Project

General

Profile

1
# Database access
2

    
3
import copy
4
import operator
5
import re
6
import warnings
7

    
8
import exc
9
import dicts
10
import iters
11
import lists
12
from Proxy import Proxy
13
import rand
14
import sql_gen
15
import strings
16
import util
17

    
18
##### Exceptions
19

    
20
def get_cur_query(cur, input_query=None):
21
    raw_query = None
22
    if hasattr(cur, 'query'): raw_query = cur.query
23
    elif hasattr(cur, '_last_executed'): raw_query = cur._last_executed
24
    
25
    if raw_query != None: return raw_query
26
    else: return '[input] '+strings.ustr(input_query)
27

    
28
def _add_cursor_info(e, *args, **kw_args):
29
    '''For params, see get_cur_query()'''
30
    exc.add_msg(e, 'query: '+strings.ustr(get_cur_query(*args, **kw_args)))
31

    
32
class DbException(exc.ExceptionWithCause):
33
    def __init__(self, msg, cause=None, cur=None):
34
        exc.ExceptionWithCause.__init__(self, msg, cause, cause_newline=True)
35
        if cur != None: _add_cursor_info(self, cur)
36

    
37
class ExceptionWithName(DbException):
38
    def __init__(self, name, cause=None):
39
        DbException.__init__(self, 'for name: '+strings.as_tt(str(name)), cause)
40
        self.name = name
41

    
42
class ExceptionWithNameValue(DbException):
43
    def __init__(self, name, value, cause=None):
44
        DbException.__init__(self, 'for name: '+strings.as_tt(str(name))
45
            +'; value: '+strings.as_tt(repr(value)), cause)
46
        self.name = name
47
        self.value = value
48

    
49
class ExceptionWithNameType(DbException):
50
    def __init__(self, type_, name, cause=None):
51
        DbException.__init__(self, 'for type: '+strings.as_tt(str(type_))
52
            +'; name: '+strings.as_tt(name), cause)
53
        self.type = type_
54
        self.name = name
55

    
56
class ConstraintException(DbException):
57
    def __init__(self, name, cols, cause=None):
58
        DbException.__init__(self, 'Violated '+strings.as_tt(name)
59
            +' constraint on columns: '+strings.as_tt(', '.join(cols)), cause)
60
        self.name = name
61
        self.cols = cols
62

    
63
class MissingCastException(DbException):
64
    def __init__(self, type_, col, cause=None):
65
        DbException.__init__(self, 'Missing cast to type '+strings.as_tt(type_)
66
            +' on column: '+strings.as_tt(col), cause)
67
        self.type = type_
68
        self.col = col
69

    
70
class NameException(DbException): pass
71

    
72
class DuplicateKeyException(ConstraintException): pass
73

    
74
class NullValueException(ConstraintException): pass
75

    
76
class FunctionValueException(ExceptionWithNameValue): pass
77

    
78
class DuplicateException(ExceptionWithNameType): pass
79

    
80
class EmptyRowException(DbException): pass
81

    
82
##### Warnings
83

    
84
class DbWarning(UserWarning): pass
85

    
86
##### Result retrieval
87

    
88
def col_names(cur): return (col[0] for col in cur.description)
89

    
90
def rows(cur): return iter(lambda: cur.fetchone(), None)
91

    
92
def consume_rows(cur):
93
    '''Used to fetch all rows so result will be cached'''
94
    iters.consume_iter(rows(cur))
95

    
96
def next_row(cur): return rows(cur).next()
97

    
98
def row(cur):
99
    row_ = next_row(cur)
100
    consume_rows(cur)
101
    return row_
102

    
103
def next_value(cur): return next_row(cur)[0]
104

    
105
def value(cur): return row(cur)[0]
106

    
107
def values(cur): return iters.func_iter(lambda: next_value(cur))
108

    
109
def value_or_none(cur):
110
    try: return value(cur)
111
    except StopIteration: return None
112

    
113
##### Escaping
114

    
115
def esc_name_by_module(module, name):
116
    if module == 'psycopg2' or module == None: quote = '"'
117
    elif module == 'MySQLdb': quote = '`'
118
    else: raise NotImplementedError("Can't escape name for "+module+' database')
119
    return sql_gen.esc_name(name, quote)
120

    
121
def esc_name_by_engine(engine, name, **kw_args):
122
    return esc_name_by_module(db_engines[engine][0], name, **kw_args)
123

    
124
def esc_name(db, name, **kw_args):
125
    return esc_name_by_module(util.root_module(db.db), name, **kw_args)
126

    
127
def qual_name(db, schema, table):
128
    def esc_name_(name): return esc_name(db, name)
129
    table = esc_name_(table)
130
    if schema != None: return esc_name_(schema)+'.'+table
131
    else: return table
132

    
133
##### Database connections
134

    
135
db_config_names = ['engine', 'host', 'user', 'password', 'database', 'schemas']
136

    
137
db_engines = {
138
    'MySQL': ('MySQLdb', {'password': 'passwd', 'database': 'db'}),
139
    'PostgreSQL': ('psycopg2', {}),
140
}
141

    
142
DatabaseErrors_set = set([DbException])
143
DatabaseErrors = tuple(DatabaseErrors_set)
144

    
145
def _add_module(module):
146
    DatabaseErrors_set.add(module.DatabaseError)
147
    global DatabaseErrors
148
    DatabaseErrors = tuple(DatabaseErrors_set)
149

    
150
def db_config_str(db_config):
151
    return db_config['engine']+' database '+db_config['database']
152

    
153
log_debug_none = lambda msg, level=2: None
154

    
155
class DbConn:
156
    def __init__(self, db_config, autocommit=True, caching=True,
157
        log_debug=log_debug_none, debug_temp=False):
158
        '''
159
        @param debug_temp Whether temporary objects should instead be permanent.
160
            This assists in debugging the internal objects used by the program.
161
        '''
162
        self.db_config = db_config
163
        self.autocommit = autocommit
164
        self.caching = caching
165
        self.log_debug = log_debug
166
        self.debug = log_debug != log_debug_none
167
        self.debug_temp = debug_temp
168
        
169
        self.__db = None
170
        self.query_results = {}
171
        self._savepoint = 0
172
        self._notices_seen = set()
173
    
174
    def __getattr__(self, name):
175
        if name == '__dict__': raise Exception('getting __dict__')
176
        if name == 'db': return self._db()
177
        else: raise AttributeError()
178
    
179
    def __getstate__(self):
180
        state = copy.copy(self.__dict__) # shallow copy
181
        state['log_debug'] = None # don't pickle the debug callback
182
        state['_DbConn__db'] = None # don't pickle the connection
183
        return state
184
    
185
    def connected(self): return self.__db != None
186
    
187
    def _db(self):
188
        if self.__db == None:
189
            # Process db_config
190
            db_config = self.db_config.copy() # don't modify input!
191
            schemas = db_config.pop('schemas', None)
192
            module_name, mappings = db_engines[db_config.pop('engine')]
193
            module = __import__(module_name)
194
            _add_module(module)
195
            for orig, new in mappings.iteritems():
196
                try: util.rename_key(db_config, orig, new)
197
                except KeyError: pass
198
            
199
            # Connect
200
            self.__db = module.connect(**db_config)
201
            
202
            # Configure connection
203
            if hasattr(self.db, 'set_isolation_level'):
204
                import psycopg2.extensions
205
                self.db.set_isolation_level(
206
                    psycopg2.extensions.ISOLATION_LEVEL_READ_COMMITTED)
207
            if schemas != None:
208
                search_path = [self.esc_name(s) for s in schemas.split(',')]
209
                search_path.append(value(run_query(self, 'SHOW search_path',
210
                    log_level=4)))
211
                run_query(self, 'SET search_path TO '+(','.join(search_path)),
212
                    log_level=3)
213
        
214
        return self.__db
215
    
216
    class DbCursor(Proxy):
217
        def __init__(self, outer):
218
            Proxy.__init__(self, outer.db.cursor())
219
            self.outer = outer
220
            self.query_results = outer.query_results
221
            self.query_lookup = None
222
            self.result = []
223
        
224
        def execute(self, query):
225
            self._is_insert = query.startswith('INSERT')
226
            self.query_lookup = query
227
            try:
228
                try:
229
                    cur = self.inner.execute(query)
230
                    self.outer.do_autocommit()
231
                finally: self.query = get_cur_query(self.inner, query)
232
            except Exception, e:
233
                _add_cursor_info(e, self, query)
234
                self.result = e # cache the exception as the result
235
                self._cache_result()
236
                raise
237
            
