changeset 0:21d312776891 draft

planemo upload for repository https://github.com/mvdbeek/docker_scriptrunner/ commit 30f8264cdd67d40dec8acde6407f32152e6a29c1-dirty
author mvdbeek
date Sat, 09 Jul 2016 16:57:13 -0400
parents
children 315a7e9ed6eb
files README.md dockerfiles/build.sh dockerfiles/deploy.sh dockerfiles/r-bioperl-python/Dockerfile dockerfiles/scriptrunner/Dockerfile macros.xml planemo_serve_script.sh scriptrunner.py scriptrunner.xml test-data/docker_images.loc.sample test-data/tf2_test.html test-data/tf2_test_in.xls test-data/tf2_test_out.xls test-data/tf2_test_runme.py test.sh tool-data/docker_images.loc.sample tool_data_table_conf.xml.sample tool_data_table_conf.xml.sample.test tool_dependencies.xml
diffstat 18 files changed, 910 insertions(+), 0 deletions(-) [+]
line wrap: on
line diff
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/README.md	Sat Jul 09 16:57:13 2016 -0400
@@ -0,0 +1,6 @@
+Docker scriptrunner for galaxy
+-----------------------------
+
+This is a [galaxy](https://github.com/galaxyproject/galaxy) tool that allows
+users to submit random scripts.
+You can install it on your galaxy server from the [galaxy toolshed]()
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/dockerfiles/build.sh	Sat Jul 09 16:57:13 2016 -0400
@@ -0,0 +1,3 @@
+#!/usr/bin/env bash
+docker build -t artbio/scriptrunner scriptrunner && \
+docker build -t artbio/scriptrunner-r-bioperl-python r-bioperl-python
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/dockerfiles/deploy.sh	Sat Jul 09 16:57:13 2016 -0400
@@ -0,0 +1,5 @@
+#!/usr/bin/env bash
+LOGIN="docker login -e=$DOCKER_EMAIL -u=$DOCKER_USERNAME -p=$DOCKER_PASSWORD" && \
+$LOGIN || (sleep 5s && $LOGIN || echo "login failed twice, quitting" && exit 1) && \
+docker push artbio/scriptrunner || (sleep 5s && docker push artbio/scriptrunner || echo "push failed twice, quitting" && exit 1) && \
+docker push artbio/scriptrunner-r-bioperl-python || (sleep 5s && docker push artbio/scriptrunner-r-bioperl-python || echo "push failed twice, quitting" && exit 1)
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/dockerfiles/r-bioperl-python/Dockerfile	Sat Jul 09 16:57:13 2016 -0400
@@ -0,0 +1,13 @@
+# Script runner base image
+#
+# VERSION       0.3
+# This Dockerfile is the base system for executing scripts with the docker scriptrunner.
+
+FROM artbio/scriptrunner
+
+MAINTAINER Marius van den Beek, m.vandenbeek@gmail.com
+
+USER root
+RUN su - galaxy -c 'bash -c ". /home/galaxy/conda/bin/activate default && \
+    conda install -y ncurses biopython perl-bioperl bedtools bioconductor-deseq2 bioconductor-edger numpy scipy r-reshape"' && \
+    chown -R galaxy:users /home/galaxy && chmod g=rwx /home/galaxy
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/dockerfiles/scriptrunner/Dockerfile	Sat Jul 09 16:57:13 2016 -0400
@@ -0,0 +1,38 @@
+# Script runner base image
+#
+# VERSION       0.3
+# This Dockerfile is the base system for executing scripts with the docker scriptrunner.
+
+FROM toolshed/requirements
+
+MAINTAINER Marius van den Beek, m.vandenbeek@gmail.com
+
+# make sure the package repository is up to date
+ENV DEBIAN_FRONTEND noninteractive
+RUN apt-get -qq update
+
+# Install all requirements that are recommend by the Galaxy project
+RUN apt-get install -y --no-install-recommends bc imagemagick libfreetype6-dev wget curl \
+libcurl4-openssl-dev libssl-dev libreadline-dev libxml2-dev liblzma-dev \
+libpcre3-dev libbz2-dev libx11-dev 
+
+RUN mkdir /var/home && \
+    useradd galaxy --create-home --home-dir /home/galaxy  --uid 1450 && \
+    chown -R galaxy:galaxy /home/galaxy/
+USER galaxy
+WORKDIR /home/galaxy
+ENV CONDA_PREFIX="/home/galaxy/conda" \
+    PATH="/home/galaxy/conda/envs/default/bin:/home/galaxy/conda/bin:$PATH"
+
+RUN curl https://repo.continuum.io/miniconda/Miniconda-latest-Linux-x86_64.sh -L -o miniconda.sh 
+RUN bash miniconda.sh -b -p "$CONDA_PREFIX" && rm miniconda.sh
+RUN bash -c "conda create -y -q -c bioconda --name default samtools==0.1.19 bcftools && \
+    conda config --add channels r && \
+    conda config --add channels bioconda && \
+    conda config --add channels iuc && \
+    . activate default && \
+    conda install ghostscript coreutils"
+USER root
+RUN chown -R galaxy:users /home/galaxy && \
+    chmod g=rwx /home/galaxy && \
+    usermod -G users galaxy
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/macros.xml	Sat Jul 09 16:57:13 2016 -0400
@@ -0,0 +1,187 @@
+
+<macro>
+    <xml name="help_macro">
+    <help>
+.. class:: warningmark
+
+**Details and attribution** Docker_toolfactory_
+
+**If you find a bug** please raise an issue at the bitbucket repository Issues_
+
+**What it does** This tool enables a user to paste and submit an arbitrary R/python/perl script to Galaxy.
