# HG changeset patch # User fubar # Date 1363914889 14400 # Node ID bf814ef1c04d4acdd69e28bd1b8d46f04d419996 testing diff -r 000000000000 -r bf814ef1c04d transposetest/test-data/transposetest_test1_input.xls --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/transposetest/test-data/transposetest_test1_input.xls Thu Mar 21 21:14:49 2013 -0400 @@ -0,0 +1,4 @@ +h1 h2 h3 h4 +r11 r12 r13 r14 +r21 r22 r23 r24 +r31 r32 r33 r34 diff -r 000000000000 -r bf814ef1c04d transposetest/test-data/transposetest_test1_output.xls --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/transposetest/test-data/transposetest_test1_output.xls Thu Mar 21 21:14:49 2013 -0400 @@ -0,0 +1,4 @@ +h1 r11 r21 r31 +h2 r12 r22 r32 +h3 r13 r23 r33 +h4 r14 r24 r34 diff -r 000000000000 -r bf814ef1c04d transposetest/transposetest.py --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/transposetest/transposetest.py Thu Mar 21 21:14:49 2013 -0400 @@ -0,0 +1,580 @@ +# transposetest.py - a self annotated version of rgToolFactory.py generated by running rgToolFactory.py +# to make a new Galaxy tool called transposetest +# User admin@here.now at 22/03/2013 01:13:57 +# rgToolFactory.py +# see https://bitbucket.org/fubar/galaxytoolfactory/wiki/Home +# +# copyright ross lazarus (ross stop lazarus at gmail stop com) May 2012 +# +# all rights reserved +# Licensed under the LGPL +# suggestions for improvement and bug fixes welcome at https://bitbucket.org/fubar/galaxytoolfactory/wiki/Home +# +# January 2013 +# problem pointed out by Carlos Borroto +# added escaping for <>$ - thought I did that ages ago... +# +# August 11 2012 +# changed to use shell=False and cl as a sequence + +# This is a Galaxy tool factory for simple scripts in python, R or whatever ails ye. +# It also serves as the wrapper for the new tool. +# +# you paste and run your script +# Only works for simple scripts that read one input from the history. +# Optionally can write one new history dataset, +# and optionally collect any number of outputs into links on an autogenerated HTML page. + +# DO NOT install on a public or important site - please. + +# installed generated tools are fine if the script is safe. +# They just run normally and their user cannot do anything unusually insecure +# but please, practice safe toolshed. +# Read the fucking code before you install any tool +# especially this one + +# After you get the script working on some test data, you can +# optionally generate a toolshed compatible gzip file +# containing your script safely wrapped as an ordinary Galaxy script in your local toolshed for +# safe and largely automated installation in a production Galaxy. + +# If you opt for an HTML output, you get all the script outputs arranged +# as a single Html history item - all output files are linked, thumbnails for all the pdfs. +# Ugly but really inexpensive. +# +# Patches appreciated please. +# +# +# long route to June 2012 product +# Behold the awesome power of Galaxy and the toolshed with the tool factory to bind them +# derived from an integrated script model +# called rgBaseScriptWrapper.py +# Note to the unwary: +# This tool allows arbitrary scripting on your Galaxy as the Galaxy user +# There is nothing stopping a malicious user doing whatever they choose +# Extremely dangerous!! +# Totally insecure. So, trusted users only +# +# preferred model is a developer using their throw away workstation instance - ie a private site. +# no real risk. The universe_wsgi.ini admin_users string is checked - only admin users are permitted to run this tool. +# + +import sys +import shutil +import subprocess +import os +import time +import tempfile +import optparse +import tarfile +import re +import shutil +import math + +progname = os.path.split(sys.argv[0])[1] +myversion = 'V000.2 June 2012' +verbose = False +debug = False +toolFactoryURL = 'https://bitbucket.org/fubar/galaxytoolfactory' + +def timenow(): + """return current time as a string + """ + return time.strftime('%d/%m/%Y %H:%M:%S', time.localtime(time.time())) + +cheetah_escape_table = { + "$": "\$" + } + +cheetah_unescape_table = { + "\$": "$" + } + +def html_escape(t): + """Unescape \$ first in case already done + cheetah barfs if any $ without \ + xml parsing is controlled with + """ + text = t + for k in cheetah_unescape_table.keys(): + text = text.replace(k,cheetah_unescape_table[k]) + for k in cheetah_escape_table.keys(): + text = text.replace(k,cheetah_escape_table[k]) + return text + + +class ScriptRunner: + """class is a wrapper for an arbitrary script + """ + + def __init__(self,opts=None,treatbashSpecial=True): + """ + cleanup inputs, setup some outputs + + """ + self.treatbashSpecial = treatbashSpecial + if opts.output_dir: # simplify for the tool tarball + os.chdir(opts.output_dir) + self.thumbformat = 'jpg' + self.opts = opts + self.toolname = re.sub('[^a-zA-Z0-9_]+', '', opts.tool_name) # a sanitizer now does this but.. + self.toolid = self.toolname + self.myname = sys.argv[0] # get our name because we write ourselves out as a tool later + self.pyfile = self.myname # crude but efficient - the cruft won't hurt much + self.xmlfile = '%s.xml' % self.toolname + 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=self.toolname,suffix=".%s" % (opts.interpreter)) + tscript = open(self.sfile,'w') # use self.sfile as script source for Popen + tscript.write(self.script) + tscript.close() + self.escapedS = [html_escape(x) for x in s] # for restructured text in help + self.escapedScript = '\n'.join(self.escapedS) + self.indentedScript = '\n'.join([' %s' % x for x in s]) # for restructured text in help + if opts.output_dir: # may not want these complexities + self.tlog = os.path.join(opts.output_dir,"%s_runner.log" % self.toolname) + art = '%s.%s' % (self.toolname,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 + a(opts.input_tab) + a(opts.output_tab) + self.outFormats = 'tabular' # TODO make this an option at tool generation time + self.inputFormats = 'tabular' # TODO make this an option at tool generation time + self.test1Input = '%s_test1_input.xls' % self.toolname + self.test1Output = '%s_test1_output.xls' % self.toolname + self.test1HTML = '%s_test1_output.html' % self.toolname + + def makeXML(self): + """ + Create a Galaxy xml tool wrapper for the new script as a string to write out + fixme - use templating or something less fugly than this example of what we produce + + + a tabular file + + reverse.py --script_path "$runMe" --interpreter "python" + --tool_name "reverse" --input_tab "$input1" --output_tab "$tab_file" + + + + + + + + + + + +**What it Does** + +Reverse the columns in a tabular file + + + + + +# reverse order of columns in a tabular file +import sys +inp = sys.argv[1] +outp = sys.argv[2] +i = open(inp,'r') +o = open(outp,'w') +for row in i: + rs = row.rstrip().split('\t') + rs.reverse() + o.write('\t'.join(rs)) + o.write('\n') +i.close() +o.close() + + + + + + + """ + newXML=""" +%(tooldesc)s +%(command)s + +%(inputs)s + + +%(outputs)s + + + + + + +%(tooltests)s + + + + +""" +# needs a dict with toolname, toolid, interpreter, scriptname, command, inputs as a multi line string ready to write, outputs ditto, help ditto + + newCommand=""" + %(toolname)s.py --script_path "$runMe" --interpreter "%(interpreter)s" + --tool_name "%(toolname)s" %(command_inputs)s %(command_outputs)s + """ # may NOT be an input or htmlout + tooltestsTabOnly = """ + + + + + """ + tooltestsHTMLOnly = """ + + + + + """ + tooltestsBoth = """ + + + + + + """ + xdict = {} + xdict['tool_version'] = self.opts.tool_version + xdict['test1Input'] = self.test1Input + xdict['test1HTML'] = self.test1HTML + xdict['test1Output'] = self.test1Output + if self.opts.make_HTML and self.opts.output_tab <> 'None': + xdict['tooltests'] = tooltestsBoth % xdict + elif self.opts.make_HTML: + xdict['tooltests'] = tooltestsHTMLOnly % xdict + else: + xdict['tooltests'] = tooltestsTabOnly % xdict + xdict['script'] = self.escapedScript + # configfile is least painful way to embed script to avoid external dependencies + # but requires escaping of <, > and $ to avoid Mako parsing + if self.opts.help_text: + xdict['help'] = open(self.opts.help_text,'r').read() + else: + xdict['help'] = 'Please ask the tool author for help as none was supplied at tool generation\n' + coda = ['**Script**','Pressing execute will run the following code over your input file and generate some outputs in your