bzr branch
http://gegoxaren.bato24.eu/bzr/brz/remove-bazaar
3641.3.29
by John Arbash Meinel
Cleanup the copyright headers |
1 |
# Copyright (C) 2008 Canonical Ltd
|
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
2 |
#
|
3 |
# This program is free software; you can redistribute it and/or modify
|
|
3641.3.29
by John Arbash Meinel
Cleanup the copyright headers |
4 |
# it under the terms of the GNU General Public License as published by
|
5 |
# the Free Software Foundation; either version 2 of the License, or
|
|
6 |
# (at your option) any later version.
|
|
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
7 |
#
|
8 |
# This program is distributed in the hope that it will be useful,
|
|
9 |
# but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
10 |
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
11 |
# GNU General Public License for more details.
|
|
12 |
#
|
|
13 |
# You should have received a copy of the GNU General Public License
|
|
14 |
# along with this program; if not, write to the Free Software
|
|
3641.3.29
by John Arbash Meinel
Cleanup the copyright headers |
15 |
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
|
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
16 |
#
|
17 |
||
18 |
"""Tests for btree indices."""
|
|
19 |
||
3641.3.6
by John Arbash Meinel
Dropping the number of nodes to 100k from 200k |
20 |
import pprint |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
21 |
import zlib |
22 |
||
23 |
from bzrlib import ( |
|
24 |
btree_index, |
|
25 |
errors, |
|
26 |
tests, |
|
27 |
)
|
|
28 |
from bzrlib.tests import ( |
|
29 |
TestCaseWithTransport, |
|
30 |
TestScenarioApplier, |
|
31 |
adapt_tests, |
|
32 |
condition_isinstance, |
|
33 |
split_suite_by_condition, |
|
34 |
)
|
|
35 |
from bzrlib.transport import get_transport |
|
36 |
||
37 |
||
38 |
def load_tests(standard_tests, module, loader): |
|
39 |
# parameterise the TestBTreeNodes tests
|
|
40 |
node_tests, others = split_suite_by_condition(standard_tests, |
|
41 |
condition_isinstance(TestBTreeNodes)) |
|
42 |
applier = TestScenarioApplier() |
|
3641.3.30
by John Arbash Meinel
Rename _parse_btree to _btree_serializer |
43 |
import bzrlib._btree_serializer_py as py_module |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
44 |
applier.scenarios = [('python', {'parse_btree': py_module})] |
45 |
if CompiledBtreeParserFeature.available(): |
|
46 |
# Is there a way to do this that gets missing feature failures rather
|
|
47 |
# than no indication to the user?
|
|
3641.3.30
by John Arbash Meinel
Rename _parse_btree to _btree_serializer |
48 |
import bzrlib._btree_serializer_c as c_module |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
49 |
applier.scenarios.append(('C', {'parse_btree': c_module})) |
50 |
adapt_tests(node_tests, applier, others) |
|
51 |
return others |
|
52 |
||
53 |
||
54 |
class _CompiledBtreeParserFeature(tests.Feature): |
|
55 |
def _probe(self): |
|
56 |
try: |
|
3641.3.30
by John Arbash Meinel
Rename _parse_btree to _btree_serializer |
57 |
import bzrlib._btree_serializer_c |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
58 |
except ImportError: |
59 |
return False |
|
60 |
return True |
|
61 |
||
62 |
def feature_name(self): |
|
3641.3.30
by John Arbash Meinel
Rename _parse_btree to _btree_serializer |
63 |
return 'bzrlib._btree_serializer_c' |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
64 |
|
65 |
CompiledBtreeParserFeature = _CompiledBtreeParserFeature() |
|
66 |
||
67 |
||
68 |
class BTreeTestCase(TestCaseWithTransport): |
|
69 |
# test names here are suffixed by the key length and reference list count
|
|
70 |
# that they test.
|
|
71 |
||
72 |
def setUp(self): |
|
73 |
TestCaseWithTransport.setUp(self) |
|
74 |
self._original_header = btree_index._RESERVED_HEADER_BYTES |
|
75 |
def restore(): |
|
76 |
btree_index._RESERVED_HEADER_BYTES = self._original_header |
|
77 |
self.addCleanup(restore) |
|
78 |
btree_index._RESERVED_HEADER_BYTES = 100 |
|
79 |
||
80 |
def make_nodes(self, count, key_elements, reference_lists): |
|
81 |
"""Generate count*key_elements sample nodes.""" |
|
82 |
keys = [] |
|
83 |
for prefix_pos in range(key_elements): |
|
84 |
if key_elements - 1: |
|
85 |
prefix = (str(prefix_pos) * 40,) |
|
86 |
else: |
|
87 |
prefix = () |
|
3641.3.6
by John Arbash Meinel
Dropping the number of nodes to 100k from 200k |
88 |
for pos in xrange(count): |
89 |
# TODO: This creates odd keys. When count == 100,000, it
|
|
90 |
# creates a 240 byte key
|
|
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
91 |
key = prefix + (str(pos) * 40,) |
92 |
value = "value:%s" % pos |
|
93 |
if reference_lists: |
|
94 |
# generate some references
|
|
95 |
refs = [] |
|
96 |
for list_pos in range(reference_lists): |
|
97 |
# as many keys in each list as its index + the key depth
|
|
98 |
# mod 2 - this generates both 0 length lists and
|
|
99 |
# ones slightly longer than the number of lists.
|
|
3641.3.6
by John Arbash Meinel
Dropping the number of nodes to 100k from 200k |
100 |
# It also ensures we have non homogeneous lists.
|
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
101 |
refs.append([]) |
102 |
for ref_pos in range(list_pos + pos % 2): |
|
103 |
if pos % 2: |
|
104 |
# refer to a nearby key
|
|
105 |
refs[-1].append(prefix + ("ref" + str(pos - 1) * 40,)) |
|
106 |
else: |
|
107 |
# serial of this ref in the ref list
|
|
108 |
refs[-1].append(prefix + ("ref" + str(ref_pos) * 40,)) |
|
109 |
refs[-1] = tuple(refs[-1]) |
|
110 |
refs = tuple(refs) |
|
111 |
else: |
|
112 |
refs = () |
|
113 |
keys.append((key, value, refs)) |
|
114 |
return keys |
|
115 |
||
3641.5.9
by John Arbash Meinel
Shrink the page size so that the three_level and iter_all |
116 |
def shrink_page_size(self): |
117 |
"""Shrink the default page size so that less fits in a page.""" |
|
118 |
old_page_size = btree_index._PAGE_SIZE |
|
119 |
def cleanup(): |
|
120 |
btree_index._PAGE_SIZE = old_page_size |
|
121 |
self.addCleanup(cleanup) |
|
122 |
btree_index._PAGE_SIZE = 2048 |
|
123 |
||
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
124 |
|
125 |
class TestBTreeBuilder(BTreeTestCase): |
|
126 |
||
127 |
def test_empty_1_0(self): |
|
128 |
builder = btree_index.BTreeBuilder(key_elements=1, reference_lists=0) |
|
129 |
# NamedTemporaryFile dies on builder.finish().read(). weird.
|
|
130 |
temp_file = builder.finish() |
|
131 |
content = temp_file.read() |
|
132 |
del temp_file |
|
133 |
self.assertEqual( |
|
3641.3.3
by John Arbash Meinel
Change the header to indicate these indexes are |
134 |
"B+Tree Graph Index 2\nnode_ref_lists=0\nkey_elements=1\nlen=0\n" |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
135 |
"row_lengths=\n", |
136 |
content) |
|
137 |
||
138 |
def test_empty_2_1(self): |
|
139 |
builder = btree_index.BTreeBuilder(key_elements=2, reference_lists=1) |
|
140 |
# NamedTemporaryFile dies on builder.finish().read(). weird.
|
|
141 |
temp_file = builder.finish() |
|
142 |
content = temp_file.read() |
|
143 |
del temp_file |
|
144 |
self.assertEqual( |
|
3641.3.3
by John Arbash Meinel
Change the header to indicate these indexes are |
145 |
"B+Tree Graph Index 2\nnode_ref_lists=1\nkey_elements=2\nlen=0\n" |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
146 |
"row_lengths=\n", |
147 |
content) |
|
148 |
||
149 |
def test_root_leaf_1_0(self): |
|
150 |
builder = btree_index.BTreeBuilder(key_elements=1, reference_lists=0) |
|
151 |
nodes = self.make_nodes(5, 1, 0) |
|
152 |
for node in nodes: |
|
153 |
builder.add_node(*node) |
|
154 |
# NamedTemporaryFile dies on builder.finish().read(). weird.
|
|
155 |
temp_file = builder.finish() |
|
156 |
content = temp_file.read() |
|
157 |
del temp_file |
|
158 |
self.assertEqual(158, len(content)) |
|
159 |
self.assertEqual( |
|
3641.3.3
by John Arbash Meinel
Change the header to indicate these indexes are |
160 |
"B+Tree Graph Index 2\nnode_ref_lists=0\nkey_elements=1\nlen=5\n" |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