+
+**Input options** This version is limited to simple transformation or reporting requiring only a single input file selected from the history.
+
+**Output options** Optional script outputs include one single new history tabular file, or for scripts that create multiple outputs,
+a new HTML report linking all the files and images created by the script can be automatically generated.
+
+.. class:: warningmark
+
+**Note to system administrators** This tool uses docker containers as protection against malicious scripts. It should only be installed on private/personnal Galaxy instances.
+
+.. class:: warningmark
+
+**Use on public servers**  is STRONGLY discouraged for obvious reasons
+
+The tools generated by this tool will run just as securely as any other normal installed Galaxy tool but like any other new tools, should always be checked carefully before installation.
+We recommend that you follow the good code hygiene practices associated with safe toolshed.
+
+**Scripting conventions** The pasted script will be executed with the path to the (optional) input tabular data file path or NONE if you do not select one, and the path to the optional
+output file or None if none is wanted, as the first and second command line parameters. The script must deal appropriately with these - see Rscript examples below.
+Note that if an optional HTML output is selected, all the output files created by the script will be nicely presented as links, with pdf images linked as thumbnails in that output.
+This can be handy for complex scripts creating lots of output.
+
+**Examples**
+<![CDATA[
+
+Each of these following trivial examples can be cut and pasted into the script box for testing.
+Please make sure you choose the appropriate interpreter and upload and select a suitable small matching test data input
+
+A simple Rscript "filter" showing how the command line parameters can be handled, takes an input file, does something (transpose in this case) and 
+writes the results to a new tabular file. Note the use of colClasses to ensure that no fiddling takes place with numeric values by treating everything
+as a string::
+
+ # transpose a tabular input file and write as a tabular output file
+ ourargs = commandArgs(TRUE)
+ inf = ourargs[1]
+ outf = ourargs[2]
+ inp = read.table(inf,head=F,row.names=NULL,sep='\t',colClasses="character")
+ outp = t(inp)
+ write.table(outp,outf, quote=FALSE, sep="\t",row.names=F,col.names=F)
+
+Calculate a multiple test adjusted p value from a column of p values - for this script to be useful, it needs the right column for the input to be specified in the code for the
+given input file type(s) specified when the tool is generated ::
+
+ # use p.adjust - assumes a HEADER row and column 1 - please fix for any real use
+ column = 1 # adjust if necessary for some other kind of input
+ ourargs = commandArgs(TRUE)
+ inf = ourargs[1]
+ outf = ourargs[2]
+ inp = read.table(inf,head=T,row.names=NULL,sep='\t')
+ p = inp[,column]
+ q = p.adjust(p,method='BH')
+ outp = cbind(inp,'BH Adjusted p-value'=q)
+ write.table(outp,outf, quote=FALSE, sep="\t",row.names=F,col.names=T) 
+
+
+A demonstration Rscript example takes no input file but generates some random data based pdf images
+You must make sure the option to create an HTML output file is
+turned on for this to work. Images (pdf) are linked via thumbnails and
+all files have a link on the resulting HTML page::
+
+ # note this script takes NO input or output because it generates random data
+ for (i in 1:10) {
+    foo = runif(100)
+    bar = rnorm(100)
+    bar = foo + 0.05*bar
+    pdf(paste('yet',i,"anotherplot.pdf",sep='_'))
+    plot(foo,bar,main=paste("Foo by Bar plot #",i),col="maroon", pch=3,cex=0.6)
+    dev.off()
+    foo = data.frame(a=runif(100),b=runif(100),c=runif(100),d=runif(100),e=runif(100),f=runif(100))
+    bar = as.matrix(foo)
+    pdf(paste('yet',i,"anotherheatmap.pdf",sep='_'))
+    heatmap(bar,main='Random Heatmap')
+    dev.off()
+ }
+
+# A slight variation taking an input tabular file from which we read the first number as nreps
+# his script takes a single parameter  
+# number of replicates  
+
+.. code-block:: R  
+
+  ourargs = commandArgs(TRUE)
+  infname = ourargs[1]
+  nreps = read.table(infname,head=F)
+  nreps = unlist(nreps)[1]
+  nreps = max(c(1,nreps))
+  nreps = min(c(20,nreps))
+  print(paste("Using nreps=",nreps))
+  for (i in 1:nreps) {
+     foo = runif(100)
+     bar = rnorm(100)
+     bar = foo + 0.2*bar
+     pdf(paste("yet",i,"anotherplot.pdf",sep="_"))
+     plot(foo,bar,main=paste("Foo by Bar plot ",i),col="maroon", pch=3,cex=0.6)
+     dev.off()
+     foo = data.frame(a=runif(100),b=runif(100),c=runif(100),d=runif(100),e=runif(100),f=runif(100))
+     bar = as.matrix(foo)
+     pdf(paste("yet",i,"anotherheatmap.pdf",sep="_"))
+     heatmap(bar,main="Random Heatmap")
+     dev.off()
+  }
+ 
+A Python example that reverses each row of a tabular file (you'll need to remove the leading spaces 
+for this to work if cut and pasted into the script box)::
+
+ # reverse order of columns in a tabular file
+ import sys
+ inp = sys.argv[1]
+ outp = sys.argv[2]
+ i = open(inp,'r')
+ o = open(outp,'w')
+ for row in i:
+     rs = row.rstrip().split('\t')
+     rs.reverse()
+     o.write('\t'.join(rs))
+     o.write('\n')
+ i.close()
+ o.close()
+ 
+A trivial shell script example to show that it works::
+
+ #!/bin/bash
+ INF=$1
+ OUTF=$2
+ cut -c2,4,6,8,10,12 $INF > $OUTF 
+
+A trivial perl script example to show that even perl works::
+
+ #
+ # change all occurances of a string in a file to another string
+ #
+ $oldfile = $ARGV[0];
+ $newfile = $ARGV[1];
+ $old = "gene";
+ $new = "foo";
+ open(OF, $oldfile);
+ open(NF, ">$newfile");
+ # read in each line of the file
+ while ($line = <OF>) {
+    $line =~ s/$old/$new/;
+    print NF $line;
+ }
+ close(OF);
+ close(NF);
+
+]]>
+
+**Citation**
+
+
+Paper_ :
+
+Creating re-usable tools from scripts: The Galaxy Tool Factory
+Ross Lazarus; Antony Kaspi; Mark Ziemann; The Galaxy Team
+Bioinformatics 2012; doi: 10.1093/bioinformatics/bts573
+
+
+**Licensing** 
+
+Copyright Ross Lazarus (ross period lazarus at gmail period com) May 2012
+All rights reserved.