238
            # Always cache certain queries
239
            if query.startswith('CREATE') or query.startswith('ALTER'):
240
                # structural changes
241
                # Rest of query must be unique in the face of name collisions,
242
                # so don't cache ADD COLUMN unless it has distinguishing comment
243
                if query.find('ADD COLUMN') < 0 or query.endswith('*/'):
244
                    self._cache_result()
245
            elif self.rowcount == 0 and query.startswith('SELECT'): # empty
246
                consume_rows(self) # fetch all rows so result will be cached
247
            
248
            return cur
249
        
250
        def fetchone(self):
251
            row = self.inner.fetchone()
252
            if row != None: self.result.append(row)
253
            # otherwise, fetched all rows
254
            else: self._cache_result()
255
            return row
256
        
257
        def _cache_result(self):
258
            # For inserts that return a result set, don't cache result set since
259
            # inserts are not idempotent. Other non-SELECT queries don't have
260
            # their result set read, so only exceptions will be cached (an
261
            # invalid query will always be invalid).
262
            if self.query_results != None and (not self._is_insert
263
                or isinstance(self.result, Exception)):
264
                
265
                assert self.query_lookup != None
266
                self.query_results[self.query_lookup] = self.CacheCursor(
267
                    util.dict_subset(dicts.AttrsDictView(self),
268
                    ['query', 'result', 'rowcount', 'description']))
269
        
270
        class CacheCursor:
271
            def __init__(self, cached_result): self.__dict__ = cached_result
272
            
273
            def execute(self, *args, **kw_args):
274
                if isinstance(self.result, Exception): raise self.result
275
                # otherwise, result is a rows list
276
                self.iter = iter(self.result)
277
            
278
            def fetchone(self):
279
                try: return self.iter.next()
280
                except StopIteration: return None
281
    
282
    def esc_value(self, value):
283
        try: str_ = self.mogrify('%s', [value])
284
        except NotImplementedError, e:
285
            module = util.root_module(self.db)
286
            if module == 'MySQLdb':
287
                import _mysql
288
                str_ = _mysql.escape_string(value)
289
            else: raise e
290
        return strings.to_unicode(str_)
291
    
292
    def esc_name(self, name): return esc_name(self, name) # calls global func
293
    
294
    def std_code(self, str_):
295
        '''Standardizes SQL code.
296
        * Ensures that string literals are prefixed by `E`
297
        '''
298
        if str_.startswith("'"): str_ = 'E'+str_
299
        return str_
300
    
301
    def can_mogrify(self):
302
        module = util.root_module(self.db)
303
        return module == 'psycopg2'
304
    
305
    def mogrify(self, query, params=None):
306
        if self.can_mogrify(): return self.db.cursor().mogrify(query, params)
307
        else: raise NotImplementedError("Can't mogrify query")
308
    
309
    def print_notices(self):
310
        if hasattr(self.db, 'notices'):
311
            for msg in self.db.notices:
312
                if msg not in self._notices_seen:
313
                    self._notices_seen.add(msg)
314
                    self.log_debug(msg, level=2)
315
    
316
    def run_query(self, query, cacheable=False, log_level=2,
317
        debug_msg_ref=None):
318
        '''
319
        @param log_ignore_excs The log_level will be increased by 2 if the query
320
            throws one of these exceptions.
321
        @param debug_msg_ref If specified, the log message will be returned in
322
            this instead of being output. This allows you to filter log messages
323
            depending on the result of the query.
324
        '''
325
        assert query != None
326
        
327
        if not self.caching: cacheable = False
328
        used_cache = False
329
        
330
        def log_msg(query):
331
            if used_cache: cache_status = 'cache hit'
332
            elif cacheable: cache_status = 'cache miss'
333
            else: cache_status = 'non-cacheable'
334
            return 'DB query: '+cache_status+':\n'+strings.as_code(query, 'SQL')
335
        
336
        try:
337
            # Get cursor
338
            if cacheable:
339
                try:
340
                    cur = self.query_results[query]
341
                    used_cache = True
342
                except KeyError: cur = self.DbCursor(self)
343
            else: cur = self.db.cursor()
344
            
345
            # Log query
346
            if self.debug and debug_msg_ref == None: # log before running
347
                self.log_debug(log_msg(query), log_level)
348
            
349
            # Run query
350
            cur.execute(query)
351
        finally:
352
            self.print_notices()
353
            if self.debug and debug_msg_ref != None: # return after running
354
                debug_msg_ref[0] = log_msg(str(get_cur_query(cur, query)))
355
        
356
        return cur
357
    
358
    def is_cached(self, query): return query in self.query_results
359
    
360
    def with_autocommit(self, func):
361
        import psycopg2.extensions
362
        
363
        prev_isolation_level = self.db.isolation_level
364
        self.db.set_isolation_level(
365
            psycopg2.extensions.ISOLATION_LEVEL_AUTOCOMMIT)
366
        try: return func()
367
        finally: self.db.set_isolation_level(prev_isolation_level)
368
    
369
    def with_savepoint(self, func):
370
        savepoint = 'level_'+str(self._savepoint)
371
        self.run_query('SAVEPOINT '+savepoint, log_level=4)
372
        self._savepoint += 1
373
        try: return func()
374
        except:
375
            self.run_query('ROLLBACK TO SAVEPOINT '+savepoint, log_level=4)
376
            raise
377
        finally:
378
            # Always release savepoint, because after ROLLBACK TO SAVEPOINT,
379
            # "The savepoint remains valid and can be rolled back to again"
380
            # (http://www.postgresql.org/docs/8.3/static/sql-rollback-to.html).
381
            self.run_query('RELEASE SAVEPOINT '+savepoint, log_level=4)
382
            
383
            self._savepoint -= 1
384
            assert self._savepoint >= 0
385
            
386
            self.do_autocommit() # OK to do this after ROLLBACK TO SAVEPOINT
387
    
388
    def do_autocommit(self):
389
        '''Autocommits if outside savepoint'''
390
        assert self._savepoint >= 0
391
        if self.autocommit and self._savepoint == 0:
392
            self.log_debug('Autocommitting', level=4)
393
            self.db.commit()
394
    
395
    def col_info(self, col):
396
        table = sql_gen.Table('columns', 'information_schema')
397
        cols = ['data_type', 'column_default',
398
            cast(self, 'boolean', 'is_nullable')]
399
        
400
        conds = [('table_name', col.table.name), ('column_name', col.name)]
401
        schema = col.table.schema
402
        if schema != None: conds.append(('table_schema', schema))
403
        
404
        type_, default, nullable = row(select(self, table, cols, conds,
405
            order_by='table_schema', limit=1, log_level=4))
406
            # TODO: order_by search_path schema order
407
        default = sql_gen.as_Code(default, self)
408
        
409
        return sql_gen.TypedCol(col.name, type_, default, nullable)
410
    
411
    def TempFunction(self, name):
412
        if self.debug_temp: schema = None
413
        else: schema = 'pg_temp'
414
        return sql_gen.Function(name, schema)
415

    
416
connect = DbConn
417

    
418
##### Recoverable querying
419

    
420
def with_savepoint(db, func): return db.with_savepoint(func)
421

    
422
def run_query(db, query, recover=None, cacheable=False, log_level=2,
423
    log_ignore_excs=None, **kw_args):
424
    '''For params, see DbConn.run_query()'''
425
    if recover == None: recover = False
426
    if log_ignore_excs == None: log_ignore_excs = ()
427
    log_ignore_excs = tuple(log_ignore_excs)
428
    
429
    debug_msg_ref = None # usually, db.run_query() logs query before running it
430
    # But if filtering with log_ignore_excs, wait until after exception parsing
431
    if log_ignore_excs != () or not db.can_mogrify(): debug_msg_ref = [None]
432
    
433
    try:
434
        try:
435
            def run(): return db.run_query(query, cacheable, log_level,
436
                debug_msg_ref, **kw_args)
437
            if recover and not db.is_cached(query):
438
                return with_savepoint(db, run)
439
            else: return run() # don't need savepoint if cached
440
        except Exception, e:
441
            msg = exc.str_(e)
442
            
443
            match = re.search(r'duplicate key value violates unique constraint '
444
                r'"((_?[^\W_]+)_.+?)"', msg)
445
            if match:
446
                constraint, table = match.groups()
447
                cols = []
448
                if recover: # need auto-rollback to run index_cols()
449
                    try: cols = index_cols(db, table, constraint)
450
                    except NotImplementedError: pass
451
                raise DuplicateKeyException(constraint, cols, e)
452
            
453
            match = re.search(r'null value in column "(.+?)" violates not-null'
454
                r' constraint', msg)
455
            if match: raise NullValueException('NOT NULL', [match.group(1)], e)
456
            
