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
|
import difflib
import re
class Differ:
@classmethod
def diff(cls, left, right):
left = cls.__remove_pg_errors(left).splitlines(keepends=True)
right = cls.__remove_ydb_errors(right).splitlines(keepends=True)
left = list(filter(lambda str: str != b'\n', left))
right = list(filter(lambda str: str != b'\n', right))
cls.__unify_tables(left, right)
return list(difflib.diff_bytes(difflib.unified_diff, left, right, n=0, fromfile=b'eth', tofile=b'out'))
__reErr = re.compile(b'(^ERROR: [^\n]+)(?:\nLINE \\d+: [^\n]+(?:\n\\s*\\^\\s*)?)?(?:\n(?:HINT|DETAIL|CONTEXT): [^\n]+)*(?:\n|$)',
re.MULTILINE)
__reYdbErr = re.compile(b'(^psql:[^\n]+\nIssues: \n)(?: *<main>:[^\n]+\n)*( *<main>:(\\d+:\\d+:)? Error: ([^\n]+)\n)(?:(?:HINT|DETAIL|CONTEXT): [^\n]+\n?)*\n?(?:\n|$)', re.MULTILINE)
@classmethod
def __remove_pg_error_msgs(cls, s):
return cls.__reErr.sub(rb"\1", s)
@classmethod
def __remove_ydb_error_msgs(cls, s):
return cls.__reYdbErr.sub(rb"ERROR: \4", s)
@classmethod
def __remove_pg_errors(cls, s):
return cls.__reErr.sub(rb"QUERY ERROR\n", s)
@classmethod
def __remove_ydb_errors(cls, s):
return cls.__reYdbErr.sub(rb"QUERY ERROR\n", s)
__reUniversalTableMarker = re.compile(rb'^-{3,100}(?:\+-{3,100})*$')
__reTableEndMarker = re.compile(rb'^\(\d+ rows?\)$')
@classmethod
def __reformat_table_row(cls, row, col_widths):
cells = [c.strip() for c in row[:-1].split(b'|')]
return b'|'.join(c.ljust(w) for (c, w) in zip(cells, col_widths))
@classmethod
def __remove_table_headers(cls, lines, header_line_numbers):
for i in reversed(header_line_numbers):
del lines[i]
del lines[i-1]
@classmethod
def __unify_tables(cls, left, right):
left_headers = []
right_headers = []
ucols = []
in_table = False
R = enumerate(right)
for (i, l) in enumerate(left):
if in_table:
if cls.__reTableEndMarker.match(l):
in_table = False
continue
j, r = next(R)
left[i] = cls.__reformat_table_row(l, ucols)
right[j] = cls.__reformat_table_row(r, ucols)
continue
if cls.__reUniversalTableMarker.match(l):
for (j, r) in R:
if cls.__reUniversalTableMarker.match(r):
break
else:
continue
lcols = [len(c) for c in l[:-1].split(b'+')]
rcols = [len(c) for c in r[:-1].split(b'+')]
if len(lcols) != len(rcols):
continue
ucols = [max(lw, rw) for lw, rw in zip(lcols, rcols)]
left_headers.append(i)
right_headers.append(j)
in_table = True
cls.__remove_table_headers(left, left_headers)
cls.__remove_table_headers(right, right_headers)
|