+Licensed under the LGPL_
+
+.. _LGPL: http://www.gnu.org/copyleft/lesser.html
+.. _Docker_toolfactory:  https://bitbucket.org/mvdbeek/dockertoolfactory
+.. _Issues:  https://bitbucket.org/mvdbeek/dockertoolfactory/issues
+.. _Paper: http://bioinformatics.oxfordjournals.org/cgi/reprint/bts573?ijkey=lczQh1sWrMwdYWJ&amp;keytype=ref
+
+    </help>
+<citations>
+  <citation type="doi">10.1093/bioinformatics/bts573</citation>
+</citations>
+    </xml>
+    <xml name="test_data_macro">
+    </xml>
+</macro>
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/planemo_serve_script.sh	Sat Jul 09 16:57:13 2016 -0400
@@ -0,0 +1,1 @@
+planemo serve --galaxy_branch dev --conda_prefix=/conda --conda_dependency_resolution --profile tf --port 80 --host 0.0.0.0 --job_config_file ~/mydisk/job_conf.xml --tool_data_table tool_data_table_conf.xml.sample.test
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/scriptrunner.py	Sat Jul 09 16:57:13 2016 -0400
@@ -0,0 +1,436 @@
+# DockerToolFactory.py
+# see https://github.com/mvdbeek/scriptrunner
+
+import sys 
+import shutil 
+import subprocess 
+import os 
+import time 
+import tempfile 
+import argparse
+import getpass
+import tarfile
+import re
+import shutil
+import math
+import fileinput
+from os.path import abspath 
+
+
+progname = os.path.split(sys.argv[0])[1] 
+verbose = False 
+debug = False
+
+def timenow():
+    """return current time as a string
+    """
+    return time.strftime('%d/%m/%Y %H:%M:%S', time.localtime(time.time()))
+
+html_escape_table = {
+     "&": "&amp;",
+     ">": "&gt;",
+     "<": "&lt;",
+     "$": "\$"
+     }
+
+def html_escape(text):
+     """Produce entities within text."""
+     return "".join(html_escape_table.get(c,c) for c in text)
+
+def cmd_exists(cmd):
+     return subprocess.call("type " + cmd, shell=True, 
+           stdout=subprocess.PIPE, stderr=subprocess.PIPE) == 0
+
+def construct_bind(host_path, container_path=False, binds=None, ro=True):
+    #TODO remove container_path if it's alwyas going to be the same as host_path
+    '''build or extend binds dictionary with container path. binds is used
+    to mount all files using the docker-py client.'''
+    if not binds:
+        binds={}
+    if isinstance(host_path, list):
+        for k,v in enumerate(host_path):
+            if not container_path:
+                container_path=host_path[k]
+            binds[host_path[k]]={'bind':container_path, 'ro':ro}
+            container_path=False #could be more elegant
+        return binds
+    else:
+        if not container_path:
+            container_path=host_path
+        binds[host_path]={'bind':container_path, 'ro':ro}
+        return binds
+
+def switch_to_docker(opts):
+    import docker #need local import, as container does not have docker-py
+    user_id = os.getuid()
+    group_id = os.getgid()
+    docker_client=docker.Client()
+    toolfactory_path=abspath(sys.argv[0])
+    binds=construct_bind(host_path=opts.script_path, ro=False)
+    binds=construct_bind(binds=binds, host_path=abspath(opts.output_dir), ro=False)
+    if len(opts.input_tab)>0:
+        binds=construct_bind(binds=binds, host_path=opts.input_tab, ro=True)
+    if not opts.output_tab == 'None':
+        binds=construct_bind(binds=binds, host_path=opts.output_tab, ro=False)
+    if opts.make_HTML:
+        binds=construct_bind(binds=binds, host_path=opts.output_html, ro=False)
+    binds=construct_bind(binds=binds, host_path=toolfactory_path)
+    volumes=binds.keys()
+    sys.argv=[abspath(opts.output_dir) if sys.argv[i-1]=='--output_dir' else arg for i,arg in enumerate(sys.argv)] ##inject absolute path of working_dir
+    cmd=['python', '-u']+sys.argv+['--dockerized', '1', "--user_id", str(user_id), "--group_id", str(group_id)]
+    image_exists = [ True for image in docker_client.images() if opts.docker_image in image['RepoTags'] ]
+    if not image_exists:
+        docker_client.pull(opts.docker_image)
+    container=docker_client.create_container(
+        image=opts.docker_image,
+        volumes=volumes,
+        command=cmd
+        )
+    docker_client.start(container=container[u'Id'], binds=binds)
+    docker_client.wait(container=container[u'Id'])
+    logs=docker_client.logs(container=container[u'Id'])
+    print "".join([log for log in logs])
+    docker_client.remove_container(container[u'Id'])
+
+class ScriptRunner:
+    """class is a wrapper for an arbitrary script
+    """
+
+    def __init__(self,opts=None,treatbashSpecial=True, image_tag='base'):
+        """
+        cleanup inputs, setup some outputs
+        
+        """
+        self.opts = opts
+        self.scriptname = 'script'
+        self.useIM = cmd_exists('convert')
+        self.useGS = cmd_exists('gs')
+        self.temp_warned = False # we want only one warning if $TMP not set
+        self.treatbashSpecial = treatbashSpecial
+        self.image_tag = image_tag
+        os.chdir(abspath(opts.output_dir))
+        self.thumbformat = 'png'
+        s = open(self.opts.script_path,'r').readlines()
+        s = [x.rstrip() for x in s] # remove pesky dos line endings if needed
+        self.script = '\n'.join(s)
+        fhandle,self.sfile = tempfile.mkstemp(prefix='script',suffix=".%s" % (opts.interpreter))
+        tscript = open(self.sfile,'w') # use self.sfile as script source for Popen
+        tscript.write(self.script)
+        tscript.close()