history::\n'] + coda.append(self.indentedScript) + coda.append('\n\n') + coda.append('**Attribution** This Galaxy tool was created by %s at %s\nusing the Galaxy Tool Factory.' % (self.opts.user_email,timenow())) + coda.append('See %s for details of that project' % (toolFactoryURL)) + coda.append('Please cite: Creating re-usable tools from scripts: The Galaxy Tool Factory. Ross Lazarus; Antony Kaspi; Mark Ziemann; The Galaxy Team. ') + coda.append('Bioinformatics 2012; doi: 10.1093/bioinformatics/bts573') + xdict['help'] = '%s\n\n%s\n\n' % (xdict['help'],'\n'.join(coda)) + if self.opts.tool_desc: + xdict['tooldesc'] = '%s' % self.opts.tool_desc + else: + xdict['tooldesc'] = '' + xdict['command_outputs'] = '' + xdict['outputs'] = '' + if self.opts.input_tab <> 'None': + xdict['command_inputs'] = '--input_tab "$input1" ' # the space may matter a lot if we append something + xdict['inputs'] = ' \n' % self.inputFormats + else: + xdict['command_inputs'] = '' # assume no input - eg a random data generator + xdict['inputs'] = '' + xdict['inputs'] += ' \n' % self.toolname + xdict['toolname'] = self.toolname + xdict['toolid'] = self.toolid + xdict['interpreter'] = self.opts.interpreter + xdict['scriptname'] = self.sfile + if self.opts.make_HTML: + xdict['command_outputs'] += ' --output_dir "$html_file.files_path" --output_html "$html_file" --make_HTML "yes" ' + xdict['outputs'] += ' \n' + if self.opts.output_tab <> 'None': + xdict['command_outputs'] += ' --output_tab "$tab_file"' + xdict['outputs'] += ' \n' % self.outFormats + xdict['command'] = newCommand % xdict + xmls = newXML % xdict + xf = open(self.xmlfile,'w') + xf.write(xmls) + xf.write('\n') + xf.close() + # ready for the tarball + + + def makeTooltar(self): + """ + a tool is a gz tarball with eg + /toolname/tool.xml /toolname/tool.py /toolname/test-data/test1_in.foo ... + """ + retval = self.run() + if retval: + print >> sys.stderr,'## Run failed. Cannot build yet. Please fix and retry' + sys.exit(1) + self.makeXML() + tdir = self.toolname + os.mkdir(tdir) + if self.opts.input_tab <> 'None': # no reproducible test otherwise? TODO: maybe.. + testdir = os.path.join(tdir,'test-data') + os.mkdir(testdir) # make tests directory + shutil.copyfile(self.opts.input_tab,os.path.join(testdir,self.test1Input)) + if self.opts.output_tab <> 'None': + shutil.copyfile(self.opts.output_tab,os.path.join(testdir,self.test1Output)) + if self.opts.make_HTML: + shutil.copyfile(self.opts.output_html,os.path.join(testdir,self.test1HTML)) + if self.opts.output_dir: + shutil.copyfile(self.tlog,os.path.join(testdir,'test1_out.log')) + op = '%s.py' % self.toolname # new name + outpiname = os.path.join(tdir,op) # path for the tool tarball + pyin = os.path.basename(self.pyfile) # our name - we rewrite ourselves (TM) + notes = ['# %s - a self annotated version of %s generated by running %s\n' % (op,pyin,pyin),] + notes.append('# to make a new Galaxy tool called %s\n' % self.toolname) + notes.append('# User %s at %s\n' % (self.opts.user_email,timenow())) + pi = open(self.pyfile,'r').readlines() # our code becomes new tool wrapper (!) - first Galaxy worm + notes += pi + outpi = open(outpiname,'w') + outpi.write(''.join(notes)) + outpi.write('\n') + outpi.close() + stname = os.path.join(tdir,self.sfile) + if not os.path.exists(stname): + shutil.copyfile(self.sfile, stname) + xtname = os.path.join(tdir,self.xmlfile) + if not os.path.exists(xtname): + shutil.copyfile(self.xmlfile,xtname) + tarpath = "%s.gz" % self.toolname + tar = tarfile.open(tarpath, "w:gz") + tar.add(tdir,arcname=self.toolname) + tar.close() + shutil.copyfile(tarpath,self.opts.new_tool) + shutil.rmtree(tdir) + ## TODO: replace with optional direct upload to local toolshed? + return retval + + def compressPDF(self,inpdf=None,thumbformat='png'): + """need absolute path to pdf + """ + assert os.path.isfile(inpdf), "## Input %s supplied to %s compressPDF not found" % (inpdf,self.myName) + hf,hlog = tempfile.mkstemp(suffix="%s.log" % self.toolname) + sto = open(hlog,'w') + outpdf = '%s_compressed' % inpdf + cl = ["gs", "-sDEVICE=pdfwrite", "-dNOPAUSE", "-dBATCH", "-sOutputFile=%s" % outpdf,inpdf] + x = subprocess.Popen(cl,stdout=sto,stderr=sto,cwd=self.opts.output_dir) + retval1 = x.wait() + if retval1 == 0: + os.unlink(inpdf) + shutil.move(outpdf,inpdf) + 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) + retval2 = x.wait() + sto.close() + 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 = """ + + + + + + + +
+ """ + galhtmlattr = """