161 |
"row_lengths=1\n", |
162 |
content[:73]) |
|
163 |
node_content = content[73:] |
|
164 |
node_bytes = zlib.decompress(node_content) |
|
165 |
expected_node = ("type=leaf\n" |
|
166 |
"0000000000000000000000000000000000000000\x00\x00value:0\n" |
|
167 |
"1111111111111111111111111111111111111111\x00\x00value:1\n" |
|
168 |
"2222222222222222222222222222222222222222\x00\x00value:2\n" |
|
169 |
"3333333333333333333333333333333333333333\x00\x00value:3\n" |
|
170 |
"4444444444444444444444444444444444444444\x00\x00value:4\n") |
|
171 |
self.assertEqual(expected_node, node_bytes) |
|
172 |
||
173 |
def test_root_leaf_2_2(self): |
|
174 |
builder = btree_index.BTreeBuilder(key_elements=2, reference_lists=2) |
|
175 |
nodes = self.make_nodes(5, 2, 2) |
|
176 |
for node in nodes: |
|
177 |
builder.add_node(*node) |
|
178 |
# NamedTemporaryFile dies on builder.finish().read(). weird.
|
|
179 |
temp_file = builder.finish() |
|
180 |
content = temp_file.read() |
|
181 |
del temp_file |
|
182 |
self.assertEqual(264, len(content)) |
|
183 |
self.assertEqual( |
|
3641.3.3
by John Arbash Meinel
Change the header to indicate these indexes are |
184 |
"B+Tree Graph Index 2\nnode_ref_lists=2\nkey_elements=2\nlen=10\n" |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
185 |
"row_lengths=1\n", |
186 |
content[:74]) |
|
187 |
node_content = content[74:] |
|
188 |
node_bytes = zlib.decompress(node_content) |
|
189 |
expected_node = ( |
|
190 |
"type=leaf\n" |
|
191 |
"0000000000000000000000000000000000000000\x000000000000000000000000000000000000000000\x00\t0000000000000000000000000000000000000000\x00ref0000000000000000000000000000000000000000\x00value:0\n" |
|
192 |
"0000000000000000000000000000000000000000\x001111111111111111111111111111111111111111\x000000000000000000000000000000000000000000\x00ref0000000000000000000000000000000000000000\t0000000000000000000000000000000000000000\x00ref0000000000000000000000000000000000000000\r0000000000000000000000000000000000000000\x00ref0000000000000000000000000000000000000000\x00value:1\n" |
|
193 |
"0000000000000000000000000000000000000000\x002222222222222222222222222222222222222222\x00\t0000000000000000000000000000000000000000\x00ref0000000000000000000000000000000000000000\x00value:2\n" |
|
194 |
"0000000000000000000000000000000000000000\x003333333333333333333333333333333333333333\x000000000000000000000000000000000000000000\x00ref2222222222222222222222222222222222222222\t0000000000000000000000000000000000000000\x00ref2222222222222222222222222222222222222222\r0000000000000000000000000000000000000000\x00ref2222222222222222222222222222222222222222\x00value:3\n" |
|
195 |
"0000000000000000000000000000000000000000\x004444444444444444444444444444444444444444\x00\t0000000000000000000000000000000000000000\x00ref0000000000000000000000000000000000000000\x00value:4\n" |
|
196 |
"1111111111111111111111111111111111111111\x000000000000000000000000000000000000000000\x00\t1111111111111111111111111111111111111111\x00ref0000000000000000000000000000000000000000\x00value:0\n" |
|
197 |
"1111111111111111111111111111111111111111\x001111111111111111111111111111111111111111\x001111111111111111111111111111111111111111\x00ref0000000000000000000000000000000000000000\t1111111111111111111111111111111111111111\x00ref0000000000000000000000000000000000000000\r1111111111111111111111111111111111111111\x00ref0000000000000000000000000000000000000000\x00value:1\n" |
|
198 |
"1111111111111111111111111111111111111111\x002222222222222222222222222222222222222222\x00\t1111111111111111111111111111111111111111\x00ref0000000000000000000000000000000000000000\x00value:2\n" |
|
199 |
"1111111111111111111111111111111111111111\x003333333333333333333333333333333333333333\x001111111111111111111111111111111111111111\x00ref2222222222222222222222222222222222222222\t1111111111111111111111111111111111111111\x00ref2222222222222222222222222222222222222222\r1111111111111111111111111111111111111111\x00ref2222222222222222222222222222222222222222\x00value:3\n" |
|
200 |
"1111111111111111111111111111111111111111\x004444444444444444444444444444444444444444\x00\t1111111111111111111111111111111111111111\x00ref0000000000000000000000000000000000000000\x00value:4\n" |
|
201 |
""
|
|
202 |
)
|
|
203 |
self.assertEqual(expected_node, node_bytes) |
|
204 |
||
205 |
def test_2_leaves_1_0(self): |
|
206 |
builder = btree_index.BTreeBuilder(key_elements=1, reference_lists=0) |
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
207 |
nodes = self.make_nodes(400, 1, 0) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
208 |
for node in nodes: |
209 |
builder.add_node(*node) |
|
210 |
# NamedTemporaryFile dies on builder.finish().read(). weird.
|
|
211 |
temp_file = builder.finish() |
|
212 |
content = temp_file.read() |
|
213 |
del temp_file |
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
214 |
self.assertEqual(9283, len(content)) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
215 |
self.assertEqual( |
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
216 |
"B+Tree Graph Index 2\nnode_ref_lists=0\nkey_elements=1\nlen=400\n" |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
217 |
"row_lengths=1,2\n", |
218 |
content[:77]) |
|
219 |
root = content[77:4096] |
|
220 |
leaf1 = content[4096:8192] |
|
221 |
leaf2 = content[8192:] |
|
222 |
root_bytes = zlib.decompress(root) |
|
223 |
expected_root = ( |
|
224 |
"type=internal\n" |
|
225 |
"offset=0\n" |
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
226 |
) + ("307" * 40) + "\n" |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
227 |
self.assertEqual(expected_root, root_bytes) |
228 |
# We already know serialisation works for leaves, check key selection:
|
|
229 |
leaf1_bytes = zlib.decompress(leaf1) |
|
230 |
sorted_node_keys = sorted(node[0] for node in nodes) |
|
231 |
node = btree_index._LeafNode(leaf1_bytes, 1, 0) |
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
232 |
self.assertEqual(231, len(node.keys)) |
233 |
self.assertEqual(sorted_node_keys[:231], sorted(node.keys)) |
|
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
234 |
leaf2_bytes = zlib.decompress(leaf2) |
235 |
node = btree_index._LeafNode(leaf2_bytes, 1, 0) |
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
236 |
self.assertEqual(400 - 231, len(node.keys)) |
237 |
self.assertEqual(sorted_node_keys[231:], sorted(node.keys)) |
|
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
238 |
|
239 |
def test_last_page_rounded_1_layer(self): |
|
240 |
builder = btree_index.BTreeBuilder(key_elements=1, reference_lists=0) |
|
241 |
nodes = self.make_nodes(10, 1, 0) |
|
242 |
for node in nodes: |
|
243 |
builder.add_node(*node) |
|
244 |
# NamedTemporaryFile dies on builder.finish().read(). weird.
|
|
245 |
temp_file = builder.finish() |
|
246 |
content = temp_file.read() |
|
247 |
del temp_file |
|
248 |
self.assertEqual(181, len(content)) |
|
249 |
self.assertEqual( |
|
3641.3.3
by John Arbash Meinel
Change the header to indicate these indexes are |
250 |
"B+Tree Graph Index 2\nnode_ref_lists=0\nkey_elements=1\nlen=10\n" |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
251 |
"row_lengths=1\n", |
252 |
content[:74]) |
|
253 |
# Check thelast page is well formed
|
|
254 |
leaf2 = content[74:] |
|
255 |
leaf2_bytes = zlib.decompress(leaf2) |
|
256 |
node = btree_index._LeafNode(leaf2_bytes, 1, 0) |
|
257 |
self.assertEqual(10, len(node.keys)) |
|
258 |
sorted_node_keys = sorted(node[0] for node in nodes) |
|
259 |
self.assertEqual(sorted_node_keys, sorted(node.keys)) |
|
260 |
||
261 |
def test_last_page_not_rounded_2_layer(self): |
|
262 |
builder = btree_index.BTreeBuilder(key_elements=1, reference_lists=0) |
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
263 |
nodes = self.make_nodes(400, 1, 0) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
264 |
for node in nodes: |
265 |
builder.add_node(*node) |
|
266 |
# NamedTemporaryFile dies on builder.finish().read(). weird.
|
|
267 |
temp_file = builder.finish() |
|
268 |
content = temp_file.read() |
|
269 |
del temp_file |
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
270 |
self.assertEqual(9283, len(content)) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
271 |
self.assertEqual( |
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
272 |
"B+Tree Graph Index 2\nnode_ref_lists=0\nkey_elements=1\nlen=400\n" |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
273 |
"row_lengths=1,2\n", |
274 |
content[:77]) |
|
3641.3.30
by John Arbash Meinel
Rename _parse_btree to _btree_serializer |
275 |
# Check the last page is well formed
|
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
276 |
leaf2 = content[8192:] |
277 |
leaf2_bytes = zlib.decompress(leaf2) |
|
278 |
node = btree_index._LeafNode(leaf2_bytes, 1, 0) |
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
279 |
self.assertEqual(400 - 231, len(node.keys)) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
280 |
sorted_node_keys = sorted(node[0] for node in nodes) |
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
281 |
self.assertEqual(sorted_node_keys[231:], sorted(node.keys)) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
282 |
|
283 |
def test_three_level_tree_details(self): |
|
284 |
# The left most pointer in the second internal node in a row should
|
|
285 |
# pointer to the second node that the internal node is for, _not_
|
|
286 |
# the first, otherwise the first node overlaps with the last node of
|
|
287 |
# the prior internal node on that row.
|
|
3641.5.9
by John Arbash Meinel
Shrink the page size so that the three_level and iter_all |
288 |
self.shrink_page_size() |
289 |
builder = btree_index.BTreeBuilder(key_elements=2, reference_lists=2) |
|
290 |
# 40K nodes is enough to create a two internal nodes on the second
|
|
291 |
# level, with a 2K page size
|
|
292 |
nodes = self.make_nodes(20000, 2, 2) |
|
3641.3.5
by John Arbash Meinel
For iter_all and three_level tests adjust spill-at. |
293 |
|
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
294 |
for node in nodes: |
295 |
builder.add_node(*node) |
|
296 |
transport = get_transport('trace+' + self.get_url('')) |
|
3641.3.7
by John Arbash Meinel
Make it easier to profile the btree writer time |
297 |
size = transport.put_file('index', self.time(builder.finish)) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
