2 |
# Authors: |
# Authors: |
3 |
# Bernhard Herzog <[email protected]> |
# Bernhard Herzog <[email protected]> |
4 |
# Jan-Oliver Wagner <[email protected]> |
# Jan-Oliver Wagner <[email protected]> |
5 |
|
# Frank Koormann <[email protected]> |
6 |
# |
# |
7 |
# This program is free software under the GPL (>=v2) |
# This program is free software under the GPL (>=v2) |
8 |
# Read the file COPYING coming with Thuban for details. |
# Read the file COPYING coming with Thuban for details. |
13 |
|
|
14 |
__version__ = "$Revision$" |
__version__ = "$Revision$" |
15 |
|
|
16 |
|
import os |
17 |
|
import inspect |
18 |
|
import warnings |
19 |
|
|
20 |
|
from base import TitledObject |
21 |
|
|
22 |
import dbflib |
import dbflib |
23 |
|
|
24 |
# the field types supported by a Table instance. |
# the field types supported by a Table instance. |
37 |
|
|
38 |
"""Mixin to implement the old table interface using the new one""" |
"""Mixin to implement the old table interface using the new one""" |
39 |
|
|
40 |
|
def __deprecation_warning(self): |
41 |
|
"""Issue a DeprecationWarning for code hat uses the old interface""" |
42 |
|
callername = inspect.currentframe().f_back.f_code.co_name |
43 |
|
warnings.warn("The %s method of the old table interface" |
44 |
|
" is deprecated" % callername, |
45 |
|
DeprecationWarning, stacklevel = 3) |
46 |
|
|
47 |
def record_count(self): |
def record_count(self): |
48 |
|
self.__deprecation_warning() |
49 |
return self.NumRows() |
return self.NumRows() |
50 |
|
|
51 |
def field_count(self): |
def field_count(self): |
52 |
|
self.__deprecation_warning() |
53 |
return self.NumColumns() |
return self.NumColumns() |
54 |
|
|
55 |
def field_info(self, field): |
def field_info(self, field): |
60 |
and prec will be zero if the information returned by the Column |
and prec will be zero if the information returned by the Column |
61 |
method doesn't provide values for them. |
method doesn't provide values for them. |
62 |
""" |
""" |
63 |
|
self.__deprecation_warning() |
64 |
col = self.Column(field) |
col = self.Column(field) |
65 |
return (col.type, col.name, |
return (col.type, col.name, |
66 |
getattr(col, "width", 0), getattr(col, "prec", 0)) |
getattr(col, "width", 0), getattr(col, "prec", 0)) |
67 |
|
|
68 |
def field_info_by_name(self, col): |
def field_info_by_name(self, col): |
69 |
|
self.__deprecation_warning() |
70 |
try: |
try: |
71 |
return self.field_info(col) |
return self.field_info(col) |
72 |
except KeyError: |
except KeyError: |
75 |
return None |
return None |
76 |
|
|
77 |
def field_range(self, fieldName): |
def field_range(self, fieldName): |
78 |
|
self.__deprecation_warning() |
79 |
min, max = self.ValueRange(fieldName) |
min, max = self.ValueRange(fieldName) |
80 |
return ((min, None), (max, None)) |
return ((min, None), (max, None)) |
81 |
|
|
82 |
def GetUniqueValues(self, field): |
def GetUniqueValues(self, field): |
83 |
|
self.__deprecation_warning() |
84 |
return self.UniqueValues(field) |
return self.UniqueValues(field) |
85 |
|
|
86 |
def read_record(self, r): |
def read_record(self, r): |
87 |
|
self.__deprecation_warning() |
88 |
return self.ReadRowAsDict(r) |
return self.ReadRowAsDict(r) |
89 |
|
|
90 |
|
|
111 |
self.index = index |
self.index = index |
112 |
|
|
113 |
|
|
114 |
class DBFTable(OldTableInterfaceMixin): |
class DBFTable(TitledObject, OldTableInterfaceMixin): |
115 |
|
|
116 |
""" |
""" |
117 |
Table interface for the data in a DBF file |
Table interface for the data in a DBF file |
133 |
|
|
134 |
def __init__(self, filename): |