457
            match = re.search(r'\b(?:invalid input (?:syntax|value)\b.*?'
458
                r'|date/time field value out of range): "(.+?)"\n'
459
                r'(?:(?s).*?)\bfunction "(.+?)"', msg)
460
            if match:
461
                value, name = match.groups()
462
                raise FunctionValueException(name, strings.to_unicode(value), e)
463
            
464
            match = re.search(r'column "(.+?)" is of type (.+?) but expression '
465
                r'is of type', msg)
466
            if match:
467
                col, type_ = match.groups()
468
                raise MissingCastException(type_, col, e)
469
            
470
            match = re.search(r'\b(\S+) "(.+?)".*? already exists', msg)
471
            if match:
472
                type_, name = match.groups()
473
                raise DuplicateException(type_, name, e)
474
            
475
            raise # no specific exception raised
476
    except log_ignore_excs:
477
        log_level += 2
478
        raise
479
    finally:
480
        if debug_msg_ref != None and debug_msg_ref[0] != None:
481
            db.log_debug(debug_msg_ref[0], log_level)
482

    
483
##### Basic queries
484

    
485
def next_version(name):
486
    version = 1 # first existing name was version 0
487
    match = re.match(r'^(.*)#(\d+)$', name)
488
    if match:
489
        name, version = match.groups()
490
        version = int(version)+1
491
    return sql_gen.concat(name, '#'+str(version))
492

    
493
def lock_table(db, table, mode):
494
    table = sql_gen.as_Table(table)
495
    run_query(db, 'LOCK TABLE '+table.to_str(db)+' IN '+mode+' MODE')
496

    
497
def run_query_into(db, query, into=None, add_indexes_=False, **kw_args):
498
    '''Outputs a query to a temp table.
499
    For params, see run_query().
500
    '''
501
    if into == None: return run_query(db, query, **kw_args)
502
    
503
    assert isinstance(into, sql_gen.Table)
504
    
505
    into.is_temp = True
506
    # "temporary tables cannot specify a schema name", so remove schema
507
    into.schema = None
508
    
509
    kw_args['recover'] = True
510
    kw_args.setdefault('log_ignore_excs', (DuplicateException,))
511
    
512
    temp = not db.debug_temp # tables are permanent in debug_temp mode
513
    
514
    # Create table
515
    while True:
516
        create_query = 'CREATE'
517
        if temp: create_query += ' TEMP'
518
        create_query += ' TABLE '+into.to_str(db)+' AS\n'+query
519
        
520
        try:
521
            cur = run_query(db, create_query, **kw_args)
522
                # CREATE TABLE AS sets rowcount to # rows in query
523
            break
524
        except DuplicateException, e:
525
            into.name = next_version(into.name)
526
            # try again with next version of name
527
    
528
    if add_indexes_: add_indexes(db, into)
529
    
530
    return cur
531

    
532
order_by_pkey = object() # tells mk_select() to order by the pkey
533

    
534
distinct_on_all = object() # tells mk_select() to SELECT DISTINCT ON all columns
535

    
536
def mk_select(db, tables, fields=None, conds=None, distinct_on=[], limit=None,
537
    start=None, order_by=order_by_pkey, default_table=None):
538
    '''
539
    @param tables The single table to select from, or a list of tables to join
540
        together, with tables after the first being sql_gen.Join objects
541
    @param fields Use None to select all fields in the table
542
    @param conds WHERE conditions: [(compare_left_side, compare_right_side),...]
543
        * container can be any iterable type
544
        * compare_left_side: sql_gen.Code|str (for col name)
545
        * compare_right_side: sql_gen.ValueCond|literal value
546
    @param distinct_on The columns to SELECT DISTINCT ON, or distinct_on_all to
547
        use all columns
548
    @return query
549
    '''
550
    # Parse tables param
551
    tables = lists.mk_seq(tables)
552
    tables = list(tables) # don't modify input! (list() copies input)
553
    table0 = sql_gen.as_Table(tables.pop(0)) # first table is separate
554
    
555
    # Parse other params
556
    if conds == None: conds = []
557
    elif dicts.is_dict(conds): conds = conds.items()
558
    conds = list(conds) # don't modify input! (list() copies input)
559
    assert limit == None or type(limit) == int
560
    assert start == None or type(start) == int
561
    if order_by is order_by_pkey:
562
        if distinct_on != []: order_by = None
563
        else: order_by = pkey(db, table0, recover=True)
564
    
565
    query = 'SELECT'
566
    
567
    def parse_col(col): return sql_gen.as_Col(col, default_table).to_str(db)
568
    
569
    # DISTINCT ON columns
570
    if distinct_on != []:
571
        query += '\nDISTINCT'
572
        if distinct_on is not distinct_on_all:
573
            query += ' ON ('+(', '.join(map(parse_col, distinct_on)))+')'
574
    
575
    # Columns
576
    if fields == None:
577
        if query.find('\n') >= 0: whitespace = '\n'
578
        else: whitespace = ' '
579
        query += whitespace+'*'
580
    else:
581
        assert fields != []
582
        query += '\n'+('\n, '.join(map(parse_col, fields)))
583
    
584
    # Main table
585
    query += '\nFROM '+table0.to_str(db)
586
    
587
    # Add joins
588
    left_table = table0
589
    for join_ in tables:
590
        table = join_.table
591
        
592
        # Parse special values
593
        if join_.type_ is sql_gen.filter_out: # filter no match
594
            conds.append((sql_gen.Col(table_not_null_col(db, table), table),
595
                sql_gen.CompareCond(None, '~=')))
596
        
597
        query += '\n'+join_.to_str(db, left_table)
598
        
599
        left_table = table
600
    
601
    missing = True
602
    if conds != []:
603
        if len(conds) == 1: whitespace = ' '
604
        else: whitespace = '\n'
605
        query += '\n'+sql_gen.combine_conds([sql_gen.ColValueCond(l, r)
606
            .to_str(db) for l, r in conds], 'WHERE')
607
        missing = False
608
    if order_by != None:
609
        query += '\nORDER BY '+sql_gen.as_Col(order_by, table0).to_str(db)
610
    if limit != None: query += '\nLIMIT '+str(limit); missing = False
611
    if start != None:
612
        if start != 0: query += '\nOFFSET '+str(start)
613
        missing = False
614
    if missing: warnings.warn(DbWarning(
615
        'SELECT statement missing a WHERE, LIMIT, or OFFSET clause: '+query))
616
    
617
    return query
618

    
619
def select(db, *args, **kw_args):
620
    '''For params, see mk_select() and run_query()'''
621
    recover = kw_args.pop('recover', None)
622
    cacheable = kw_args.pop('cacheable', True)
623
    log_level = kw_args.pop('log_level', 2)
624
    
625
    return run_query(db, mk_select(db, *args, **kw_args), recover, cacheable,
626
        log_level=log_level)
627

    
628
def mk_insert_select(db, table, cols=None, select_query=None, returning=None,
629
    embeddable=False, ignore=False):
630
    '''
631
    @param returning str|None An inserted column (such as pkey) to return
632
    @param embeddable Whether the query should be embeddable as a nested SELECT.
633
        Warning: If you set this and cacheable=True when the query is run, the
634
        query will be fully cached, not just if it raises an exception.
635
    @param ignore Whether to ignore duplicate keys.
636
    '''
637
    table = sql_gen.remove_table_rename(sql_gen.as_Table(table))
638
    if cols == []: cols = None # no cols (all defaults) = unknown col names
639
    if cols != None: cols = [sql_gen.to_name_only_col(c, table) for c in cols]
640
    if select_query == None: select_query = 'DEFAULT VALUES'
641
    if returning != None: returning = sql_gen.as_Col(returning, table)
642
    
643
    first_line = 'INSERT INTO '+table.to_str(db)
644
    
645
    def mk_insert(select_query):
646
        query = first_line
647
        if cols != None:
648
            query += '\n('+(', '.join((c.to_str(db) for c in cols)))+')'
649
        query += '\n'+select_query
650
        
651
        if returning != None:
652
            returning_name_col = sql_gen.to_name_only_col(returning)
653
            query += '\nRETURNING '+returning_name_col.to_str(db)
654
        
655
        return query
656
    
657
    return_type = 'unknown'
658
    if returning != None: return_type = returning.to_str(db)+'%TYPE'
659
    
660
    lang = 'sql'
661
    if ignore:
662
        assert cols != None
663
        # Always return something to set the correct rowcount
664
        if returning == None: returning = sql_gen.NamedCol('NULL', None)
665
        
