# HG changeset patch # User fubar # Date 1409206959 14400 # Node ID fb3fa6a2874de422b6f877d37f0e25ade45359d6 # Parent bff4c9bfabc7981608ece5ffffd893ad0fdc2d14 Citations added (thanks John!) and a few more output formats for Alistair Chilcott diff -r bff4c9bfabc7 -r fb3fa6a2874d README.txt --- a/README.txt Thu Aug 07 22:11:02 2014 -0400 +++ b/README.txt Thu Aug 28 02:22:39 2014 -0400 @@ -16,6 +16,15 @@ freeze the supplied script into a new, ordinary Galaxy tool that runs it over a single input file, working just like any other Galaxy tool for your users. +If you use the Html output option, please ensure that sanitize_all_html is set to False and +uncommented in universe_wsgi.ini - it should show: + +# By default, all tool output served as 'text/html' will be sanitized +sanitize_all_html = False + + +*More Detail* + To use the ToolFactory, you should have prepared a script to paste into a text box, and a small test input example ready to select from your history to test your new script. There is an example in each scripting language on the Tool Factory form. You can just @@ -40,31 +49,6 @@ Tool factory tools are perfect for workflow components. One input, one output, no variables. -*Reasons to read further* - -If you use Galaxy to support your research; - -You and fellow users are sometimes forced to take data out of Galaxy, process it with ugly -little perl/awk/sed/R... scripts and put it back; - -You do this when you can't do some transformation in Galaxy (the 90/10 rule); - -You don't have enough developer resources for wrapping dozens of even relatively simple tools; - -Your research and your institution would be far better off if those feral scripts were all tucked -safely in your local toolshed and Galaxy histories. - -*The good news* If it can be trivially scripted, it can be running safely in your -local Galaxy via your own local toolshed in a few minutes - with functional tests. - - -*Value proposition* The ToolFactory allows Galaxy to efficiently take over most of your lab's -dark script matter, making it reproducible in Galaxy and shareable through the ToolShed. - -That's what this tool does. You paste a simple script and the tool returns -a new, real Galaxy tool, ready to be installed from the local toolshed to local servers. -Scripts can be wrapped and online literally within minutes. - *To fully and safely exploit the awesome power* of this tool, Galaxy and the ToolShed, you should be a developer installing this tool on a private/personal/scratch local instance where you are an admin_user. Then, if you break it, you get to keep all the pieces @@ -72,9 +56,10 @@ ** Installation ** This is a Galaxy tool. You can install it most conveniently using the administrative "Search and browse tool sheds" link. -Find the Galaxy Test toolshed (not main) and search for the toolfactory repository. +Find the Galaxy Main toolshed at https://toolshed.g2.bx.psu.edu/ and search for the toolfactory repository. Open it and review the code and select the option to install it. +( If you can't get the tool that way, the xml and py files here need to be copied into a new tools subdirectory such as tools/toolfactory Your tool_conf.xml needs a new entry pointing to the xml file - something like:: @@ -85,11 +70,8 @@ If not already there (I just added it to datatypes_conf.xml.sample), please add: -to your local data_types_conf.xml. - -Ensure that html sanitization is set to False and uncommented in universe_wsgi.ini - -You'll have to restart the server for the new tool to be available. +to your local data_types_conf.xml. +) Of course, R, python, perl etc are needed on your path if you want to test scripts using those interpreters. Adding new ones to this tool code should be easy enough. Please make suggestions as bitbucket issues and code. @@ -137,18 +119,6 @@ Patches and suggestions welcome as bitbucket issues please? -long route to June 2012 product -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 - - - - copyright ross lazarus (ross stop lazarus at gmail stop com) May 2012 all rights reserved diff -r bff4c9bfabc7 -r fb3fa6a2874d rgToolFactory.py --- a/rgToolFactory.py Thu Aug 07 22:11:02 2014 -0400 +++ b/rgToolFactory.py Thu Aug 28 02:22:39 2014 -0400 @@ -134,6 +134,19 @@ stdout=subprocess.PIPE, stderr=subprocess.PIPE) == 0 +def parse_citations(citations_text): + """ + """ + citations = [c for c in citations_text.split("**ENTRY**") if c.strip()] + citation_tuples = [] + for citation in citations: + if citation.startswith("doi"): + citation_tuples.append( ("doi", citation[len("doi"):].strip() ) ) + else: + citation_tuples.append( ("bibtex", citation[len("bibtex"):].strip() ) ) + return citation_tuples + + class ScriptRunner: """class is a wrapper for