|
48
|
1 #!/usr/bin/env python
|
|
|
2 # rgToolFactory.py
|
|
|
3 # see https://github.com/fubar2/toolfactory
|
|
|
4 #
|
|
|
5 # copyright ross lazarus (ross stop lazarus at gmail stop com) May 2012
|
|
|
6 #
|
|
|
7 # all rights reserved
|
|
|
8 # Licensed under the LGPL
|
|
49
|
9 # suggestions for improvement and bug fixes welcome at
|
|
|
10 # https://github.com/fubar2/toolfactory
|
|
48
|
11 #
|
|
|
12 # July 2020: BCC was fun and I feel like rip van winkle after 5 years.
|
|
|
13 # Decided to
|
|
|
14 # 1. Fix the toolfactory so it works - done for simplest case
|
|
|
15 # 2. Fix planemo so the toolfactory function works
|
|
|
16 # 3. Rewrite bits using galaxyxml functions where that makes sense - done
|
|
|
17 #
|
|
|
18 # removed all the old complications including making the new tool use this same script
|
|
|
19 # galaxyxml now generates the tool xml https://github.com/hexylena/galaxyxml
|
|
|
20 # No support for automatic HTML file creation from arbitrary outputs
|
|
|
21 # essential problem is to create two command lines - one for the tool xml and a different
|
|
|
22 # one to run the executable with the supplied test data and settings
|
|
|
23 # Be simpler to write the tool, then run it with planemo and soak up the test outputs.
|
|
|
24
|
|
|
25
|
|
|
26
|
|
|
27 import argparse
|
|
|
28 import logging
|
|
|
29 import os
|
|
|
30 import re
|
|
|
31 import shutil
|
|
|
32 import subprocess
|
|
|
33 import sys
|
|
|
34 import tarfile
|
|
|
35 import tempfile
|
|
|
36 import time
|
|
|
37
|
|
|
38 import galaxyxml.tool as gxt
|
|
|
39 import galaxyxml.tool.parameters as gxtp
|
|
|
40
|
|
|
41 import lxml
|
|
|
42
|
|
|
43 import yaml
|
|
|
44
|
|
|
45 myversion = "V2.1 July 2020"
|
|
|
46 verbose = True
|
|
|
47 debug = True
|
|
|
48 toolFactoryURL = "https://github.com/fubar2/toolfactory"
|
|
|
49 ourdelim = "~~~"
|
|
49
|
50 ALOT = 10000000 # srsly. command or test overrides use read() so just in case
|
|
|
51 STDIOXML = """<stdio>
|
|
|
52 <exit_code range="100:" level="debug" description="shite happens" />
|
|
|
53 </stdio>"""
|
|
48
|
54
|
|
|
55 # --input_files="$input_files~~~$CL~~~$input_formats~~~$input_label
|
|
|
56 # ~~~$input_help"
|
|
|
57 IPATHPOS = 0
|
|
|
58 ICLPOS = 1
|
|
|
59 IFMTPOS = 2
|
|
|
60 ILABPOS = 3
|
|
|
61 IHELPOS = 4
|
|
|
62 IOCLPOS = 5
|
|
|
63
|
|
49
|
64 # --output_files "$otab.history_name~~~$otab.history_format~~~$otab.CL~~~otab.history_test
|
|
48
|
65 ONAMEPOS = 0
|
|
|
66 OFMTPOS = 1
|
|
|
67 OCLPOS = 2
|
|
49
|
68 OTESTPOS = 3
|
|
|
69 OOCLPOS = 4
|
|
|
70
|
|
48
|
71
|
|
|
72 # --additional_parameters="$i.param_name~~~$i.param_value~~~
|
|
|
73 # $i.param_label~~~$i.param_help~~~$i.param_type~~~$i.CL~~~i$.param_CLoverride"
|
|
|
74 ANAMEPOS = 0
|
|
|
75 AVALPOS = 1
|
|
|
76 ALABPOS = 2
|
|
|
77 AHELPPOS = 3
|
|
|
78 ATYPEPOS = 4
|
|
|
79 ACLPOS = 5
|
|
|
80 AOVERPOS = 6
|
|
|
81 AOCLPOS = 7
|
|
|
82
|
|
|
83
|
|
|
84 foo = len(lxml.__version__)
|
|
|
85 # fug you, flake8. Say my name!
|
|
49
|
86 FAKEEXE = "~~~REMOVE~~~ME~~~"
|
|
|
87 # need this until a PR/version bump to fix galaxyxml prepending the exe even
|
|
|
88 # with override.
|
|
|
89
|
|
48
|
90
|
|
|
91 def timenow():
|
|
|
92 """return current time as a string
|
|
|
93 """
|
|
|
94 return time.strftime("%d/%m/%Y %H:%M:%S", time.localtime(time.time()))
|
|
|
95
|
|
|
96
|
|
|
97 def quote_non_numeric(s):
|
|
|
98 """return a prequoted string for non-numerics
|
|
|
99 useful for perl and Rscript parameter passing?
|
|
|
100 """
|
|
|
101 try:
|
|
|
102 _ = float(s)
|
|
|
103 return s
|
|
|
104 except ValueError:
|
|
|
105 return '"%s"' % s
|
|
|
106
|
|
|
107
|
|
|
108 html_escape_table = {"&": "&", ">": ">", "<": "<", "$": r"\$"}
|
|
|
109
|
|
|
110
|
|
|
111 def html_escape(text):
|
|
|
112 """Produce entities within text."""