666
        embeddable = True # must use function
667
        lang = 'plpgsql'
668
        row = [sql_gen.Col(c.name, 'row') for c in cols]
669
        
670
        query = '''\
671
DECLARE
672
    row '''+table.to_str(db)+'''%ROWTYPE;
673
BEGIN
674
    /* Need an EXCEPTION block for each individual row because "When an error is
675
    caught by an EXCEPTION clause, [...] all changes to persistent database
676
    state within the block are rolled back."
677
    This is unfortunate because "A block containing an EXCEPTION clause is
678
    significantly more expensive to enter and exit than a block without one."
679
    (http://www.postgresql.org/docs/8.3/static/plpgsql-control-structures.html\
680
#PLPGSQL-ERROR-TRAPPING)
681
    */
682
    FOR '''+(', '.join((c.to_str(db) for c in row)))+''' IN
683
'''+select_query+'''
684
    LOOP
685
        BEGIN
686
            RETURN QUERY
687
'''+mk_insert(sql_gen.Values(row).to_str(db))+'''
688
;
689
        EXCEPTION
690
            WHEN unique_violation THEN NULL; -- continue to next row
691
        END;
692
    END LOOP;
693
END;\
694
'''
695
    else: query = mk_insert(select_query)
696
    
697
    if embeddable:
698
        # Create function
699
        function_name = sql_gen.clean_name(first_line)
700
        while True:
701
            try:
702
                function = db.TempFunction(function_name)
703
                
704
                function_query = '''\
705
CREATE FUNCTION '''+function.to_str(db)+'''()
706
RETURNS SETOF '''+return_type+'''
707
LANGUAGE '''+lang+'''
708
AS $$
709
'''+query+'''
710
$$;
711
'''
712
                run_query(db, function_query, recover=True, cacheable=True,
713
                    log_ignore_excs=(DuplicateException,))
714
                break # this version was successful
715
            except DuplicateException, e:
716
                function_name = next_version(function_name)
717
                # try again with next version of name
718
        
719
        # Return query that uses function
720
        cols = None
721
        if returning != None: cols = [returning]
722
        func_table = sql_gen.NamedTable('f', sql_gen.FunctionCall(function),
723
            cols) # AS clause requires function alias
724
        return mk_select(db, func_table, start=0, order_by=None)
725
    
726
    return query
727

    
728
def insert_select(db, *args, **kw_args):
729
    '''For params, see mk_insert_select() and run_query_into()
730
    @param into sql_gen.Table with suggested name of temp table to put RETURNING
731
        values in
732
    '''
733
    into = kw_args.pop('into', None)
734
    if into != None: kw_args['embeddable'] = True
735
    recover = kw_args.pop('recover', None)
736
    if kw_args.get('ignore', False): recover = True
737
    cacheable = kw_args.pop('cacheable', True)
738
    log_level = kw_args.pop('log_level', 2)
739
    
740
    return run_query_into(db, mk_insert_select(db, *args, **kw_args), into,
741
        recover=recover, cacheable=cacheable, log_level=log_level)
742

    
743
default = sql_gen.default # tells insert() to use the default value for a column
744

    
745
def insert(db, table, row, *args, **kw_args):
746
    '''For params, see insert_select()'''
747
    if lists.is_seq(row): cols = None
748
    else:
749
        cols = row.keys()
750
        row = row.values()
751
    row = list(row) # ensure that "== []" works
752
    
753
    if row == []: query = None
754
    else: query = sql_gen.Values(row).to_str(db)
755
    
756
    return insert_select(db, table, cols, query, *args, **kw_args)
757

    
758
def mk_update(db, table, changes=None, cond=None):
759
    '''
760
    @param changes [(col, new_value),...]
761
        * container can be any iterable type
762
        * col: sql_gen.Code|str (for col name)
763
        * new_value: sql_gen.Code|literal value
764
    @param cond sql_gen.Code WHERE condition. e.g. use sql_gen.*Cond objects.
765
    @return str query
766
    '''
767
    query = 'UPDATE '+sql_gen.as_Table(table).to_str(db)+'\nSET\n'
768
    query += ',\n'.join((sql_gen.to_name_only_col(col, table).to_str(db)+' = '
769
        +sql_gen.as_Value(new_value).to_str(db) for col, new_value in changes))
770
    if cond != None: query += '\nWHERE\n'+cond.to_str(db)
771
    
772
    return query
773

    
774
def update(db, *args, **kw_args):
775
    '''For params, see mk_update() and run_query()'''
776
    recover = kw_args.pop('recover', None)
777
    cacheable = kw_args.pop('cacheable', False)
778
    log_level = kw_args.pop('log_level', 2)
779
    
780
    return run_query(db, mk_update(db, *args, **kw_args), recover, cacheable,
781
        log_level=log_level)
782

    
783
def last_insert_id(db):
784
    module = util.root_module(db.db)
785
    if module == 'psycopg2': return value(run_query(db, 'SELECT lastval()'))
786
    elif module == 'MySQLdb': return db.insert_id()
787
    else: return None
788

    
789
def mk_flatten_mapping(db, into, cols, preserve=[], as_items=False):
790
    '''Creates a mapping from original column names (which may have collisions)
791
    to names that will be distinct among the columns' tables.
792
    This is meant to be used for several tables that are being joined together.
793
    @param cols The columns to combine. Duplicates will be removed.
794
    @param into The table for the new columns.
795
    @param preserve [sql_gen.Col...] Columns not to rename. Note that these
796
        columns will be included in the mapping even if they are not in cols.
797
        The tables of the provided Col objects will be changed to into, so make
798
        copies of them if you want to keep the original tables.
799
    @param as_items Whether to return a list of dict items instead of a dict
800
    @return dict(orig_col=new_col, ...)
801
        * orig_col: sql_gen.Col(orig_col_name, orig_table)
802
        * new_col: sql_gen.Col(orig_col_name, into)
803
        * All mappings use the into table so its name can easily be
804
          changed for all columns at once
805
    '''
806
    cols = lists.uniqify(cols)
807
    
808
    items = []
809
    for col in preserve:
810
        orig_col = copy.copy(col)
811
        col.table = into
812
        items.append((orig_col, col))
813
    preserve = set(preserve)
814
    for col in cols:
815
        if col not in preserve:
816
            items.append((col, sql_gen.Col(str(col), into, col.srcs)))
817
    
818
    if not as_items: items = dict(items)
819
    return items
820

    
821
def flatten(db, into, joins, cols, limit=None, start=None, **kw_args):
822
    '''For params, see mk_flatten_mapping()
823
    @return See return value of mk_flatten_mapping()
824
    '''
825
    items = mk_flatten_mapping(db, into, cols, as_items=True, **kw_args)
826
    cols = [sql_gen.NamedCol(new.name, old) for old, new in items]
827
    run_query_into(db, mk_select(db, joins, cols, limit=limit, start=start),
828
        into=into, add_indexes_=True)
829
    return dict(items)
830

    
831
def mk_track_data_error(db, errors_table, cols, value, error_code, error):
832
    assert cols != ()
833
    
834
    cols = map(sql_gen.to_name_only_col, cols)
835
    
836
    columns_cols = ['column']
837
    columns = sql_gen.NamedValues('columns', columns_cols,
838
        [[c.name] for c in cols])
839
    values_cols = ['value', 'error_code', 'error']
840
    values = sql_gen.NamedValues('values', values_cols,
841
        [value, error_code, error])
842
    
843
    select_cols = columns_cols+values_cols
844
    name_only_cols = map(sql_gen.to_name_only_col, select_cols)
845
    errors_table = sql_gen.NamedTable('errors', errors_table)
846
    joins = [columns, sql_gen.Join(values, type_='CROSS'),
847
        sql_gen.Join(errors_table, dict(zip(name_only_cols, select_cols)),
848
        sql_gen.filter_out)]
849
    
850
    return mk_insert_select(db, errors_table, name_only_cols,
851
        mk_select(db, joins, select_cols, order_by=None))
852

    
853
def track_data_error(db, errors_table, cols, value, error_code, error):
854
    '''
855
    @param errors_table If None, does nothing.
856
    '''
857
    if errors_table == None or cols == (): return
858
    
859
    for col in cols:
860
        try:
861
            insert(db, errors_table, dict(column=col.name, value=value,
862
                error_code=error_code, error=error), recover=True,
863
                cacheable=True, log_level=4)
864
        except DuplicateKeyException: pass
865

