<tool id="regexColumn1" name="Column Regex Find And Replace" version="0.1.0">
  <description></description>
  <command interpreter="python">regex.py --input $input --output $out_file1 --column $field
    #for $check in $checks:
    --pattern='$check.pattern' --replacement='$check.replacement'
    #end for
  </command>
  <inputs>
    <param format="tabular" name="input" type="data" label="Select cells from"/>
    <param name="field" label="using column" type="data_column" data_ref="input" />
    <repeat name="checks" title="Check">
      <param name="pattern" size="40" type="text" value="chr([0-9A-Za-z])+" label="Find Regex" help="here you can enter text or regular expression (for syntax check lower part of this frame)">
        <sanitizer>
          <valid>
            <add preset="string.printable"/>
            <remove value="&#92;" />
            <remove value="&apos;" />
          </valid>
          <mapping initial="none">
            <add source="&#92;" target="__backslash__" />
            <add source="&apos;" target="__sq__"/>
          </mapping>
        </sanitizer>
      </param>
      <param name="replacement" size="40" type="text" value="newchr\1" label="Replacement">
        <sanitizer>
          <valid>
            <add preset="string.printable"/>
            <remove value="&#92;" />
            <remove value="&apos;" />
          </valid>
          <mapping initial="none">
            <add source="&#92;" target="__backslash__" />
            <add source="&apos;" target="__sq__"/>
          </mapping>
        </sanitizer>      
      </param>
    </repeat>
  </inputs>
  <outputs>
    <data format="input" name="out_file1" metadata_source="input" />
  </outputs>
  <tests>
    <test>
      <param name="input" value="find_tabular_1.txt" ftype="tabular" />
      <param name="field" value="1" />
      <param name="pattern" value="moo"/>
      <param name="replacement" value="cow" />
      <output name="out_file1" file="replace_tabular_1.txt"/>
    </test>
  </tests>
  <help>

.. class:: warningmark

**This tool will attempt to reuse the metadata from your first input.** To change metadata assignments click on the "edit attributes" link of the history item generated by this tool.

.. class:: infomark

**TIP:** If your data is not TAB delimited, use *Text Manipulation-&gt;Convert*

-----

This tool goes line by line through the specified input file and
if the text in the selected column matches a specified regular expression pattern
replaces the text with the corresponding specified replacement.

This tool can be used to change between the chromosome naming conventions of UCSC and Ensembl.  

For example to remove the **chr** part of the reference sequence name in the first column of this GFF file::

 ##gff-version 2
 ##Date: Thu Mar 23 11:21:17 2006
 ##bed2gff.pl $Rev: 601 $
 ##Input file: ./database/files/61c6c604e0ef50b280e2fd9f1aa7da61.dat
 chr1	bed2gff	CCDS1000.1_cds_0_0_chr1_148325916_f	148325916	148325975	.	+	.	score "0";
 chr21	bed2gff	CCDS13614.1_cds_0_0_chr21_32707033_f	32707033	32707192	.	+	.	score "0";
 chrX	bed2gff	CCDS14606.1_cds_0_0_chrX_122745048_f	122745048	122745924	.	+	.	score "0";

Setting::

 using column: c1 
 Find Regex: chr([0-9]+|X|Y|M[Tt]?) 
 Replacement: \1 

produces::

 ##gff-version 2
 ##Date: Thu Mar 23 11:21:17 2006
 ##bed2gff.pl $Rev: 601 $
 ##Input file: ./database/files/61c6c604e0ef50b280e2fd9f1aa7da61.dat
 1    bed2gff CCDS1000.1_cds_0_0_chr1_148325916_f     148325916       148325975       .       +       .       score "0";
 21   bed2gff CCDS13614.1_cds_0_0_chr21_32707033_f    32707033        32707192        .       +       .       score "0";
 X    bed2gff CCDS14606.1_cds_0_0_chrX_122745048_f    122745048       122745924       .       +       .       score "0";


This tool uses Python regular expressions with the **re.sub()** function. 
More information about Python regular expressions can be found here:
http://docs.python.org/library/re.html.

The regex **chr([0-9]+|X|Y|M)** means start with text **chr** followed by either: one or more digits, or the letter X, or the letter Y, or the letter M (optionally followed by a single letter T or t).  
Note that the parentheses **()** capture patterns in the text that can be used in the replacement text by using a backslash-number reference:  **\\1**



Galaxy aggressively escapes input supplied to tools, so if something
is not working please let us know and we can look into whether this is
the cause. Also if you would like help constructing regular
expressions for your inputs, please let us know at help@msi.umn.edu.

</help>
</tool>