def __init__(self, filename): |
135 |
self.filename = filename |
self.filename = filename |
136 |
|
|
137 |
|
# Omit the extension in the title as it's not really needed and |
138 |
|
# it can be confusing because dbflib removes extensions and |
139 |
|
# appends some variations of '.dbf' before it tries to open the |
140 |
|
# file. So the title could be e.g. myshapefile.shp when the real |
141 |
|
# filename is myshapefile.dbf |
142 |
|
title = os.path.splitext(os.path.basename(self.filename))[0] |
143 |
|
TitledObject.__init__(self, title) |
144 |
|
|
145 |
self.dbf = dbflib.DBFFile(filename) |
self.dbf = dbflib.DBFFile(filename) |
146 |
|
|
147 |
# If true, self.dbf is open for writing. |
# If true, self.dbf is open for writing. |
182 |
""" |
""" |
183 |
return self.column_map[col] |
return self.column_map[col] |
184 |
|
|
185 |
|
def HasColumn(self, col): |
186 |
|
"""Return whether the table has a column with the given name or index |
187 |
|
""" |
188 |
|
return self.column_map.has_key(col) |
189 |
|
|
190 |
def ReadRowAsDict(self, row): |
def ReadRowAsDict(self, row): |
191 |
"""Return the entire row as a dictionary with column names as keys""" |
"""Return the entire row as a dictionary with column names as keys""" |
192 |
return self.dbf.read_record(row) |
return self.dbf.read_record(row) |
231 |
values.sort() |
values.sort() |
232 |
return values |
return values |
233 |
|
|
234 |
|
def Dependencies(self): |
235 |
|
"""Return an empty sequence. The DBFTable doesn't depend on anything""" |
236 |
|
return () |
237 |
|
|
238 |
# DBF specific interface parts. |
# DBF specific interface parts. |
239 |
|
|
240 |
|
def Width(self, col): |
241 |
|
"""Return column width""" |
242 |
|
return self.column_map[col].width |
243 |
|
|
244 |
def Destroy(self): |
def Destroy(self): |
245 |
self.dbf.close() |
self.dbf.close() |
246 |
self.dbf = None |
self.dbf = None |
266 |
self.dbf.write_record(record, values) |
self.dbf.write_record(record, values) |
267 |
self.dbf.commit() |
self.dbf.commit() |
268 |
|
|
269 |
|
def FileName(self): |
270 |
|
"""Return the filename the DBFTable was instantiated with""" |
271 |
# Temporary backwards compatibility |
return self.filename |
|
Table = DBFTable |
|
|
|
|
272 |
|
|
273 |
|
|
274 |
class MemoryColumn: |
class MemoryColumn: |
278 |
self.type = type |
self.type = type |
279 |
self.index = index |
self.index = index |
280 |
|
|
281 |
class MemoryTable(OldTableInterfaceMixin): |
class MemoryTable(TitledObject, OldTableInterfaceMixin): |
282 |
|
|
283 |
"""Very simple table implementation that operates on a list of tuples""" |
"""Very simple table implementation that operates on a list of tuples""" |
284 |
|
|
290 |
data -- List of tuples, one for each row of data |
data -- List of tuples, one for each row of data |
291 |
""" |
""" |
292 |
self.data = data |
self.data = data |
293 |
|
title = 'MemoryTable' |
294 |
|
TitledObject.__init__(self, title) |
295 |
|
|
296 |
# Create the column information objects |
# Create the column information objects |
297 |
self.columns = [] |
self.columns = [] |
322 |
""" |
""" |
323 |
return self.columns |
return self.columns |
324 |
|
|
325 |
|
def HasColumn(self, col): |
326 |
|
"""Return whether the table has a column with the given name or index |
327 |
|
""" |
328 |
|
return self.column_map.has_key(col) |
329 |
|
|
330 |
def NumRows(self): |
def NumRows(self): |
331 |
"""Return the number of rows in the table""" |
"""Return the number of rows in the table""" |
332 |
return len(self.data) |
return len(self.data) |
358 |