|
|
|
113 return "".join(html_escape_table.get(c, c) for c in text)
|
|
|
114
|
|
|
115
|
|
|
116 def html_unescape(text):
|
|
|
117 """Revert entities within text. Multiple character targets so use replace"""
|
|
|
118 t = text.replace("&", "&")
|
|
|
119 t = t.replace(">", ">")
|
|
|
120 t = t.replace("<", "<")
|
|
|
121 t = t.replace("\\$", "$")
|
|
|
122 return t
|
|
|
123
|
|
|
124
|
|
|
125 def parse_citations(citations_text):
|
|
|
126 """
|
|
|
127 """
|
|
|
128 citations = [c for c in citations_text.split("**ENTRY**") if c.strip()]
|
|
|
129 citation_tuples = []
|
|
|
130 for citation in citations:
|
|
|
131 if citation.startswith("doi"):
|
|
|
132 citation_tuples.append(("doi", citation[len("doi") :].strip()))
|
|
|
133 else:
|
|
49
|
134 citation_tuples.append(("bibtex", citation[len("bibtex") :].strip()))
|
|
48
|
135 return citation_tuples
|
|
|
136
|
|
|
137
|
|
|
138 class ScriptRunner:
|
|
|
139 """Wrapper for an arbitrary script
|
|
|
140 uses galaxyxml
|
|
|
141
|
|
|
142 """
|
|
|
143
|
|
|
144 def __init__(self, args=None):
|
|
|
145 """
|
|
|
146 prepare command line cl for running the tool here
|
|
|
147 and prepare elements needed for galaxyxml tool generation
|
|
|
148 """
|
|
|
149 self.infiles = [x.split(ourdelim) for x in args.input_files]
|
|
|
150 self.outfiles = [x.split(ourdelim) for x in args.output_files]
|
|
|
151 self.addpar = [x.split(ourdelim) for x in args.additional_parameters]
|
|
|
152 self.args = args
|
|
|
153 self.cleanuppar()
|
|
|
154 self.lastclredirect = None
|
|
|
155 self.lastxclredirect = None
|
|
|
156 self.cl = []
|
|
|
157 self.xmlcl = []
|
|
|
158 self.is_positional = self.args.parampass == "positional"
|
|
49
|
159 if self.args.packages:
|
|
|
160 self.executeme = self.args.packages.split(",")[0].split(":")[0]
|
|
|
161 else:
|
|
|
162 self.executeme = self.args.sysexe
|
|
|
163 assert (
|
|
|
164 self.executeme is not None
|
|
|
165 ), "No system or managed executable passed in. Cannot build"
|
|
48
|
166 aCL = self.cl.append
|
|
49
|
167 aXCL = self.xmlcl.append
|
|
48
|
168 assert args.parampass in [
|
|
|
169 "0",
|
|
|
170 "argparse",
|
|
|
171 "positional",
|
|
49
|
172 ], 'args.parampass must be "0","positional" or "argparse"'
|
|
48
|
173 self.tool_name = re.sub("[^a-zA-Z0-9_]+", "", args.tool_name)
|
|
|
174 self.tool_id = self.tool_name
|
|
|
175 self.tool = gxt.Tool(
|
|
|
176 self.args.tool_name,
|
|
|
177 self.tool_id,
|
|
|
178 self.args.tool_version,
|
|
|
179 self.args.tool_desc,
|
|
49
|
180 FAKEEXE,
|
|
48
|
181 )
|
|
49
|
182 if self.args.script_path:
|
|
|
183 self.tool.interpreter = self.executeme
|
|
48
|
184 self.tooloutdir = "tfout"
|
|
|
185 self.repdir = "TF_run_report_tempdir"
|
|
|
186 self.testdir = os.path.join(self.tooloutdir, "test-data")
|
|
|
187 if not os.path.exists(self.tooloutdir):
|
|
|
188 os.mkdir(self.tooloutdir)
|
|
|
189 if not os.path.exists(self.testdir):
|
|
|
190 os.mkdir(self.testdir) # make tests directory
|
|
|
191 if not os.path.exists(self.repdir):
|
|
|
192 os.mkdir(self.repdir)
|
|
|
193 self.tinputs = gxtp.Inputs()
|
|
|
194 self.toutputs = gxtp.Outputs()
|
|
|
195 self.testparam = []
|
|
49
|
196 if self.args.script_path:
|
|
|
197 self.prepScript()
|
|
|
198 if self.args.command_override:
|
|
|
199 scos = open(self.args.command_override, "r").readlines()
|
|
|
200 self.command_override = [x.rstrip() for x in scos]
|
|
|
201 else:
|
|
|
202 self.command_override = None
|
|
|
203 if self.args.test_override:
|
|
|
204 stos = open(self.args.test_override, "r").readlines()
|
|
|
205 self.test_override = [x.rstrip() for x in stos]
|
|
|
206 else:
|
|
|
207 self.test_override = None
|
|
|
208 if self.args.cl_prefix: # DIY CL start
|
|
|
209 clp = self.args.cl_prefix.split(" ")
|
|
|
210 for c in clp:
|
|
|
211 aCL(c)
|
|
|
212 aXCL(c)
|
|
|
213 else:
|
|
|
214 if self.args.runmode == "Executable":
|
|
|
215 if self.args.script_path:
|
|
|
216 aCL(self.executeme)
|
|
|
217 aCL(self.sfile)
|
|
|
218 aXCL("$runme")
|
|
|
219 else:
|
|
|
220 aCL(self.executeme) # this little CL will just run
|
|
|
221 aXCL(self.executeme)
|
|
48
|
222 else:
|
|
49
|
223 if self.args.script_path:
|
|
|
224 aCL(self.executeme)
|
|
|
225 aCL(self.sfile)