    
866
def cast(db, type_, col, errors_table=None):
867
    '''Casts an (unrenamed) column or value.
868
    If errors_table set and col has srcs, saves errors in errors_table (using
869
    col's srcs attr as the source columns) and converts errors to warnings.
870
    @param col str|sql_gen.Col|sql_gen.Literal
871
    @param errors_table None|sql_gen.Table|str
872
    '''
873
    col = sql_gen.as_Col(col)
874
    save_errors = (errors_table != None and isinstance(col, sql_gen.Col)
875
        and col.srcs != ())
876
    if not save_errors: return sql_gen.Cast(type_, col) # can't save errors
877
    
878
    assert not isinstance(col, sql_gen.NamedCol)
879
    
880
    errors_table = sql_gen.as_Table(errors_table)
881
    srcs = map(sql_gen.to_name_only_col, col.srcs)
882
    function_name = str(sql_gen.FunctionCall(type_, *srcs))
883
    function = db.TempFunction(function_name)
884
    
885
    while True:
886
        # Create function definition
887
        errors_table_cols = map(sql_gen.Col,
888
            ['column', 'value', 'error_code', 'error'])
889
        query = '''\
890
CREATE FUNCTION '''+function.to_str(db)+'''(value text)
891
RETURNS '''+type_+'''
892
LANGUAGE plpgsql
893
STRICT
894
AS $$
895
BEGIN
896
    /* The explicit cast to the return type is needed to make the cast happen
897
    inside the try block. (Implicit casts to the return type happen at the end
898
    of the function, outside any block.) */
899
    RETURN value::'''+type_+''';
900
EXCEPTION
901
    WHEN data_exception THEN
902
        -- Save error in errors table.
903
        DECLARE
904
            error_code text := SQLSTATE;
905
            error text := SQLERRM;
906
            "column" text;
907
        BEGIN
908
            -- Insert the value and error for *each* source column.
909
            FOR "column" IN
910
'''+mk_select(db, sql_gen.NamedValues('c', None, [[c.name] for c in srcs]),
911
    order_by=None)+'''
912
            LOOP
913
                BEGIN
914
'''+mk_insert_select(db, errors_table, errors_table_cols,
915
    sql_gen.Values(errors_table_cols).to_str(db))+''';
916
                EXCEPTION
917
                    WHEN unique_violation THEN NULL; -- continue to next row
918
                END;
919
            END LOOP;
920
        END;
921
        
922
        RAISE WARNING '%', SQLERRM;
923
        RETURN NULL;
924
END;
925
$$;
926
'''
927
        
928
        # Create function
929
        try:
930
            run_query(db, query, recover=True, cacheable=True,
931
                log_ignore_excs=(DuplicateException,))
932
            break # successful
933
        except DuplicateException:
934
            function.name = next_version(function.name)
935
            # try again with next version of name
936
    
937
    return sql_gen.FunctionCall(function, col)
938

    
939
##### Database structure queries
940

    
941
def table_row_count(db, table, recover=None):
942
    return value(run_query(db, mk_select(db, table, [sql_gen.row_count],
943
        order_by=None, start=0), recover=recover, log_level=3))
944

    
945
def table_cols(db, table, recover=None):
946
    return list(col_names(select(db, table, limit=0, order_by=None,
947
        recover=recover, log_level=4)))
948

    
949
def pkey(db, table, recover=None):
950
    '''Assumed to be first column in table'''
951
    return table_cols(db, table, recover)[0]
952

    
953
not_null_col = 'not_null_col'
954

    
955
def table_not_null_col(db, table, recover=None):
956
    '''Name assumed to be the value of not_null_col. If not found, uses pkey.'''
957
    if not_null_col in table_cols(db, table, recover): return not_null_col
958
    else: return pkey(db, table, recover)
959

    
960
def index_cols(db, table, index):
961
    '''Can also use this for UNIQUE constraints, because a UNIQUE index is
962
    automatically created. When you don't know whether something is a UNIQUE
963
    constraint or a UNIQUE index, use this function.'''
964
    module = util.root_module(db.db)
965
    if module == 'psycopg2':
966
        return list(values(run_query(db, '''\
967
SELECT attname
968
FROM
969
(
970
        SELECT attnum, attname
971
        FROM pg_index
972
        JOIN pg_class index ON index.oid = indexrelid
973
        JOIN pg_class table_ ON table_.oid = indrelid
974
        JOIN pg_attribute ON attrelid = indrelid AND attnum = ANY (indkey)
975
        WHERE
976
            table_.relname = '''+db.esc_value(table)+'''
977
            AND index.relname = '''+db.esc_value(index)+'''
978
    UNION
979
        SELECT attnum, attname
980
        FROM
981
        (
982
            SELECT
983
                indrelid
984
                , (regexp_matches(indexprs, E':varattno (\\\\d+)', 'g'))[1]::int
985
                    AS indkey
986
            FROM pg_index
987
            JOIN pg_class index ON index.oid = indexrelid
988
            JOIN pg_class table_ ON table_.oid = indrelid
989
            WHERE
990
                table_.relname = '''+db.esc_value(table)+'''
991
                AND index.relname = '''+db.esc_value(index)+'''
992
        ) s
993
        JOIN pg_attribute ON attrelid = indrelid AND attnum = indkey
994
) s
995
ORDER BY attnum
996
'''
997
            , cacheable=True, log_level=4)))
998
    else: raise NotImplementedError("Can't list index columns for "+module+
999
        ' database')
1000

    
1001
def constraint_cols(db, table, constraint):
1002
    module = util.root_module(db.db)
1003
    if module == 'psycopg2':
1004
        return list(values(run_query(db, '''\
1005
SELECT attname
1006
FROM pg_constraint
1007
JOIN pg_class ON pg_class.oid = conrelid
1008
JOIN pg_attribute ON attrelid = conrelid AND attnum = ANY (conkey)
1009
WHERE
1010
    relname = '''+db.esc_value(table)+'''
1011
    AND conname = '''+db.esc_value(constraint)+'''
1012
ORDER BY attnum
1013
'''
1014
            )))
1015
    else: raise NotImplementedError("Can't list constraint columns for "+module+
1016
        ' database')
1017

    
1018
row_num_col = '_row_num'
1019

    
1020
def add_index(db, exprs, table=None, unique=False, ensure_not_null_=True):
1021
    '''Adds an index on column(s) or expression(s) if it doesn't already exist.
1022
    Currently, only function calls are supported as expressions.
1023
    @param ensure_not_null_ If set, translates NULL values to sentinel values.
1024
        This allows indexes to be used for comparisons where NULLs are equal.
1025
    '''
1026
    exprs = lists.mk_seq(exprs)
1027
    
1028
    # Parse exprs
1029
    old_exprs = exprs[:]
1030
    exprs = []
1031
    cols = []
1032
    for i, expr in enumerate(old_exprs):
1033
        expr = sql_gen.as_Col(expr, table)
1034
        
1035
        # Handle nullable columns
1036
        if ensure_not_null_:
1037
            try: expr = ensure_not_null(db, expr)
1038
            except KeyError: pass # unknown type, so just create plain index
1039
        
1040
        # Extract col
1041
        expr = copy.deepcopy(expr) # don't modify input!
1042
        if isinstance(expr, sql_gen.FunctionCall):
1043
            col = expr.args[0]
1044
            expr = sql_gen.Expr(expr)
1045
        else: col = expr
1046
        assert isinstance(col, sql_gen.Col)
1047
        
1048
        # Extract table
1049
        if table == None:
1050
            assert sql_gen.is_table_col(col)
1051
            table = col.table
1052
        
1053
        col.table = None
1054
        
1055
        exprs.append(expr)
1056
        cols.append(col)
1057
    
1058
    table = sql_gen.as_Table(table)
1059
    index = sql_gen.Table(str(sql_gen.Col(','.join(map(str, cols)), table)))
1060
    
1061
    # Add index
1062
    while True:
1063
        str_ = 'CREATE'
1064
        if unique: str_ += ' UNIQUE'
1065
        str_ += ' INDEX '+index.to_str(db)+' ON '+table.to_str(db)+' ('+(
1066
            ', '.join((v.to_str(db) for v in exprs)))+')'
1067
        
1068
        try:
1069
            run_query(db, str_, recover=True, cacheable=True, log_level=3,
1070
                log_ignore_excs=(DuplicateException,))
1071
            break
1072
        except DuplicateException:
1073
            index.name = next_version(index.name)
1074
            # try again with next version of name
1075

    
1076
def add_pkey(db, table, cols=None, recover=None):
1077
    '''Adds a primary key.
1078
    @param cols [sql_gen.Col,...] The columns in the primary key.
1079
        Defaults to the first column in the table.
1080
    @pre The table must not already have a primary key.
1081
    '''
1082
    table = sql_gen.as_Table(table)
1083
    if cols == None: cols = [pkey(db, table, recover)]
1084
    col_strs = [sql_gen.to_name_only_col(v).to_str(db) for v in cols]
1085
    