an arbitrary script """ @@ -264,6 +277,10 @@ %(help)s + + %(citations)s + 10.1093/bioinformatics/bts573 + """ # needs a dict with toolname, toolid, interpreter, scriptname, command, inputs as a multi line string ready to write, outputs ditto, help ditto newCommand=""" @@ -303,7 +320,8 @@ xdict = {} xdict['requirements'] = '' if self.opts.make_HTML: - xdict['requirements'] = protorequirements + if self.opts.include_dependencies == "yes": + xdict['requirements'] = protorequirements xdict['tool_version'] = self.opts.tool_version xdict['test1Input'] = self.test1Input xdict['test1HTML'] = self.test1HTML @@ -323,6 +341,16 @@ xdict['help'] = ''.join([x for x in helptext]) else: xdict['help'] = 'Please ask the tool author (%s) for help as none was supplied at tool generation\n' % (self.opts.user_email) + if self.opts.citations: + citationstext = open(self.opts.citations,'r').read() + citation_tuples = parse_citations(citationstext) + citations_xml = "" + for citation_type, citation_content in citation_tuples: + citation_xml = """%s""" % (citation_type, html_escape(citation_content)) + citations_xml += citation_xml + xdict['citations'] = citations_xml + else: + xdict['citations'] = "" coda = ['**Script**','Pressing execute will run the following code over your input file and generate some outputs in your history::'] coda.append('\n') coda.append(self.indentedScript) @@ -671,6 +699,7 @@ a('--make_Tool',default=None) a('--make_HTML',default=None) a('--help_text',default=None) + a('--citations',default=None) a('--tool_desc',default=None) a('--new_tool',default=None) a('--tool_version',default=None) diff -r bff4c9bfabc7 -r fb3fa6a2874d rgToolFactory.xml --- a/rgToolFactory.xml Thu Aug 07 22:11:02 2014 -0400 +++ b/rgToolFactory.xml Thu Aug 28 02:22:39 2014 -0400 @@ -1,5 +1,5 @@ - Makes scripts into tools + Run a script; make a tool! ghostscript graphicsmagick @@ -20,8 +20,11 @@ --new_tool "$new_tool" --help_text "$helpme" #if $make_HTML.value=="yes": - --include_dependencies "${makeMode.include_deps}" + #if makeMode.include.deps.value=="yes": + --include_dependencies "yes" + #end if #end if + --citations "$citeme" #end if #if $make_HTML.value=="yes": --output_dir "$html_file.files_path" --output_html "$html_file" --make_HTML "yes" @@ -66,7 +69,28 @@ - + + + + + + + + + + + + + + + + + + + + @@ -86,11 +110,17 @@ + + + + + + + + + + @@ -125,6 +159,19 @@ ${makeMode.help_text} #end if + +#if $makeMode.make_Tool == "yes": +#for $citation in $makeMode.citations: +#if $citation.citation_type.type == "bibtex": +**ENTRY**bibtex +${citation.citation_type.bibtex} +#else +**ENTRY**doi +${citation.citation_type.doi} +#end if +#end for +#end if + @@ -310,7 +357,9 @@ - + + 10.1093/bioinformatics/bts573 + diff -r bff4c9bfabc7 -r fb3fa6a2874d rgToolFactoryMultIn.py --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/rgToolFactoryMultIn.py Thu Aug 28 02:22:39 2014 -0400 @@ -0,0 +1,736 @@ +# rgToolFactoryMultIn.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 +# +# august 2014 +# Allows arbitrary number of input files +# NOTE positional parameters are now passed to script +# and output (may be "None") is *before* arbitrary number of inputs +# +# march 2014 +# had to remove dependencies because cross toolshed dependencies are not possible - can't pre-specify a toolshed url for graphicsmagick and ghostscript +# grrrrr - night before a demo +# added dependencies to a tool_dependencies.xml if html page generated so generated tool is properly portable +# +# added ghostscript and graphicsmagick as dependencies +# fixed a wierd problem where gs was trying to use the new_files_path from universe (database/tmp) as ./database/tmp +# errors ensued +# +# august 2013 +# found a problem with GS if $TMP or $TEMP missing - now inject /tmp and warn +# +# july 2013 +# added ability to combine images and individual log files into html output +# just make sure there's a log file foo.log and it will be output +# together with all images named like "foo_*.pdf +# otherwise old format for html +# +# January 2013 +# problem pointed out by Carlos Borroto +# added escaping for <>$ - 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 = 'V001.1 March 2014' +verbose = False +debug = False +toolFactoryURL = 'https://bitbucket.org/fubar/galaxytoolfactory' + +# if we do html we need these dependencies specified in a tool_dependencies.xml file and referred to in the generated +# tool xml +toolhtmldepskel = """ + + + + + + + + + %s + + +""" + +protorequirements = """ + ghostscript + graphicsmagick + """ + +def timenow(): + """return current time as a string + """ + return time.strftime('%d/%m/%Y %H:%M:%S', time.localtime(time.time())) + +html_escape_table = { + "&": "&", + ">": ">", + "<": "<", + "$": "\$" + } + +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 + + +class ScriptRunner: + """class is a wrapper for an arbitrary script + """ + + def __init__(self,opts=None,treatbashSpecial=True): + """ + cleanup inputs, setup some outputs + + """ + self.useGM = cmd_exists('gm') + 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 + if opts.output_dir: # simplify for the tool tarball + os.chdir(opts.output_dir) + self.thumbformat = 'png' + 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.