.debug("format is %s", format)
 
         if format in ("fasta", "fastq"):
             self.klass = "seqio"
@@ -179,7 +173,6 @@ def __init__(self, filename, outputdir=None, format="fasta", mode="cycle"):
         mkdir(outputdir)
 
     def _open(self, filename):
-
         if self.klass == "seqio":
             handle = SeqIO.parse(open(filename), self.format)
         elif self.klass == "clust":
@@ -260,16 +253,14 @@ def split(self, N, force=False):
         """
         mode = self.mode
         assert mode in ("batch", "cycle", "optimal")
-        logging.debug("set split mode=%s" % mode)
+        logger.debug("set split mode=%s", mode)
 
         self.names = self.__class__.get_names(self.filename, N)
         if self.outputdir:
             self.names = [op.join(self.outputdir, x) for x in self.names]
 
         if not need_update(self.filename, self.names) and not force:
-            logging.error(
-                "file %s already existed, skip file splitting" % self.names[0]
-            )
+            logger.error("file %s already existed, skip file splitting", self.names[0])
             return
 
         filehandles = [open(x, "w") for x in self.names]
@@ -277,7 +268,7 @@ def split(self, N, force=False):
         if mode == "batch":
             for batch, fw in zip(self._batch_iterator(N), filehandles):
                 count = self.write(fw, batch)
-                logging.debug("write %d records to %s" % (count, fw.name))
+                logger.debug("write %d records to %s", count, fw.name)
 
         elif mode == "cycle":
             handle = self._open(self.filename)
@@ -415,7 +406,7 @@ def must_open(filename, mode="r", checkexists=False, skipcheck=False, oappend=Fa
                 else:
                     fp = open(filename, "w")
             else:
-                logging.debug("File `{0}` already exists. Skipped.".format(filename))
+                logger.debug("File `%s` already exists. Skipped.", filename)
                 return None
         else:
             fp = open(filename, mode)
@@ -462,7 +453,7 @@ def write_file(filename, contents, meta=None, skipcheck=False, append=False, tee
 
     fileop = "appended" if append else "written"
     message = "{0} {1} to `{2}`.".format(meta, fileop, filename)
-    logging.debug(message.capitalize())
+    logger.debug(message.capitalize())
     if meta == "run script" and not append:
         sh("chmod u+x {0}".format(filename))
 
@@ -542,7 +533,6 @@ def flexible_cast(s):
 
 
 def main():
-
     actions = (
         ("pairwise", "convert a list of IDs into all pairs"),
         ("split", "split large file into N chunks"),
@@ -821,7 +811,7 @@ def group(args):
             if len(cols) < len(atoms):
                 cols = [x for x in range(len(atoms))]
             if groupby not in cols:
-                logging.error("groupby col index `{0}` is out of range".format(groupby))
+                logger.error("groupby col index `%s` is out of range", groupby)
                 sys.exit()
 
             key = atoms[groupby]
@@ -922,13 +912,13 @@ def split(args):
     fs = FileSplitter(filename, outputdir=outdir, format=opts.format, mode=opts.mode)
 
     if opts.all:
-        logging.debug("option -all override N")
+        logger.debug("option -all override N")
         N = fs.num_records
     else:
         N = min(fs.num_records, int(N))
         assert N > 0, "N must be > 0"
 
-    logging.debug("split file into %d chunks" % N)
+    logger.debug("split file into %d chunks", N)
     fs.split(N)
 
     return fs
diff --git a/jcvi/graphics/base.py b/jcvi/graphics/base.py
index 0a140998..5569a70a 100644
--- a/jcvi/graphics/base.py
+++ b/jcvi/graphics/base.py
@@ -11,6 +11,7 @@
 logging.getLogger("numexpr").setLevel(logging.WARNING)
 logging.getLogger("PIL").setLevel(logging.INFO)
 