1086
    run_query(db, 'ALTER TABLE '+table.to_str(db)+' ADD PRIMARY KEY ('
1087
        +(', '.join(col_strs))+')', recover=True, cacheable=True, log_level=3,
1088
        log_ignore_excs=(DuplicateException,))
1089

    
1090
def add_not_null(db, col):
1091
    table = col.table
1092
    col = sql_gen.to_name_only_col(col)
1093
    run_query(db, 'ALTER TABLE '+table.to_str(db)+' ALTER COLUMN '
1094
        +col.to_str(db)+' SET NOT NULL', cacheable=True, log_level=3)
1095

    
1096
def add_index_col(db, col, suffix, expr, nullable=True):
1097
    if sql_gen.index_col(col) != None: return # already has index col
1098
    
1099
    new_col = sql_gen.suffixed_col(col, suffix)
1100
    
1101
    # Add column
1102
    new_typed_col = sql_gen.TypedCol(new_col.name, db.col_info(col).type)
1103
    add_col(db, col.table, new_typed_col, comment='src: '+repr(col),
1104
        log_level=3)
1105
    new_col.name = new_typed_col.name # propagate any renaming
1106
    
1107
    update(db, col.table, [(new_col, expr)], cacheable=True, log_level=3)
1108
    if not nullable: add_not_null(db, new_col)
1109
    add_index(db, new_col)
1110
    
1111
    col.table.index_cols[col.name] = new_col
1112

    
1113
# Controls when ensure_not_null() will use index columns
1114
not_null_index_cols_min_rows = 0 # rows; initially always use index columns
1115

    
1116
def ensure_not_null(db, col):
1117
    '''For params, see sql_gen.ensure_not_null()'''
1118
    expr = sql_gen.ensure_not_null(db, col)
1119
    
1120
    # If a nullable column in a temp table, add separate index column instead.
1121
    # Note that for small datasources, this adds 6-25% to the total import time.
1122
    if (sql_gen.is_temp_col(col) and isinstance(expr, sql_gen.EnsureNotNull)
1123
        and table_row_count(db, col.table) >= not_null_index_cols_min_rows):
1124
        add_index_col(db, col, '::NOT NULL', expr, nullable=False)
1125
        expr = sql_gen.index_col(col)
1126
    
1127
    return expr
1128

    
1129
already_indexed = object() # tells add_indexes() the pkey has already been added
1130

    
1131
def add_indexes(db, table, has_pkey=True):
1132
    '''Adds an index on all columns in a table.
1133
    @param has_pkey bool|already_indexed Whether a pkey instead of a regular
1134
        index should be added on the first column.
1135
        * If already_indexed, the pkey is assumed to have already been added
1136
    '''
1137
    cols = table_cols(db, table)
1138
    if has_pkey:
1139
        if has_pkey is not already_indexed: add_pkey(db, table)
1140
        cols = cols[1:]
1141
    for col in cols: add_index(db, col, table)
1142

    
1143
def add_col(db, table, col, comment=None, **kw_args):
1144
    '''
1145
    @param col TypedCol Name may be versioned, so be sure to propagate any
1146
        renaming back to any source column for the TypedCol.
1147
    @param comment None|str SQL comment used to distinguish columns of the same
1148
        name from each other when they contain different data, to allow the
1149
        ADD COLUMN query to be cached. If not set, query will not be cached.
1150
    '''
1151
    assert isinstance(col, sql_gen.TypedCol)
1152
    
1153
    while True:
1154
        str_ = 'ALTER TABLE '+table.to_str(db)+' ADD COLUMN '+col.to_str(db)
1155
        if comment != None: str_ += ' '+sql_gen.esc_comment(comment)
1156
        
1157
        try:
1158
            run_query(db, str_, recover=True, cacheable=True, **kw_args)
1159
            break
1160
        except DuplicateException:
1161
            col.name = next_version(col.name)
1162
            # try again with next version of name
1163

    
1164
row_num_typed_col = sql_gen.TypedCol(row_num_col, 'serial', nullable=False,
1165
    constraints='PRIMARY KEY')
1166

    
1167
def add_row_num(db, table):
1168
    '''Adds a row number column to a table. Its name is in row_num_col. It will
1169
    be the primary key.'''
1170
    add_col(db, table, row_num_typed_col, log_level=3)
1171

    
1172
def cast_temp_col(db, type_, col, errors_table=None):
1173
    '''Like cast(), but creates a new column with the cast values if the input
1174
    is a column.
1175
    @return The new column or cast value
1176
    '''
1177
    def cast_(col): return cast(db, type_, col, errors_table)
1178
    
1179
    try: col = sql_gen.underlying_col(col)
1180
    except sql_gen.NoUnderlyingTableException: return sql_gen.wrap(cast_, col)
1181
    
1182
    table = col.table
1183
    new_col = sql_gen.Col(sql_gen.concat(col.name, '::'+type_), table, col.srcs)
1184
    expr = cast_(col)
1185
    
1186
    # Add column
1187
    new_typed_col = sql_gen.TypedCol(new_col.name, type_)
1188
    add_col(db, table, new_typed_col, comment='src: '+repr(col))
1189
    new_col.name = new_typed_col.name # propagate any renaming
1190
    
1191
    update(db, table, [(new_col, expr)], cacheable=True)
1192
    add_index(db, new_col)
1193
    
1194
    return new_col
1195

    
1196
def drop_table(db, table):
1197
    table = sql_gen.as_Table(table)
1198
    return run_query(db, 'DROP TABLE IF EXISTS '+table.to_str(db)+' CASCADE')
1199

    
1200
def create_table(db, table, cols, has_pkey=True, col_indexes=True):
1201
    '''Creates a table.
1202
    @param cols [sql_gen.TypedCol,...] The column names and types
1203
    @param has_pkey If set, the first column becomes the primary key.
1204
    @param col_indexes bool|[ref]
1205
        * If True, indexes will be added on all non-pkey columns.
1206
        * If a list reference, [0] will be set to a function to do this.
1207
          This can be used to delay index creation until the table is populated.
1208
    '''
1209
    table = sql_gen.as_Table(table)
1210
    
1211
    if has_pkey:
1212
        cols[0] = pkey = copy.copy(cols[0]) # don't modify input!
1213
        pkey.constraints = 'PRIMARY KEY'
1214
    
1215
    str_ = 'CREATE TABLE '+table.to_str(db)+' (\n'
1216
    str_ += '\n, '.join(v.to_str(db) for v in cols)
1217
    str_ += '\n);\n'
1218
    run_query(db, str_, cacheable=True, log_level=2)
1219
    
1220
    # Add indexes
1221
    if has_pkey: has_pkey = already_indexed
1222
    def add_indexes_(): add_indexes(db, table, has_pkey)
1223
    if isinstance(col_indexes, list): col_indexes[0] = add_indexes_ # defer
1224
    elif col_indexes: add_indexes_() # add now
1225

    
1226
def vacuum(db, table):
1227
    table = sql_gen.as_Table(table)
1228
    db.with_autocommit(lambda: run_query(db, 'VACUUM ANALYZE '+table.to_str(db),
1229
        log_level=3))
1230

    
1231
def truncate(db, table, schema='public', **kw_args):
1232
    '''For params, see run_query()'''
1233
    table = sql_gen.as_Table(table, schema)
1234
    return run_query(db, 'TRUNCATE '+table.to_str(db)+' CASCADE', **kw_args)
1235

    
1236
def empty_temp(db, tables):
1237
    if db.debug_temp: return # leave temp tables there for debugging
1238
    tables = lists.mk_seq(tables)
1239
    for table in tables: truncate(db, table, log_level=3)
1240

    
1241
def tables(db, schema_like='public', table_like='%', exact=False):
1242
    if exact: compare = '='
1243
    else: compare = 'LIKE'
1244
    
1245
    module = util.root_module(db.db)
1246
    if module == 'psycopg2':
1247
        conds = [('schemaname', sql_gen.CompareCond(schema_like, compare)),
1248
            ('tablename', sql_gen.CompareCond(table_like, compare))]
1249
        return values(select(db, 'pg_tables', ['tablename'], conds,
1250
            order_by='tablename', log_level=4))
1251
    elif module == 'MySQLdb':
1252
        return values(run_query(db, 'SHOW TABLES LIKE '+db.esc_value(table_like)
1253
            , cacheable=True, log_level=4))
1254
    else: raise NotImplementedError("Can't list tables for "+module+' database')
1255