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.toolname) + if opts.output_dir: # may not want these complexities + self.tlog = os.path.join(self.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 = [] + self.test1Inputs = [] # now a list + a = self.cl.append + a(opts.interpreter) + if self.treatbashSpecial and opts.interpreter in ['bash','sh']: + a(self.sfile) + else: + a('-') # stdin + # if multiple inputs - positional or need to distinguish them with cl params + if opts.output_tab: + a('%s' % opts.output_tab) + if opts.input_tab: + tests = [] + for i,intab in enumerate(opts.input_tab): # if multiple, make tests + if intab.find(',') <> -1: + (gpath,uname) = intab.split(',') + else: + gpath = uname = intab + a('"%s"' % (intab)) + tests.append(os.path.basename(gpath)) + self.test1Inputs = '' % (','.join(tests)) + else: + self.test1Inputs = '' + self.outFormats = opts.output_format + self.inputFormats = opts.input_formats + 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 +%(requirements)s + +%(command)s + + +%(inputs)s + + +%(outputs)s + + + +%(script)s + + + +%(tooltests)s + + + +%(help)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 - appended later + tooltestsTabOnly = """ + + %(test1Inputs)s + + + + + + """ + tooltestsHTMLOnly = """ + + %(test1Inputs)s + + + + + + """ + tooltestsBoth = """ + + %(test1Inputs)s + + + + + + + """ + xdict = {} + xdict['requirements'] = '' + if self.opts.make_HTML: + if self.opts.include_dependencies == "yes": + xdict['requirements'] = protorequirements + xdict['tool_version'] = self.opts.tool_version + xdict['test1HTML'] = self.test1HTML + xdict['test1Output'] = self.test1Output + xdict['test1Inputs'] = self.test1Inputs + 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: + helptext = open(self.opts.help_text,'r').readlines() + helptext = [html_escape(x) for x in helptext] # must html escape here too - thanks to Marius van den Beek + xdict['help'] = ''.join([x for x in helptext]) + else: + xdict['help'] = 'Please ask the tool author (%s) for help as none was supplied at tool generation\n' % (self.opts.user_email) + coda = ['**Script**','Pressing execute will run the following code over your input file and generate some outputs in your history::'] + coda.append('\n') + coda.append(self.indentedScript) + coda.append('\n**Attribution**\nThis Galaxy tool was created by %s at %s\nusing the Galaxy Tool Factory.\n' % (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\n') + xdict['help'] = '%s\n%s' % (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': + cins = ['\n',] + cins.append('#for intab in $input1:') + cins.append('--input_tab "$intab"') + cins.append('#end for\n') + xdict['command_inputs'] = '\n'.join(cins) + 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' + else: + xdict['command_outputs'] += ' --output_dir "./"' + 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) + tdir = self.toolname + os.mkdir(tdir) + self.makeXML() + if self.opts.make_HTML: + if self.opts.help_text: + hlp = open(self.opts.help_text,'r').read() + else: + hlp = 'Please ask the tool author for help as none was supplied at tool generation\n' + if self.opts.include_dependencies == "yes": + tooldepcontent = toolhtmldepskel % hlp + depf = open(os.path.join(tdir,'tool_dependencies.xml'),'w') + depf.write(tooldepcontent) + depf.write('\n') + depf.close() + 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 + for i,intab in enumerate(self.opts.input_tab): + si = self.opts.input_tab[i] + if si.find(',') <> -1: + s = si.split(',')[0] + si = s + dest = os.path.join(testdir,os.path.basename(si)) + if si <> dest: + shutil.copyfile(si,dest) + 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')) + outpif = '%s.py' % self.toolname # new name + outpiname = os.path.join(tdir,outpif) # 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' % (outpiname,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 + 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) + if self.useGM: + cl2 = ['gm', 'convert', inpdf, outpng] + else: # assume imagemagick + 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 = """ + + + + + + + +
+ """ + 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: + 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 os.path.abspath(x) <> os.path.abspath(self.tlog)] + logfiles.append(os.path.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('%s%s' % (fname,fname,sfsize)) + else: + fhtml.append('%s%s' % (fname,fname,sfsize)) + for logfname in logfiles: # expect at least tlog - if more + if os.path.abspath(logfname) == os.path.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('