+        self.indentedScript = '\n'.join([' %s' % html_escape(x) for x in s]) # for restructured text in help
+        self.escapedScript = '\n'.join([html_escape(x) for x in s])
+        self.elog = os.path.join(self.opts.output_dir,"%s_error.log" % self.scriptname)
+        if opts.output_dir: # may not want these complexities
+            self.tlog = os.path.join(self.opts.output_dir,"%s_runner.log" % self.scriptname)
+            art = '%s.%s' % (self.scriptname,opts.interpreter)
+            artpath = os.path.join(self.opts.output_dir,art) # need full path
+            artifact = open(artpath,'w') # use self.sfile as script source for Popen
+            artifact.write(self.script)
+            artifact.close()
+        self.cl = []
+        self.html = []
+        a = self.cl.append
+        a(opts.interpreter)
+        if self.treatbashSpecial and opts.interpreter in ['bash','sh']:
+            a(self.sfile)
+        else:
+            a('-') # stdin
+	for input in opts.input_tab:
+	  a(input) 
+        if opts.output_tab == 'None': #If tool generates only HTML, set output name to toolname
+            a(str(self.scriptname)+'.out')
+        a(opts.output_tab)
+	for param in opts.additional_parameters:
+          param, value=param.split(',')
+          a('--'+param)
+          a(value)
+        self.outFormats = opts.output_format
+        self.inputFormats = [formats for formats in opts.input_formats]
+        self.test1Input = '%s_test1_input.xls' % self.scriptname
+        self.test1Output = '%s_test1_output.xls' % self.scriptname
+        self.test1HTML = '%s_test1_output.html' % self.scriptname
+
+
+    def compressPDF(self,inpdf=None,thumbformat='png'):
+        """need absolute path to pdf
+           note that GS gets confoozled if no $TMP or $TEMP
+           so we set it
+        """
+        assert os.path.isfile(inpdf), "## Input %s supplied to %s compressPDF not found" % (inpdf,self.myName)
+        hlog = os.path.join(self.opts.output_dir,"compress_%s.txt" % os.path.basename(inpdf))
+        sto = open(hlog,'a')
+        our_env = os.environ.copy()
+        our_tmp = our_env.get('TMP',None)
+        if not our_tmp:
+            our_tmp = our_env.get('TEMP',None)
+        if not (our_tmp and os.path.exists(our_tmp)):
+            newtmp = os.path.join(self.opts.output_dir,'tmp')
+            try:
+                os.mkdir(newtmp)
+            except:
+                sto.write('## WARNING - cannot make %s - it may exist or permissions need fixing\n' % newtmp)
+            our_env['TEMP'] = newtmp
+            if not self.temp_warned:
+               sto.write('## WARNING - no $TMP or $TEMP!!! Please fix - using %s temporarily\n' % newtmp)
+               self.temp_warned = True          
+        outpdf = '%s_compressed' % inpdf
+        cl = ["gs", "-sDEVICE=pdfwrite", "-dNOPAUSE", "-dUseCIEColor", "-dBATCH","-dPDFSETTINGS=/printer", "-sOutputFile=%s" % outpdf,inpdf]
+        x = subprocess.Popen(cl,stdout=sto,stderr=sto,cwd=self.opts.output_dir,env=our_env)
+        retval1 = x.wait()
+        sto.close()
+        if retval1 == 0:
+            os.unlink(inpdf)
+            shutil.move(outpdf,inpdf)
+            os.unlink(hlog)
+        hlog = os.path.join(self.opts.output_dir,"thumbnail_%s.txt" % os.path.basename(inpdf))
+        sto = open(hlog,'w')
+        outpng = '%s.%s' % (os.path.splitext(inpdf)[0],thumbformat)
+        cl2 = ['convert', inpdf, outpng]
+        x = subprocess.Popen(cl2,stdout=sto,stderr=sto,cwd=self.opts.output_dir,env=our_env)
+        retval2 = x.wait()
+        sto.close()
+        if retval2 == 0:
+             os.unlink(hlog)
+        retval = retval1 or retval2
+        return retval
+
+
+    def getfSize(self,fpath,outpath):
+        """
+        format a nice file size string
+        """
+        size = ''
+        fp = os.path.join(outpath,fpath)
+        if os.path.isfile(fp):
+            size = '0 B'
+            n = float(os.path.getsize(fp))
+            if n > 2**20:
+                size = '%1.1f MB' % (n/2**20)
+            elif n > 2**10:
+                size = '%1.1f KB' % (n/2**10)
+            elif n > 0:
+                size = '%d B' % (int(n))
+        return size
+
+    def makeHtml(self):
+        """ Create an HTML file content to list all the artifacts found in the output_dir
+        """
+
+        galhtmlprefix = """<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> 
+        <html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en"> 
+        <head> <meta http-equiv="Content-Type" content="text/html; charset=utf-8" /> 
+        <meta name="generator" content="Galaxy %s tool output - see http://g2.trac.bx.psu.edu/" /> 
+        <title></title> 
+        <link rel="stylesheet" href="/static/style/base.css" type="text/css" /> 
+        </head> 
+        <body> 
+        <div class="toolFormBody"> 
+        """ 
+        galhtmlattr = """<hr/><div class="infomessage">This tool (%s) was generated by the <a href="https://bitbucket.org/fubar/galaxytoolfactory/overview">Galaxy Tool Factory</a></div><br/>""" 
+        galhtmlpostfix = """</div></body></html>\n"""
+
+        flist = os.listdir(self.opts.output_dir)
+        flist = [x for x in flist if x <> 'Rplots.pdf']
+        flist.sort()
+        html = []
+        html.append(galhtmlprefix % progname)