|
|
|
226 aXCL("$runme")
|
|
|
227 else:
|
|
|
228 aCL(self.executeme) # this little CL will just run
|
|
|
229 aXCL(self.executeme)
|
|
|
230 self.elog = os.path.join(self.repdir,"%s_error_log.txt" % self.tool_name)
|
|
|
231 self.tlog = os.path.join(self.repdir,"%s_runner_log.txt" % self.tool_name)
|
|
48
|
232
|
|
|
233 if self.args.parampass == "0":
|
|
|
234 self.clsimple()
|
|
|
235 else:
|
|
|
236 clsuffix = []
|
|
|
237 xclsuffix = []
|
|
|
238 for i, p in enumerate(self.infiles):
|
|
|
239 if p[IOCLPOS] == "STDIN":
|
|
|
240 appendme = [
|
|
|
241 p[IOCLPOS],
|
|
|
242 p[ICLPOS],
|
|
|
243 p[IPATHPOS],
|
|
|
244 "< %s" % p[IPATHPOS],
|
|
|
245 ]
|
|
|
246 xappendme = [
|
|
|
247 p[IOCLPOS],
|
|
|
248 p[ICLPOS],
|
|
|
249 p[IPATHPOS],
|
|
|
250 "< $%s" % p[ICLPOS],
|
|
|
251 ]
|
|
|
252 else:
|
|
|
253 appendme = [p[IOCLPOS], p[ICLPOS], p[IPATHPOS], ""]
|
|
|
254 xappendme = [p[IOCLPOS], p[ICLPOS], "$%s" % p[ICLPOS], ""]
|
|
|
255 clsuffix.append(appendme)
|
|
|
256 xclsuffix.append(xappendme)
|
|
|
257 for i, p in enumerate(self.outfiles):
|
|
|
258 if p[OOCLPOS] == "STDOUT":
|
|
|
259 self.lastclredirect = [">", p[ONAMEPOS]]
|
|
|
260 self.lastxclredirect = [">", "$%s" % p[OCLPOS]]
|
|
|
261 else:
|
|
|
262 clsuffix.append([p[OOCLPOS], p[OCLPOS], p[ONAMEPOS], ""])
|
|
49
|
263 xclsuffix.append([p[OOCLPOS], p[OCLPOS], "$%s" % p[ONAMEPOS], ""])
|
|
48
|
264 for p in self.addpar:
|
|
49
|
265 clsuffix.append([p[AOCLPOS], p[ACLPOS], p[AVALPOS], p[AOVERPOS]])
|
|
48
|
266 xclsuffix.append(
|
|
|
267 [p[AOCLPOS], p[ACLPOS], '"$%s"' % p[ANAMEPOS], p[AOVERPOS]]
|
|
|
268 )
|
|
|
269 clsuffix.sort()
|
|
|
270 xclsuffix.sort()
|
|
|
271 self.xclsuffix = xclsuffix
|
|
|
272 self.clsuffix = clsuffix
|
|
|
273 if self.args.parampass == "positional":
|
|
|
274 self.clpositional()
|
|
|
275 else:
|
|
|
276 self.clargparse()
|
|
|
277
|
|
|
278 def prepScript(self):
|
|
|
279 rx = open(self.args.script_path, "r").readlines()
|
|
|
280 rx = [x.rstrip() for x in rx]
|
|
|
281 rxcheck = [x.strip() for x in rx if x.strip() > ""]
|
|
|
282 assert len(rxcheck) > 0, "Supplied script is empty. Cannot run"
|
|
|
283 self.script = "\n".join(rx)
|
|
|
284 fhandle, self.sfile = tempfile.mkstemp(
|
|
49
|
285 prefix=self.tool_name, suffix="_%s" % (self.executeme)
|
|
48
|
286 )
|
|
|
287 tscript = open(self.sfile, "w")
|
|
|
288 tscript.write(self.script)
|
|
|
289 tscript.close()
|
|
49
|
290 self.indentedScript = " %s" % "\n".join([" %s" % html_escape(x) for x in rx])
|
|
|
291 self.escapedScript = "%s" % "\n".join([" %s" % html_escape(x) for x in rx])
|
|
|
292 art = "%s.%s" % (self.tool_name, self.executeme)
|
|
48
|
293 artifact = open(art, "wb")
|
|
|
294 artifact.write(bytes(self.script, "utf8"))
|
|
|
295 artifact.close()
|
|
|
296
|
|
|
297 def cleanuppar(self):
|
|
|
298 """ positional parameters are complicated by their numeric ordinal"""
|
|
|
299 for i, p in enumerate(self.infiles):
|
|
|
300 if self.args.parampass == "positional":
|
|
|
301 assert p[ICLPOS].isdigit(), (
|
|
|
302 "Positional parameters must be ordinal integers - got %s for %s"
|
|
|
303 % (p[ICLPOS], p[ILABPOS])
|
|
|
304 )
|
|
|
305 p.append(p[ICLPOS])
|
|
|
306 if p[ICLPOS].isdigit() or self.args.parampass == "0":
|
|
|
307 scl = "input%d" % (i + 1)
|
|
|
308 p[ICLPOS] = scl
|
|
|
309 self.infiles[i] = p
|
|
|
310 for i, p in enumerate(
|
|
|
311 self.outfiles
|
|
|
312 ): # trying to automagically gather using extensions
|
|
|
313 if self.args.parampass == "positional" and p[OCLPOS] != "STDOUT":
|
|
|
314 assert p[OCLPOS].isdigit(), (
|
|
|
315 "Positional parameters must be ordinal integers - got %s for %s"
|
|
|
316 % (p[OCLPOS], p[ONAMEPOS])
|
|
|
317 )
|
|
|
318 p.append(p[OCLPOS])
|
|
|
319 if p[OCLPOS].isdigit() or p[OCLPOS] == "STDOUT":
|
|
|
320 scl = p[ONAMEPOS]
|
|
|
321 p[OCLPOS] = scl
|
|
|
322 self.outfiles[i] = p
|
|
|
323 for i, p in enumerate(self.addpar):
|
|
|
324 if self.args.parampass == "positional":
|
|
|
325 assert p[ACLPOS].isdigit(), (
|
|
|
326 "Positional parameters must be ordinal integers - got %s for %s"
|
|
|
327 % (p[ACLPOS], p[ANAMEPOS])
|
|
|
328 )
|
|
|
329 p.append(p[ACLPOS])
|
|
|
330 if p[ACLPOS].isdigit():
|
|
|
