|
1
|
|
|
""" |
|
2
|
|
|
PyStratum |
|
3
|
|
|
""" |
|
4
|
|
|
import os |
|
5
|
|
|
import re |
|
6
|
|
|
from configparser import ConfigParser |
|
7
|
|
|
from typing import Any, Dict |
|
8
|
|
|
|
|
9
|
|
|
from pystratum_backend.StratumStyle import StratumStyle |
|
10
|
|
|
from pystratum_common.backend.CommonConstantWorker import CommonConstantWorker |
|
11
|
|
|
from pystratum_common.Util import Util |
|
12
|
|
|
|
|
13
|
|
|
from pystratum_mssql.backend.MsSqlWorker import MsSqlWorker |
|
14
|
|
|
|
|
15
|
|
|
|
|
16
|
|
|
class MsSqlConstantWorker(MsSqlWorker, CommonConstantWorker): |
|
17
|
|
|
""" |
|
18
|
|
|
Class for creating constants based on column widths, and auto increment columns and labels for SQL Server |
|
19
|
|
|
databases. |
|
20
|
|
|
""" |
|
21
|
|
|
|
|
22
|
|
|
# ------------------------------------------------------------------------------------------------------------------ |
|
23
|
|
|
def __init__(self, io: StratumStyle, config: ConfigParser): |
|
24
|
|
|
""" |
|
25
|
|
|
Object constructor. |
|
26
|
|
|
|
|
27
|
|
|
:param io: The output decorator. |
|
28
|
|
|
""" |
|
29
|
|
|
MsSqlWorker.__init__(self, io, config) |
|
30
|
|
|
CommonConstantWorker.__init__(self, io, config) |
|
31
|
|
|
|
|
32
|
|
|
self._columns: Dict[str, Any] = {} |
|
33
|
|
|
""" |
|
34
|
|
|
All columns in the database. |
|
35
|
|
|
""" |
|
36
|
|
|
|
|
37
|
|
|
# ------------------------------------------------------------------------------------------------------------------ |
|
38
|
|
|
def _get_old_columns(self) -> None: |
|
39
|
|
|
""" |
|
40
|
|
|
Reads from file constants_filename the previous table and column names, the width of the column, |
|
41
|
|
|
and the constant name (if assigned) and stores this data in old_columns. |
|
42
|
|
|
""" |
|
43
|
|
|
if os.path.exists(self._constants_filename): |
|
44
|
|
|
with open(self._constants_filename, 'r') as f: |
|
45
|
|
|
line_number = 0 |
|
46
|
|
|
for line in f: |
|
47
|
|
|
line_number += 1 |
|
48
|
|
|
if line != "\n": |
|
49
|
|
|
p = re.compile(r'\s*(?:([a-zA-Z0-9_]+)\.)?([a-zA-Z0-9_]+)\.' |
|
50
|
|
|
r'([a-zA-Z0-9_]+)\s+(\d+)\s*(\*|[a-zA-Z0-9_]+)?\s*') |
|
51
|
|
|
matches = p.findall(line) |
|
52
|
|
|
|
|
53
|
|
|
if matches: |
|
54
|
|
|
matches = matches[0] |
|
55
|
|
|
schema_name = str(matches[0]) |
|
56
|
|
|
table_name = str(matches[1]) |
|
57
|
|
|
column_name = str(matches[2]) |
|
58
|
|
|
length = str(matches[3]) |
|
59
|
|
|
constant_name = str(matches[4]) |
|
60
|
|
|
|
|
61
|
|
|
if constant_name: |
|
62
|
|
|
column_info = {'schema_name': schema_name, |
|
63
|
|
|
'table_name': table_name, |
|
64
|
|
|
'column_name': column_name, |
|
65
|
|
|
'length': length, |
|
66
|
|
|
'constant_name': constant_name} |
|
67
|
|
|
else: |
|
68
|
|
|
column_info = {'schema_name': schema_name, |
|
69
|
|
|
'table_name': table_name, |
|
70
|
|
|
'column_name': column_name, |
|
71
|
|
|
'length': length} |
|
72
|
|
|
|
|
73
|
|
|
if schema_name in self._old_columns: |
|
74
|
|
|
if table_name in self._old_columns[schema_name]: |
|
75
|
|
|
if column_name in self._old_columns[schema_name][table_name]: |
|
76
|
|
|
pass |
|
77
|
|
|
else: |
|
78
|
|
|
self._old_columns[schema_name][table_name][column_name] = column_info |
|
79
|
|
|
else: |
|
80
|
|
|
self._old_columns[schema_name][table_name] = {column_name: column_info} |
|
81
|
|
|
else: |
|
82
|
|
|
self._old_columns[schema_name] = {table_name: {column_name: column_info}} |
|
83
|
|
|
|
|
84
|
|
|
# ------------------------------------------------------------------------------------------------------------------ |
|