+        html.append('<div class="infomessage">Galaxy Tool "%s" run at %s</div><br/>' % (self.scriptname,timenow()))
+        fhtml = []
+        if len(flist) > 0:
+            logfiles = [x for x in flist if x.lower().endswith('.log')] # log file names determine sections
+            logfiles.sort()
+            logfiles = [x for x in logfiles if abspath(x) <> abspath(self.tlog)]
+            logfiles.append(abspath(self.tlog)) # make it the last one
+            pdflist = []
+            npdf = len([x for x in flist if os.path.splitext(x)[-1].lower() == '.pdf'])
+            for rownum,fname in enumerate(flist):
+                dname,e = os.path.splitext(fname)
+                sfsize = self.getfSize(fname,self.opts.output_dir)
+                if e.lower() == '.pdf' : # compress and make a thumbnail
+                    thumb = '%s.%s' % (dname,self.thumbformat)
+                    pdff = os.path.join(self.opts.output_dir,fname)
+                    retval = self.compressPDF(inpdf=pdff,thumbformat=self.thumbformat)
+                    if retval == 0:
+                        pdflist.append((fname,thumb))
+                    else:
+                        pdflist.append((fname,fname))
+                if (rownum+1) % 2 == 0:
+                    fhtml.append('<tr class="odd_row"><td><a href="%s">%s</a></td><td>%s</td></tr>' % (fname,fname,sfsize))
+                else:
+                    fhtml.append('<tr><td><a href="%s">%s</a></td><td>%s</td></tr>' % (fname,fname,sfsize))
+            for logfname in logfiles: # expect at least tlog - if more
+                if abspath(logfname) == abspath(self.tlog): # handled later
+                    sectionname = 'All tool run'
+                    if (len(logfiles) > 1):
+                        sectionname = 'Other'
+                    ourpdfs = pdflist
+                else:
+                    realname = os.path.basename(logfname)
+                    sectionname = os.path.splitext(realname)[0].split('_')[0] # break in case _ added to log
+                    ourpdfs = [x for x in pdflist if os.path.basename(x[0]).split('_')[0] == sectionname]
+                    pdflist = [x for x in pdflist if os.path.basename(x[0]).split('_')[0] <> sectionname] # remove
+                nacross = 1
+                npdf = len(ourpdfs)
+
+                if npdf > 0:
+                    nacross = math.sqrt(npdf) ## int(round(math.log(npdf,2)))
+                    if int(nacross)**2 != npdf:
+                        nacross += 1
+                    nacross = int(nacross)
+                    width = min(400,int(1200/nacross))
+                    html.append('<div class="toolFormTitle">%s images and outputs</div>' % sectionname)
+                    html.append('(Click on a thumbnail image to download the corresponding original PDF image)<br/>')
+                    ntogo = nacross # counter for table row padding with empty cells
+                    html.append('<div><table class="simple" cellpadding="2" cellspacing="2">\n<tr>')
+                    for i,paths in enumerate(ourpdfs): 
+                        fname,thumb = paths
+                        s= """<td><a href="%s"><img src="%s" title="Click to download a PDF of %s" hspace="5" width="%d" 
+                           alt="Image called %s"/></a></td>\n""" % (fname,thumb,fname,width,fname)
+                        if ((i+1) % nacross == 0):
+                            s += '</tr>\n'
+                            ntogo = 0
+                            if i < (npdf - 1): # more to come
+                               s += '<tr>'
+                               ntogo = nacross
+                        else:
+                            ntogo -= 1
+                        html.append(s)
+                    if html[-1].strip().endswith('</tr>'):
+                        html.append('</table></div>\n')
+                    else:
+                        if ntogo > 0: # pad
+                           html.append('<td>&nbsp;</td>'*ntogo)
+                        html.append('</tr></table></div>\n')
+                logt = open(logfname,'r').readlines()
+                logtext = [x for x in logt if x.strip() > '']
+                html.append('<div class="toolFormTitle">%s log output</div>' % sectionname)
+                if len(logtext) > 1:
+                    html.append('\n<pre>\n')
+                    html += logtext
+                    html.append('\n</pre>\n')
+                else:
+                    html.append('%s is empty<br/>' % logfname)
+        if len(fhtml) > 0:
+           fhtml.insert(0,'<div><table class="colored" cellpadding="3" cellspacing="3"><tr><th>Output File Name (click to view)</th><th>Size</th></tr>\n')
+           fhtml.append('</table></div><br/>')
+           html.append('<div class="toolFormTitle">All output files available for downloading</div>\n')
+           html += fhtml # add all non-pdf files to the end of the display
+        else:
+            html.append('<div class="warningmessagelarge">### Error - %s returned no files - please confirm that parameters are sane</div>' % self.opts.interpreter)
+        html.append(galhtmlpostfix)
+        htmlf = file(self.opts.output_html,'w')
+        htmlf.write('\n'.join(html))
+        htmlf.write('\n')
+        htmlf.close()
+        self.html = html
+
+
+    def run(self):
+        """
+        scripts must be small enough not to fill the pipe!