331 scl = "input%s" % p[ACLPOS]
|
|
|
332 p[ACLPOS] = scl
|
|
|
333 self.addpar[i] = p
|
|
|
334
|
|
|
335 def clsimple(self):
|
|
|
336 """ no parameters - uses < and > for i/o
|
|
|
337 """
|
|
|
338 aCL = self.cl.append
|
|
|
339 aCL("<")
|
|
|
340 aCL(self.infiles[0][IPATHPOS])
|
|
|
341 aCL(">")
|
|
|
342 aCL(self.outfiles[0][OCLPOS])
|
|
|
343 aXCL = self.xmlcl.append
|
|
|
344 aXCL("<")
|
|
|
345 aXCL("$%s" % self.infiles[0][ICLPOS])
|
|
|
346 aXCL(">")
|
|
|
347 aXCL("$%s" % self.outfiles[0][ONAMEPOS])
|
|
|
348
|
|
|
349 def clpositional(self):
|
|
|
350 # inputs in order then params
|
|
|
351 aCL = self.cl.append
|
|
|
352 for (o_v, k, v, koverride) in self.clsuffix:
|
|
|
353 if " " in v:
|
|
|
354 aCL("%s" % v)
|
|
|
355 else:
|
|
|
356 aCL(v)
|
|
|
357 aXCL = self.xmlcl.append
|
|
|
358 for (o_v, k, v, koverride) in self.xclsuffix:
|
|
|
359 aXCL(v)
|
|
|
360 if self.lastxclredirect:
|
|
|
361 aXCL(self.lastxclredirect[0])
|
|
|
362 aXCL(self.lastxclredirect[1])
|
|
|
363
|
|
|
364 def clargparse(self):
|
|
|
365 """ argparse style
|
|
|
366 """
|
|
|
367 aCL = self.cl.append
|
|
|
368 aXCL = self.xmlcl.append
|
|
|
369 # inputs then params in argparse named form
|
|
|
370 for (o_v, k, v, koverride) in self.xclsuffix:
|
|
|
371 if koverride > "":
|
|
|
372 k = koverride
|
|
|
373 elif len(k.strip()) == 1:
|
|
|
374 k = "-%s" % k
|
|
|
375 else:
|
|
|
376 k = "--%s" % k
|
|
|
377 aXCL(k)
|
|
|
378 aXCL(v)
|
|
|
379 for (o_v, k, v, koverride) in self.clsuffix:
|
|
|
380 if koverride > "":
|
|
|
381 k = koverride
|
|
|
382 elif len(k.strip()) == 1:
|
|
|
383 k = "-%s" % k
|
|
|
384 else:
|
|
|
385 k = "--%s" % k
|
|
|
386 aCL(k)
|
|
|
387 aCL(v)
|
|
|
388
|
|
|
389 def getNdash(self, newname):
|
|
|
390 if self.is_positional:
|
|
|
391 ndash = 0
|
|
|
392 else:
|
|
|
393 ndash = 2
|
|
|
394 if len(newname) < 2:
|
|
|
395 ndash = 1
|
|
|
396 return ndash
|
|
|
397
|
|
|
398 def doXMLparam(self):
|
|
|
399 """flake8 made me do this..."""
|
|
|
400 for p in self.outfiles:
|
|
49
|
401 newname, newfmt, newcl, test, oldcl = p
|
|
48
|
402 ndash = self.getNdash(newcl)
|
|
|
403 aparm = gxtp.OutputData(newcl, format=newfmt, num_dashes=ndash)
|
|
|
404 aparm.positional = self.is_positional
|
|
|
405 if self.is_positional:
|
|
|
406 if oldcl == "STDOUT":
|
|
|
407 aparm.positional = 9999999
|
|
|
408 aparm.command_line_override = "> $%s" % newcl
|
|
|
409 else:
|
|
|
410 aparm.positional = int(oldcl)
|
|
|
411 aparm.command_line_override = "$%s" % newcl
|
|
|
412 self.toutputs.append(aparm)
|
|
49
|
413 usetest = None
|
|
|
414 ld = None
|
|
|
415 if test > '':
|
|
|
416 if test.startswith('diff'):
|
|
|
417 usetest = 'diff'
|
|
|
418 if test.split(':')[1].isdigit:
|
|
|
419 ld = int(test.split(':')[1])
|
|
|
420 else:
|
|
|
421 usetest = test
|
|
|
422 tp = gxtp.TestOutput(name=newcl, value="%s_sample" % newcl, format=newfmt,
|
|
|
423 compare=usetest, lines_diff=ld, delta=None,)
|
|
48
|
424 self.testparam.append(tp)
|
|
|
425 for p in self.infiles:
|
|
|
426 newname = p[ICLPOS]
|
|
|
427 newfmt = p[IFMTPOS]
|
|
|
428 ndash = self.getNdash(newname)
|
|
|
429 if not len(p[ILABPOS]) > 0:
|
|
|
430 alab = p[ICLPOS]
|
|
|
431 else:
|
|
|
432 alab = p[ILABPOS]
|
|
|
433 aninput = gxtp.DataParam(
|
|
|
434 newname,
|
|
|
435 optional=False,
|
|
|
436 label=alab,
|
|
|
437 help=p[IHELPOS],
|
|
|
438 format=newfmt,
|
|
|
439 multiple=False,
|
|
|
440 num_dashes=ndash,
|
|
|
441 )
|
|
|
442 aninput.positional = self.is_positional
|
|
|
443 self.tinputs.append(aninput)
|
|
|
444 tparm = gxtp.TestParam(name=newname, value="%s_sample" % newname)
|
|
|
445 self.testparam.append(tparm)
|
|
|
446 for p in self.addpar:
|
|
|
447 newname, newval, newlabel, newhelp, newtype, newcl, override, oldcl = p
|
|
|
448 if not len(newlabel) > 0:
|
|
|
449 newlabel = newname
|
|
|
450 ndash = self.getNdash(newname)
|
|
|
451 if newtype == "text":
|
|
|
452 aparm = gxtp.TextParam(
|
|
|
453 newname,
|
|
|
454 label=newlabel,
|
|
|
455 help=newhelp,
|
|
|
456 value=newval,
|
|
|
457 num_dashes=ndash,
|
|
|
458 )
|
|
|
459 elif newtype == "integer":
|
|
|
460 aparm = gxtp.IntegerParam(
|
|
|
461 newname,
|
|
|
462 label=newname,
|
|
|
463 help=newhelp,
|
|
|