return min(values), max(values) |
return min(values), max(values) |
359 |
|
|
360 |
def UniqueValues(self, col): |
def UniqueValues(self, col): |
361 |
"""Return a sorted list of all unique values in the column col""" |
"""Return a sorted list of all unique values in the column col |
362 |
|
|
363 |
|
col can be either column index or name. |
364 |
|
""" |
365 |
dict = {} |
dict = {} |
366 |
|
|
367 |
for i in range(self.NumRows()): |
for i in range(self.NumRows()): |
372 |
values.sort() |
values.sort() |
373 |
return values |
return values |
374 |
|
|
375 |
|
def Width(self, col): |
376 |
|
"""Return the maximum width of values in the column |
377 |
|
|
378 |
|
The return value is the the maximum length of string |
379 |
|
representation of the values in the column (represented by index |
380 |
|
or name). |
381 |
|
""" |
382 |
|
max = 0 |
383 |
|
|
384 |
|
type = self.column_map[col].type |
385 |
|
index = self.column_map[col].index |
386 |
|
values = [row[index] for row in self.data] |
387 |
|
if not values: |
388 |
|
return None |
389 |
|
|
390 |
|
if type == FIELDTYPE_DOUBLE: |
391 |
|
format = "%.12f" |
392 |
|
elif type == FIELDTYPE_INT: |
393 |
|
format = "%d" |
394 |
|
else: |
395 |
|
format = "%s" |
396 |
|
for value in values: |
397 |
|
l = len(format % value) |
398 |
|
if l > max: |
399 |
|
max = l |
400 |
|
|
401 |
|
return max |
402 |
|
|
403 |
|
def Dependencies(self): |
404 |
|
"""Return an empty sequence. The MemoryTable doesn't depend on anything |
405 |
|
""" |
406 |
|
return () |
407 |
|
|
408 |
def write_record(self, record, values): |
def write_record(self, record, values): |
409 |
# TODO: Check for correct lenght and perhaps also |
# TODO: Check for correct lenght and perhaps also |
411 |
# TODO: Allow values to be a dictionary and write the single |
# TODO: Allow values to be a dictionary and write the single |
412 |
# fields that are specified. |
# fields that are specified. |
413 |
self.data[record] = values |
self.data[record] = values |
414 |
|
|
415 |
|
|
416 |
|
def table_to_dbf(table, filename): |
417 |
|
"""Create the dbf file filename from the table""" |
418 |
|
dbf = dbflib.create(filename) |
419 |
|
|
420 |
|
dbflib_fieldtypes = {FIELDTYPE_STRING: dbflib.FTString, |
421 |
|
FIELDTYPE_INT: dbflib.FTInteger, |
422 |
|
FIELDTYPE_DOUBLE: dbflib.FTDouble} |
423 |
|
|
424 |
|
# Initialise the header. Distinguish between DBFTable and others. |
425 |
|
for col in table.Columns(): |
426 |
|
width = table.Width(col.name) |
427 |
|
if col.type == FIELDTYPE_DOUBLE: |
428 |
|
prec = getattr(col, "prec", 12) |
429 |
|
else: |
430 |
|
prec = 0 |
431 |
|
dbf.add_field(col.name, dbflib_fieldtypes[col.type], width, prec) |
432 |
|
|
433 |
|
for i in range(table.NumRows()): |
434 |
|
record = table.ReadRowAsDict(i) |
435 |
|
dbf.write_record(i, record) |
436 |
|
dbf.close() |
437 |
|
|
438 |
|
def table_to_csv(table, filename): |
439 |
|
"""Export table to csv file.""" |
440 |
|
|
441 |
|
file = open(filename,"w") |
442 |
|
columns = table.Columns() |
443 |
|
if columns: |
444 |
|
header = "#%s" % columns[0].name |
445 |
|
for col in columns[1:]: |
446 |
|
header = header + ",%s" % col.name |
447 |
|
header = header + "\n" |
448 |
|
file.write(header) |
449 |
|
|
450 |
|
for i in range(table.NumRows()): |
451 |
|
record = table.ReadRowAsDict(i) |
452 |
|
if len(record): |
453 |
|
line = "%s" % record[columns[0].name] |
454 |
|
for col in columns[1:]: |
455 |
|
line = line + ",%s" % record[col.name] |
456 |
|
line = line + "\n" |
457 |
|
file.write(line) |
458 |
|
file.close() |
459 |
|
|