+        """
+        if self.treatbashSpecial and self.opts.interpreter in ['bash','sh']:
+          retval = self.runBash()
+        else:
+            if self.opts.output_dir:
+                ste = open(self.elog,'w')
+                sto = open(self.tlog,'w')
+                sto.write('## Toolfactory generated command line = %s\n' % ' '.join(self.cl))
+                sto.flush()
+                p = subprocess.Popen(self.cl,shell=False,stdout=sto,stderr=ste,stdin=subprocess.PIPE,cwd=self.opts.output_dir)
+            else:
+                p = subprocess.Popen(self.cl,shell=False,stdin=subprocess.PIPE)
+            p.stdin.write(self.script)
+            p.stdin.close()
+            retval = p.wait()
+            if self.opts.output_dir:
+                sto.close()
+                ste.close()
+                err = open(self.elog,'r').readlines()
+                if retval <> 0 and err: # problem
+                    print >> sys.stderr,err #same problem, need to capture docker stdin/stdout
+            if self.opts.make_HTML:
+                self.makeHtml()
+        return retval
+
+    def runBash(self):
+        """
+        cannot use - for bash so use self.sfile
+        """
+        if self.opts.output_dir:
+            s = '## Toolfactory generated command line = %s\n' % ' '.join(self.cl)
+            sto = open(self.tlog,'w')
+            sto.write(s)
+            sto.flush()
+            p = subprocess.Popen(self.cl,shell=False,stdout=sto,stderr=sto,cwd=self.opts.output_dir)
+        else:
+            p = subprocess.Popen(self.cl,shell=False)            
+        retval = p.wait()
+        if self.opts.output_dir:
+            sto.close()
+        if self.opts.make_HTML:
+            self.makeHtml()
+        return retval
+  
+
+def change_user_id(new_uid, new_gid):
+    """
+    To avoid issues with wrong user ids, we change the user id of the 'galaxy' user in the container
+    to the user id with which the script has been called initially.
+    """
+    cmd1 = ["/usr/sbin/usermod", "-d", "/var/home/galaxy", "galaxy"]
+    cmd2 = ["/usr/sbin/usermod", "-u", new_uid, "galaxy"]
+    cmd3 = ["/usr/sbin/groupmod", "-g", new_gid, "galaxy"]
+    cmd4 = ["/usr/sbin/usermod", "-d", "/home/galaxy", "galaxy"]
+    [subprocess.call(cmd) for cmd in [cmd1, cmd2, cmd3, cmd4]]
+
+
+def main():
+    u = """
+    This is a Galaxy wrapper. It expects to be called by a special purpose tool.xml as:
+    <command interpreter="python">rgBaseScriptWrapper.py --script_path "$scriptPath" --tool_name "foo" --interpreter "Rscript"
+    </command>
+    """
+    op = argparse.ArgumentParser()
+    a = op.add_argument
+    a('--docker_image',default=None)
+    a('--script_path',default=None)
+    a('--tool_name',default=None)
+    a('--interpreter',default=None)
+    a('--output_dir',default='./')
+    a('--output_html',default=None)
+    a('--input_tab',default='None', nargs='*')
+    a('--output_tab',default='None')
+    a('--user_email',default='Unknown')
+    a('--bad_user',default=None)
+    a('--make_HTML',default=None)
+    a('--new_tool',default=None)
+    a('--dockerized',default=0)
+    a('--group_id',default=None)
+    a('--user_id',default=None)
+    a('--output_format', default='tabular')
+    a('--input_format', dest='input_formats', action='append', default=[])
+    a('--additional_parameters', dest='additional_parameters', action='append', default=[])
+    opts = op.parse_args()
+    assert not opts.bad_user,'UNAUTHORISED: %s is NOT authorized to use this tool until Galaxy admin adds %s to admin_users in universe_wsgi.ini' % (opts.bad_user,opts.bad_user)
+    assert os.path.isfile(opts.script_path),'## Tool Factory wrapper expects a script path - eg --script_path=foo.R'
+    if opts.output_dir:
+        try:
+            os.makedirs(opts.output_dir)
+        except:
+            pass
+    if opts.dockerized==0:
+      switch_to_docker(opts)
+      return
+    change_user_id(opts.user_id, opts.group_id)
+    os.setgid(int(opts.group_id))
+    os.setuid(int(opts.user_id))
+    r = ScriptRunner(opts)
+    retcode = r.run()
+    os.unlink(r.sfile)
+    if retcode:
+        sys.exit(retcode) # indicate failure to job runner
+
+
+if __name__ == "__main__":
+    main()
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/scriptrunner.xml	Sat Jul 09 16:57:13 2016 -0400
@@ -0,0 +1,127 @@
+<tool id="docker_scriptrunner" name="docker scriptrunner" version="0.1.6">
+  <description>Runs scripts using docker</description>
+  <macros>
+    <import>macros.xml</import>
+  </macros>
+   <requirements>
+      <requirement type="package" version="0.4.0">docker-py</requirement>
+  </requirements>
+  <command>
+    python "$__tool_directory__/scriptrunner.py"
+    --script_path "$runme"
+    --interpreter "$interpreter"
+    --user_email "$__user_email__"
+    #if $generate_simple_output.make_TAB=="yes":
+       --output_tab "$tab_file"
+    #end if
+    #if $make_HTML.value=="yes":
+      --output_dir "$html_file.files_path" --output_html "$html_file" --make_HTML "yes"
+    #else:
+       --output_dir "."