298 |
del builder |
299 |
index = btree_index.BTreeGraphIndex(transport, 'index', size) |
|
300 |
# Seed the metadata, we're using internal calls now.
|
|
301 |
index.key_count() |
|
302 |
self.assertEqual(3, len(index._row_lengths), |
|
303 |
"Not enough rows: %r" % index._row_lengths) |
|
304 |
self.assertEqual(4, len(index._row_offsets)) |
|
305 |
self.assertEqual(sum(index._row_lengths), index._row_offsets[-1]) |
|
306 |
internal_nodes = index._get_internal_nodes([0, 1, 2]) |
|
307 |
root_node = internal_nodes[0] |
|
308 |
internal_node1 = internal_nodes[1] |
|
309 |
internal_node2 = internal_nodes[2] |
|
310 |
# The left most node node2 points at should be one after the right most
|
|
311 |
# node pointed at by node1.
|
|
312 |
self.assertEqual(internal_node2.offset, 1 + len(internal_node1.keys)) |
|
313 |
# The left most key of the second node pointed at by internal_node2
|
|
314 |
# should be its first key. We can check this by looking for its first key
|
|
315 |
# in the second node it points at
|
|
316 |
pos = index._row_offsets[2] + internal_node2.offset + 1 |
|
317 |
leaf = index._get_leaf_nodes([pos])[pos] |
|
318 |
self.assertTrue(internal_node2.keys[0] in leaf.keys) |
|
319 |
||
320 |
def test_2_leaves_2_2(self): |
|
321 |
builder = btree_index.BTreeBuilder(key_elements=2, reference_lists=2) |
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
322 |
nodes = self.make_nodes(100, 2, 2) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
323 |
for node in nodes: |
324 |
builder.add_node(*node) |
|
325 |
# NamedTemporaryFile dies on builder.finish().read(). weird.
|
|
326 |
temp_file = builder.finish() |
|
327 |
content = temp_file.read() |
|
328 |
del temp_file |
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
329 |
self.assertEqual(12643, len(content)) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
330 |
self.assertEqual( |
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
331 |
"B+Tree Graph Index 2\nnode_ref_lists=2\nkey_elements=2\nlen=200\n" |
332 |
"row_lengths=1,3\n", |
|
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
333 |
content[:77]) |
334 |
root = content[77:4096] |
|
335 |
leaf1 = content[4096:8192] |
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
336 |
leaf2 = content[8192:12288] |
337 |
leaf3 = content[12288:] |
|
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
338 |
root_bytes = zlib.decompress(root) |
339 |
expected_root = ( |
|
340 |
"type=internal\n" |
|
341 |
"offset=0\n" |
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
342 |
+ ("0" * 40) + "\x00" + ("91" * 40) + "\n" |
343 |
+ ("1" * 40) + "\x00" + ("81" * 40) + "\n" |
|
344 |
)
|
|
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
345 |
self.assertEqual(expected_root, root_bytes) |
346 |
# We assume the other leaf nodes have been written correctly - layering
|
|
347 |
# FTW.
|
|
348 |
||
349 |
def test_spill_index_stress_1_1(self): |
|
350 |
builder = btree_index.BTreeBuilder(key_elements=1, spill_at=2) |
|
351 |
nodes = [node[0:2] for node in self.make_nodes(16, 1, 0)] |
|
352 |
builder.add_node(*nodes[0]) |
|
353 |
# Test the parts of the index that take up memory are doing so
|
|
354 |
# predictably.
|
|
355 |
self.assertEqual(1, len(builder._nodes)) |
|
356 |
self.assertEqual(1, len(builder._keys)) |
|
3644.2.5
by John Arbash Meinel
Restore the exact old tests, only assert that |
357 |
self.assertIs(None, builder._nodes_by_key) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
358 |
builder.add_node(*nodes[1]) |
359 |
self.assertEqual(0, len(builder._nodes)) |
|
360 |
self.assertEqual(0, len(builder._keys)) |
|
3644.2.5
by John Arbash Meinel
Restore the exact old tests, only assert that |
361 |
self.assertIs(None, builder._nodes_by_key) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
362 |
self.assertEqual(1, len(builder._backing_indices)) |
363 |
self.assertEqual(2, builder._backing_indices[0].key_count()) |
|
364 |
# now back to memory
|
|
365 |
builder.add_node(*nodes[2]) |
|
366 |
self.assertEqual(1, len(builder._nodes)) |
|
367 |
self.assertEqual(1, len(builder._keys)) |
|
3644.2.5
by John Arbash Meinel
Restore the exact old tests, only assert that |
368 |
self.assertIs(None, builder._nodes_by_key) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
369 |
# And spills to a second backing index combing all
|
370 |
builder.add_node(*nodes[3]) |
|
371 |
self.assertEqual(0, len(builder._nodes)) |
|
372 |
self.assertEqual(0, len(builder._keys)) |
|
3644.2.5
by John Arbash Meinel
Restore the exact old tests, only assert that |
373 |
self.assertIs(None, builder._nodes_by_key) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
374 |
self.assertEqual(2, len(builder._backing_indices)) |
375 |
self.assertEqual(None, builder._backing_indices[0]) |
|
376 |
self.assertEqual(4, builder._backing_indices[1].key_count()) |
|
377 |
# The next spills to the 2-len slot
|
|
378 |
builder.add_node(*nodes[4]) |
|
379 |
builder.add_node(*nodes[5]) |
|
380 |
self.assertEqual(0, len(builder._nodes)) |
|
381 |
self.assertEqual(0, len(builder._keys)) |
|
3644.2.5
by John Arbash Meinel
Restore the exact old tests, only assert that |
382 |
self.assertIs(None, builder._nodes_by_key) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
383 |
self.assertEqual(2, len(builder._backing_indices)) |
384 |
self.assertEqual(2, builder._backing_indices[0].key_count()) |
|
385 |
self.assertEqual(4, builder._backing_indices[1].key_count()) |
|
386 |
# Next spill combines
|
|
387 |
builder.add_node(*nodes[6]) |
|
388 |
builder.add_node(*nodes[7]) |
|
389 |
self.assertEqual(3, len(builder._backing_indices)) |
|
390 |
self.assertEqual(None, builder._backing_indices[0]) |
|
391 |
self.assertEqual(None, builder._backing_indices[1]) |
|
392 |
self.assertEqual(8, builder._backing_indices[2].key_count()) |
|
393 |
# And so forth - counting up in binary.
|
|
394 |
builder.add_node(*nodes[8]) |
|
395 |
builder.add_node(*nodes[9]) |
|
396 |
self.assertEqual(3, len(builder._backing_indices)) |
|
397 |
self.assertEqual(2, builder._backing_indices[0].key_count()) |
|
398 |
self.assertEqual(None, builder._backing_indices[1]) |
|
399 |
self.assertEqual(8, builder._backing_indices[2].key_count()) |
|
400 |
builder.add_node(*nodes[10]) |
|
401 |
builder.add_node(*nodes[11]) |
|
402 |
self.assertEqual(3, len(builder._backing_indices)) |
|
403 |
self.assertEqual(None, builder._backing_indices[0]) |
|
404 |
self.assertEqual(4, builder._backing_indices[1].key_count()) |
|
405 |
self.assertEqual(8, builder._backing_indices[2].key_count()) |
|
406 |
builder.add_node(*nodes[12]) |
|
407 |
# Test that memory and disk are both used for query methods; and that
|
|
408 |
# None is skipped over happily.
|
|
409 |
self.assertEqual([(builder,) + node for node in sorted(nodes[:13])], |
|
410 |
list(builder.iter_all_entries())) |
|
411 |
# Two nodes - one memory one disk
|
|
412 |
self.assertEqual(set([(builder,) + node for node in nodes[11:13]]), |
|
413 |
set(builder.iter_entries([nodes[12][0], nodes[11][0]]))) |
|
414 |
self.assertEqual(13, builder.key_count()) |
|
415 |
self.assertEqual(set([(builder,) + node for node in nodes[11:13]]), |
|
416 |
set(builder.iter_entries_prefix([nodes[12][0], nodes[11][0]]))) |
|
417 |
builder.add_node(*nodes[13]) |
|
418 |
self.assertEqual(3, len(builder._backing_indices)) |
|
419 |
self.assertEqual(2, builder._backing_indices[0].key_count()) |
|
420 |
self.assertEqual(4, builder._backing_indices[1].key_count()) |
|
421 |
self.assertEqual(8, builder._backing_indices[2].key_count()) |
|
422 |
builder.add_node(*nodes[14]) |
|
423 |
builder.add_node(*nodes[15]) |
|
424 |
self.assertEqual(4, len(builder._backing_indices)) |
|
425 |
self.assertEqual(None, builder._backing_indices[0]) |
|
426 |
self.assertEqual(None, builder._backing_indices[1]) |
|
427 |
self.assertEqual(None, builder._backing_indices[2]) |
|
428 |
self.assertEqual(16, builder._backing_indices[3].key_count()) |
|
429 |
# Now finish, and check we got a correctly ordered tree
|
|
430 |
transport = self.get_transport('') |
|
431 |
size = transport.put_file('index', builder.finish()) |
|
432 |
index = btree_index.BTreeGraphIndex(transport, 'index', size) |
|
433 |
nodes = list(index.iter_all_entries()) |
|
434 |
self.assertEqual(sorted(nodes), nodes) |
|
435 |
self.assertEqual(16, len(nodes)) |
|
436 |
||
3777.5.3
by John Arbash Meinel
Add Builder.set_optimize(for_size=True) for GraphIndexBuilder and BTreeBuilder. |
