1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233
|
#
# Tests for the indexed_gzip module.
#
# Author: Paul McCarthy <pauldmccarthy@gmail.com>
#
from __future__ import print_function
import os
import os.path as op
import itertools as it
import functools as ft
import subprocess as sp
import multiprocessing as mp
import copy as cp
import sys
import time
import gzip
import random
import shutil
import pickle
import hashlib
import textwrap
import tempfile
import contextlib
import numpy as np
from io import BytesIO
import pytest
import indexed_gzip as igzip
from . import gen_test_data
from . import check_data_valid
from . import tempdir
from . import touch
from . import compress
SEEK_SET = os.SEEK_SET
SEEK_CUR = os.SEEK_CUR
SEEK_END = os.SEEK_END
def error_fn(*args, **kwargs):
raise Exception("Error")
def read_element(gzf, element, seek=True):
if seek:
gzf.seek(int(element) * 8)
bytes = gzf.read(8)
val = np.ndarray(1, np.uint64, buffer=bytes)
return val[0]
def write_text_to_gzip_file(fname, lines):
with gzip.open(fname, mode='wb') as f:
for line in lines:
f.write('{}\n'.format(line).encode())
@pytest.mark.parametrize('drop', [False, True])
def test_open_close(testfile, nelems, seed, drop):
f = igzip._IndexedGzipFile(filename=testfile, drop_handles=drop)
assert not f.closed
element = np.random.randint(0, nelems, 1)[0]
readval = read_element(f, element)
assert readval == element
f.close()
assert f.closed
with pytest.raises(IOError):
f.close()
def test_open_function(testfile, nelems):
f1 = None
f2 = None
try:
f1 = igzip.IndexedGzipFile(testfile)
f2 = igzip.open( testfile)
element = np.random.randint(0, nelems, 1)[0]
readval1 = read_element(f1, element)
readval2 = read_element(f2, element)
assert readval1 == element
assert readval2 == element
finally:
if f1 is not None: f1.close()
if f2 is not None: f2.close()
@pytest.mark.parametrize('drop', [False, True])
def test_open_close_ctxmanager(testfile, nelems, seed, drop):
with igzip._IndexedGzipFile(filename=testfile, drop_handles=drop) as f:
element = np.random.randint(0, nelems, 1)[0]
readval = read_element(f, element)
assert readval == element
assert f.closed
def test_open_mode():
modes = [('r', True),
('rb', True),
(None, True),
('rt', False),
('w', False),
('wt', False)]
# open from file
with tempdir():
with gzip.open('f.gz', 'wb') as f:
f.write(b'12345')
for mode, expect in modes:
if expect:
gzf = igzip.IndexedGzipFile('f.gz', mode=mode)
assert gzf.read() == b'12345'
else:
with pytest.raises(ValueError):
igzip.IndexedGzipFile('f.gz', mode=mode)
# open from fileobj
class BytesIOWithMode(BytesIO):
pass
# accept file-like without mode attribute
modes.append(('del', True))
for mode, expect in modes:
if mode == 'del' and hasattr(BytesIOWithMode, 'mode'):
delattr(BytesIOWithMode, 'mode')
else:
BytesIOWithMode.mode = mode
fileobj = BytesIOWithMode()
with gzip.GzipFile(fileobj=fileobj, mode='wb') as f:
f.write(b'12345')
print(mode, expect)
if expect:
assert igzip.IndexedGzipFile(fileobj=fileobj).read() == b'12345'
else:
with pytest.raises(ValueError):
igzip.IndexedGzipFile(fileobj=fileobj).read()
@pytest.mark.parametrize('drop', [False, True])
def test_atts(testfile, drop):
modes = [None, 'rb', 'r']
for m in modes:
with igzip._IndexedGzipFile(filename=testfile,
mode=m,
drop_handles=drop) as f:
assert not f.closed
assert f.readable()
assert f.seekable()
assert not f.writable()
assert f.mode == 'rb'
assert f.tell() == 0
if not drop:
assert f.fileobj() is not None
assert f.fileno() == f.fileobj().fileno()
else:
with pytest.raises(igzip.NoHandleError):
f.fileobj()
with pytest.raises(igzip.NoHandleError):
f.fileno()
@pytest.mark.parametrize('drop', [False, True])
def test_init_failure_cases(concat, drop):
with tempdir() as td:
testfile = op.join(td, 'test.gz')
gen_test_data(testfile, 65536, concat)
# No writing
with pytest.raises(ValueError):
gf = igzip._IndexedGzipFile(filename=testfile,
mode='w',
drop_handles=drop)
with pytest.raises(ValueError):
gf = igzip._IndexedGzipFile(filename=testfile,