+    #end if
+    #if $additional_parameters != 'None':
+      #for i in $additional_parameters:
+	--additional_parameters
+        "$i.param_name, $i.param_value"
+      #end for
+    #end if
+    #if $input_files != 'None':
+       --input_tab 
+	#for i in $input_files:
+	  $i.input
+	#end for
+        #for i in $input_files:
+          --input_format "Any"
+        #end for
+    #end if
+    --docker_image "$docker_image" 
+  </command>
+<configfiles>
+  <configfile name="runme">$dynScript</configfile>
+</configfiles>
+  <inputs>
+    <repeat name="input_files" title="Select input files">
+      <param name="input" format="data" type="data"  label="Select an input file from your history" optional="true" size="120"
+         help="Most scripts will need an input - your script MUST be ready for whatever format you choose"/>
+    </repeat>
+    <repeat name="additional_parameters" title="Set additional parameters">
+      <param name="param_name" type="text" value="parameter_name" label="Choose a parameter name" size="20">
+        <sanitizer invalid_char="">
+          <valid initial="string.letters,string.digits"><add value="@_."/></valid>>
+        </sanitizer>
+      </param>
+     <param name="param_value" type="text" value="parameter_value" label="Enter a parameter value" size="50">
+       <sanitizer>
+         <valid initial="string.printable">
+         </valid>
+         <mapping initial="none"/>
+       </sanitizer>
+     </param>
+    </repeat> 
+    <param name="make_HTML" type="select" label="Create an HTML report with links to all output files and thumbnail links to PDF images" size="60"
+         help="Recommended for presenting complex outputs in an accessible manner. Turn off for simple tools so they just create one output">
+        <option value="yes">Yes, arrange all outputs produced by my script as an HTML output</option>
+        <option value="" selected="true">No, no HTML output file thanks</option>
+    </param>
+    <conditional name="generate_simple_output">
+      <param name="make_TAB" type="select" label="Create a new history output alongside the HTML file specified above" 
+           help="This is useful if your script creates a single new tabular file you want to appear in the history after the tool executes">
+          <option value="yes" selected="true">My script writes to a new history output</option>
+          <option value="">I do not want a new history output file</option>
+      </param>
+      <when value="yes">
+        <param name="out_format" type="select" label="Select the datatype that your tool/script produces" help="If your datatype is not listed here, it has to be added in galaxy's datatypes_conf.xml">
+          <options from_parameter="tool.app.datatypes_registry.upload_file_formats">
+            <column name="value" index="0"/>
+          </options>
+        </param>
+      </when>
+      <when value="">
+      </when>
+    </conditional>
+    <param name="interpreter" type="select" label="Select the interpreter for your code">
+        <option value="Rscript" selected="true">Rscript</option>
+        <option value="python">python</option>
+        <option value="perl">perl</option>
+        <option value="sh">sh</option>
+    </param>   
+    <param help="The script you are pasting will be run in this image. Make sure it contains the necessary tools" label="Select a docker image" name="docker_image" type="select">
+        <options from_data_table="docker_scriptrunner_images"/>
+    </param>
+    <param name="dynScript" label="Copy and paste the script to be executed here" type="text" value="" area="True" size="8x120"  
+      help="Script must deal with two command line parameters: Path to input tabular file path (or 'None' if none selected) and path to output tabular history file (or 'None').">
+      <sanitizer>
+         <valid initial="string.printable">
+         </valid>
+         <mapping initial="none"/>
+      </sanitizer>
+     </param>
+  </inputs>
+  <outputs>
+    <data format_source="input" name="tab_file">
+      <filter>generate_simple_output['make_TAB'] == "yes"</filter>
+        <actions>
+          <action type="format">
+                <option type="from_param" name="generate_simple_output.out_format" />
+          </action>
+        </actions>
+    </data>
+    <data format="html" name="html_file" >
+        <filter>make_HTML == "yes"</filter>
+    </data>
+ </outputs>
+<tests>
+    <test>
+        <param name='input_tab' value='tf2_test_in.xls' ftype='tabular' />
+        <param name="make_TAB" value="yes" />
+        <param name="make_HTML" value="yes" />
+        <param name="out_format" value="tabular" />
+        <param name="interpreter" value='python' />
+        <param name="runme" value="tf2_test_runme.py"/>
+        <output name='output1' file='tf2_test_out.xls' compare='diff' lines_diff = '10'/>
+        <output name='html_file' file="tf2_test.html" compare='diff' lines_diff = '10'/>
+    </test>
+</tests>
+<expand macro="help_macro" />
+<expand macro="test_data_macro" />
+</tool>