    
1256
def table_exists(db, table):
1257
    table = sql_gen.as_Table(table)
1258
    return list(tables(db, table.schema, table.name, exact=True)) != []
1259

    
1260
def function_exists(db, function):
1261
    function = sql_gen.as_Function(function)
1262
    
1263
    info_table = sql_gen.Table('routines', 'information_schema')
1264
    conds = [('routine_name', function.name)]
1265
    schema = function.schema
1266
    if schema != None: conds.append(('routine_schema', schema))
1267
    # Exclude trigger functions, since they cannot be called directly
1268
    conds.append(('data_type', sql_gen.CompareCond('trigger', '!=')))
1269
    
1270
    return list(values(select(db, info_table, ['routine_name'], conds,
1271
        order_by='routine_schema', limit=1, log_level=4))) != []
1272
        # TODO: order_by search_path schema order
1273

    
1274
def errors_table(db, table, if_exists=True):
1275
    '''
1276
    @param if_exists If set, returns None if the errors table doesn't exist
1277
    @return None|sql_gen.Table
1278
    '''
1279
    table = sql_gen.as_Table(table)
1280
    if table.srcs != (): table = table.srcs[0]
1281
    
1282
    errors_table = sql_gen.suffixed_table(table, '.errors')
1283
    if if_exists and not table_exists(db, errors_table): return None
1284
    return errors_table
1285

    
1286
##### Database management
1287

    
1288
def empty_db(db, schema='public', **kw_args):
1289
    '''For kw_args, see tables()'''
1290
    for table in tables(db, schema, **kw_args): truncate(db, table, schema)
1291

    
1292
##### Heuristic queries
1293

    
1294
def put(db, table, row, pkey_=None, row_ct_ref=None):
1295
    '''Recovers from errors.
1296
    Only works under PostgreSQL (uses INSERT RETURNING).
1297
    '''
1298
    row = sql_gen.ColDict(db, table, row)
1299
    if pkey_ == None: pkey_ = pkey(db, table, recover=True)
1300
    
1301
    try:
1302
        cur = insert(db, table, row, pkey_, recover=True)
1303
        if row_ct_ref != None and cur.rowcount >= 0:
1304
            row_ct_ref[0] += cur.rowcount
1305
        return value(cur)
1306
    except DuplicateKeyException, e:
1307
        row = sql_gen.ColDict(db, table,
1308
            util.dict_subset_right_join(row, e.cols))
1309
        return value(select(db, table, [pkey_], row, recover=True))
1310

    
1311
def get(db, table, row, pkey, row_ct_ref=None, create=False):
1312
    '''Recovers from errors'''
1313
    try: return value(select(db, table, [pkey], row, limit=1, recover=True))
1314
    except StopIteration:
1315
        if not create: raise
1316
        return put(db, table, row, pkey, row_ct_ref) # insert new row
1317

    
1318
def is_func_result(col):
1319
    return col.table.name.find('(') >= 0 and col.name == 'result'
1320

    
1321
def into_table_name(out_table, in_tables0, mapping, is_func):
1322
    def in_col_str(in_col):
1323
        in_col = sql_gen.remove_col_rename(in_col)
1324
        if isinstance(in_col, sql_gen.Col):
1325
            table = in_col.table
1326
            if table == in_tables0:
1327
                in_col = sql_gen.to_name_only_col(in_col)
1328
            elif is_func_result(in_col): in_col = table # omit col name
1329
        return str(in_col)
1330
    
1331
    str_ = str(out_table)
1332
    if is_func:
1333
        str_ += '('
1334
        
1335
        try: value_in_col = mapping['value']
1336
        except KeyError:
1337
            str_ += ', '.join((str(k)+'='+in_col_str(v)
1338
                for k, v in mapping.iteritems()))
1339
        else: str_ += in_col_str(value_in_col)
1340
        
1341
        str_ += ')'
1342
    else:
1343
        out_col = 'rank'
1344
        try: in_col = mapping[out_col]
1345
        except KeyError: str_ += '_pkeys'
1346
        else: # has a rank column, so hierarchical
1347
            str_ += '['+str(out_col)+'='+in_col_str(in_col)+']'
1348
    return str_
1349

    
1350
def put_table(db, out_table, in_tables, mapping, row_ct_ref=None, into=None,
1351
    default=None, is_func=False, on_error=exc.raise_):
1352
    '''Recovers from errors.
1353
    Only works under PostgreSQL (uses INSERT RETURNING).
1354
    IMPORTANT: Must be run at the *beginning* of a transaction.
1355
    @param in_tables The main input table to select from, followed by a list of
1356
        tables to join with it using the main input table's pkey
1357
    @param mapping dict(out_table_col=in_table_col, ...)
1358
        * out_table_col: str (*not* sql_gen.Col)
1359
        * in_table_col: sql_gen.Col|literal-value
1360
    @param into The table to contain the output and input pkeys.
1361
        Defaults to `out_table.name+'_pkeys'`.
1362
    @param default The *output* column to use as the pkey for missing rows.
1363
        If this output column does not exist in the mapping, uses None.
1364
    @param is_func Whether out_table is the name of a SQL function, not a table
1365
    @return sql_gen.Col Where the output pkeys are made available
1366
    '''
1367
    out_table = sql_gen.as_Table(out_table)
1368
    
1369
    def log_debug(msg): db.log_debug(msg, level=1.5)
1370
    def col_ustr(str_):
1371
        return strings.repr_no_u(sql_gen.remove_col_rename(str_))
1372
    
1373
    log_debug('********** New iteration **********')
1374
    log_debug('Inserting these input columns into '+strings.as_tt(
1375
        out_table.to_str(db))+':\n'+strings.as_table(mapping, ustr=col_ustr))
1376
    
1377
    is_function = function_exists(db, out_table)
1378
    
1379
    if is_function: out_pkey = 'result'
1380
    else: out_pkey = pkey(db, out_table, recover=True)
1381
    out_pkey_col = sql_gen.as_Col(out_pkey, out_table)
1382
    
1383
    if mapping == {}: # need at least one column for INSERT SELECT
1384
        mapping = {out_pkey: None} # ColDict will replace with default value
1385
    
1386
    # Create input joins from list of input tables
1387
    in_tables_ = in_tables[:] # don't modify input!
1388
    in_tables0 = in_tables_.pop(0) # first table is separate
1389
    errors_table_ = errors_table(db, in_tables0)
1390
    in_pkey = pkey(db, in_tables0, recover=True)
1391
    in_pkey_col = sql_gen.as_Col(in_pkey, in_tables0)
1392
    input_joins = [in_tables0]+[sql_gen.Join(v,
1393
        {in_pkey: sql_gen.join_same_not_null}) for v in in_tables_]
1394
    
1395
    if into == None:
1396
        into = into_table_name(out_table, in_tables0, mapping, is_func)
1397
    into = sql_gen.as_Table(into)
1398
    
1399
    # Set column sources
1400
    in_cols = filter(sql_gen.is_table_col, mapping.values())
1401
    for col in in_cols:
1402
        if col.table == in_tables0: col.set_srcs(sql_gen.src_self)
1403
    
1404
    log_debug('Joining together input tables into temp table')
1405
    # Place in new table for speed and so don't modify input if values edited
1406
    in_table = sql_gen.Table('in')
1407
    mapping = dicts.join(mapping, flatten(db, in_table, input_joins, in_cols,
1408
        preserve=[in_pkey_col], start=0))
1409
    input_joins = [in_table]
1410
    db.log_debug('Temp table: '+strings.as_tt(in_table.to_str(db)), level=2)
1411
    
1412
    mapping = sql_gen.ColDict(db, out_table, mapping)
1413
        # after applying dicts.join() because that returns a plain dict
1414
    
1415
    # Resolve default value column
1416
    if default != None:
1417
        try: default = mapping[default]
1418
        except KeyError:
1419
            db.log_debug('Default value column '
1420
                +strings.as_tt(strings.repr_no_u(default))
1421
                +' does not exist in mapping, falling back to None', level=2.1)
1422
            default = None
1423
    
1424
    pkeys_names = [in_pkey, out_pkey]
1425
    pkeys_cols = [in_pkey_col, out_pkey_col]
1426
    
1427
    pkeys_table_exists_ref = [False]
1428
    def insert_into_pkeys(joins, cols, distinct=False):
1429
        kw_args = {}
1430
        if distinct: kw_args.update(dict(distinct_on=[in_pkey_col]))
1431
        query = mk_select(db, joins, cols, order_by=None, start=0, **kw_args)
1432
        