+
 from functools import partial
 
 import numpy as np
@@ -36,10 +37,8 @@
 from matplotlib.path import Path
 from typing import Optional
 
-from jcvi.formats.base import LineFile
-from jcvi.apps.base import glob, listify, datadir, sample_N, which
-
-logging.getLogger().setLevel(logging.DEBUG)
+from ..apps.base import datadir, glob, listify, logger, sample_N, which
+from ..formats.base import LineFile
 
 
 CHARS = {
@@ -95,10 +94,8 @@ def __init__(self, fig, usetex: bool = True):
         try:
             self.build_height_array(fig, usetex=usetex)
         except ValueError as e:
-            logging.debug(
-                "Failed to init heights (error: {}). Variable label sizes skipped.".format(
-                    e
-                )
+            logger.debug(
+                "Failed to init heights (error: %s). Variable label sizes skipped.", e
             )
 
     @classmethod
@@ -204,8 +201,8 @@ def load_image(filename):
         ret[:, :, 2] = ret[:, :, 1] = ret[:, :, 0] = img
         img = ret
     else:
-        h, w, c = img.shape
-    logging.debug("Image `{0}` loaded ({1}px x {2}px).".format(filename, w, h))
+        h, w, _ = img.shape
+    logger.debug("Image `%s` loaded (%dpx x %dpx).", filename, w, h)
     return img
 
 
@@ -315,24 +312,22 @@ def savefig(figname, dpi=150, iopts=None, cleanup=True):
     except:
         format = "pdf"
     try:
-        logging.debug(f"Matplotlib backend is: {mpl.get_backend()}")
-        logging.debug(f"Attempting save as: {figname}")
+        logger.debug("Matplotlib backend is: %s", mpl.get_backend())
+        logger.debug("Attempting save as: %s", figname)
         plt.savefig(figname, dpi=dpi, format=format)
     except Exception as e:
-        message = "savefig failed with message:"
-        message += "\n{0}".format(str(e))
-        logging.error(message)
-        logging.info("Try running again with --notex option to disable latex.")
+        logger.error("savefig failed with message:\n%s", e)
+        logger.info("Try running again with --notex option to disable latex.")
         if op.exists(figname):
             if op.getsize(figname) < 1000:
-                logging.debug(f"Cleaning up empty file: {figname}")
+                logger.debug("Cleaning up empty file: %s", figname)
                 remove(figname)
         sys.exit(1)
 
     msg = "Figure saved to `{0}`".format(figname)
     if iopts:
         msg += " {0}".format(iopts)
-    logging.debug(msg)
+    logger.debug(msg)
 
     if cleanup:
         plt.rcdefaults()
@@ -392,7 +387,7 @@ def fontprop(ax, name, size=12):
     fname = op.join(datadir, name)
     prop = fm.FontProperties(fname=fname, size=size)
 
-    logging.debug("Set font to `{0}` (`{1}`).".format(name, prop.get_file()))
+    logger.debug("Set font to `%s` (`%s`)", name, prop.get_file())
     for text in ax.texts:
         text.set_fontproperties(prop)
 
@@ -439,9 +434,7 @@ def setup_theme(
     if usetex:
         rc("text", usetex=True)
     else:
-        logging.info(
-            "Set text.usetex={}. Font styles may be inconsistent.".format(usetex)
-        )
+        logger.info("Set text.usetex=%s. Font styles may be inconsistent.", usetex)
         rc("text", usetex=False)
 
     if font == "Helvetica":

From b2f009cdd66214985c0f5221e855e747bc532636 Mon Sep 17 00:00:00 2001
From: Haibao Tang <tanghaibao@gmail.com>
Date: Thu, 5 Oct 2023 15:53:31 -0700
Subject: [PATCH 3/3] Extra opts in gff.bed() (#602)

* Add two options to gff.bed()