437 |
def test_set_optimize(self): |
438 |
builder = btree_index.BTreeBuilder(key_elements=2, reference_lists=2) |
|
439 |
builder.set_optimize(for_size=True) |
|
440 |
self.assertTrue(builder._optimize_for_size) |
|
441 |
builder.set_optimize(for_size=False) |
|
442 |
self.assertFalse(builder._optimize_for_size) |
|
443 |
||
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
444 |
def test_spill_index_stress_2_2(self): |
445 |
# test that references and longer keys don't confuse things.
|
|
446 |
builder = btree_index.BTreeBuilder(key_elements=2, reference_lists=2, |
|
447 |
spill_at=2) |
|
448 |
nodes = self.make_nodes(16, 2, 2) |
|
449 |
builder.add_node(*nodes[0]) |
|
450 |
# Test the parts of the index that take up memory are doing so
|
|
451 |
# predictably.
|
|
452 |
self.assertEqual(1, len(builder._keys)) |
|
3644.2.1
by John Arbash Meinel
Change the IndexBuilders to not generate the nodes_by_key unless needed. |
453 |
self.assertEqual(1, len(builder._nodes)) |
3644.2.5
by John Arbash Meinel
Restore the exact old tests, only assert that |
454 |
self.assertIs(None, builder._nodes_by_key) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
455 |
builder.add_node(*nodes[1]) |
456 |
self.assertEqual(0, len(builder._keys)) |
|
457 |
self.assertEqual(0, len(builder._nodes)) |
|
3644.2.5
by John Arbash Meinel
Restore the exact old tests, only assert that |
458 |
self.assertIs(None, builder._nodes_by_key) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
459 |
self.assertEqual(1, len(builder._backing_indices)) |
460 |
self.assertEqual(2, builder._backing_indices[0].key_count()) |
|
461 |
# now back to memory
|
|
3644.2.6
by John Arbash Meinel
Restore the exact old tests, only assert that |
462 |
old = dict(builder._get_nodes_by_key()) #Build up the nodes by key dict |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
463 |
builder.add_node(*nodes[2]) |
3644.2.1
by John Arbash Meinel
Change the IndexBuilders to not generate the nodes_by_key unless needed. |
464 |
self.assertEqual(1, len(builder._nodes)) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
465 |
self.assertEqual(1, len(builder._keys)) |
3644.2.6
by John Arbash Meinel
Restore the exact old tests, only assert that |
466 |
self.assertIsNot(None, builder._nodes_by_key) |
467 |
self.assertNotEqual({}, builder._nodes_by_key) |
|
468 |
# We should have a new entry
|
|
469 |
self.assertNotEqual(old, builder._nodes_by_key) |
|
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
470 |
# And spills to a second backing index combing all
|
471 |
builder.add_node(*nodes[3]) |
|
472 |
self.assertEqual(0, len(builder._nodes)) |
|
473 |
self.assertEqual(0, len(builder._keys)) |
|
3644.2.5
by John Arbash Meinel
Restore the exact old tests, only assert that |
474 |
self.assertIs(None, builder._nodes_by_key) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
475 |
self.assertEqual(2, len(builder._backing_indices)) |
476 |
self.assertEqual(None, builder._backing_indices[0]) |
|
477 |
self.assertEqual(4, builder._backing_indices[1].key_count()) |
|
478 |
# The next spills to the 2-len slot
|
|
479 |
builder.add_node(*nodes[4]) |
|
480 |
builder.add_node(*nodes[5]) |
|
481 |
self.assertEqual(0, len(builder._nodes)) |
|
482 |
self.assertEqual(0, len(builder._keys)) |
|
3644.2.5
by John Arbash Meinel
Restore the exact old tests, only assert that |
483 |
self.assertIs(None, builder._nodes_by_key) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
484 |
self.assertEqual(2, len(builder._backing_indices)) |
485 |
self.assertEqual(2, builder._backing_indices[0].key_count()) |
|
486 |
self.assertEqual(4, builder._backing_indices[1].key_count()) |
|
487 |
# Next spill combines
|
|
488 |
builder.add_node(*nodes[6]) |
|
489 |
builder.add_node(*nodes[7]) |
|
490 |
self.assertEqual(3, len(builder._backing_indices)) |
|
491 |
self.assertEqual(None, builder._backing_indices[0]) |
|
492 |
self.assertEqual(None, builder._backing_indices[1]) |
|
493 |
self.assertEqual(8, builder._backing_indices[2].key_count()) |
|
494 |
# And so forth - counting up in binary.
|
|
495 |
builder.add_node(*nodes[8]) |
|
496 |
builder.add_node(*nodes[9]) |
|
497 |
self.assertEqual(3, len(builder._backing_indices)) |
|
498 |
self.assertEqual(2, builder._backing_indices[0].key_count()) |
|
499 |
self.assertEqual(None, builder._backing_indices[1]) |
|
500 |
self.assertEqual(8, builder._backing_indices[2].key_count()) |
|
501 |
builder.add_node(*nodes[10]) |
|
502 |
builder.add_node(*nodes[11]) |
|
503 |
self.assertEqual(3, len(builder._backing_indices)) |
|
504 |
self.assertEqual(None, builder._backing_indices[0]) |
|
505 |
self.assertEqual(4, builder._backing_indices[1].key_count()) |
|
506 |
self.assertEqual(8, builder._backing_indices[2].key_count()) |
|
507 |
builder.add_node(*nodes[12]) |
|
508 |
# Test that memory and disk are both used for query methods; and that
|
|
509 |
# None is skipped over happily.
|
|
510 |
self.assertEqual([(builder,) + node for node in sorted(nodes[:13])], |
|
511 |
list(builder.iter_all_entries())) |
|
512 |
# Two nodes - one memory one disk
|
|
3644.2.5
by John Arbash Meinel
Restore the exact old tests, only assert that |
513 |
self.assertEqual(set([(builder,) + node for node in nodes[11:13]]), |
514 |
set(builder.iter_entries([nodes[12][0], nodes[11][0]]))) |
|
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
515 |
self.assertEqual(13, builder.key_count()) |
3644.2.5
by John Arbash Meinel
Restore the exact old tests, only assert that |
516 |
self.assertEqual(set([(builder,) + node for node in nodes[11:13]]), |
517 |
set(builder.iter_entries_prefix([nodes[12][0], nodes[11][0]]))) |
|
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
518 |
builder.add_node(*nodes[13]) |
519 |
self.assertEqual(3, len(builder._backing_indices)) |
|
520 |
self.assertEqual(2, builder._backing_indices[0].key_count()) |
|
521 |
self.assertEqual(4, builder._backing_indices[1].key_count()) |
|
522 |
self.assertEqual(8, builder._backing_indices[2].key_count()) |
|
523 |
builder.add_node(*nodes[14]) |
|
524 |
builder.add_node(*nodes[15]) |
|
525 |
self.assertEqual(4, len(builder._backing_indices)) |
|
526 |
self.assertEqual(None, builder._backing_indices[0]) |
|
527 |
self.assertEqual(None, builder._backing_indices[1]) |
|
528 |
self.assertEqual(None, builder._backing_indices[2]) |
|
529 |
self.assertEqual(16, builder._backing_indices[3].key_count()) |
|
530 |
# Now finish, and check we got a correctly ordered tree
|
|
531 |
transport = self.get_transport('') |
|
532 |
size = transport.put_file('index', builder.finish()) |
|
533 |
index = btree_index.BTreeGraphIndex(transport, 'index', size) |
|
534 |
nodes = list(index.iter_all_entries()) |
|
535 |
self.assertEqual(sorted(nodes), nodes) |
|
536 |
self.assertEqual(16, len(nodes)) |
|
537 |
||
538 |
def test_spill_index_duplicate_key_caught_on_finish(self): |
|
539 |
builder = btree_index.BTreeBuilder(key_elements=1, spill_at=2) |
|
540 |
nodes = [node[0:2] for node in self.make_nodes(16, 1, 0)] |
|
541 |
builder.add_node(*nodes[0]) |
|
542 |
builder.add_node(*nodes[1]) |
|
543 |
builder.add_node(*nodes[0]) |
|
544 |
self.assertRaises(errors.BadIndexDuplicateKey, builder.finish) |
|
545 |
||
546 |
||
547 |
class TestBTreeIndex(BTreeTestCase): |
|
548 |
||
549 |
def make_index(self, ref_lists=0, key_elements=1, nodes=[]): |
|
550 |
builder = btree_index.BTreeBuilder(reference_lists=ref_lists, |
|
551 |
key_elements=key_elements) |
|
552 |
for key, value, references in nodes: |
|
553 |
builder.add_node(key, value, references) |
|
554 |
stream = builder.finish() |
|
555 |
trans = get_transport('trace+' + self.get_url()) |
|
556 |
size = trans.put_file('index', stream) |
|
557 |
return btree_index.BTreeGraphIndex(trans, 'index', size) |
|
558 |
||
559 |
def test_trivial_constructor(self): |
|
560 |
transport = get_transport('trace+' + self.get_url('')) |
|
561 |
index = btree_index.BTreeGraphIndex(transport, 'index', None) |
|
562 |
# Checks the page size at load, but that isn't logged yet.
|
|
563 |
self.assertEqual([], transport._activity) |
|
564 |
||
565 |
def test_with_size_constructor(self): |
|
566 |
transport = get_transport('trace+' + self.get_url('')) |
|
567 |
index = btree_index.BTreeGraphIndex(transport, 'index', 1) |
|
568 |
# Checks the page size at load, but that isn't logged yet.
|
|
569 |
self.assertEqual([], transport._activity) |
|
570 |
||
571 |
def test_empty_key_count_no_size(self): |
|
572 |
builder = btree_index.BTreeBuilder(key_elements=1, reference_lists=0) |
|
573 |
transport = get_transport('trace+' + self.get_url('')) |
|
574 |
transport.put_file('index', builder.finish()) |
|
575 |
index = btree_index.BTreeGraphIndex(transport, 'index', None) |
|
576 |
del transport._activity[:] |
|
577 |
self.assertEqual([], transport._activity) |
|
578 |
self.assertEqual(0, index.key_count()) |
|
579 |
# The entire index should have been requested (as we generally have the
|
|
580 |
# size available, and doing many small readvs is inappropriate).
|
|
581 |
# We can't tell how much was actually read here, but - check the code.
|
|
3824.1.2
by John Arbash Meinel
iter_all_entries() shouldn't need to re-read the page. |
582 |
self.assertEqual([('get', 'index')], transport._activity) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