85
|
|
|
def _get_columns(self) -> None: |
|
86
|
|
|
""" |
|
87
|
|
|
Retrieves metadata all columns in the database. |
|
88
|
|
|
""" |
|
89
|
|
|
rows = self._dl.get_all_table_columns() |
|
90
|
|
|
for row in rows: |
|
91
|
|
|
row['length'] = MsSqlConstantWorker.derive_field_length(row) |
|
92
|
|
|
|
|
93
|
|
|
if row['schema_name'] in self._columns: |
|
94
|
|
|
if row['table_name'] in self._columns[row['schema_name']]: |
|
95
|
|
|
if row['column_name'] in self._columns[row['schema_name']][row['table_name']]: |
|
96
|
|
|
pass |
|
97
|
|
|
else: |
|
98
|
|
|
self._columns[row['schema_name']][row['table_name']][row['column_name']] = row |
|
99
|
|
|
else: |
|
100
|
|
|
self._columns[row['schema_name']][row['table_name']] = {row['column_name']: row} |
|
101
|
|
|
else: |
|
102
|
|
|
self._columns[row['schema_name']] = {row['table_name']: {row['column_name']: row}} |
|
103
|
|
|
|
|
104
|
|
|
# ------------------------------------------------------------------------------------------------------------------ |
|
105
|
|
|
def _enhance_columns(self) -> None: |
|
106
|
|
|
""" |
|
107
|
|
|
Enhances old_columns as follows: |
|
108
|
|
|
If the constant name is *, is is replaced with the column name prefixed by prefix in uppercase. |
|
109
|
|
|
Otherwise the constant name is set to uppercase. |
|
110
|
|
|
""" |
|
111
|
|
|
if self._old_columns: |
|
112
|
|
|
for schema_name, schema in sorted(self._old_columns.items()): |
|
113
|
|
|
for table_name, table in sorted(schema.items()): |
|
114
|
|
|
for column_name, column in sorted(table.items()): |
|
115
|
|
|
if 'constant_name' in column: |
|
116
|
|
|
if column['constant_name'].strip() == '*': |
|
117
|
|
|
constant_name = str(self._prefix + column['column_name']).upper() |
|
118
|
|
|
self._old_columns[schema_name][table_name][column_name]['constant_name'] = constant_name |
|
119
|
|
|
else: |
|
120
|
|
|
constant_name = str( |
|
121
|
|
|
self._old_columns[schema_name][table_name][column_name][ |
|
122
|
|
|
'constant_name']).upper() |
|
123
|
|
|
self._old_columns[schema_name][table_name][column_name]['constant_name'] = constant_name |
|
124
|
|
|
|
|
125
|
|
|
# ------------------------------------------------------------------------------------------------------------------ |
|
126
|
|
|
def _merge_columns(self) -> None: |
|
127
|
|
|
""" |
|
128
|
|
|
Preserves relevant data in old_columns into columns. |
|
129
|
|
|
""" |
|
130
|
|
|
if self._old_columns: |
|
131
|
|
|
for schema_name, schema in sorted(self._old_columns.items()): |
|
132
|
|
|
for table_name, table in sorted(schema.items()): |
|
133
|
|
|
for column_name, column in sorted(table.items()): |
|
134
|
|
|
if 'constant_name' in column: |
|
135
|
|
|
try: |
|
136
|
|
|
self._columns[schema_name][table_name][column_name]['constant_name'] = \ |
|
137
|
|
|
column['constant_name'] |
|
138
|
|
|
except KeyError: |
|
139
|
|
|
# Either the column or table is not present anymore. |
|
140
|
|
|
self._io.warning('Dropping constant {0} because column is not present anymore'. |
|
141
|
|
|
format(column['constant_name'])) |
|
142
|
|
|
|
|
143
|
|
|
# ------------------------------------------------------------------------------------------------------------------ |
|
144
|
|
|
def _write_columns(self) -> None: |
|
145
|
|
|
""" |
|
146
|
|
|
Writes table and column names, the width of the column, and the constant name (if assigned) to |
|
147
|
|
|
constants_filename. |
|
148
|
|
|
""" |
|
149
|
|
|
content = '' |
|
150
|
|
|
|
|
151
|
|
|
for schema_name, schema in sorted(self._columns.items()): |
