From fbcf3e2bbc3c70fd154f713dcfef186406c301e0 Mon Sep 17 00:00:00 2001
From: Nicola Soranzo <soranzo@crs4.it>
Date: Tue, 28 Oct 2014 16:57:49 +0100
Subject: [PATCH] Make stripping and condensing optional.

---
 tools/filters/convert_characters.py  | 60 +++++++++++++---------------
 tools/filters/convert_characters.xml | 26 ++++++++++--
 2 files changed, 49 insertions(+), 37 deletions(-)

diff --git a/tools/filters/convert_characters.py b/tools/filters/convert_characters.py
index 76574adc430..e93b2f223be 100644
--- a/tools/filters/convert_characters.py
+++ b/tools/filters/convert_characters.py
@@ -1,30 +1,18 @@
 #!/usr/bin/env python
 #By, Guruprasad Ananda.
 
+import optparse
 import re
-import sys
 
-
-def stop_err(msg):
-    sys.stderr.write(msg)
-    sys.exit()
-
-
-def main():
-    if len(sys.argv) != 4:
-        stop_err("usage: convert_characters infile from_char outfile")
-
-    try:
-        fin = open(sys.argv[1], 'r')
-    except:
-        stop_err("Input file cannot be opened for reading.")
-
-    from_char = sys.argv[2]
-
-    try:
-        fout = open(sys.argv[3], 'w')
-    except:
-        stop_err("Output file cannot be opened for writing.")
+def __main__():
+    parser = optparse.OptionParser()
+    parser.add_option('--strip', action='store_true',
+                      help='strip leading and trailing whitespaces')
+    parser.add_option('--condense', action='store_true',
+                      help='condense consecutive delimiters')
+    (options, args) = parser.parse_args()
+    if len(args) != 3:
+        parser.error("usage: convert_characters.py infile from_char outfile")
 
     char_dict = {
         'T': '\t',
@@ -38,20 +26,26 @@ def main():
         'Sc': ';'
     }
     # regexp to match 1 or more occurences.
-    from_ch = char_dict[from_char] + '+'
+    from_char = args[1]
+    from_ch = char_dict[from_char]
+    if options.condense:
+        from_ch += '+'
+
     skipped = 0
+    with open(args[0], 'rU') as fin:
+        with open(args[2], 'w') as fout:
+            for line in fin:
+                if options.strip:
+                    line = line.strip()
+                else:
+                    line = line.rstrip('\n')
+                try:
+                    fout.write("%s\n" % (re.sub(from_ch, '\t', line)))
+                except:
+                    skipped += 1
 
-    for line in fin:
-        line = line.strip()
-        try:
-            fout.write("%s\n" % (re.sub(from_ch, '\t', line)))
-        except:
-            skipped += 1
-
-    fin.close()
-    fout.close()
     if skipped:
         print "Skipped %d lines as invalid." % skipped
 
 if __name__ == "__main__":
-    main()
+    __main__()
diff --git a/tools/filters/convert_characters.xml b/tools/filters/convert_characters.xml
index 9507f792d8d..2de7602b9c4 100644
--- a/tools/filters/convert_characters.xml
+++ b/tools/filters/convert_characters.xml
@@ -1,6 +1,15 @@
 <tool id="Convert characters1" name="Convert">
   <description>delimiters to TAB</description>
-  <command interpreter="python">convert_characters.py $input $convert_from $out_file1</command>
+  <command interpreter="python">
+convert_characters.py
+#if $strip
+  --strip
+#end if
+#if $condense
+  --condense
+#end if
+$input $convert_from $out_file1
+  </command>
   <inputs>
     <param name="convert_from" type="select" label="Convert all">
       <option value="s">Whitespaces</option>
@@ -15,19 +24,28 @@
       <option value="Sc">Semicolons</option>
     </param>
     <param format="txt" name="input" type="data" label="in Dataset"/>
+    <param name="strip" type="boolean" checked="true" label="Strip leading and trailing whitespaces" />
+    <param name="condense" type="boolean" checked="true" label="Condense consecutive delimiters in one TAB" />
   </inputs>
   <outputs>
     <data format="tabular" name="out_file1" />
   </outputs>
+  <stdio>
+    <exit_code range="1:" level="fatal" />
+  </stdio>
   <tests>
     <test>
       <param name="convert_from" value="s"/>
       <param name="input" value="1.bed"/>
+      <param name="strip" value="true" />
+      <param name="condense" value="true" />
       <output name="out_file1" file="eq-convert.dat"/>
     </test>
     <test>
       <param name="convert_from" value="s"/>
       <param name="input" value="a.txt"/>
+      <param name="strip" value="true" />
+      <param name="condense" value="true" />
       <output name="out_file1" file="a.tab"/>
     </test>
   </tests>
@@ -35,7 +53,7 @@
 
 **What it does**
 
-Converts all delimiters of a specified type into TABs.  Consecutive characters are condensed. For example, if columns are separated by 5 spaces they will converted into 1 tab.
+Converts all delimiters of a specified type into TABs. Consecutive delimiters can be condensed in a single TAB.
 
 -----
 
@@ -48,12 +66,12 @@ Converts all delimiters of a specified type into TABs.  Consecutive characters a
     chrX|151559494|151559583|NM_018558_exon_1_0_chrX_151559495_f|0|+
     chrX|151564643|151564711|NM_018558_exon_2_0_chrX_151564644_f||||0|+
 
-- Converting all pipe delimiters of the above file to TABs will get::
+- Converting all pipe delimiters of the above file to TABs and condensing delimiters will get::
 
     chrX  151283558  151283724  NM_000808_exon_8_0_chrX_151283559_r  0  -
     chrX  151370273  151370486  NM_000808_exon_9_0_chrX_151370274_r  0  -
     chrX  151559494  151559583  NM_018558_exon_1_0_chrX_151559495_f  0  +
     chrX  151564643  151564711  NM_018558_exon_2_0_chrX_151564644_f  0  +
 
-</help>
+  </help>
 </tool>