583 |
|
584 |
def test_empty_key_count(self): |
|
585 |
builder = btree_index.BTreeBuilder(key_elements=1, reference_lists=0) |
|
586 |
transport = get_transport('trace+' + self.get_url('')) |
|
587 |
size = transport.put_file('index', builder.finish()) |
|
588 |
self.assertEqual(72, size) |
|
589 |
index = btree_index.BTreeGraphIndex(transport, 'index', size) |
|
590 |
del transport._activity[:] |
|
591 |
self.assertEqual([], transport._activity) |
|
592 |
self.assertEqual(0, index.key_count()) |
|
593 |
# The entire index should have been read, as 4K > size
|
|
594 |
self.assertEqual([('readv', 'index', [(0, 72)], False, None)], |
|
595 |
transport._activity) |
|
596 |
||
597 |
def test_non_empty_key_count_2_2(self): |
|
598 |
builder = btree_index.BTreeBuilder(key_elements=2, reference_lists=2) |
|
599 |
nodes = self.make_nodes(35, 2, 2) |
|
600 |
for node in nodes: |
|
601 |
builder.add_node(*node) |
|
602 |
transport = get_transport('trace+' + self.get_url('')) |
|
603 |
size = transport.put_file('index', builder.finish()) |
|
604 |
index = btree_index.BTreeGraphIndex(transport, 'index', size) |
|
605 |
del transport._activity[:] |
|
606 |
self.assertEqual([], transport._activity) |
|
607 |
self.assertEqual(70, index.key_count()) |
|
608 |
# The entire index should have been read, as it is one page long.
|
|
609 |
self.assertEqual([('readv', 'index', [(0, size)], False, None)], |
|
610 |
transport._activity) |
|
3641.5.8
by John Arbash Meinel
Fix up the test suite, now that we pack more efficiently. |
611 |
self.assertEqual(1199, size) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
612 |
|
3824.1.1
by John Arbash Meinel
Fix _read_nodes() to only issue a single read if there is no known size. |
613 |
def test__read_nodes_no_size_one_page_reads_once(self): |
614 |
self.make_index(nodes=[(('key',), 'value', ())]) |
|
615 |
trans = get_transport('trace+' + self.get_url()) |
|
616 |
index = btree_index.BTreeGraphIndex(trans, 'index', None) |
|
617 |
del trans._activity[:] |
|
618 |
nodes = dict(index._read_nodes([0])) |
|
619 |
self.assertEqual([0], nodes.keys()) |
|
620 |
node = nodes[0] |
|
621 |
self.assertEqual([('key',)], node.keys.keys()) |
|
622 |
self.assertEqual([('get', 'index')], trans._activity) |
|
623 |
||
624 |
def test__read_nodes_no_size_multiple_pages(self): |
|
625 |
index = self.make_index(2, 2, nodes=self.make_nodes(160, 2, 2)) |
|
626 |
index.key_count() |
|
627 |
num_pages = index._row_offsets[-1] |
|
628 |
# Reopen with a traced transport and no size
|
|
629 |
trans = get_transport('trace+' + self.get_url()) |
|
630 |
index = btree_index.BTreeGraphIndex(trans, 'index', None) |
|
631 |
del trans._activity[:] |
|
632 |
nodes = dict(index._read_nodes([0])) |
|
633 |
self.assertEqual(range(num_pages), nodes.keys()) |
|
634 |
||
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
635 |
def test_2_levels_key_count_2_2(self): |
636 |
builder = btree_index.BTreeBuilder(key_elements=2, reference_lists=2) |
|
637 |
nodes = self.make_nodes(160, 2, 2) |
|
638 |
for node in nodes: |
|
639 |
builder.add_node(*node) |
|
640 |
transport = get_transport('trace+' + self.get_url('')) |
|
641 |
size = transport.put_file('index', builder.finish()) |
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
642 |
self.assertEqual(17692, size) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
643 |
index = btree_index.BTreeGraphIndex(transport, 'index', size) |
644 |
del transport._activity[:] |
|
645 |
self.assertEqual([], transport._activity) |
|
646 |
self.assertEqual(320, index.key_count()) |
|
647 |
# The entire index should not have been read.
|
|
648 |
self.assertEqual([('readv', 'index', [(0, 4096)], False, None)], |
|
649 |
transport._activity) |
|
650 |
||
651 |
def test_validate_one_page(self): |
|
652 |
builder = btree_index.BTreeBuilder(key_elements=2, reference_lists=2) |
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
653 |
nodes = self.make_nodes(45, 2, 2) |
654 |
for node in nodes: |
|
655 |
builder.add_node(*node) |
|
656 |
transport = get_transport('trace+' + self.get_url('')) |
|
657 |
size = transport.put_file('index', builder.finish()) |
|
658 |
index = btree_index.BTreeGraphIndex(transport, 'index', size) |
|
659 |
del transport._activity[:] |
|
660 |
self.assertEqual([], transport._activity) |
|
661 |
index.validate() |
|
662 |
# The entire index should have been read linearly.
|
|
663 |
self.assertEqual([('readv', 'index', [(0, size)], False, None)], |
|
664 |
transport._activity) |
|
665 |
self.assertEqual(1514, size) |
|
666 |
||
667 |
def test_validate_two_pages(self): |
|
668 |
builder = btree_index.BTreeBuilder(key_elements=2, reference_lists=2) |
|
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
669 |
nodes = self.make_nodes(80, 2, 2) |
670 |
for node in nodes: |
|
671 |
builder.add_node(*node) |
|
672 |
transport = get_transport('trace+' + self.get_url('')) |
|
673 |
size = transport.put_file('index', builder.finish()) |
|
674 |
# Root page, 2 leaf pages
|
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
675 |
self.assertEqual(9339, size) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
676 |
index = btree_index.BTreeGraphIndex(transport, 'index', size) |
677 |
del transport._activity[:] |
|
678 |
self.assertEqual([], transport._activity) |
|
679 |
index.validate() |
|
680 |
# The entire index should have been read linearly.
|
|
681 |
self.assertEqual([('readv', 'index', [(0, 4096)], False, None), |
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
682 |
('readv', 'index', [(4096, 4096), (8192, 1147)], False, None)], |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
683 |
transport._activity) |
684 |
# XXX: TODO: write some badly-ordered nodes, and some pointers-to-wrong
|
|
685 |
# node and make validate find them.
|
|
686 |
||
687 |
def test_eq_ne(self): |
|
688 |
# two indices are equal when constructed with the same parameters:
|
|
689 |
transport1 = get_transport('trace+' + self.get_url('')) |
|
690 |
transport2 = get_transport(self.get_url('')) |
|
691 |
self.assertTrue( |
|
692 |
btree_index.BTreeGraphIndex(transport1, 'index', None) == |
|
693 |
btree_index.BTreeGraphIndex(transport1, 'index', None)) |
|
694 |
self.assertTrue( |
|
695 |
btree_index.BTreeGraphIndex(transport1, 'index', 20) == |
|
696 |
btree_index.BTreeGraphIndex(transport1, 'index', 20)) |
|
697 |
self.assertFalse( |
|
698 |
btree_index.BTreeGraphIndex(transport1, 'index', 20) == |
|
699 |
btree_index.BTreeGraphIndex(transport2, 'index', 20)) |
|
700 |
self.assertFalse( |
|
701 |
btree_index.BTreeGraphIndex(transport1, 'inde1', 20) == |
|
702 |
btree_index.BTreeGraphIndex(transport1, 'inde2', 20)) |
|
703 |
self.assertFalse( |
|
704 |
btree_index.BTreeGraphIndex(transport1, 'index', 10) == |
|
705 |
btree_index.BTreeGraphIndex(transport1, 'index', 20)) |
|
706 |
self.assertFalse( |
|
707 |
btree_index.BTreeGraphIndex(transport1, 'index', None) != |
|
708 |
btree_index.BTreeGraphIndex(transport1, 'index', None)) |
|
709 |
self.assertFalse( |
|
710 |
btree_index.BTreeGraphIndex(transport1, 'index', 20) != |
|
711 |
btree_index.BTreeGraphIndex(transport1, 'index', 20)) |
|
712 |
self.assertTrue( |
|
713 |
btree_index.BTreeGraphIndex(transport1, 'index', 20) != |
|
714 |
btree_index.BTreeGraphIndex(transport2, 'index', 20)) |
|
715 |
self.assertTrue( |
|
716 |
btree_index.BTreeGraphIndex(transport1, 'inde1', 20) != |
|
717 |
btree_index.BTreeGraphIndex(transport1, 'inde2', 20)) |
|
718 |
self.assertTrue( |
|
719 |
btree_index.BTreeGraphIndex(transport1, 'index', 10) != |
|
720 |
btree_index.BTreeGraphIndex(transport1, 'index', 20)) |
|
721 |
||
3824.1.2
by John Arbash Meinel
iter_all_entries() shouldn't need to re-read the page. |
722 |
def test_iter_all_only_root_no_size(self): |
723 |
self.make_index(nodes=[(('key',), 'value', ())]) |
|
724 |
trans = get_transport('trace+' + self.get_url('')) |
|
725 |
index = btree_index.BTreeGraphIndex(trans, 'index', None) |
|
726 |
del trans._activity[:] |
|
727 |
self.assertEqual([(('key',), 'value')], |
|
728 |
[x[1:] for x in index.iter_all_entries()]) |
|
729 |
self.assertEqual([('get', 'index')], trans._activity) |
|
730 |
||
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
731 |
def test_iter_all_entries_reads(self): |
732 |
# iterating all entries reads the header, then does a linear
|
|
733 |
# read.
|
|
3641.5.9
by John Arbash Meinel
Shrink the page size so that the three_level and iter_all |
734 |
self.shrink_page_size() |
3641.5.11
by John Arbash Meinel
Some small test cleanup |
735 |
builder = btree_index.BTreeBuilder(key_elements=2, reference_lists=2) |
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
736 |
# 20k nodes is enough to create a two internal nodes on the second
|
3641.5.9
by John Arbash Meinel
Shrink the page size so that the three_level and iter_all |
737 |
# level, with a 2K page size
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
738 |
nodes = self.make_nodes(10000, 2, 2) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
739 |
for node in nodes: |
740 |
builder.add_node(*node) |
|
741 |
transport = get_transport('trace+' + self.get_url('')) |
|
742 |
size = transport.put_file('index', builder.finish()) |
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
743 |
self.assertEqual(1303220, size, 'number of expected bytes in the' |
744 |
' output changed') |
|
3641.5.9
by John Arbash Meinel
Shrink the page size so that the three_level and iter_all |
745 |
page_size = btree_index._PAGE_SIZE |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
746 |
del builder |
747 |
index = btree_index.BTreeGraphIndex(transport, 'index', size) |
|
748 |
del transport._activity[:] |
|
749 |
self.assertEqual([], transport._activity) |
|
3641.5.8
by John Arbash Meinel
Fix up the test suite, now that we pack more efficiently. |
750 |
found_nodes = self.time(list, index.iter_all_entries()) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
751 |
bare_nodes = [] |
752 |
for node in found_nodes: |
|
753 |
self.assertTrue(node[0] is index) |
|
754 |
bare_nodes.append(node[1:]) |
|
755 |
self.assertEqual(3, len(index._row_lengths), |
|
756 |
"Not enough rows: %r" % index._row_lengths) |
|
757 |
# Should be as long as the nodes we supplied
|
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
758 |
self.assertEqual(20000, len(found_nodes)) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
759 |
# Should have the same content
|
760 |
self.assertEqual(set(nodes), set(bare_nodes)) |
|
761 |
# Should have done linear scan IO up the index, ignoring
|
|
762 |
# the internal nodes:
|
|
763 |
# The entire index should have been read
|
|
764 |
total_pages = sum(index._row_lengths) |
|
765 |
self.assertEqual(total_pages, index._row_offsets[-1]) |
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
766 |
self.assertEqual(1303220, size) |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
767 |
# The start of the leaves
|
3641.5.9
by John Arbash Meinel
Shrink the page size so that the three_level and iter_all |
768 |
first_byte = index._row_offsets[-2] * page_size |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
769 |
readv_request = [] |
3641.5.9
by John Arbash Meinel
Shrink the page size so that the three_level and iter_all |
770 |
for offset in range(first_byte, size, page_size): |
771 |
readv_request.append((offset, page_size)) |
|
3641.3.6
by John Arbash Meinel
Dropping the number of nodes to 100k from 200k |
772 |
# The last page is truncated
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
773 |
readv_request[-1] = (readv_request[-1][0], 1303220 % page_size) |
3641.5.9
by John Arbash Meinel
Shrink the page size so that the three_level and iter_all |
774 |
expected = [('readv', 'index', [(0, page_size)], False, None), |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
