| 1 |
205
|
aaronmk
|
#!/usr/bin/env python
|
| 2 |
|
|
# Modifies a map spreadsheet A->B or any file using a replacements spreadsheet
|
| 3 |
|
|
# A->C or B->C
|
| 4 |
|
|
|
| 5 |
|
|
import csv
|
| 6 |
9785
|
aaronmk
|
import HTMLParser
|
| 7 |
1705
|
aaronmk
|
import os.path
|
| 8 |
205
|
aaronmk
|
import re
|
| 9 |
|
|
import sys
|
| 10 |
|
|
|
| 11 |
1705
|
aaronmk
|
sys.path.append(os.path.dirname(__file__)+"/../lib")
|
| 12 |
|
|
|
| 13 |
|
|
import maps
|
| 14 |
4357
|
aaronmk
|
import opts
|
| 15 |
9784
|
aaronmk
|
import strings
|
| 16 |
1705
|
aaronmk
|
|
| 17 |
10348
|
aaronmk
|
quote_re = '[\'"`]'
|
| 18 |
12746
|
aaronmk
|
excluded_prefix_re = (
|
| 19 |
12897
|
aaronmk
|
'(?<![a-z] )' #not if it's a word in a sentence (uppercase SQL keywords OK)
|
| 20 |
12750
|
aaronmk
|
+'(?<!-)' # not if it's part of a '-'-separated identifier
|
| 21 |
12749
|
aaronmk
|
+'(?<!\*)' # don't double leading *
|
| 22 |
|
|
)
|
| 23 |
|
|
excluded_suffix_re = (
|
| 24 |
12897
|
aaronmk
|
'(?! [a-z])' # not if it's a word in a sentence (uppercase SQL keywords OK)
|
| 25 |
12750
|
aaronmk
|
+'(?!-)' # not if it's part of a '-'-separated identifier
|
| 26 |
12749
|
aaronmk
|
)
|
| 27 |
10348
|
aaronmk
|
|
| 28 |
9785
|
aaronmk
|
def unescape_html(str_): return HTMLParser.HTMLParser().unescape(str_)
|
| 29 |
|
|
|
| 30 |
|
|
def repl_unescape_html(match): return unescape_html(match.group(0))
|
| 31 |
|
|
|
| 32 |
205
|
aaronmk
|
def main():
|
| 33 |
4357
|
aaronmk
|
env_names = []
|
| 34 |
|
|
def usage_err():
|
| 35 |
|
|
raise SystemExit('Usage: '+opts.env_usage(env_names, True)+' '
|
| 36 |
|
|
+sys.argv[0]+' <map repl [col_num] [| '+sys.argv[0]
|
| 37 |
|
|
+' repl_1 [col_num_1]]... >new_map')
|
| 38 |
|
|
|
| 39 |
205
|
aaronmk
|
col_num = None
|
| 40 |
|
|
try: col_num = sys.argv[2]
|
| 41 |
211
|
aaronmk
|
except IndexError: pass
|
| 42 |
205
|
aaronmk
|
if col_num != None: col_num = int(col_num) # 0-based
|
| 43 |
12914
|
aaronmk
|
# whether all patterns are plain text
|
| 44 |
|
|
# defaults to on if matching entire cells in a spreadsheet (w/ col_num)
|
| 45 |
|
|
text = opts.env_flag('text', col_num != None, env_names)
|
| 46 |
205
|
aaronmk
|
|
| 47 |
12914
|
aaronmk
|
try: _prog_name, repl_path = sys.argv[:2]
|
| 48 |
|
|
except ValueError: usage_err()
|
| 49 |
|
|
|
| 50 |
205
|
aaronmk
|
# Get replacements
|
| 51 |
|
|
repls = []
|
| 52 |
|
|
stream = open(repl_path, 'rb')
|
| 53 |
|
|
reader = csv.reader(stream)
|
| 54 |
8740
|
aaronmk
|
reader.next() # skip header
|
| 55 |
205
|
aaronmk
|
for row in reader:
|
| 56 |
210
|
aaronmk
|
in_, out = row[:2]
|
| 57 |
216
|
aaronmk
|
if in_ != '':
|
| 58 |
10348
|
aaronmk
|
is_word = re.match(r'^\w+$', in_)
|
| 59 |
|
|
if text or is_word: # match as whole-word text (like SQL identifier)
|
| 60 |
13076
|
aaronmk
|
# this can also be done in Postgres with expression substitution
|
| 61 |
|
|
# (wiki.vegpath.org/Postgres_queries#expression-substitution)
|
| 62 |
10348
|
aaronmk
|
in_str_re = re.escape(in_)
|
| 63 |
|
|
q = quote_re
|
| 64 |
|
|
in_ = '(?<='+q+')'+in_str_re+'(?='+q+')' # require quotes
|
| 65 |
12742
|
aaronmk
|
if is_word: # also match with quotes optional
|
| 66 |
12748
|
aaronmk
|
# don't try to match word w/ suffix, because there are cases
|
| 67 |
|
|
# where a mapping adds a suffix which would cause the same
|
| 68 |
|
|
# replacement to be performed repeatedly
|
| 69 |
12754
|
aaronmk
|
in_word_re = r'\b'+in_str_re+r'\b'
|
| 70 |
|
|
|
| 71 |
|
|
# only use excluded_prefix_re/excluded_suffix_re in text
|
| 72 |
|
|
# mode (used in renaming columns in SQL scripts), to prevent
|
| 73 |
|
|
# the special coding for column renames from also affecting
|
| 74 |
|
|
# regular regexp/word replacements
|
| 75 |
|
|
if text: in_word_re = (excluded_prefix_re+in_word_re
|
| 76 |
12749
|
aaronmk
|
+excluded_suffix_re)
|
| 77 |
12754
|
aaronmk
|
|
| 78 |
10348
|
aaronmk
|
in_ = '(?:'+in_+'|'+in_word_re+')'
|
| 79 |
6713
|
aaronmk
|
repls.append((r'(?m)'+in_, out))
|
| 80 |
205
|
aaronmk
|
stream.close()
|
| 81 |
|
|
def repl_all(str_):
|
| 82 |
9784
|
aaronmk
|
str_ = strings.ustr(str_)
|
| 83 |
9785
|
aaronmk
|
for repl, with_ in repls:
|
| 84 |
|
|
if with_ == 'unescape_html()': with_ = repl_unescape_html
|
| 85 |
|
|
str_ = re.sub(repl, with_, str_)
|
| 86 |
205
|
aaronmk
|
return str_
|
| 87 |
|
|
|
| 88 |
|
|
# Modify map or file
|
| 89 |
|
|
if col_num != None:
|
| 90 |
|
|
reader = csv.reader(sys.stdin)
|
| 91 |
|
|
writer = csv.writer(sys.stdout)
|
| 92 |
|
|
cols = reader.next()
|
| 93 |
|
|
writer.writerow(cols)
|
| 94 |
|
|
for row in reader:
|
| 95 |
|
|
row[col_num] = repl_all(row[col_num])
|
| 96 |
|
|
writer.writerow(row)
|
| 97 |
9784
|
aaronmk
|
else: sys.stdout.write(strings.to_raw_str(repl_all(sys.stdin.read())))
|
| 98 |
205
|
aaronmk
|
|
| 99 |
|
|
main()
|
