Mercurial > repos > jjohnson > query_tabular

comparison filters.py @ 20:ab27c4bd14b9 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author jjohnson
date Fri, 14 Jul 2017 11:39:27 -0400
parents
children bed5018e7ae3
comparison
equal deleted inserted replaced
19:9d9ab2c69014 20:ab27c4bd14b9
1 #!/usr/binsenv python
2
3 from __future__ import print_function
4
5 import re
6 import sys
7
8
9 class LineFilter(object):
10 def __init__(self, source, filter_dict):
11 self.source = source
12 self.filter_dict = filter_dict
13 self.func = lambda i, l: l.rstrip('\r\n') if l else None
14 self.src_lines = []
15 self.src_line_cnt = 0
16 if not filter_dict:
17 return
18 if filter_dict['filter'] == 'regex':
19 rgx = re.compile(filter_dict['pattern'])
20 if filter_dict['action'] == 'exclude_match':
21 self.func = lambda i, l: l if not rgx.match(l) else None
22 elif filter_dict['action'] == 'include_match':
23 self.func = lambda i, l: l if rgx.match(l) else None
24 elif filter_dict['action'] == 'exclude_find':
25 self.func = lambda i, l: l if not rgx.search(l) else None
26 elif filter_dict['action'] == 'include_find':
27 self.func = lambda i, l: l if rgx.search(l) else None
28 elif filter_dict['filter'] == 'select_columns':
29 cols = [int(c) - 1 for c in filter_dict['columns']]
30 self.func = lambda i, l: self.select_columns(l, cols)
31 elif filter_dict['filter'] == 'replace':
32 p = filter_dict['pattern']
33 r = filter_dict['replace']
34 c = int(filter_dict['column']) - 1
35 self.func = lambda i, l: '\t'.join(
36 [x if j != c else re.sub(p, r, x) for j, x in enumerate(l.split('\t'))])
37 elif filter_dict['filter'] == 'prepend_line_num':
38 self.func = lambda i, l: '%d\t%s' % (i, l)
39 elif filter_dict['filter'] == 'append_line_num':
40 self.func = lambda i, l: '%s\t%d' % (l.rstrip('\r\n'), i)
41 elif filter_dict['filter'] == 'prepend_text':
42 s = filter_dict['column_text']
43 self.func = lambda i, l: '%s\t%s' % (s, l)
44 elif filter_dict['filter'] == 'append_text':
45 s = filter_dict['column_text']
46 self.func = lambda i, l: '%s\t%s' % (l.rstrip('\r\n'), s)
47 elif filter_dict['filter'] == 'skip':
48 cnt = filter_dict['count']
49 self.func = lambda i, l: l if i > cnt else None
50 elif filter_dict['filter'] == 'normalize':
51 cols = [int(c) - 1 for c in filter_dict['columns']]
52 sep = filter_dict['separator']
53 self.func = lambda i, l: self.normalize(l, cols, sep)
54
55 def __iter__(self):
56 return self
57
58 def __next__(self):
59 return next(self)
60
61 def next(self):
62 if not self.src_lines:
63 self.get_lines()
64 if self.src_lines:
65 return self.src_lines.pop(0)
66 raise StopIteration
67
68 def select_columns(self, line, cols):
69 fields = line.split('\t')
70 return '\t'.join([fields[x] for x in cols])
71
72 def normalize(self, line, split_cols, sep):
73 lines = []
74 fields = line.rstrip('\r\n').split('\t')
75 split_fields = dict()
76 cnt = 0
77 for c in split_cols:
78 if c < len(fields):
79 split_fields[c] = fields[c].split(sep)
80 cnt = max(cnt, len(split_fields[c]))
81 if cnt == 0:
82 lines.append('\t'.join(fields))
83 else:
84 for n in range(0, cnt):
85 flds = [x if c not in split_cols else split_fields[c][n]
86 if n < len(split_fields[c])
87 else '' for (c, x) in enumerate(fields)]
88 lines.append('\t'.join(flds))
89 return lines
90
91 def get_lines(self):
92 for i, next_line in enumerate(self.source):
93 self.src_line_cnt += 1
94 line = self.func(self.src_line_cnt, next_line)
95 if line:
96 if isinstance(line, list):
97 self.src_lines.extend(line)
98 else:
99 self.src_lines.append(line)
100 return
101
102
103 class TabularReader:
104 """
105 Tabular file iterator. Returns a list
106 """
107 def __init__(self, input_file, skip=0, comment_char=None, col_idx=None,
108 filters=None):
109 self.skip = skip
110 self.comment_char = comment_char
111 self.col_idx = col_idx
112 self.filters = filters
113 self.tsv_file = \
114 input_file if isinstance(input_file, file) else open(input_file)
115 if skip and skip > 0:
116 for i in range(skip):
117 if not self.tsv_file.readline():
118 break
119 source = LineFilter(self.tsv_file, None)
120 if comment_char:
121 source = LineFilter(source,
122 {"filter": "regex", "pattern": comment_char,
123 "action": "exclude_match"})
124 if filters:
125 for f in filters:
126 source = LineFilter(source, f)
127 self.source = source
128
129 def __iter__(self):
130 return self
131
132 def __next__(self):
133 return next(self)
134
135 def next(self):
136 ''' Iteration '''
137 for i, line in enumerate(self.source):
138 fields = line.rstrip('\r\n').split('\t')
139 if self.col_idx:
140 fields = [fields[i] for i in self.col_idx]
141 return fields
142 raise StopIteration
143
144
145 def filter_file(input_file, output, skip=0, comment_char='#', filters=None):
146 data_lines = 0
147 try:
148 tr = TabularReader(input_file, skip=skip, comment_char=comment_char,
149 filters=filters)
150 for linenum, fields in enumerate(tr):
151 data_lines += 1
152 try:
153 output.write('%s\n' % '\t'.join(fields))
154 except Exception as e:
155 print('Failed at line: %d err: %s' % (linenum, e),
156 file=sys.stderr)
157 except Exception as e:
158 print('Failed: %s' % (e), file=sys.stderr)
159 exit(1)