|
152
|
|
|
for table_name, table in sorted(schema.items()): |
|
153
|
|
|
width1 = 0 |
|
154
|
|
|
width2 = 0 |
|
155
|
|
|
|
|
156
|
|
|
key_map = {} |
|
157
|
|
|
for column_name, column in table.items(): |
|
158
|
|
|
key_map[column['column_id']] = column_name |
|
159
|
|
|
width1 = max(len(str(column['column_name'])), width1) |
|
160
|
|
|
width2 = max(len(str(column['length'])), width2) |
|
161
|
|
|
|
|
162
|
|
|
for col_id, column_name in sorted(key_map.items()): |
|
163
|
|
|
if table[column_name]['length'] is not None: |
|
164
|
|
|
if 'constant_name' in table[column_name]: |
|
165
|
|
|
line_format = "%s.%s.%-{0:d}s %{1:d}d %s\n".format(int(width1), int(width2)) |
|
166
|
|
|
content += line_format % (schema_name, |
|
167
|
|
|
table[column_name]['table_name'], |
|
168
|
|
|
table[column_name]['column_name'], |
|
169
|
|
|
table[column_name]['length'], |
|
170
|
|
|
table[column_name]['constant_name']) |
|
171
|
|
|
else: |
|
172
|
|
|
line_format = "%s.%s.%-{0:d}s %{1:d}d\n".format(int(width1), int(width2)) |
|
173
|
|
|
content += line_format % (schema_name, |
|
174
|
|
|
table[column_name]['table_name'], |
|
175
|
|
|
table[column_name]['column_name'], |
|
176
|
|
|
table[column_name]['length']) |
|
177
|
|
|
|
|
178
|
|
|
content += "\n""" |
|
179
|
|
|
|
|
180
|
|
|
# Save the columns, width, and constants to the filesystem. |
|
181
|
|
|
Util.write_two_phases(self._constants_filename, content, self._io) |
|
182
|
|
|
|
|
183
|
|
|
# ------------------------------------------------------------------------------------------------------------------ |
|
184
|
|
|
def _get_labels(self) -> None: |
|
185
|
|
|
""" |
|
186
|
|
|
Gets all primary key labels from the database. |
|
187
|
|
|
""" |
|
188
|
|
|
tables = self._dl.get_label_tables(self._label_regex) |
|
189
|
|
|
|
|
190
|
|
|
for table in tables: |
|
191
|
|
|
rows = self._dl.get_labels_from_table(table['database'], |
|
192
|
|
|
table['schema_name'], |
|
193
|
|
|
table['table_name'], |
|
194
|
|
|
table['id'], |
|
195
|
|
|
table['label']) |
|
196
|
|
|
for row in rows: |
|
197
|
|
|
if row['label'] not in self._labels: |
|
198
|
|
|
self._labels[row['label']] = row['id'] |
|
199
|
|
|
else: |
|
200
|
|
|
# todo improve exception. |
|
201
|
|
|
Exception("Duplicate label '%s'") |
|
202
|
|
|
|
|
203
|
|
|
# ------------------------------------------------------------------------------------------------------------------ |
|
204
|
|
|
def _fill_constants(self) -> None: |
|
205
|
|
|
""" |
|
206
|
|
|
Merges columns and labels (i.e. all known constants) into constants. |
|
207
|
|
|
""" |
|
208
|
|
|
for schema_name, schema in sorted(self._columns.items()): |
|
209
|
|
|
for table_name, table in sorted(schema.items()): |
|
210
|
|
|
for column_name, column in sorted(table.items()): |
|
211
|
|
|
if 'constant_name' in column: |
|
212
|
|
|
self._constants[column['constant_name']] = column['length'] |
|
213
|
|
|
|
|
214
|
|
|
for label, label_id in sorted(self._labels.items()): |
|
215
|
|
|
self._constants[label] = label_id |
|
216
|
|
|
|
|
217
|
|
|
# ------------------------------------------------------------------------------------------------------------------ |
|
218
|
|
|
@staticmethod |
|
219
|
|
|
def derive_field_length(column: Dict[str, Any]) -> int: |
|
220
|
|
|
""" |
|
221
|
|
|
Returns the width of a field based based on the data type of column. |
|
222
|
|
|
|
|
223
|
|
|
:param dict column: Info about the column. |
|
224
|
|
|
|
|
225
|
|
|
:rtype: int |
|
226
|
|
|
""" |
|
227
|
|
|
data_type = column['data_type'] |