464 value=newval,
|
|
|
465 num_dashes=ndash,
|
|
|
466 )
|
|
|
467 elif newtype == "float":
|
|
|
468 aparm = gxtp.FloatParam(
|
|
|
469 newname,
|
|
|
470 label=newname,
|
|
|
471 help=newhelp,
|
|
|
472 value=newval,
|
|
|
473 num_dashes=ndash,
|
|
|
474 )
|
|
|
475 else:
|
|
|
476 raise ValueError(
|
|
|
477 'Unrecognised parameter type "%s" for\
|
|
|
478 additional parameter %s in makeXML'
|
|
|
479 % (newtype, newname)
|
|
|
480 )
|
|
|
481 aparm.positional = self.is_positional
|
|
|
482 if self.is_positional:
|
|
|
483 aninput.positional = int(oldcl)
|
|
|
484 self.tinputs.append(aparm)
|
|
|
485 self.tparm = gxtp.TestParam(newname, value=newval)
|
|
|
486 self.testparam.append(tparm)
|
|
|
487
|
|
|
488 def doNoXMLparam(self):
|
|
49
|
489 """filter style package - stdin to stdout"""
|
|
48
|
490 alab = self.infiles[0][ILABPOS]
|
|
|
491 if len(alab) == 0:
|
|
|
492 alab = self.infiles[0][ICLPOS]
|
|
|
493 max1s = (
|
|
49
|
494 "Maximum one input if parampass is 0 but multiple input files supplied - %s"
|
|
48
|
495 % str(self.infiles)
|
|
|
496 )
|
|
|
497 assert len(self.infiles) == 1, max1s
|
|
|
498 newname = self.infiles[0][ICLPOS]
|
|
|
499 aninput = gxtp.DataParam(
|
|
|
500 newname,
|
|
|
501 optional=False,
|
|
|
502 label=alab,
|
|
|
503 help=self.infiles[0][IHELPOS],
|
|
|
504 format=self.infiles[0][IFMTPOS],
|
|
|
505 multiple=False,
|
|
|
506 num_dashes=0,
|
|
|
507 )
|
|
|
508 aninput.command_line_override = "< $%s" % newname
|
|
|
509 aninput.positional = self.is_positional
|
|
|
510 self.tinputs.append(aninput)
|
|
|
511 tp = gxtp.TestParam(name=newname, value="%s_sample" % newname)
|
|
|
512 self.testparam.append(tp)
|
|
|
513 newname = self.outfiles[0][OCLPOS]
|
|
|
514 newfmt = self.outfiles[0][OFMTPOS]
|
|
|
515 anout = gxtp.OutputData(newname, format=newfmt, num_dashes=0)
|
|
|
516 anout.command_line_override = "> $%s" % newname
|
|
|
517 anout.positional = self.is_positional
|
|
|
518 self.toutputs.append(anout)
|
|
49
|
519 tp = gxtp.TestOutput(name=newname, value="%s_sample" % newname, format=newfmt)
|
|
48
|
520 self.testparam.append(tp)
|
|
|
521
|
|
|
522 def makeXML(self):
|
|
|
523 """
|
|
|
524 Create a Galaxy xml tool wrapper for the new script
|
|
|
525 Uses galaxyhtml
|
|
|
526 Hmmm. How to get the command line into correct order...
|
|
|
527 """
|
|
49
|
528 if self.command_override:
|
|
|
529 self.tool.command_line_override = self.command_override # config file
|
|
48
|
530 else:
|
|
|
531 self.tool.command_line_override = self.xmlcl
|
|
49
|
532 # if self.args.interpreter_name:
|
|
|
533 # self.tool.interpreter = self.args.interpreter_name
|
|
48
|
534 if self.args.help_text:
|
|
|
535 helptext = open(self.args.help_text, "r").readlines()
|
|
|
536 helptext = [html_escape(x) for x in helptext]
|
|
|
537 self.tool.help = "".join([x for x in helptext])
|
|
|
538 else:
|
|
|
539 self.tool.help = (
|
|
|
540 "Please ask the tool author (%s) for help \
|
|
|
541 as none was supplied at tool generation\n"
|
|
|
542 % (self.args.user_email)
|
|
|
543 )
|
|
|
544 self.tool.version_command = None # do not want
|
|
|
545 requirements = gxtp.Requirements()
|
|
49
|
546 if self.args.packages:
|
|
|
547 for d in self.args.packages.split(","):
|
|
|
548 if ":" in d:
|
|
|
549 packg, ver = d.split(":")
|
|
|
550 else:
|
|
|
551 packg = d
|
|
|
552 ver = ""
|
|
|
553 requirements.append(gxtp.Requirement("package", packg.strip(), ver.strip()))
|
|
48
|
554 self.tool.requirements = requirements
|
|
|
555 if self.args.parampass == "0":
|
|
|
556 self.doNoXMLparam()
|
|
|
557 else:
|
|
|
558 self.doXMLparam()
|
|
|
559 self.tool.outputs = self.toutputs
|
|
|
560 self.tool.inputs = self.tinputs
|
|
49
|
561 if (
|
|
|
562 self.args.script_path
|
|
|
563 ):
|
|
48
|
564 configfiles = gxtp.Configfiles()
|
|
49
|
565 configfiles.append(gxtp.Configfile(name="runme", text=self.script))
|
|
48
|
566 self.tool.configfiles = configfiles
|
|
|
567 tests = gxtp.Tests()
|
|
|
568 test_a = gxtp.Test()
|
|
|
569 for tp in self.testparam:
|
|
|
570 test_a.append(tp)
|
|
|
571 tests.append(test_a)
|
|
|
572 self.tool.tests = tests
|
|
|
573 self.tool.add_comment(
|
|
|
574 "Created by %s at %s using the Galaxy Tool Factory."