775 |
('readv', 'index', readv_request, False, None)] |
776 |
if expected != transport._activity: |
|
777 |
self.assertEqualDiff(pprint.pformat(expected), |
|
778 |
pprint.pformat(transport._activity)) |
|
779 |
||
780 |
def _test_iter_entries_references_resolved(self): |
|
781 |
index = self.make_index(1, nodes=[ |
|
782 |
(('name', ), 'data', ([('ref', ), ('ref', )], )), |
|
783 |
(('ref', ), 'refdata', ([], ))]) |
|
784 |
self.assertEqual(set([(index, ('name', ), 'data', ((('ref',),('ref',)),)), |
|
785 |
(index, ('ref', ), 'refdata', ((), ))]), |
|
786 |
set(index.iter_entries([('name',), ('ref',)]))) |
|
787 |
||
788 |
def test_iter_entries_references_2_refs_resolved(self): |
|
789 |
# iterating some entries reads just the pages needed. For now, to
|
|
790 |
# get it working and start measuring, only 4K pages are read.
|
|
791 |
builder = btree_index.BTreeBuilder(key_elements=2, reference_lists=2) |
|
792 |
# 80 nodes is enough to create a two-level index.
|
|
793 |
nodes = self.make_nodes(160, 2, 2) |
|
794 |
for node in nodes: |
|
795 |
builder.add_node(*node) |
|
796 |
transport = get_transport('trace+' + self.get_url('')) |
|
797 |
size = transport.put_file('index', builder.finish()) |
|
798 |
del builder |
|
799 |
index = btree_index.BTreeGraphIndex(transport, 'index', size) |
|
800 |
del transport._activity[:] |
|
801 |
self.assertEqual([], transport._activity) |
|
802 |
# search for one key
|
|
803 |
found_nodes = list(index.iter_entries([nodes[30][0]])) |
|
804 |
bare_nodes = [] |
|
805 |
for node in found_nodes: |
|
806 |
self.assertTrue(node[0] is index) |
|
807 |
bare_nodes.append(node[1:]) |
|
808 |
# Should be as long as the nodes we supplied
|
|
809 |
self.assertEqual(1, len(found_nodes)) |
|
810 |
# Should have the same content
|
|
811 |
self.assertEqual(nodes[30], bare_nodes[0]) |
|
812 |
# Should have read the root node, then one leaf page:
|
|
813 |
self.assertEqual([('readv', 'index', [(0, 4096)], False, None), |
|
3641.5.17
by John Arbash Meinel
Fix up the test suite now that things don't pack as well. |
814 |
('readv', 'index', [(8192, 4096), ], False, None)], |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
815 |
transport._activity) |
816 |
||
817 |
def test_iter_key_prefix_1_element_key_None(self): |
|
818 |
index = self.make_index() |
|
819 |
self.assertRaises(errors.BadIndexKey, list, |
|
820 |
index.iter_entries_prefix([(None, )])) |
|
821 |
||
822 |
def test_iter_key_prefix_wrong_length(self): |
|
823 |
index = self.make_index() |
|
824 |
self.assertRaises(errors.BadIndexKey, list, |
|
825 |
index.iter_entries_prefix([('foo', None)])) |
|
826 |
index = self.make_index(key_elements=2) |
|
827 |
self.assertRaises(errors.BadIndexKey, list, |
|
828 |
index.iter_entries_prefix([('foo', )])) |
|
829 |
self.assertRaises(errors.BadIndexKey, list, |
|
830 |
index.iter_entries_prefix([('foo', None, None)])) |
|
831 |
||
832 |
def test_iter_key_prefix_1_key_element_no_refs(self): |
|
833 |
index = self.make_index( nodes=[ |
|
834 |
(('name', ), 'data', ()), |
|
835 |
(('ref', ), 'refdata', ())]) |
|
836 |
self.assertEqual(set([(index, ('name', ), 'data'), |
|
837 |
(index, ('ref', ), 'refdata')]), |
|
838 |
set(index.iter_entries_prefix([('name', ), ('ref', )]))) |
|
839 |
||
840 |
def test_iter_key_prefix_1_key_element_refs(self): |
|
841 |
index = self.make_index(1, nodes=[ |
|
842 |
(('name', ), 'data', ([('ref', )], )), |
|
843 |
(('ref', ), 'refdata', ([], ))]) |
|
844 |
self.assertEqual(set([(index, ('name', ), 'data', ((('ref',),),)), |
|
845 |
(index, ('ref', ), 'refdata', ((), ))]), |
|
846 |
set(index.iter_entries_prefix([('name', ), ('ref', )]))) |
|
847 |
||
848 |
def test_iter_key_prefix_2_key_element_no_refs(self): |
|
849 |
index = self.make_index(key_elements=2, nodes=[ |
|
850 |
(('name', 'fin1'), 'data', ()), |
|
851 |
(('name', 'fin2'), 'beta', ()), |
|
852 |
(('ref', 'erence'), 'refdata', ())]) |
|
853 |
self.assertEqual(set([(index, ('name', 'fin1'), 'data'), |
|
854 |
(index, ('ref', 'erence'), 'refdata')]), |
|
855 |
set(index.iter_entries_prefix([('name', 'fin1'), ('ref', 'erence')]))) |
|
856 |
self.assertEqual(set([(index, ('name', 'fin1'), 'data'), |
|
857 |
(index, ('name', 'fin2'), 'beta')]), |
|
858 |
set(index.iter_entries_prefix([('name', None)]))) |
|
859 |
||
860 |
def test_iter_key_prefix_2_key_element_refs(self): |
|
861 |
index = self.make_index(1, key_elements=2, nodes=[ |
|
862 |
(('name', 'fin1'), 'data', ([('ref', 'erence')], )), |
|
863 |
(('name', 'fin2'), 'beta', ([], )), |
|
864 |
(('ref', 'erence'), 'refdata', ([], ))]) |
|
865 |
self.assertEqual(set([(index, ('name', 'fin1'), 'data', ((('ref', 'erence'),),)), |
|
866 |
(index, ('ref', 'erence'), 'refdata', ((), ))]), |
|
867 |
set(index.iter_entries_prefix([('name', 'fin1'), ('ref', 'erence')]))) |
|
868 |
self.assertEqual(set([(index, ('name', 'fin1'), 'data', ((('ref', 'erence'),),)), |
|
869 |
(index, ('name', 'fin2'), 'beta', ((), ))]), |
|
870 |
set(index.iter_entries_prefix([('name', None)]))) |
|
871 |
||
872 |
||
873 |
class TestBTreeNodes(BTreeTestCase): |
|
874 |
||
875 |
def restore_parser(self): |
|
3641.3.30
by John Arbash Meinel
Rename _parse_btree to _btree_serializer |
876 |
btree_index._btree_serializer = self.saved_parser |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
877 |
|
878 |
def setUp(self): |
|
879 |
BTreeTestCase.setUp(self) |
|
3641.3.30
by John Arbash Meinel
Rename _parse_btree to _btree_serializer |
880 |
self.saved_parser = btree_index._btree_serializer |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
881 |
self.addCleanup(self.restore_parser) |
3641.3.30
by John Arbash Meinel
Rename _parse_btree to _btree_serializer |
882 |
btree_index._btree_serializer = self.parse_btree |
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
883 |
|
884 |
def test_LeafNode_1_0(self): |
|
885 |
node_bytes = ("type=leaf\n" |
|
886 |
"0000000000000000000000000000000000000000\x00\x00value:0\n" |
|
887 |
"1111111111111111111111111111111111111111\x00\x00value:1\n" |
|
888 |
"2222222222222222222222222222222222222222\x00\x00value:2\n" |
|
889 |
"3333333333333333333333333333333333333333\x00\x00value:3\n" |
|
890 |
"4444444444444444444444444444444444444444\x00\x00value:4\n") |
|
891 |
node = btree_index._LeafNode(node_bytes, 1, 0) |
|
892 |
# We do direct access, or don't care about order, to leaf nodes most of
|
|
893 |
# the time, so a dict is useful:
|
|
894 |
self.assertEqual({ |
|
895 |
("0000000000000000000000000000000000000000",): ("value:0", ()), |
|
896 |
("1111111111111111111111111111111111111111",): ("value:1", ()), |
|
897 |
("2222222222222222222222222222222222222222",): ("value:2", ()), |
|
898 |
("3333333333333333333333333333333333333333",): ("value:3", ()), |
|
899 |
("4444444444444444444444444444444444444444",): ("value:4", ()), |
|
900 |
}, node.keys) |
|
901 |
||
902 |
def test_LeafNode_2_2(self): |
|
903 |
node_bytes = ("type=leaf\n" |
|
904 |
"00\x0000\x00\t00\x00ref00\x00value:0\n" |
|
905 |
"00\x0011\x0000\x00ref00\t00\x00ref00\r01\x00ref01\x00value:1\n" |
|
906 |
"11\x0033\x0011\x00ref22\t11\x00ref22\r11\x00ref22\x00value:3\n" |
|
907 |
"11\x0044\x00\t11\x00ref00\x00value:4\n" |
|
908 |
""
|
|
909 |
)
|
|
910 |
node = btree_index._LeafNode(node_bytes, 2, 2) |
|
911 |
# We do direct access, or don't care about order, to leaf nodes most of
|
|
912 |
# the time, so a dict is useful:
|
|
913 |
self.assertEqual({ |
|
914 |
('00', '00'): ('value:0', ((), (('00', 'ref00'),))), |
|
915 |
('00', '11'): ('value:1', |
|
916 |
((('00', 'ref00'),), (('00', 'ref00'), ('01', 'ref01')))), |
|
917 |
('11', '33'): ('value:3', |
|
918 |
((('11', 'ref22'),), (('11', 'ref22'), ('11', 'ref22')))), |
|
919 |
('11', '44'): ('value:4', ((), (('11', 'ref00'),))) |
|
920 |
}, node.keys) |
|
921 |
||
922 |
def test_InternalNode_1(self): |
|
923 |
node_bytes = ("type=internal\n" |
|
924 |
"offset=1\n" |
|
925 |
"0000000000000000000000000000000000000000\n" |
|
926 |
"1111111111111111111111111111111111111111\n" |
|
927 |
"2222222222222222222222222222222222222222\n" |
|
928 |
"3333333333333333333333333333333333333333\n" |
|
929 |
"4444444444444444444444444444444444444444\n" |
|
930 |
)