1433
        if pkeys_table_exists_ref[0]:
1434
            insert_select(db, into, pkeys_names, query)
1435
        else:
1436
            run_query_into(db, query, into=into)
1437
            pkeys_table_exists_ref[0] = True
1438
    
1439
    limit_ref = [None]
1440
    conds = set()
1441
    distinct_on = sql_gen.ColDict(db, out_table)
1442
    def mk_main_select(joins, cols):
1443
        distinct_on_cols = [c.to_Col() for c in distinct_on.values()]
1444
        return mk_select(db, joins, cols, conds, distinct_on_cols,
1445
            limit=limit_ref[0], start=0)
1446
    
1447
    exc_strs = set()
1448
    def log_exc(e):
1449
        e_str = exc.str_(e, first_line_only=True)
1450
        log_debug('Caught exception: '+e_str)
1451
        assert e_str not in exc_strs # avoid infinite loops
1452
        exc_strs.add(e_str)
1453
    
1454
    def remove_all_rows():
1455
        log_debug('Ignoring all rows')
1456
        limit_ref[0] = 0 # just create an empty pkeys table
1457
    
1458
    def ignore(in_col, value, e):
1459
        track_data_error(db, errors_table_, in_col.srcs, value, e.cause.pgcode,
1460
            e.cause.pgerror)
1461
        log_debug('Ignoring rows with '+strings.as_tt(repr(in_col))+' = '
1462
            +strings.as_tt(repr(value)))
1463
    
1464
    def remove_rows(in_col, value, e):
1465
        ignore(in_col, value, e)
1466
        cond = (in_col, sql_gen.CompareCond(value, '!='))
1467
        assert cond not in conds # avoid infinite loops
1468
        conds.add(cond)
1469
    
1470
    def invalid2null(in_col, value, e):
1471
        ignore(in_col, value, e)
1472
        update(db, in_table, [(in_col, None)],
1473
            sql_gen.ColValueCond(in_col, value))
1474
    
1475
    def insert_pkeys_table(which):
1476
        return sql_gen.Table(sql_gen.concat(in_table.name,
1477
            '_insert_'+which+'_pkeys'))
1478
    insert_out_pkeys = insert_pkeys_table('out')
1479
    insert_in_pkeys = insert_pkeys_table('in')
1480
    
1481
    # Do inserts and selects
1482
    join_cols = sql_gen.ColDict(db, out_table)
1483
    while True:
1484
        if limit_ref[0] == 0: # special case
1485
            log_debug('Creating an empty pkeys table')
1486
            cur = run_query_into(db, mk_select(db, out_table, [out_pkey],
1487
                limit=limit_ref[0]), into=insert_out_pkeys)
1488
            break # don't do main case
1489
        
1490
        has_joins = join_cols != {}
1491
        
1492
        log_debug('Trying to insert new rows')
1493
        
1494
        # Prepare to insert new rows
1495
        insert_joins = input_joins[:] # don't modify original!
1496
        insert_args = dict(recover=True, cacheable=False)
1497
        if has_joins:
1498
            insert_args.update(dict(ignore=True))
1499
        else:
1500
            insert_args.update(dict(returning=out_pkey, into=insert_out_pkeys))
1501
        main_select = mk_main_select(insert_joins, mapping.values())
1502
        
1503
        def main_insert():
1504
            if is_function:
1505
                log_debug('Calling function on input rows')
1506
                args = dict(((k.name, v) for k, v in mapping.iteritems()))
1507
                func_call = sql_gen.NamedCol(out_pkey,
1508
                    sql_gen.FunctionCall(out_table, **args))
1509
                insert_into_pkeys(input_joins, [in_pkey_col, func_call])
1510
                return None
1511
            else:
1512
                return insert_select(db, out_table, mapping.keys(), main_select,
1513
                    **insert_args)
1514
        
1515
        try:
1516
            cur = with_savepoint(db, main_insert)
1517
            break # insert successful
1518
        except MissingCastException, e:
1519
            log_exc(e)
1520
            
1521
            out_col = e.col
1522
            type_ = e.type
1523
            
1524
            log_debug('Casting '+strings.as_tt(out_col)+' input to '
1525
                +strings.as_tt(type_))
1526
            mapping[out_col] = cast_temp_col(db, type_, mapping[out_col],
1527
                errors_table_)
1528
        except DuplicateKeyException, e:
1529
            log_exc(e)
1530
            
1531
            old_join_cols = join_cols.copy()
1532
            distinct_on.update(util.dict_subset(mapping, e.cols))
1533
            join_cols.update(util.dict_subset_right_join(mapping, e.cols))
1534
            log_debug('Ignoring existing rows, comparing on these columns:\n'
1535
                +strings.as_inline_table(join_cols, ustr=col_ustr))
1536
            assert join_cols != old_join_cols # avoid infinite loops
1537
        except NullValueException, e:
1538
            log_exc(e)
1539
            
1540
            out_col, = e.cols
1541
            try: in_col = mapping[out_col]
1542
            except KeyError:
1543
                log_debug('Missing mapping for NOT NULL column '+out_col)
1544
                remove_all_rows()
1545
            else: remove_rows(in_col, None, e)
1546
        except FunctionValueException, e:
1547
            log_exc(e)
1548
            
1549
            func_name = e.name
1550
            value = e.value
1551
            for out_col, in_col in mapping.iteritems():
1552
                in_col = sql_gen.unwrap_func_call(in_col, func_name)
1553
                invalid2null(in_col, value, e)
1554
        except DatabaseErrors, e:
1555
            log_exc(e)
1556
            
1557
            log_debug('No handler for exception')
1558
            on_error(e)
1559
            remove_all_rows()
1560
        # after exception handled, rerun loop with additional constraints
1561
    
1562
    if cur != None and row_ct_ref != None and cur.rowcount >= 0:
1563
        row_ct_ref[0] += cur.rowcount
1564
    
1565
    if is_function: pass # pkeys table already created
1566
    elif has_joins:
1567
        select_joins = input_joins+[sql_gen.Join(out_table, join_cols)]
1568
        log_debug('Getting output table pkeys of existing/inserted rows')
1569
        insert_into_pkeys(select_joins, pkeys_cols, distinct=True)
1570
    else:
1571
        add_row_num(db, insert_out_pkeys) # for joining with input pkeys
1572
        
1573
        log_debug('Getting input table pkeys of inserted rows')
1574
        run_query_into(db, mk_main_select(input_joins, [in_pkey]),
1575
            into=insert_in_pkeys)
1576
        add_row_num(db, insert_in_pkeys) # for joining with output pkeys
1577
        
1578
        assert table_row_count(db, insert_out_pkeys) == table_row_count(db,
1579
            insert_in_pkeys)
1580
        
1581
        log_debug('Combining output and input pkeys in inserted order')
1582
        pkey_joins = [insert_in_pkeys, sql_gen.Join(insert_out_pkeys,
1583
            {row_num_col: sql_gen.join_same_not_null})]
1584
        insert_into_pkeys(pkey_joins, pkeys_names)
1585
        
1586
        empty_temp(db, [insert_out_pkeys, insert_in_pkeys])
1587
    
1588
    db.log_debug('Adding pkey on pkeys table to enable fast joins', level=2.5)
1589
    add_pkey(db, into)
1590
    
1591
    log_debug('Setting pkeys of missing rows to '+strings.as_tt(repr(default)))
1592
    missing_rows_joins = input_joins+[sql_gen.Join(into,
1593
        {in_pkey: sql_gen.join_same_not_null}, sql_gen.filter_out)]
1594
        # must use join_same_not_null or query will take forever
1595
    insert_into_pkeys(missing_rows_joins,
1596
        [in_pkey_col, sql_gen.NamedCol(out_pkey, default)])
1597
    
1598
    assert table_row_count(db, into) == table_row_count(db, in_table)
1599
    
1600
    empty_temp(db, in_table)
1601
    
1602
    srcs = []
1603
    if is_func: srcs = sql_gen.cols_srcs(in_cols)
1604
    return sql_gen.Col(out_pkey, into, srcs)
1605

    
1606
##### Data cleanup
1607

    
1608
def cleanup_table(db, table, cols):
1609
    table = sql_gen.as_Table(table)
1610
    cols = map(sql_gen.as_Col, cols)
1611
    
1612
    expr = ('nullif(nullif(trim(both from %s), '+db.esc_value('')+'), '
1613
        +db.esc_value(r'\N')+')')
1614
    changes = [(v, sql_gen.CustomCode(expr % v.to_str(db)))
1615
        for v in cols]
1616
    
1617
    update(db, table, changes)
(24-24/36)