|
|
|
575 % (self.args.user_email, timenow())
|
|
|
576 )
|
|
|
577 self.tool.add_comment("Source in git at: %s" % (toolFactoryURL))
|
|
|
578 self.tool.add_comment(
|
|
|
579 "Cite: Creating re-usable tools from scripts doi: \
|
|
|
580 10.1093/bioinformatics/bts573"
|
|
|
581 )
|
|
49
|
582 exml0 = self.tool.export()
|
|
|
583 exml = exml0.replace(FAKEEXE, "") # temporary work around until PR accepted
|
|
|
584 if self.test_override: # cannot do this inside galaxyxml as it expects lxml objects for tests
|
|
|
585 part1 = exml.split('<tests>')[0]
|
|
|
586 part2 = exml.split('</tests>')[1]
|
|
|
587 fixed = '%s\n%s\n%s' % (part1,self.test_override,part2)
|
|
|
588 exml = fixed
|
|
|
589 xf = open("%s.xml" % self.tool_name, "w")
|
|
48
|
590 xf.write(exml)
|
|
|
591 xf.write("\n")
|
|
|
592 xf.close()
|
|
|
593 # ready for the tarball
|
|
|
594
|
|
49
|
595 def makeTool(self):
|
|
|
596 """write xmls and samples into place
|
|
48
|
597 """
|
|
|
598 self.makeXML()
|
|
49
|
599 if self.args.script_path:
|
|
|
600 stname = os.path.join(self.tooloutdir, "%s" % (self.sfile))
|
|
|
601 if not os.path.exists(stname):
|
|
|
602 shutil.copyfile(self.sfile, stname)
|
|
|
603 xreal = "%s.xml" % self.tool_name
|
|
|
604 xout = os.path.join(self.tooloutdir, xreal)
|
|
|
605 shutil.copyfile(xreal, xout)
|
|
48
|
606 for p in self.infiles:
|
|
|
607 pth = p[IPATHPOS]
|
|
|
608 dest = os.path.join(self.testdir, "%s_sample" % p[ICLPOS])
|
|
|
609 shutil.copyfile(pth, dest)
|
|
49
|
610
|
|
|
611 def makeToolTar(self):
|
|
48
|
612 self.newtarpath = "toolfactory_%s.tgz" % self.tool_name
|
|
|
613 tf = tarfile.open(self.newtarpath, "w:gz")
|
|
|
614 tf.add(name=self.tooloutdir, arcname=self.tool_name)
|
|
|
615 tf.close()
|
|
|
616 shutil.copyfile(self.newtarpath, self.args.new_tool)
|
|
49
|
617 if os.path.exists(self.tlog) and os.stat(self.tlog).st_size > 0:
|
|
|
618 shutil.copyfile(
|
|
|
619 self.tlog, os.path.join(self.tooloutdir, "test1_log_outfiletxt")
|
|
|
620 )
|
|
|
621
|
|
|
622 def moveRunOutputs(self):
|
|
|
623 """need to move files into toolfactory collection after any run - planemo or not
|
|
|
624 """
|
|
|
625 for p in self.outfiles:
|
|
|
626 naym = p[ONAMEPOS]
|
|
|
627 src = os.path.join(self.tooloutdir,naym)
|
|
|
628 if os.path.isfile(src):
|
|
|
629 dest = os.path.join(self.testdir, "%s_sample" % naym)
|
|
|
630 shutil.copyfile(naym, dest)
|
|
|
631 else:
|
|
|
632 print('### problem - output file %s not found in tooloutdir %s' % (src,self.tooloutdir))
|
|
|
633 with os.scandir(self.tooloutdir) as outs:
|
|
48
|
634 for entry in outs:
|
|
49
|
635 if not entry.is_file() or entry.name.startswith('.'):
|
|
48
|
636 continue
|
|
49
|
637 if "." in entry.name:
|
|
|
638 nayme,ext = os.path.splitext(entry.name)
|
|
|
639 else:
|
|
|
640 ext = ".txt"
|
|
|
641 ofn = "%s%s" % (entry.name.replace(".", "_"), ext)
|
|
|
642 dest = os.path.join(self.repdir, ofn)
|
|
|
643 src = os.path.join(self.tooloutdir,entry.name)
|
|
|
644 shutil.copyfile(src, dest)
|
|
48
|
645
|
|
|
646 def run(self):
|
|
|
647 """
|
|
|
648
|
|
|
649 """
|
|
|
650 s = "run cl=%s" % str(self.cl)
|
|
|
651
|
|
|
652 logging.debug(s)
|
|
|
653 scl = " ".join(self.cl)
|
|
|
654 err = None
|
|
|
655 if self.args.parampass != "0":
|
|
|
656 ste = open(self.elog, "wb")
|
|
|
657 if self.lastclredirect:
|
|
49
|
658 sto = open(self.lastclredirect[1], "wb") # is name of an output file
|
|
48
|
659 else:
|
|
|
660 sto = open(self.tlog, "wb")
|
|
|
661 sto.write(
|
|
|
662 bytes(
|
|
49
|
663 "## Executing Toolfactory generated command line = %s\n" % scl,
|
|
48
|
664 "utf8",
|
|
|
665 )
|
|
|
666 )
|
|
|
667 sto.flush()
|
|
|
668 p = subprocess.run(self.cl, shell=False, stdout=sto, stderr=ste)
|
|
|
669 sto.close()
|
|
|
670 ste.close()
|
|
|
671 tmp_stderr = open(self.elog, "rb")
|
|
|
672 err = ""
|
|
|
673 buffsize = 1048576
|
|
|
674 try:
|
|
|
675 while True:
|
|
|
676 err += str(tmp_stderr.read(buffsize))
|
|
|
677 if not err or len(err) % buffsize != 0:
|
|
|
678 break
|
|
|
679 except OverflowError:
|
|
|
680 pass
|
|
|
681 tmp_stderr.close()
|
|
|
682 retval = p.returncode
|
|
49
|
683 else: # work around special case - stdin and write to stdout
|
|
48
|
684 sti = open(self.infiles[0][IPATHPOS], "rb")
|
|
|
685 sto = open(self.outfiles[0][ONAMEPOS], "wb")
|
|
|
686 # must use shell to redirect