* automatically find the other opts when --ensembl_cds
---
 jcvi/formats/gff.py | 39 +++++++++++++++++++++++++++++++++------
 1 file changed, 33 insertions(+), 6 deletions(-)

diff --git a/jcvi/formats/gff.py b/jcvi/formats/gff.py
index 9eb20d16..8c4a68b0 100644
--- a/jcvi/formats/gff.py
+++ b/jcvi/formats/gff.py
@@ -497,7 +497,6 @@ def to_range(obj, score=None, id=None, strand=None):
 
 
 def main():
-
     actions = (
         ("addparent", "merge sister features and infer their parent"),
         ("bed", "parse gff and produce bed file for particular feature type"),
@@ -3022,6 +3021,18 @@ def bed(args):
         default="Parent",
         help="Parent gene key to group with --primary_only",
     )
+    p.add_option(
+        "--add_chr",
+        default=False,
+        action="store_true",
+        help="Add `chr` prefix to seqid",
+    )
+    p.add_option(
+        "--ensembl_cds",
+        default=False,
+        action="store_true",
+        help="Use transcript_name.exon_number as accn",
+    )
     p.set_outfile()
 
     opts, args = p.parse_args(args)
@@ -3036,11 +3047,15 @@ def bed(args):
     span = opts.span
     primary_only = opts.primary_only
     parent_key = opts.parent_key
-
+    add_chr = opts.add_chr
+    ensembl_cds = opts.ensembl_cds
     if opts.type:
         type = set(x.strip() for x in opts.type.split(","))
     if opts.source:
         source = set(x.strip() for x in opts.source.split(","))
+    if ensembl_cds:
+        type = {"CDS"}
+        source = {"protein_coding"}
 
     gff = Gff(
         gffile,
@@ -3053,6 +3068,8 @@ def bed(args):
     )
     b = Bed()
     seen_parents = set()  # used with --primary_only
+    seen = set()  # used with --ensembl_cds
+    skipped_identical_range = 0
     skipped_non_primary = 0
 
     for g in gff:
@@ -3072,6 +3089,18 @@ def bed(args):
             bl.accn = accn
         if span:
             bl.score = bl.span
+        if add_chr:
+            bl.seqid = "chr" + bl.seqid
+        if ensembl_cds:
+            bl.accn = "{0}.{1}".format(
+                g.get_attr("transcript_name"), g.get_attr("exon_number")
+            )
+            position = (bl.seqid, bl.start, bl.end)
+            if position in seen:
+                skipped_identical_range += 1
+                continue
+            seen.add(position)
+
         b.append(bl)
 
     sorted = not opts.nosort
@@ -3083,6 +3112,8 @@ def bed(args):
     )
     if primary_only:
         logging.debug("Skipped non-primary: %d", skipped_non_primary)
+    if ensembl_cds:
+        logging.debug("Skipped due to identical range: %d", skipped_identical_range)
 
 
 def make_index(gff_file):
@@ -3134,7 +3165,6 @@ def children(args):
     parents = set(opts.parents.split(","))
 
     for feat in get_parents(gff_file, parents):
-
         cc = [c.id for c in g.children(feat.id, 1)]
         if len(cc) <= 1:
             continue
@@ -3513,7 +3543,6 @@ def get_coords(feature, site, fLen, seqlen, feat, children_list, gffdb):
     elif site in ["TrSS", "TrES"]:
         children = []
         for c in gffdb.children(feat.id, 1):
-
             if c.featuretype not in children_list:
                 continue
             children.append((c.start, c.stop))
@@ -3686,7 +3715,6 @@ def bed12(args):
     fw = must_open(outfile, "w")
 
     for f in g.features_of_type(parent):
-
         chrom = f.chrom
         chromStart = f.start - 1
         chromEnd = f.stop
@@ -3701,7 +3729,6 @@ def bed12(args):
         blocks = []
 
         for c in g.children(name, 1):
-
             cstart, cend = c.start - 1, c.stop
 
             if c.featuretype == block: