From f9cf7e640c0fdf052bb7edeb2cf84def862a6207 Mon Sep 17 00:00:00 2001
From: ccedmendoza <ccedmendoza@users.noreply.github.com>
Date: Mon, 22 Jul 2019 13:37:58 +0200
Subject: [PATCH 1/9] Make filter work with file containing header

---
 tools/filters/gff/gtf_filter_by_attribute_values_list.py | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/tools/filters/gff/gtf_filter_by_attribute_values_list.py b/tools/filters/gff/gtf_filter_by_attribute_values_list.py
index f1754f1bf3a..6097a1f191d 100644
--- a/tools/filters/gff/gtf_filter_by_attribute_values_list.py
+++ b/tools/filters/gff/gtf_filter_by_attribute_values_list.py
@@ -55,10 +55,11 @@ def gff_filter(gff_file, attribute_name, ids_file, output_file):
     # Filter GFF file using ids.
     output = open(output_file, 'w')
     for line in open(gff_file):
-        fields = line.split('\t')
-        attributes = parse_gff_attributes(fields[8])
-        if (attribute_name in attributes) and (attributes[attribute_name] in ids_dict):
-            output.write(line)
+        if (line[0] != '#'):
+            fields = line.split('\t')
+            attributes = parse_gff_attributes(fields[8])
+            if (attribute_name in attributes) and (attributes[attribute_name] in ids_dict):
+                output.write(line)
     output.close()
 
 

From f44c0b311c2b2f25b969ee8e01a6a3dd5698d152 Mon Sep 17 00:00:00 2001
From: ccedmendoza <cedricmendoza@gmail.com>
Date: Tue, 23 Jul 2019 12:52:27 +0200
Subject: [PATCH 2/9] Simplify script while adding empty lines and comments
 handling

---
 .../gtf_filter_by_attribute_values_list.py    | 65 +++++--------------
 1 file changed, 17 insertions(+), 48 deletions(-)

diff --git a/tools/filters/gff/gtf_filter_by_attribute_values_list.py b/tools/filters/gff/gtf_filter_by_attribute_values_list.py
index 6097a1f191d..008fedf5908 100644
--- a/tools/filters/gff/gtf_filter_by_attribute_values_list.py
+++ b/tools/filters/gff/gtf_filter_by_attribute_values_list.py
@@ -7,60 +7,29 @@
 from __future__ import print_function
 
 import sys
-
-
-def parse_gff_attributes(attr_str):
-    """
-    Parses a GFF/GTF attribute string and returns a dictionary of name-value
-    pairs. The general format for a GFF3 attributes string is
-        name1=value1;name2=value2
-    The general format for a GTF attribute string is
-        name1 "value1" ; name2 "value2"
-    The general format for a GFF attribute string is a single string that
-    denotes the interval's group; in this case, method returns a dictionary
-    with a single key-value pair, and key name is 'group'
-    """
-    attributes_list = attr_str.split(";")
-    attributes = {}
-    for name_value_pair in attributes_list:
-        # Try splitting by space and, if necessary, by '=' sign.
-        pair = name_value_pair.strip().split(" ")
-        if len(pair) == 1:
-            pair = name_value_pair.strip().split("=")
-        if len(pair) == 1:
-            # Could not split for some reason -- raise exception?
-            continue
-        if pair == '':
-            continue
-        name = pair[0].strip()
-        if name == '':
-            continue
-        # Need to strip double quote from values
-        value = pair[1].strip(" \"")
-        attributes[name] = value
-
-    if len(attributes) == 0:
-        # Could not split attributes string, so entire string must be
-        # 'group' attribute. This is the case for strictly GFF files.
-        attributes['group'] = attr_str
-    return attributes
-
+import re
 
 def gff_filter(gff_file, attribute_name, ids_file, output_file):
     # Put ids in dict for quick lookup.
     ids_dict = {}
-    for line in open(ids_file):
-        ids_dict[line.split('\t')[0].strip()] = True
+    with open(ids_file) as ids:
+        for line in ids:
+            ids_dict[line.split('\t')[0].strip()] = True
 
     # Filter GFF file using ids.
-    output = open(output_file, 'w')
-    for line in open(gff_file):
-        if (line[0] != '#'):
-            fields = line.split('\t')
-            attributes = parse_gff_attributes(fields[8])
-            if (attribute_name in attributes) and (attributes[attribute_name] in ids_dict):
-                output.write(line)
-    output.close()
+    with open(output_file, 'w') as output:
+        with open(gff_file) as gff:
+            for line in gff:
+                # Keep commented lines
+                if line.startswith('#'):
+                    output.write(line)
+                else:
+                    # Create pattern using attribute to filter
+                    prog = re.compile(r".*"+re.escape(attribute_name)+r" \"(.+?)\"\;")
+                    line_match = prog.match(line)
+                    # If there is a match and the id after the filtered attribute (attribut "id";) is in the list. We print it
+                    if line_match and line_match.group(1) in ids_dict:
+                            output.write(line)
 
 
 if __name__ == "__main__":

From 8491730b834aa3fff3947d12b7d8cc27bdbcb549 Mon Sep 17 00:00:00 2001
From: ccedmendoza <ccedmendoza@users.noreply.github.com>
Date: Tue, 23 Jul 2019 13:25:52 +0200
Subject: [PATCH 3/9] Corrent linting

---
 .../gff/gtf_filter_by_attribute_values_list.py        | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/tools/filters/gff/gtf_filter_by_attribute_values_list.py b/tools/filters/gff/gtf_filter_by_attribute_values_list.py
index 008fedf5908..8ae704297bb 100644
--- a/tools/filters/gff/gtf_filter_by_attribute_values_list.py
+++ b/tools/filters/gff/gtf_filter_by_attribute_values_list.py
@@ -6,8 +6,9 @@
 #
 from __future__ import print_function
 
-import sys
 import re
+import sys
+
 
 def gff_filter(gff_file, attribute_name, ids_file, output_file):
     # Put ids in dict for quick lookup.
@@ -24,12 +25,12 @@ def gff_filter(gff_file, attribute_name, ids_file, output_file):
                 if line.startswith('#'):
                     output.write(line)
                 else:
-                    # Create pattern using attribute to filter
-                    prog = re.compile(r".*"+re.escape(attribute_name)+r" \"(.+?)\"\;")
+                    # Create pattern using attribute to filter 
+                    prog = re.compile(r".*" + re.escape(attribute_name) + r" \"(.+?)\"\;")
                     line_match = prog.match(line)
-                    # If there is a match and the id after the filtered attribute (attribut "id";) is in the list. We print it
+                    # If there is a match and the id after the filtered attribute (attribut "id";) is in the list, we print it
                     if line_match and line_match.group(1) in ids_dict:
-                            output.write(line)
+                        output.write(line)
 
 
 if __name__ == "__main__":

From 765e70080383e4b5bc09e51d3cd4c268f96f09de Mon Sep 17 00:00:00 2001
From: ccedmendoza <ccedmendoza@users.noreply.github.com>
Date: Tue, 23 Jul 2019 13:35:31 +0200
Subject: [PATCH 4/9] remove trailing white space

---
 tools/filters/gff/gtf_filter_by_attribute_values_list.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/filters/gff/gtf_filter_by_attribute_values_list.py b/tools/filters/gff/gtf_filter_by_attribute_values_list.py
index 8ae704297bb..52db30dcd4c 100644
--- a/tools/filters/gff/gtf_filter_by_attribute_values_list.py
+++ b/tools/filters/gff/gtf_filter_by_attribute_values_list.py
@@ -25,7 +25,7 @@ def gff_filter(gff_file, attribute_name, ids_file, output_file):
                 if line.startswith('#'):
                     output.write(line)
                 else:
-                    # Create pattern using attribute to filter 
+                    # Create pattern using attribute to filter
                     prog = re.compile(r".*" + re.escape(attribute_name) + r" \"(.+?)\"\;")
                     line_match = prog.match(line)
                     # If there is a match and the id after the filtered attribute (attribut "id";) is in the list, we print it

From cec6e6926b0980a54ed6f1536ead5844275b680d Mon Sep 17 00:00:00 2001
From: Matthias Bernt <m.bernt@ufz.de>
Date: Sat, 13 Feb 2021 14:54:26 +0100
Subject: [PATCH 5/9] revert to a less intrusive change

---
 .../gtf_filter_by_attribute_values_list.py    | 66 ++++++++++++++-----
 .../gtf_filter_by_attribute_values_list.xml   |  6 +-
 2 files changed, 52 insertions(+), 20 deletions(-)

diff --git a/tools/filters/gff/gtf_filter_by_attribute_values_list.py b/tools/filters/gff/gtf_filter_by_attribute_values_list.py
index 52db30dcd4c..cb641f40e05 100644
--- a/tools/filters/gff/gtf_filter_by_attribute_values_list.py
+++ b/tools/filters/gff/gtf_filter_by_attribute_values_list.py
@@ -6,31 +6,63 @@
 #
 from __future__ import print_function
 
-import re
 import sys
 
 
+def parse_gff_attributes(attr_str):
+    """
+    Parses a GFF/GTF attribute string and returns a dictionary of name-value
+    pairs. The general format for a GFF3 attributes string is
+        name1=value1;name2=value2
+    The general format for a GTF attribute string is
+        name1 "value1" ; name2 "value2"
+    The general format for a GFF attribute string is a single string that
+    denotes the interval's group; in this case, method returns a dictionary
+    with a single key-value pair, and key name is 'group'
+    """
+    attributes_list = attr_str.split(";")
+    attributes = {}
+    for name_value_pair in attributes_list:
+        # Try splitting by space and, if necessary, by '=' sign.
+        pair = name_value_pair.strip().split(" ")
+        if len(pair) == 1:
+            pair = name_value_pair.strip().split("=")
+        if len(pair) == 1:
+            # Could not split for some reason -- raise exception?
+            continue
+        if pair == '':
+            continue
+        name = pair[0].strip()
+        if name == '':
+            continue
+        # Need to strip double quote from values
+        value = pair[1].strip(" \"")
+        attributes[name] = value
+
+    if len(attributes) == 0:
+        # Could not split attributes string, so entire string must be
+        # 'group' attribute. This is the case for strictly GFF files.
+        attributes['group'] = attr_str
+    return attributes
+
+
 def gff_filter(gff_file, attribute_name, ids_file, output_file):
     # Put ids in dict for quick lookup.
     ids_dict = {}
-    with open(ids_file) as ids:
-        for line in ids:
-            ids_dict[line.split('\t')[0].strip()] = True
+    for line in open(ids_file):
+        ids_dict[line.split('\t')[0].strip()] = True
 
     # Filter GFF file using ids.
-    with open(output_file, 'w') as output:
-        with open(gff_file) as gff:
-            for line in gff:
-                # Keep commented lines
-                if line.startswith('#'):
-                    output.write(line)
-                else:
-                    # Create pattern using attribute to filter
-                    prog = re.compile(r".*" + re.escape(attribute_name) + r" \"(.+?)\"\;")
-                    line_match = prog.match(line)
-                    # If there is a match and the id after the filtered attribute (attribut "id";) is in the list, we print it
-                    if line_match and line_match.group(1) in ids_dict:
-                        output.write(line)
+    with open(output_file, 'w') as output, open(gff_file) as ingff:
+        for line in ingff:
+            if not line or line.startswith('#'):
+                output.write(line)
+                continue
+            fields = line.split('\t')
+            attributes = parse_gff_attributes(fields[8])
+            if (attribute_name in attributes) and (attributes[attribute_name] in ids_dict):
+                output.write(line)
+        output.close()
 
 
 if __name__ == "__main__":
diff --git a/tools/filters/gff/gtf_filter_by_attribute_values_list.xml b/tools/filters/gff/gtf_filter_by_attribute_values_list.xml
index 0f5d0dadabc..99251dffb1d 100644
--- a/tools/filters/gff/gtf_filter_by_attribute_values_list.xml
+++ b/tools/filters/gff/gtf_filter_by_attribute_values_list.xml
@@ -1,7 +1,7 @@
 <tool id="gtf_filter_by_attribute_values_list" name="Filter GTF data by attribute values_list" version="0.1">
     <description></description>
     <command interpreter="python">
-      gtf_filter_by_attribute_values_list.py $input $attribute_name $ids $output
+      gtf_filter_by_attribute_values_list.py '$input' '$attribute_name' '$ids' '$output'
     </command>
     <inputs>
         <param format="gtf" name="input" type="data" label="Filter"/>
@@ -19,10 +19,10 @@
     <tests>
         <!-- Test filtering with a simple list of values. -->
         <test>
-            <param name="input" value="gops_subtract_in1.gff"/>
+            <param name="input" value="gops_subtract_in1_wcomment.gff"/>
             <param name="attribute_name" value="gene_id"/>
             <param name="ids" value="gtf_filter_by_attribute_values_list_in1.txt"/>
-            <output name="output" file="gtf_filter_by_attribute_values_list_out1.gtf"/>
+            <output name="output" file="gtf_filter_by_attribute_values_list_out1_wcomment.gtf"/>
         </test>
         <!-- Test filtering with a more complex tabular file. -->
         <test>

From faed88a1bdafaaada5549d2b4e26ba8b3a26d6bb Mon Sep 17 00:00:00 2001
From: Matthias Bernt <m.bernt@ufz.de>
Date: Sat, 13 Feb 2021 16:34:09 +0100
Subject: [PATCH 6/9] remove interpreter

---
 tools/filters/gff/gtf_filter_by_attribute_values_list.xml | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/filters/gff/gtf_filter_by_attribute_values_list.xml b/tools/filters/gff/gtf_filter_by_attribute_values_list.xml
index 99251dffb1d..9dee10a0dca 100644
--- a/tools/filters/gff/gtf_filter_by_attribute_values_list.xml
+++ b/tools/filters/gff/gtf_filter_by_attribute_values_list.xml
@@ -1,7 +1,7 @@
 <tool id="gtf_filter_by_attribute_values_list" name="Filter GTF data by attribute values_list" version="0.1">
     <description></description>
-    <command interpreter="python">
-      gtf_filter_by_attribute_values_list.py '$input' '$attribute_name' '$ids' '$output'
+    <command>
+        python '$__tool_directory__/gtf_filter_by_attribute_values_list.py' '$input' '$attribute_name' '$ids' '$output'
     </command>
     <inputs>
         <param format="gtf" name="input" type="data" label="Filter"/>

From e4d9e1a29e898af13fcf66e7a4b809d10016bd8e Mon Sep 17 00:00:00 2001
From: M Bernt <m.bernt@ufz.de>
Date: Sat, 13 Feb 2021 20:38:05 +0100
Subject: [PATCH 7/9] Apply suggestions from code review

Co-authored-by: Nicola Soranzo <nicola.soranzo@gmail.com>
---
 tools/filters/gff/gtf_filter_by_attribute_values_list.py  | 3 +--
 tools/filters/gff/gtf_filter_by_attribute_values_list.xml | 6 +++---
 2 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/tools/filters/gff/gtf_filter_by_attribute_values_list.py b/tools/filters/gff/gtf_filter_by_attribute_values_list.py
index cb641f40e05..8be10348557 100644
--- a/tools/filters/gff/gtf_filter_by_attribute_values_list.py
+++ b/tools/filters/gff/gtf_filter_by_attribute_values_list.py
@@ -60,9 +60,8 @@ def gff_filter(gff_file, attribute_name, ids_file, output_file):
                 continue
             fields = line.split('\t')
             attributes = parse_gff_attributes(fields[8])
-            if (attribute_name in attributes) and (attributes[attribute_name] in ids_dict):
+            if attribute_name in attributes and attributes[attribute_name] in ids_dict:
                 output.write(line)
-        output.close()
 