This tool (%s) was generated by the Galaxy Tool Factory

""" + galhtmlpostfix = """
\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('
Galaxy Tool "%s" run at %s

' % (self.toolname,timenow())) + fhtml = [] + if len(flist) > 0: + pdflist = [] + npdf = len([x for x in flist if os.path.splitext(x)[-1].lower() == '.pdf']) + nacross = 1 + if npdf > 0: + nacross = int(round(math.log(npdf,2))) + nacross = max(1,nacross) + width = min(400,int(1200/nacross)) + 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)) + if (rownum+1) % 2 == 0: + fhtml.append('%s%s' % (fname,fname,sfsize)) + else: + fhtml.append('%s%s' % (fname,fname,sfsize)) + ntogo = nacross # counter for table row padding with empty cells + if len(pdflist) > 0: + html.append('
\n') + for i,paths in enumerate(pdflist): + fname,thumb = paths + s= """\n""" % (fname,thumb,fname,width,fname) + if ((i+1) % nacross == 0): + s += '\n' + ntogo = 0 + if i < (npdf - 1): # more to come + s += '' + ntogo = nacross + else: + ntogo -= 1 + html.append(s) + if html[-1].strip().endswith(''): + html.append('
Image called %s
\n') + else: + if ntogo > 0: # pad + html.append(' '*ntogo) + html.append('\n') + if len(fhtml) > 0: + fhtml.insert(0,'
\n') + fhtml.append('
Output File Name (click to view)Size

') + html += fhtml # add all non-pdf files to the end of the display + else: + html.append('
### Error - %s returned no files - please confirm that parameters are sane
' % self.opts.interpreter) + rlog = open(self.tlog,'r').readlines() + rlog = [x for x in rlog if x.strip() > ''] + if len(rlog) > 1: + html.append('
%s log
\n' % self.opts.interpreter)
+            html += rlog
+            html.append('
\n') + html.append(galhtmlattr % (self.toolname)) + 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: + 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=sto,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() + 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 main(): + u = """ + This is a Galaxy wrapper. It expects to be called by a special purpose tool.xml as: + rgBaseScriptWrapper.py --script_path "$scriptPath" --tool_name "foo" --interpreter "Rscript" + + """ + op = optparse.OptionParser() + a = op.add_option + a('--script_path',default=None) + a('--tool_name',default=None) + a('--interpreter',default=None) + a('--output_dir',default=None) + a('--output_html',default=None) + a('--input_tab',default="None") + a('--output_tab',default="None") + a('--user_email',default='Unknown') + a('--bad_user',default=None) + a('--make_Tool',default=None) + a('--make_HTML',default=None) + a('--help_text',default=None) + a('--tool_desc',default=None) + a('--new_tool',default=None) + a('--tool_version',default=None) + opts, args = op.parse_args() + assert not opts.bad_user,'UNAUTHORISED: %s is NOT authorized to use this tool until Galaxy admin adds %s to admin_users in universe_wsgi.ini' % (opts.bad_user,opts.bad_user) + assert opts.tool_name,'## Tool Factory expects a tool name - eg --tool_name=DESeq' + assert opts.interpreter,'## Tool Factory wrapper expects an interpreter - eg --interpreter=Rscript' + assert os.path.isfile(opts.script_path),'## Tool Factory wrapper expects a script path - eg --script_path=foo.R' + if opts.output_dir: + try: + os.makedirs(opts.output_dir) + except: + pass + r = ScriptRunner(opts) + if opts.make_Tool: + retcode = r.makeTooltar() + else: + retcode = r.run() + os.unlink(r.sfile) + if retcode: + sys.exit(retcode) # indicate failure to job runner + + +if __name__ == "__main__": + main() + + + diff -r 000000000000 -r bf814ef1c04d transposetest/transposetest.xml --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/transposetest/transposetest.xml Thu Mar 21 21:14:49 2013 -0400 @@ -0,0 +1,66 @@ + +Transpose test + + transposetest.py --script_path "$runMe" --interpreter "Rscript" + --tool_name "transposetest" --input_tab "$input1" --output_tab "$tab_file" + + + + + + + + + + + + + + + + + + + + + + + + + +