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/test-data/docker_images.loc.sample	Sat Jul 09 16:57:13 2016 -0400
@@ -0,0 +1,3 @@
+# Place here your docker images that usrs can select in the docker scriptrunner
+artbio/scriptrunner	artbio/scriptrunner
+artbio/scriptrunner-r-bioperl-python	artbio/scriptrunner-r-bioperl-python
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/test-data/tf2_test.html	Sat Jul 09 16:57:13 2016 -0400
@@ -0,0 +1,25 @@
+<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> 
+        <html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en"> 
+        <head> <meta http-equiv="Content-Type" content="text/html; charset=utf-8" /> 
+        <meta name="generator" content="Galaxy scriptrunner.py tool output - see http://g2.trac.bx.psu.edu/" /> 
+        <title></title> 
+        <link rel="stylesheet" href="/static/style/base.css" type="text/css" /> 
+        </head> 
+        <body> 
+        <div class="toolFormBody"> 
+        
+<div class="infomessage">Galaxy Tool "script" run at 09/07/2016 11:37:54</div><br/>
+<div class="toolFormTitle">script log output</div>
+script_error.log is empty<br/>
+<div class="toolFormTitle">Other log output</div>
+/tmp/tmpG7m9zp/job_working_directory/000/1/dataset_2_files/script_runner.log is empty<br/>
+<div class="toolFormTitle">All output files available for downloading</div>
+
+<div><table class="colored" cellpadding="3" cellspacing="3"><tr><th>Output File Name (click to view)</th><th>Size</th></tr>
+
+<tr><td><a href="script.python">script.python</a></td><td>0 B</td></tr>
+<tr class="odd_row"><td><a href="script_error.log">script_error.log</a></td><td>0 B</td></tr>
+<tr><td><a href="script_runner.log">script_runner.log</a></td><td>100 B</td></tr>
+</table></div><br/>
+</div></body></html>
+
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/test-data/tf2_test_in.xls	Sat Jul 09 16:57:13 2016 -0400
@@ -0,0 +1,3 @@
+col1	col2	col3
+r11	r12	r13
+r21	r22	r23
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/test-data/tf2_test_runme.py	Sat Jul 09 16:57:13 2016 -0400
@@ -0,0 +1,25 @@
+# bog standard argparse for 3 possible comma separated lists
+# followed by some silly reverse each row code provided as an example
+# you're supposed to replace it with your great code..
+import argparse
+import copy
+
+argp = argparse.ArgumentParser()
+argp.add_argument('--INNAMES', default=None)
+argp.add_argument('--INPATHS', default=None)
+argp.add_argument('--OUTPATH', default=None)
+argp.add_argument('--additional_parameters', default=[], action="append")
+argp.add_argument('otherargs', nargs=argparse.REMAINDER)
+args = argp.parse_args()
+fout = open(args.OUTPATH, 'w')
+sins = open(args.INPATHS.split(',')[0]).readlines()
+for i, sin in enumerate(sins):
+    row = sin.strip().split('\t')
+    rrow = copy.copy(row)
+    lrow = len(row)
+    if (lrow > 1):
+        for j in range(lrow):
+            rrow[j] = row[lrow - j - 1]
+        fout.write('\t'.join(rrow))
+        fout.write('\n')
+fout.close()
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/test.sh	Sat Jul 09 16:57:13 2016 -0400
@@ -0,0 +1,7 @@
+#!/usr/bin/env bash
+planemo test --galaxy_branch dev \
+             --conda_auto_init \
+             --conda_dependency_resolution \
+             --conda_auto_install \
+             --conda_ensure_channels scrapinghub \
+             --tool_data_table tool_data_table_conf.xml.sample.test
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/tool-data/docker_images.loc.sample	Sat Jul 09 16:57:13 2016 -0400
@@ -0,0 +1,3 @@
+# Place here your docker images that usrs can select in the docker scriptrunner
+artbio/scriptrunner	artbio/scriptrunner
+artbio/scriptrunner-r-bioperl-python	artbio/scriptrunner-r-bioperl-python
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/tool_data_table_conf.xml.sample	Sat Jul 09 16:57:13 2016 -0400
@@ -0,0 +1,7 @@
+<tables>
+    <!-- Place available docker images here -->
+    <table name="docker_scriptrunner_images" comment_char="#">
+        <columns>name, value</columns>
+        <file path="tool-data/docker_images.loc" />
+    </table>
+</tables>
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/tool_data_table_conf.xml.sample.test	Sat Jul 09 16:57:13 2016 -0400
@@ -0,0 +1,7 @@
+<tables>
+    <!-- Place available docker images here -->
+    <table name="docker_scriptrunner_images" comment_char="#">
+        <columns>name, value</columns>
+        <file path="${__HERE__}/test-data/docker_images.loc.sample" />
+    </table>
+</tables>
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/tool_dependencies.xml	Sat Jul 09 16:57:13 2016 -0400
@@ -0,0 +1,14 @@
+<?xml version="1.0"?>
+<tool_dependency>
+    <package name="docker-py" version="0.4.0">
+        <install version="1.0">
+            <actions>
+                <action type="setup_virtualenv">docker-py==0.4.0
+                </action>
+            </actions>
+        </install>
+    </package>
+        <readme>
+            Only Admins can use this tool generator but please do NOT install o
+       </readme>
+</tool_dependency>