 
 if __name__ == "__main__":
diff --git a/tools/filters/gff/gtf_filter_by_attribute_values_list.xml b/tools/filters/gff/gtf_filter_by_attribute_values_list.xml
index 9dee10a0dca..c75b12d65a2 100644
--- a/tools/filters/gff/gtf_filter_by_attribute_values_list.xml
+++ b/tools/filters/gff/gtf_filter_by_attribute_values_list.xml
@@ -1,8 +1,8 @@
 <tool id="gtf_filter_by_attribute_values_list" name="Filter GTF data by attribute values_list" version="0.1">
     <description></description>
-    <command>
-        python '$__tool_directory__/gtf_filter_by_attribute_values_list.py' '$input' '$attribute_name' '$ids' '$output'
-    </command>
+    <command><![CDATA[
+python '$__tool_directory__/gtf_filter_by_attribute_values_list.py' '$input' $attribute_name '$ids' '$output'
+    ]]></command>
     <inputs>
         <param format="gtf" name="input" type="data" label="Filter"/>
         <param name="attribute_name" type="select" label="Using attribute name">

From 4be9bc6430ea4fcf849ffee2cce0740c70a26029 Mon Sep 17 00:00:00 2001
From: Matthias Bernt <m.bernt@ufz.de>
Date: Sat, 13 Feb 2021 22:02:57 +0100
Subject: [PATCH 8/9] bump tool version

---
 tools/filters/gff/gtf_filter_by_attribute_values_list.xml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/filters/gff/gtf_filter_by_attribute_values_list.xml b/tools/filters/gff/gtf_filter_by_attribute_values_list.xml
index 9dee10a0dca..ef7461f3d4f 100644
--- a/tools/filters/gff/gtf_filter_by_attribute_values_list.xml
+++ b/tools/filters/gff/gtf_filter_by_attribute_values_list.xml
@@ -1,4 +1,4 @@
-<tool id="gtf_filter_by_attribute_values_list" name="Filter GTF data by attribute values_list" version="0.1">
+<tool id="gtf_filter_by_attribute_values_list" name="Filter GTF data by attribute values_list" version="0.2">
     <description></description>
     <command>
         python '$__tool_directory__/gtf_filter_by_attribute_values_list.py' '$input' '$attribute_name' '$ids' '$output'

From 610e31147a4e29ee1a4c5f5cdb3ba8005eeb2471 Mon Sep 17 00:00:00 2001
From: Matthias Bernt <m.bernt@ufz.de>
Date: Sat, 13 Feb 2021 23:04:17 +0100
Subject: [PATCH 9/9] add python requirement

---
 tools/filters/gff/gtf_filter_by_attribute_values_list.xml | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/tools/filters/gff/gtf_filter_by_attribute_values_list.xml b/tools/filters/gff/gtf_filter_by_attribute_values_list.xml
index bf014ae8b7c..b1063b78827 100644
--- a/tools/filters/gff/gtf_filter_by_attribute_values_list.xml
+++ b/tools/filters/gff/gtf_filter_by_attribute_values_list.xml
@@ -1,5 +1,8 @@
 <tool id="gtf_filter_by_attribute_values_list" name="Filter GTF data by attribute values_list" version="0.2">
     <description></description>
+    <requirements>
+        <requirement type="package" version="3.6">python</requirement>
+    </requirements>
     <command><![CDATA[
 python '$__tool_directory__/gtf_filter_by_attribute_values_list.py' '$input' $attribute_name '$ids' '$output'
     ]]></command>