|
228
|
|
|
|
|
229
|
|
|
if data_type == 'bigint': |
|
230
|
|
|
return column['precision'] |
|
231
|
|
|
|
|
232
|
|
|
if data_type == 'int': |
|
233
|
|
|
return column['precision'] |
|
234
|
|
|
|
|
235
|
|
|
if data_type == 'smallint': |
|
236
|
|
|
return column['precision'] |
|
237
|
|
|
|
|
238
|
|
|
if data_type == 'tinyint': |
|
239
|
|
|
return column['precision'] |
|
240
|
|
|
|
|
241
|
|
|
if data_type == 'bit': |
|
242
|
|
|
return column['max_length'] |
|
243
|
|
|
|
|
244
|
|
|
if data_type == 'money': |
|
245
|
|
|
return column['precision'] |
|
246
|
|
|
|
|
247
|
|
|
if data_type == 'smallmoney': |
|
248
|
|
|
return column['precision'] |
|
249
|
|
|
|
|
250
|
|
|
if data_type == 'decimal': |
|
251
|
|
|
return column['precision'] |
|
252
|
|
|
|
|
253
|
|
|
if data_type == 'numeric': |
|
254
|
|
|
return column['precision'] |
|
255
|
|
|
|
|
256
|
|
|
if data_type == 'float': |
|
257
|
|
|
return column['precision'] |
|
258
|
|
|
|
|
259
|
|
|
if data_type == 'real': |
|
260
|
|
|
return column['precision'] |
|
261
|
|
|
|
|
262
|
|
|
if data_type == 'date': |
|
263
|
|
|
return column['precision'] |
|
264
|
|
|
|
|
265
|
|
|
if data_type == 'datetime': |
|
266
|
|
|
return column['precision'] |
|
267
|
|
|
|
|
268
|
|
|
if data_type == 'datetime2': |
|
269
|
|
|
return column['precision'] |
|
270
|
|
|
|
|
271
|
|
|
if data_type == 'datetimeoffset': |
|
272
|
|
|
return column['precision'] |
|
273
|
|
|
|
|
274
|
|
|
if data_type == 'smalldatetime': |
|
275
|
|
|
return column['precision'] |
|
276
|
|
|
|
|
277
|
|
|
if data_type == 'time': |
|
278
|
|
|
return column['precision'] |
|
279
|
|
|
|
|
280
|
|
|
if data_type == 'char': |
|
281
|
|
|
return column['max_length'] |
|
282
|
|
|
|
|
283
|
|
|
if data_type == 'varchar': |
|
284
|
|
|
if column['max_length'] == -1: |
|
285
|
|
|
# This is a varchar(max) data type. |
|
286
|
|
|
return 2147483647 |
|
287
|
|
|
|
|
288
|
|
|
return column['max_length'] |
|
289
|
|
|
|
|
290
|
|
|
if data_type == 'text': |
|
291
|
|
|
return 2147483647 |
|
292
|
|
|
|
|
293
|
|
|
if data_type == 'nchar': |
|
294
|
|
|
return column['max_length'] / 2 |
|
295
|
|
|
|
|
296
|
|
|
if data_type == 'nvarchar': |
|
297
|
|
|
if column['max_length'] == -1: |
|
298
|
|
|
# This is a nvarchar(max) data type. |
|
299
|
|
|
return 1073741823 |
|
300
|
|
|
|
|
301
|
|
|
return column['max_length'] / 2 |
|
302
|
|
|
|
|
303
|
|
|
if data_type == 'ntext': |
|
304
|
|
|
return 1073741823 |
|
305
|
|
|
|
|
306
|
|
|
if data_type == 'binary': |
|
307
|
|
|
return column['max_length'] |
|
308
|
|
|
|
|
309
|
|
|
if data_type == 'varbinary': |
|
310
|
|
|
return column['max_length'] |
|
311
|
|
|
|
|
312
|
|
|
if data_type == 'image': |
|
313
|
|
|
return 2147483647 |
|
314
|
|
|
|
|
315
|
|
|
if data_type == 'xml': |
|
316
|
|
|
return 2147483647 |
|
317
|
|
|
|
|
318
|
|
|
if data_type == 'geography': |
|
319
|
|
|
if column['max_length'] == -1: |
|
320
|
|
|
# This is a varchar(max) data type. |
|
321
|
|
|
return 2147483647 |
|
322
|
|
|
|
|
323
|
|
|
if data_type == 'geometry': |
|
324
|
|
|
if column['max_length'] == -1: |
|
325
|
|
|
# This is a varchar(max) data type. |
|
326
|
|
|
return 2147483647 |
|
327
|
|
|
|
|
328
|
|
|
if data_type == 'sysname': |
|
329
|
|
|
return column['max_length'] |
|
330
|
|
|
|
|
331
|
|
|
print(column) |
|
332
|
|
|
raise Exception("Unexpected data type '{0}'".format(data_type)) |
|
333
|
|
|
|
|
334
|
|
|
# ---------------------------------------------------------------------------------------------------------------------- |
|
335
|
|
|
|