|
|
|
687 p = subprocess.run(self.cl, shell=False, stdout=sto, stdin=sti)
|
|
|
688 retval = p.returncode
|
|
|
689 sto.close()
|
|
|
690 sti.close()
|
|
|
691 if os.path.isfile(self.tlog) and os.stat(self.tlog).st_size == 0:
|
|
|
692 os.unlink(self.tlog)
|
|
|
693 if os.path.isfile(self.elog) and os.stat(self.elog).st_size == 0:
|
|
|
694 os.unlink(self.elog)
|
|
|
695 if retval != 0 and err: # problem
|
|
|
696 sys.stderr.write(err)
|
|
|
697 logging.debug("run done")
|
|
|
698 return retval
|
|
|
699
|
|
|
700 def planemo_shedload(self):
|
|
|
701 """
|
|
|
702 planemo shed_create --shed_target testtoolshed
|
|
|
703 planemo shed_update --check_diff --shed_target testtoolshed
|
|
|
704 """
|
|
49
|
705 if os.path.exists(self.tlog):
|
|
|
706 tout = open(self.tlog,'a')
|
|
|
707 else:
|
|
|
708 tout = open(self.tlog,'w')
|
|
|
709 cll = ["planemo", "shed_create", "--shed_target", "local"]
|
|
|
710 try:
|
|
|
711 p = subprocess.run(cll, shell=False, cwd=self.tooloutdir, stdout=tout, stderr = tout)
|
|
|
712 except:
|
|
|
713 pass
|
|
48
|
714 if p.returncode != 0:
|
|
49
|
715 print("Repository %s exists" % self.args.tool_name)
|
|
48
|
716 else:
|
|
49
|
717 print("initiated %s" % self.args.tool_name)
|
|
|
718 cll = [
|
|
|
719 "planemo",
|
|
|
720 "shed_upload",
|
|
|
721 "--shed_target",
|
|
|
722 "local",
|
|
|
723 "--owner",
|
|
|
724 "fubar",
|
|
|
725 "--name",
|
|
|
726 self.args.tool_name,
|
|
|
727 "--shed_key",
|
|
|
728 self.args.toolshed_api_key,
|
|
|
729 "--tar",
|
|
|
730 self.newtarpath,
|
|
|
731 ]
|
|
|
732 print("Run", " ".join(cll))
|
|
|
733 p = subprocess.run(cll, shell=False)
|
|
|
734 print("Ran", " ".join(cll), "got", p.returncode)
|
|
|
735 tout.close()
|
|
48
|
736 return p.returncode
|
|
|
737
|
|
49
|
738 def planemo_test(self, genoutputs=True):
|
|
48
|
739 """planemo is a requirement so is available
|
|
|
740 """
|
|
49
|
741 xreal = "%s.xml" % self.tool_name
|
|
|
742 if os.path.exists(self.tlog):
|
|
|
743 tout = open(self.tlog,'a')
|
|
|
744 else:
|
|
|
745 tout = open(self.tlog,'w')
|
|
|
746 if genoutputs:
|
|
|
747 cll = [
|
|
|
748 "planemo",
|
|
|
749 "test",
|
|
|
750 "--galaxy_root",
|
|
|
751 self.args.galaxy_root,
|
|
|
752 "--update_test_data",
|
|
|
753 xreal,
|
|
|
754 ]
|
|
|
755 else:
|
|
|
756 cll = ["planemo", "test", "--galaxy_root", self.args.galaxy_root, xreal]
|
|
|
757 try:
|
|
|
758 p = subprocess.run(cll, shell=False, cwd=self.tooloutdir, stderr=tout, stdout=tout)
|
|
|
759 except:
|
|
|
760 pass
|
|
|
761 tout.close()
|
|
48
|
762 return p.returncode
|
|
|
763
|
|
|
764 def eph_galaxy_load(self):
|
|
|
765 """
|
|
|
766 """
|
|
49
|
767 if os.path.exists(self.tlog):
|
|
|
768 tout = open(self.tlog,'a')
|
|
|
769 else:
|
|
|
770 tout = open(self.tlog,'w')
|
|
|
771 cll = [
|
|
|
772 "shed-tools",
|
|
|
773 "install",
|
|
|
774 "-g",
|
|
|
775 self.args.galaxy_url,
|
|
|
776 "--latest",
|
|
|
777 "-a",
|
|
|
778 self.args.galaxy_api_key,
|
|
|
779 "--name",
|
|
|
780 self.args.tool_name,
|
|
|
781 "--owner",
|
|
|
782 "fubar",
|
|
|
783 "--toolshed",
|
|
|
784 self.args.toolshed_url,
|
|
|
785 "--section_label",
|
|
|
786 "Generated Tools",
|
|
|
787 "--install_tool_dependencies",
|
|
|
788 ]
|
|
|
789 print("running\n", " ".join(cll))
|
|
|
790 p = subprocess.run(cll, shell=False, stderr=tout, stdout=tout)
|
|
48
|
791 if p.returncode != 0:
|
|
49
|
792 print(
|
|
|
793 "Repository %s installation returned %d"
|
|
|
794 % (self.args.tool_name, p.returncode)
|
|
|
795 )
|
|
48
|
796 else:
|
|
49
|
797 print("installed %s" % self.args.tool_name)
|
|
|
798 tout.close()
|
|
48
|
799 return p.returncode
|
|
|
800
|
|
|
801 def writeShedyml(self):
|
|
49
|
802 yuser = self.args.user_email.split("@")[0]
|
|
|
803 yfname = os.path.join(self.tooloutdir, ".shed.yml")
|
|
|
804 yamlf = open(yfname, "w")
|
|
|
805 odict = {
|
|
|
806 "name": self.tool_name,
|
|
|
807 "owner": yuser,
|
|
|
808 "type": "unrestricted",
|
|
|
809 "description": self.args.tool_desc,
|
|
|
810 }
|
|
48
|
811 yaml.dump(odict, yamlf, allow_unicode=True)
|
|
|
812 yamlf.close()
|
|
|
813
|
|
49
|
814
|
|
|
815 def install_load(self):
|
|
|
816 _ = self.planemo_test(genoutputs=True)
|
|
|
817 testres = self.planemo_test(genoutputs=False)
|
|
|
818 if testres == 0:
|
|
|
819 if self.args.make_Tool == "install":
|
|
|
