summaryrefslogtreecommitdiff
path: root/third_party/waf/waflib/extras/run_do_script.py
blob: f3c58122c9b73cee10a955449ed4a55003ab7c70 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
#!/usr/bin/env python
# encoding: utf-8
# Hans-Martin von Gaudecker, 2012

"""
Run a Stata do-script in the directory specified by **ctx.bldnode**. The
first and only argument will be the name of the do-script (no extension),
which can be accessed inside the do-script by the local macro `1'. Useful
for keeping a log file.

The tool uses the log file that is automatically kept by Stata only 
for error-catching purposes, it will be destroyed if the task finished
without error. In case of an error in **some_script.do**, you can inspect
it as **some_script.log** in the **ctx.bldnode** directory.

Note that Stata will not return an error code if it exits abnormally -- 
catching errors relies on parsing the log file mentioned before. Should
the parser behave incorrectly please send an email to hmgaudecker [at] gmail.

**WARNING**

    The tool will not work if multiple do-scripts of the same name---but in
    different directories---are run at the same time! Avoid this situation.

Usage::

    ctx(features='run_do_script', 
        source='some_script.do',
        target=['some_table.tex', 'some_figure.eps'],
        deps='some_data.csv')
"""


import os, re, sys
from waflib import Task, TaskGen, Logs

if sys.platform == 'darwin':
	STATA_COMMANDS = ['Stata64MP', 'StataMP',
								'Stata64SE', 'StataSE', 
								'Stata64', 'Stata']
	STATAFLAGS = '-e -q do'
	STATAENCODING = 'MacRoman'
elif sys.platform.startswith('linux'):
	STATA_COMMANDS = ['stata-mp', 'stata-se', 'stata']
	STATAFLAGS = '-b -q do'
	# Not sure whether this is correct...
	STATAENCODING = 'Latin-1'
elif sys.platform.lower().startswith('win'):
	STATA_COMMANDS = ['StataMP-64', 'StataMP-ia',
								'StataMP', 'StataSE-64',
								'StataSE-ia', 'StataSE',
								'Stata-64', 'Stata-ia',
								'Stata.e', 'WMPSTATA',
								'WSESTATA', 'WSTATA']
	STATAFLAGS = '/e do'
	STATAENCODING = 'Latin-1'
else:
	raise Exception("Unknown sys.platform: %s " % sys.platform)

def configure(ctx):
	ctx.find_program(STATA_COMMANDS, var='STATACMD', errmsg="""\n
No Stata executable found!\n\n
If Stata is needed:\n
	1) Check the settings of your system path.
	2) Note we are looking for Stata executables called: %s
	   If yours has a different name, please report to hmgaudecker [at] gmail\n
Else:\n
	Do not load the 'run_do_script' tool in the main wscript.\n\n""" % STATA_COMMANDS)
	ctx.env.STATAFLAGS = STATAFLAGS
	ctx.env.STATAENCODING = STATAENCODING

class run_do_script_base(Task.Task):
	"""Run a Stata do-script from the bldnode directory."""
	run_str = '"${STATACMD}" ${STATAFLAGS} "${SRC[0].abspath()}" "${DOFILETRUNK}"'
	shell = True

class run_do_script(run_do_script_base):
	"""Use the log file automatically kept by Stata for error-catching.
	Erase it if the task finished without error. If not, it will show 
	up as do_script.log in the bldnode directory.
	"""
	def run(self):
		run_do_script_base.run(self)
		ret, log_tail  = self.check_erase_log_file()
		if ret:
			Logs.error("""Running Stata on %r failed with code %r.\n\nCheck the log file %s, last 10 lines\n\n%s\n\n\n""",
				self.inputs[0], ret, self.env.LOGFILEPATH, log_tail)
		return ret

	def check_erase_log_file(self):
		"""Parse Stata's default log file and erase it if everything okay.

		Parser is based on Brendan Halpin's shell script found here:
			http://teaching.sociology.ul.ie/bhalpin/wordpress/?p=122
		"""

		if sys.version_info.major >= 3:
			kwargs = {'file': self.env.LOGFILEPATH, 'mode': 'r', 'encoding': self.env.STATAENCODING}
		else:
			kwargs = {'name': self.env.LOGFILEPATH, 'mode': 'r'}
		with open(**kwargs) as log:
			log_tail = log.readlines()[-10:]
			for line in log_tail:
				error_found = re.match("r\(([0-9]+)\)", line)
				if error_found:
					return error_found.group(1), ''.join(log_tail)
				else:
					pass
		# Only end up here if the parser did not identify an error.
		os.remove(self.env.LOGFILEPATH)
		return None, None


@TaskGen.feature('run_do_script')
@TaskGen.before_method('process_source')
def apply_run_do_script(tg):
	"""Task generator customising the options etc. to call Stata in batch
	mode for running a do-script.
	"""

	# Convert sources and targets to nodes
	src_node = tg.path.find_resource(tg.source)
	tgt_nodes = [tg.path.find_or_declare(t) for t in tg.to_list(tg.target)]

	tsk = tg.create_task('run_do_script', src=src_node, tgt=tgt_nodes)
	tsk.env.DOFILETRUNK = os.path.splitext(src_node.name)[0]
	tsk.env.LOGFILEPATH = os.path.join(tg.bld.bldnode.abspath(), '%s.log' % (tsk.env.DOFILETRUNK))

	# dependencies (if the attribute 'deps' changes, trigger a recompilation)
	for x in tg.to_list(getattr(tg, 'deps', [])):
		node = tg.path.find_resource(x)
		if not node:
			tg.bld.fatal('Could not find dependency %r for running %r' % (x, src_node.abspath()))
		tsk.dep_nodes.append(node)
	Logs.debug('deps: found dependencies %r for running %r', tsk.dep_nodes, src_node.abspath())

	# Bypass the execution of process_source by setting the source to an empty list
	tg.source = []