Project

General

Profile

1
#!/usr/bin/env python
2
# Scrubs the taxonpaths in VegBIEN using TNRS.
3
# Runs continuously until no new rows are added after max_pause.
4

    
5
import os.path
6
import StringIO
7
import sys
8
import time
9

    
10
sys.path.append(os.path.dirname(__file__)+"/../lib")
11

    
12
import csvs
13
import opts
14
import profiling
15
import sql
16
import sql_gen
17
import sql_io
18
import streams
19
import strings
20
import tnrs
21

    
22
# Config
23
pause = 60 # sec
24
max_pause = 2*60*60 # sec; = 2 hr; must be >= max import time of one partition
25
assert pause <= max_pause
26

    
27
tnrs_data = sql_gen.Table('tnrs')
28

    
29
def main():
30
    # Input
31
    env_names = []
32
    db_config = opts.get_env_vars(sql.db_config_names, None, env_names)
33
    verbosity = float(opts.get_env_var('verbosity', 3, env_names))
34
    if not 'engine' in db_config: raise SystemExit('Usage: '
35
        +opts.env_usage(env_names)+' '+sys.argv[0]+' 2>>log')
36
    
37
    def log(msg, level=1):
38
        '''Higher level -> more verbose'''
39
        if level <= verbosity:
40
            sys.stderr.write(strings.to_raw_str(msg.rstrip('\n')+'\n'))
41
    
42
    # Connect to DB
43
    db = sql.connect(db_config, log_debug=log)
44
    
45
    tnrs_profiler = profiling.ItersProfiler(iter_text='row')
46
    try:
47
        # Iterate over unscrubbed verbatim taxonpaths
48
        start = 0
49
        total_pause = 0
50
        while True:
51
            # Fetch next set
52
            cur = sql.select(db, 'taxonpath', ['taxonomicnamewithauthor'],
53
                [('canon_taxonpath_id', None)], limit=tnrs.max_names,
54
                start=start, cacheable=False)
55
            this_ct = cur.rowcount
56
            start += this_ct # advance start to fetch next set
57
            if this_ct == 0:
58
                total_pause += pause
59
                if total_pause > max_pause: break
60
                log('Waited '+str(total_pause)+' sec. Waiting...')
61
                time.sleep(pause) # wait for more rows
62
                continue # try again
63
            # otherwise, rows found
64
            total_pause = 0
65
            names = list(sql.values(cur))
66
            
67
            # Run TNRS
68
            log('Processing '+str(this_ct)+' taxonpaths')
69
            log('Making TNRS request')
70
            tnrs_profiler.start()
71
            try:
72
                try: stream = tnrs.repeated_tnrs_request(names)
73
                finally: tnrs_profiler.stop(iter_ct=this_ct)
74
            except tnrs.InvalidResponse: pass # skip set in case it caused error
75
            else:
76
                log('Storing TNRS response data')
77
                stream_info = csvs.stream_info(stream, parse_header=True)
78
                stream = streams.ProgressInputStream(stream, sys.stderr, n=1000)
79
                sql_io.append_csv(db, tnrs_data, stream_info, stream)
80
    finally:
81
        log(tnrs_profiler.msg())
82

    
83
main()
(56-56/61)