820 self.planemo_shedload()
|
|
|
821 self.eph_galaxy_load()
|
|
|
822 else:
|
|
|
823 os.stderr.write(
|
|
|
824 "Planemo test failed - tool %s was not installed" % self.args.tool_name
|
|
|
825 )
|
|
|
826
|
|
48
|
827 def main():
|
|
|
828 """
|
|
|
829 This is a Galaxy wrapper. It expects to be called by a special purpose tool.xml as:
|
|
49
|
830 <command interpreter="python">rgBaseScriptWrapper.py --script_path "$scriptPath"
|
|
|
831 --tool_name "foo" --interpreter "Rscript"
|
|
48
|
832 </command>
|
|
|
833 """
|
|
|
834 parser = argparse.ArgumentParser()
|
|
|
835 a = parser.add_argument
|
|
49
|
836 a("--script_path", default=None)
|
|
|
837 a("--history_test", default=None)
|
|
|
838 a("--cl_prefix", default=None)
|
|
|
839 a("--sysexe", default=None)
|
|
|
840 a("--packages", default=None)
|
|
48
|
841 a("--tool_name", default=None)
|
|
|
842 a("--input_files", default=[], action="append")
|
|
|
843 a("--output_files", default=[], action="append")
|
|
|
844 a("--user_email", default="Unknown")
|
|
|
845 a("--bad_user", default=None)
|
|
49
|
846 a("--make_Tool", default="runonly")
|
|
48
|
847 a("--help_text", default=None)
|
|
|
848 a("--tool_desc", default=None)
|
|
|
849 a("--tool_version", default=None)
|
|
|
850 a("--citations", default=None)
|
|
49
|
851 a("--command_override", default=None)
|
|
|
852 a("--test_override", default=None)
|
|
48
|
853 a("--additional_parameters", action="append", default=[])
|
|
|
854 a("--edit_additional_parameters", action="store_true", default=False)
|
|
|
855 a("--parampass", default="positional")
|
|
|
856 a("--tfout", default="./tfout")
|
|
|
857 a("--new_tool", default="new_tool")
|
|
|
858 a("--runmode", default=None)
|
|
49
|
859 a("--galaxy_url", default="http://localhost:8080")
|
|
|
860 a("--galaxy_api_key", default="fbdd3c2eecd191e88939fffc02eeeaf8")
|
|
48
|
861 a("--toolshed_url", default="http://localhost:9009")
|
|
49
|
862 a("--toolshed_api_key", default="d46e5ed0e242ed52c6e1f506b5d7f9f7")
|
|
|
863 a("--galaxy_root", default="/home/ross/galaxy")
|
|
48
|
864
|
|
|
865 args = parser.parse_args()
|
|
|
866 assert not args.bad_user, (
|
|
|
867 'UNAUTHORISED: %s is NOT authorized to use this tool until Galaxy admin adds %s to "admin_users" in the Galaxy configuration file'
|
|
|
868 % (args.bad_user, args.bad_user)
|
|
|
869 )
|
|
49
|
870 assert args.tool_name, "## Tool Factory expects a tool name - eg --tool_name=DESeq"
|
|
48
|
871 assert (
|
|
49
|
872 args.sysexe or args.packages
|
|
48
|
873 ), "## Tool Factory wrapper expects an interpreter or an executable package"
|
|
49
|
874 args.input_files = [x.replace('"', "").replace("'", "") for x in args.input_files]
|
|
48
|
875 # remove quotes we need to deal with spaces in CL params
|
|
|
876 for i, x in enumerate(args.additional_parameters):
|
|
49
|
877 args.additional_parameters[i] = args.additional_parameters[i].replace('"', "")
|
|
48
|
878 r = ScriptRunner(args)
|
|
49
|
879 r.writeShedyml()
|
|
|
880 r.makeTool()
|
|
|
881 if args.make_Tool == "runonly":
|
|
|
882 retcode = r.run()
|
|
|
883 if retcode:
|
|
|
884 sys.stderr.write(
|
|
|
885 "## Run failed with return code %d. Cannot build yet. Please fix and retry"
|
|
|
886 % retcode
|
|
|
887 )
|
|
|
888 sys.exit(1)
|
|
|
889 else:
|
|
|
890 r.moveRunOutputs()
|
|
|
891 elif args.make_Tool in ["gentestinstall", "generate", "gentest"]:
|
|
|
892 retcode = r.run()
|
|
|
893 if retcode:
|
|
|
894 sys.stderr.write(
|
|
|
895 "## Run failed with return code %d. Cannot build yet. Please fix and retry"
|
|
|
896 % retcode
|
|
|
897 )
|
|
|
898 sys.exit(1)
|
|
|
899 r.moveRunOutputs()
|
|
|
900 r.makeToolTar()
|
|
|
901 if args.make_Tool in ["gentestinstall","gentest"]:
|
|
|
902 r.planemo_test(genoutputs=False)
|
|
|
903 r.moveRunOutputs()
|
|
|
904 r.planemo_shedload()
|
|
|
905 r.eph_galaxy_load()
|
|
48
|
906 else:
|
|
49
|
907 retcode = r.planemo_test(genoutputs=True) # this fails :(
|
|
|
908 r.moveRunOutputs()
|
|
|
909 r.makeToolTar()
|
|
|
910 retcode = r.planemo_test(genoutputs=False)
|
|
|
911 r.moveRunOutputs()
|
|
|
912 if args.make_Tool == "planemotestinstall":
|
|
|
913 r.planemo_shedload()
|
|
|
914 r.eph_galaxy_load()
|
|
|
915 # if retcode:
|
|
48
|
916 # sys.exit(retcode) # indicate failure to job runner
|
|
|
917
|
|
|
918
|
|
|
919 if __name__ == "__main__":
|
|
|
920 main()
|