|
|
931 |
node = btree_index._InternalNode(node_bytes) |
|
932 |
# We want to bisect to find the right children from this node, so a
|
|
933 |
# vector is most useful.
|
|
934 |
self.assertEqual([ |
|
935 |
("0000000000000000000000000000000000000000",), |
|
936 |
("1111111111111111111111111111111111111111",), |
|
937 |
("2222222222222222222222222222222222222222",), |
|
938 |
("3333333333333333333333333333333333333333",), |
|
939 |
("4444444444444444444444444444444444444444",), |
|
940 |
], node.keys) |
|
941 |
self.assertEqual(1, node.offset) |
|
942 |
||
943 |
def test_LeafNode_2_2(self): |
|
944 |
node_bytes = ("type=leaf\n" |
|
945 |
"00\x0000\x00\t00\x00ref00\x00value:0\n" |
|
946 |
"00\x0011\x0000\x00ref00\t00\x00ref00\r01\x00ref01\x00value:1\n" |
|
947 |
"11\x0033\x0011\x00ref22\t11\x00ref22\r11\x00ref22\x00value:3\n" |
|
948 |
"11\x0044\x00\t11\x00ref00\x00value:4\n" |
|
949 |
""
|
|
950 |
)
|
|
951 |
node = btree_index._LeafNode(node_bytes, 2, 2) |
|
952 |
# We do direct access, or don't care about order, to leaf nodes most of
|
|
953 |
# the time, so a dict is useful:
|
|
954 |
self.assertEqual({ |
|
955 |
('00', '00'): ('value:0', ((), (('00', 'ref00'),))), |
|
956 |
('00', '11'): ('value:1', |
|
957 |
((('00', 'ref00'),), (('00', 'ref00'), ('01', 'ref01')))), |
|
958 |
('11', '33'): ('value:3', |
|
959 |
((('11', 'ref22'),), (('11', 'ref22'), ('11', 'ref22')))), |
|
960 |
('11', '44'): ('value:4', ((), (('11', 'ref00'),))) |
|
961 |
}, node.keys) |
|
962 |
||
3641.3.19
by John Arbash Meinel
The flatten code now handles the no-ref-list case. |
963 |
def assertFlattened(self, expected, key, value, refs): |
3641.3.18
by John Arbash Meinel
Start working on a compiled function for transforming |
964 |
flat_key, flat_line = self.parse_btree._flatten_node( |
3641.3.19
by John Arbash Meinel
The flatten code now handles the no-ref-list case. |
965 |
(None, key, value, refs), bool(refs)) |
3641.3.18
by John Arbash Meinel
Start working on a compiled function for transforming |
966 |
self.assertEqual('\x00'.join(key), flat_key) |
967 |
self.assertEqual(expected, flat_line) |
|
968 |
||
3641.3.19
by John Arbash Meinel
The flatten code now handles the no-ref-list case. |
969 |
def test__flatten_node(self): |
970 |
self.assertFlattened('key\0\0value\n', ('key',), 'value', []) |
|
971 |
self.assertFlattened('key\0tuple\0\0value str\n', |
|
972 |
('key', 'tuple'), 'value str', []) |
|
973 |
self.assertFlattened('key\0tuple\0triple\0\0value str\n', |
|
974 |
('key', 'tuple', 'triple'), 'value str', []) |
|
975 |
self.assertFlattened('k\0t\0s\0ref\0value str\n', |
|
976 |
('k', 't', 's'), 'value str', [[('ref',)]]) |
|
977 |
self.assertFlattened('key\0tuple\0ref\0key\0value str\n', |
|
978 |
('key', 'tuple'), 'value str', [[('ref', 'key')]]) |
|
979 |
self.assertFlattened("00\x0000\x00\t00\x00ref00\x00value:0\n", |
|
980 |
('00', '00'), 'value:0', ((), (('00', 'ref00'),))) |
|
981 |
self.assertFlattened( |
|
982 |
"00\x0011\x0000\x00ref00\t00\x00ref00\r01\x00ref01\x00value:1\n", |
|
983 |
('00', '11'), 'value:1', |
|
984 |
((('00', 'ref00'),), (('00', 'ref00'), ('01', 'ref01')))) |
|
985 |
self.assertFlattened( |
|
986 |
"11\x0033\x0011\x00ref22\t11\x00ref22\r11\x00ref22\x00value:3\n", |
|
987 |
('11', '33'), 'value:3', |
|
988 |
((('11', 'ref22'),), (('11', 'ref22'), ('11', 'ref22')))) |
|
989 |
self.assertFlattened( |
|
990 |
"11\x0044\x00\t11\x00ref00\x00value:4\n", |
|
991 |
('11', '44'), 'value:4', ((), (('11', 'ref00'),))) |
|
3641.3.18
by John Arbash Meinel
Start working on a compiled function for transforming |
992 |
|
3641.3.1
by John Arbash Meinel
Bring in the btree_index and chunk_writer code and their tests. |
993 |
|
994 |
class TestCompiledBtree(tests.TestCase): |
|
995 |
||
996 |
def test_exists(self): |
|
997 |
# This is just to let the user know if they don't have the feature
|
|
998 |
# available
|
|
999 |
self.requireFeature(CompiledBtreeParserFeature) |
|
1000 |
||
1001 |
||
1002 |
class TestMultiBisectRight(tests.TestCase): |
|
1003 |
||
1004 |
def assertMultiBisectRight(self, offsets, search_keys, fixed_keys): |
|
1005 |
self.assertEqual(offsets, |
|
1006 |
btree_index.BTreeGraphIndex._multi_bisect_right( |
|
1007 |
search_keys, fixed_keys)) |
|
1008 |
||
1009 |
def test_after(self): |
|
1010 |
self.assertMultiBisectRight([(1, ['b'])], ['b'], ['a']) |
|
1011 |
self.assertMultiBisectRight([(3, ['e', 'f', 'g'])], |
|
1012 |
['e', 'f', 'g'], ['a', 'b', 'c']) |
|
1013 |
||
1014 |
def test_before(self): |
|
1015 |
self.assertMultiBisectRight([(0, ['a'])], ['a'], ['b']) |
|
1016 |
self.assertMultiBisectRight([(0, ['a', 'b', 'c', 'd'])], |
|
1017 |
['a', 'b', 'c', 'd'], ['e', 'f', 'g']) |
|
1018 |
||
1019 |
def test_exact(self): |
|
1020 |
self.assertMultiBisectRight([(1, ['a'])], ['a'], ['a']) |
|
1021 |
self.assertMultiBisectRight([(1, ['a']), (2, ['b'])], ['a', 'b'], ['a', 'b']) |
|
1022 |
self.assertMultiBisectRight([(1, ['a']), (3, ['c'])], |
|
1023 |
['a', 'c'], ['a', 'b', 'c']) |
|
1024 |
||
1025 |
def test_inbetween(self): |
|
1026 |
self.assertMultiBisectRight([(1, ['b'])], ['b'], ['a', 'c']) |
|
1027 |
self.assertMultiBisectRight([(1, ['b', 'c', 'd']), (2, ['f', 'g'])], |
|
1028 |
['b', 'c', 'd', 'f', 'g'], ['a', 'e', 'h']) |
|
1029 |
||
1030 |
def test_mixed(self): |
|
1031 |
self.assertMultiBisectRight([(0, ['a', 'b']), (2, ['d', 'e']), |
|
1032 |
(4, ['g', 'h'])], |
|
1033 |
['a', 'b', 'd', 'e', 'g', 'h'], |
|
1034 |
['c', 'd', 'f', 'g']) |
|
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1035 |
|
1036 |
||
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1037 |
class TestExpandOffsets(tests.TestCase): |
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1038 |
|
1039 |
def make_index(self, size, recommended_pages=None): |
|
1040 |
"""Make an index with a generic size. |
|
1041 |
||
1042 |
This doesn't actually create anything on disk, it just primes a
|
|
1043 |
BTreeGraphIndex with the recommended information.