%s images and outputs
' % sectionname) + html.append('(Click on a thumbnail image to download the corresponding original PDF image)
') + ntogo = nacross # counter for table row padding with empty cells + html.append('
\n') + for i,paths in enumerate(ourpdfs): + 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') + logt = open(logfname,'r').readlines() + logtext = [x for x in logt if x.strip() > ''] + html.append('
%s log output
' % sectionname) + if len(logtext) > 1: + html.append('\n
\n')
+                    html += logtext
+                    html.append('\n
\n') + else: + html.append('%s is empty
' % logfname) + if len(fhtml) > 0: + fhtml.insert(0,'
\n') + fhtml.append('
Output File Name (click to view)Size

') + html.append('
All output files available for downloading
\n') + 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) + 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 + 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='./') + a('--output_html',default=None) + a('--input_tab',default=[], action="append") + a("--input_formats",default="tabular") + a('--output_tab',default="None") + a('--output_format',default='tabular') + 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) + a('--include_dependencies',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 + opts.input_tab = [x.replace('"','').replace("'",'') for x in opts.input_tab] + 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 bff4c9bfabc7 -r fb3fa6a2874d rgToolFactoryMultIn.xml --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/rgToolFactoryMultIn.xml Thu Aug 28 02:22:39 2014 -0400 @@ -0,0 +1,343 @@ + + Makes scripts into tools + + ghostscript + graphicsmagick + + +#if ( $__user_email__ not in $__admin_users__ ): + rgToolFactoryMultIn.py --bad_user $__user_email__ +#else: + rgToolFactoryMultIn.py --script_path "$runme" --interpreter "$interpreter" + --tool_name "$tool_name" --user_email "$__user_email__" + #if $make_TAB.value=="yes": + --output_tab "$output1" + --output_format "$output_format" + #end if + #if $makeMode.make_Tool=="yes": + --make_Tool "$makeMode.make_Tool" + --tool_desc "$makeMode.tool_desc" + --tool_version "$makeMode.tool_version" + --new_tool "$new_tool" + --help_text "$helpme" + #if $make_HTML.value=="yes": + #if $makeMode.include_deps.value=="yes": + --include_dependencies "yes" + #end if + #end if + #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 $input1 != 'None': + --input_formats "$input_formats" + #for intab in $input1: + #if $add_names.value == "yes": + --input_tab "$intab,$intab.name" + #else: + --input_tab "$intab" + #end if + #end for + --input_formats = "$input_formats" + #end if +#end if + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + make_TAB=="yes" + + + + + + + + make_HTML == "yes" + + + makeMode['make_Tool'] == "yes" + + + +$dynScript + +#if $makeMode.make_Tool == "yes": +${makeMode.help_text} +#end if + + + + +.. class:: warningmark + +**Details and attribution** GTF_ + +**Local Admins ONLY** Only users whose IDs found in the local admin_user configuration setting in universe_wsgi.ini can run this tool. + +**If you find a bug** please raise an issue at the bitbucket repository GTFI_ + +**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. + +**Tool Generation option** Once the script is working with test data, this tool will optionally generate a new Galaxy tool in a gzip file +ready to upload to your local toolshed for sharing and installation. Provide a small sample input when you run generate the tool because +it will become the input for the generated functional test. + +.. class:: warningmark + +**Note to system administrators** This tool offers *NO* built in protection against malicious scripts. It should only be installed on private/personnal Galaxy instances. +Admin_users will have the power to do anything they want as the Galaxy user if you install this tool. + +.. 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** + $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 = ) { + $line =~ s/$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 +.. _GTF: https://bitbucket.org/fubar/galaxytoolfactory +.. _GTFI: https://bitbucket.org/fubar/galaxytoolfactory/issues +.. _Paper: http://bioinformatics.oxfordjournals.org/cgi/reprint/bts573?ijkey=lczQh1sWrMwdYWJ&keytype=ref + + + + + + +