|
|
1044 |
"""
|
|
1045 |
index = btree_index.BTreeGraphIndex(get_transport('memory:///'), |
|
1046 |
'test-index', size=size) |
|
1047 |
if recommended_pages is not None: |
|
1048 |
index._recommended_pages = recommended_pages |
|
1049 |
return index |
|
1050 |
||
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1051 |
def set_cached_offsets(self, index, cached_offsets): |
1052 |
"""Monkeypatch to give a canned answer for _get_offsets_for...().""" |
|
1053 |
def _get_offsets_to_cached_pages(): |
|
1054 |
cached = set(cached_offsets) |
|
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1055 |
return cached |
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1056 |
index._get_offsets_to_cached_pages = _get_offsets_to_cached_pages |
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1057 |
|
1058 |
def prepare_index(self, index, node_ref_lists, key_length, key_count, |
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1059 |
row_lengths, cached_offsets): |
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1060 |
"""Setup the BTreeGraphIndex with some pre-canned information.""" |
1061 |
index.node_ref_lists = node_ref_lists |
|
1062 |
index._key_length = key_length |
|
1063 |
index._key_count = key_count |
|
1064 |
index._row_lengths = row_lengths |
|
1065 |
index._compute_row_offsets() |
|
1066 |
index._root_node = btree_index._InternalNode('internal\noffset=0\n') |
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1067 |
self.set_cached_offsets(index, cached_offsets) |
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1068 |
|
1069 |
def make_100_node_index(self): |
|
1070 |
index = self.make_index(4096*100, 6) |
|
3763.8.14
by John Arbash Meinel
Add in a shortcut when we haven't cached much yet. |
1071 |
# Consider we've already made a single request at the middle
|
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1072 |
self.prepare_index(index, node_ref_lists=0, key_length=1, |
1073 |
key_count=1000, row_lengths=[1, 99], |
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1074 |
cached_offsets=[0, 50]) |
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1075 |
return index |
1076 |
||
3763.8.9
by John Arbash Meinel
Finish up the algorithm to stay within a given layer. |
1077 |
def make_1000_node_index(self): |
1078 |
index = self.make_index(4096*1000, 6) |
|
3763.8.14
by John Arbash Meinel
Add in a shortcut when we haven't cached much yet. |
1079 |
# Pretend we've already made a single request in the middle
|
3763.8.9
by John Arbash Meinel
Finish up the algorithm to stay within a given layer. |
1080 |
self.prepare_index(index, node_ref_lists=0, key_length=1, |
1081 |
key_count=90000, row_lengths=[1, 9, 990], |
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1082 |
cached_offsets=[0, 5, 500]) |
3763.8.9
by John Arbash Meinel
Finish up the algorithm to stay within a given layer. |
1083 |
return index |
1084 |
||
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1085 |
def assertNumPages(self, expected_pages, index, size): |
1086 |
index._size = size |
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1087 |
self.assertEqual(expected_pages, index._compute_total_pages_in_index()) |
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1088 |
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1089 |
def assertExpandOffsets(self, expected, index, offsets): |
1090 |
self.assertEqual(expected, index._expand_offsets(offsets), |
|
1091 |
'We did not get the expected value after expanding'
|
|
1092 |
' %s' % (offsets,)) |
|
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1093 |
|
1094 |
def test_default_recommended_pages(self): |
|
1095 |
index = self.make_index(None) |
|
1096 |
# local transport recommends 4096 byte reads, which is 1 page
|
|
1097 |
self.assertEqual(1, index._recommended_pages) |
|
1098 |
||
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1099 |
def test__compute_total_pages_in_index(self): |
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1100 |
index = self.make_index(None) |
1101 |
self.assertNumPages(1, index, 1024) |
|
1102 |
self.assertNumPages(1, index, 4095) |
|
1103 |
self.assertNumPages(1, index, 4096) |
|
1104 |
self.assertNumPages(2, index, 4097) |
|
1105 |
self.assertNumPages(2, index, 8192) |
|
1106 |
self.assertNumPages(76, index, 4096*75 + 10) |
|
1107 |
||
3763.8.9
by John Arbash Meinel
Finish up the algorithm to stay within a given layer. |
1108 |
def test__find_layer_start_and_stop(self): |
1109 |
index = self.make_1000_node_index() |
|
1110 |
self.assertEqual((0, 1), index._find_layer_first_and_end(0)) |
|
1111 |
self.assertEqual((1, 10), index._find_layer_first_and_end(1)) |
|
1112 |
self.assertEqual((1, 10), index._find_layer_first_and_end(9)) |
|
1113 |
self.assertEqual((10, 1000), index._find_layer_first_and_end(10)) |
|
1114 |
self.assertEqual((10, 1000), index._find_layer_first_and_end(99)) |
|
1115 |
self.assertEqual((10, 1000), index._find_layer_first_and_end(999)) |
|
1116 |
||
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1117 |
def test_unknown_size(self): |
1118 |
# We should not expand if we don't know the file size
|
|
1119 |
index = self.make_index(None, 10) |
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1120 |
self.assertExpandOffsets([0], index, [0]) |
1121 |
self.assertExpandOffsets([1, 4, 9], index, [1, 4, 9]) |
|
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1122 |
|
1123 |
def test_more_than_recommended(self): |
|
1124 |
index = self.make_index(4096*100, 2) |
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1125 |
self.assertExpandOffsets([1, 10], index, [1, 10]) |
1126 |
self.assertExpandOffsets([1, 10, 20], index, [1, 10, 20]) |
|
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1127 |
|
1128 |
def test_read_all_from_root(self): |
|
1129 |
index = self.make_index(4096*10, 20) |
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1130 |
self.assertExpandOffsets(range(10), index, [0]) |
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1131 |
|
1132 |
def test_read_all_when_cached(self): |
|
1133 |
# We've read enough that we can grab all the rest in a single request
|
|
1134 |
index = self.make_index(4096*10, 5) |
|
1135 |
self.prepare_index(index, node_ref_lists=0, key_length=1, |
|
1136 |
key_count=1000, row_lengths=[1, 9], |
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1137 |
cached_offsets=[0, 1, 2, 5, 6]) |
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1138 |
# It should fill the remaining nodes, regardless of the one requested
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1139 |
self.assertExpandOffsets([3, 4, 7, 8, 9], index, [3]) |
1140 |
self.assertExpandOffsets([3, 4, 7, 8, 9], index, [8]) |
|
1141 |
self.assertExpandOffsets([3, 4, 7, 8, 9], index, [9]) |
|
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1142 |
|
1143 |
def test_no_root_node(self): |
|
1144 |
index = self.make_index(4096*10, 5) |
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1145 |
self.assertExpandOffsets([0], index, [0]) |
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1146 |
|
1147 |
def test_include_neighbors(self): |
|
1148 |
index = self.make_100_node_index() |
|
1149 |
# We expand in both directions, until we have at least 'recommended'
|
|
1150 |
# pages
|
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1151 |
self.assertExpandOffsets([9, 10, 11, 12, 13, 14, 15], index, [12]) |
1152 |
self.assertExpandOffsets([88, 89, 90, 91, 92, 93, 94], index, [91]) |
|
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1153 |
# If we hit an 'edge' we continue in the other direction
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1154 |
self.assertExpandOffsets([1, 2, 3, 4, 5, 6], index, [2]) |
1155 |
self.assertExpandOffsets([94, 95, 96, 97, 98, 99], index, [98]) |
|
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1156 |
|
1157 |
# Requesting many nodes will expand all locations equally
|
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1158 |
self.assertExpandOffsets([1, 2, 3, 80, 81, 82], index, [2, 81]) |
1159 |
self.assertExpandOffsets([1, 2, 3, 9, 10, 11, 80, 81, 82], index, |
|
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1160 |
[2, 10, 81]) |
1161 |
||
3763.8.8
by John Arbash Meinel
simple test of overlapped behavior |
1162 |
def test_stop_at_cached(self): |
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1163 |
index = self.make_100_node_index() |
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1164 |
self.set_cached_offsets(index, [0, 10, 19]) |
1165 |
self.assertExpandOffsets([11, 12, 13, 14, 15, 16], index, [11]) |
|
1166 |
self.assertExpandOffsets([11, 12, 13, 14, 15, 16], index, [12]) |
|
1167 |
self.assertExpandOffsets([12, 13, 14, 15, 16, 17, 18], index, [15]) |
|
1168 |
self.assertExpandOffsets([13, 14, 15, 16, 17, 18], index, [16]) |
|
1169 |
self.assertExpandOffsets([13, 14, 15, 16, 17, 18], index, [17]) |
|
1170 |
self.assertExpandOffsets([13, 14, 15, 16, 17, 18], index, [18]) |
|
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1171 |
|
3763.8.8
by John Arbash Meinel
simple test of overlapped behavior |
1172 |
def test_cannot_fully_expand(self): |
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1173 |
index = self.make_100_node_index() |
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1174 |
self.set_cached_offsets(index, [0, 10, 12]) |
3763.8.7
by John Arbash Meinel
A bit of doc updates, start putting in tests for current behavior. |
1175 |
# We don't go into an endless loop if we are bound by cached nodes
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1176 |
self.assertExpandOffsets([11], index, [11]) |
3763.8.8
by John Arbash Meinel
simple test of overlapped behavior |
1177 |
|
1178 |
def test_overlap(self): |
|
1179 |
index = self.make_100_node_index() |
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1180 |
self.assertExpandOffsets([10, 11, 12, 13, 14, 15], index, [12, 13]) |
1181 |
self.assertExpandOffsets([10, 11, 12, 13, 14, 15], index, [11, 14]) |
|
3763.8.9
by John Arbash Meinel
Finish up the algorithm to stay within a given layer. |
1182 |
|
1183 |
def test_stay_within_layer(self): |
|
1184 |
index = self.make_1000_node_index() |
|
1185 |
# When expanding a request, we won't read nodes from the next layer
|
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1186 |
self.assertExpandOffsets([1, 2, 3, 4], index, [2]) |
1187 |
self.assertExpandOffsets([6, 7, 8, 9], index, [6]) |
|
1188 |
self.assertExpandOffsets([6, 7, 8, 9], index, [9]) |
|
1189 |
self.assertExpandOffsets([10, 11, 12, 13, 14, 15], index, [10]) |
|
1190 |
self.assertExpandOffsets([10, 11, 12, 13, 14, 15, 16], index, [13]) |
|
3763.8.9
by John Arbash Meinel
Finish up the algorithm to stay within a given layer. |
1191 |
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1192 |
self.set_cached_offsets(index, [0, 4, 12]) |
1193 |
self.assertExpandOffsets([5, 6, 7, 8, 9], index, [7]) |
|
1194 |
self.assertExpandOffsets([10, 11], index, [11]) |
|
3763.8.14
by John Arbash Meinel
Add in a shortcut when we haven't cached much yet. |
1195 |
|
1196 |
def test_small_requests_unexpanded(self): |
|
1197 |
index = self.make_100_node_index() |
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1198 |
self.set_cached_offsets(index, [0]) |
1199 |
self.assertExpandOffsets([1], index, [1]) |
|
1200 |
self.assertExpandOffsets([50], index, [50]) |
|
3763.8.14
by John Arbash Meinel
Add in a shortcut when we haven't cached much yet. |
1201 |
# If we request more than one node, then we'll expand
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1202 |
self.assertExpandOffsets([49, 50, 51, 59, 60, 61], index, [50, 60]) |
3763.8.14
by John Arbash Meinel
Add in a shortcut when we haven't cached much yet. |
1203 |
|
1204 |
# The first pass does not expand
|
|
1205 |
index = self.make_1000_node_index() |
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1206 |
self.set_cached_offsets(index, [0]) |
1207 |
self.assertExpandOffsets([1], index, [1]) |
|
1208 |
self.set_cached_offsets(index, [0, 1]) |
|
1209 |
self.assertExpandOffsets([100], index, [100]) |
|
1210 |
self.set_cached_offsets(index, [0, 1, 100]) |
|
3763.8.14
by John Arbash Meinel
Add in a shortcut when we haven't cached much yet. |
1211 |
# But after the first depth, we will expand
|
3763.8.15
by John Arbash Meinel
Review comments from Martin. Code clarity/variable name/docstring updates. |
1212 |
self.assertExpandOffsets([2, 3, 4, 5, 6, 7], index, [2]) |
1213 |
self.assertExpandOffsets([2, 3, 4, 5, 6, 7], index, [4]) |
|
1214 |
self.set_cached_offsets(index, [0, 1, 2, 3, 4, 5, 6, 7, 100]) |
|
1215 |
self.assertExpandOffsets([102, 103, 104, 105, 106, 107, 108], index, |
|
1216 |
[105]) |