mode='wb',
drop_handles=drop)
# No writing
f = open(testfile, mode='wb')
with pytest.raises(ValueError):
gf = igzip._IndexedGzipFile(fileobj=f, drop_handles=drop)
f.close()
# No writing
f = open(testfile, mode='w')
with pytest.raises(ValueError):
gf = igzip._IndexedGzipFile(fileobj=f, drop_handles=drop)
f.close()
# Need a filename or fid
with pytest.raises(ValueError):
f = igzip._IndexedGzipFile(drop_handles=drop)
# Doesn't exist
with pytest.raises(FileNotFoundError):
f = igzip._IndexedGzipFile(filename='no_file.gz',
drop_handles=drop)
# can only specify one of filename/fid
with pytest.raises(ValueError):
with open(testfile, mode='rb'):
f = igzip._IndexedGzipFile(filename=testfile,
fileobj=f,
drop_handles=drop)
@pytest.mark.parametrize('drop', [False, True])
def test_init_success_cases(concat, drop):
with tempdir() as td:
testfile = op.join(td, 'test.gz')
gen_test_data(testfile, 65536, concat)
gf1 = igzip._IndexedGzipFile(filename=testfile,
drop_handles=drop)
gf2 = igzip._IndexedGzipFile(filename=testfile,
mode='r',
drop_handles=drop)
gf3 = igzip._IndexedGzipFile(filename=testfile,
mode='rb',
drop_handles=drop)
gf1.close()
gf2.close()
gf3.close()
del gf1
del gf2
del gf3
@pytest.mark.parametrize('drop,file_like_object', [[False, False],
[False, True],
[True, False],
[True, True]])
def test_create_from_open_handle(testfile, nelems, seed, drop, file_like_object):
f = open(testfile, 'rb')
if file_like_object:
f = BytesIO(f.read())
gzf = igzip._IndexedGzipFile(fileobj=f, drop_handles=drop)
assert gzf.fileobj() is f
assert not gzf.drop_handles
element = np.random.randint(0, nelems, 1)[0]
readval = read_element(gzf, element)
gzf.close()
try:
assert readval == element
assert gzf.closed
assert not f.closed
finally:
f.close()
del gzf
del f
def test_accept_filename_or_fileobj(testfile, nelems):
f = None
gzf1 = None
gzf2 = None
gzf3 = None
try:
f = open(testfile, 'rb')
gzf1 = igzip._IndexedGzipFile(testfile)
gzf2 = igzip._IndexedGzipFile(f)
gzf3 = igzip._IndexedGzipFile(fileobj=BytesIO(open(testfile, 'rb').read()))
element = np.random.randint(0, nelems, 1)[0]
readval1 = read_element(gzf1, element)
readval2 = read_element(gzf2, element)
readval3 = read_element(gzf3, element)
assert readval1 == element
assert readval2 == element
assert readval3 == element
finally:
if gzf3 is not None: gzf3.close()
if gzf2 is not None: gzf2.close()
if gzf1 is not None: gzf1.close()
if f is not None: f .close()
del f
del gzf1
del gzf2
del gzf3
def test_prioritize_fd_over_f(testfile, nelems):
"""When a fileobj with an associated fileno is passed to IndexedGzipFile,
the fileobj's file descriptor (fd) should be utilized by zran.c
instead of the file-like object specified by fileobj (f).
"""
if sys.version_info[0] < 3:
# We can't set the .read attribute in Python 2
# because it's read-only, so skip it.
return
f = None
gzf = None
try:
f = open(testfile, 'rb')
f.read = error_fn # If the file-like object were directly used by zran.c, reading would raise an error.
gzf = igzip._IndexedGzipFile(fileobj=f)
element = np.random.randint(0, nelems, 1)[0]
readval = read_element(gzf, element)
assert readval == element
finally:
if gzf is not None: gzf.close()
if f is not None: f .close()
del f
del gzf
def test_handles_not_dropped(testfile, nelems, seed):
# When drop_handles is False
with igzip._IndexedGzipFile(filename=testfile, drop_handles=False) as f:
fid = f.fileobj()
assert fid is not None
# Check that the file object
# doesn't change across reads
for i in range(5):
element = np.random.randint(0, nelems, 1)[0]
readval = read_element(f, element)
assert readval == element
assert f.fileobj() is fid
# Also when given an open stream
with open(testfile, 'rb') as f:
with igzip._IndexedGzipFile(fileobj=f) as gzf:
assert gzf.fileobj() is f
for i in range(5):
element = np.random.randint(0, nelems, 1)[0]
readval = read_element(gzf, element)
assert readval == element
assert gzf.fileobj() is f
def test_manual_build():
with tempdir() as td:
nelems = 65536
fname = op.join(td, 'test.gz')
gen_test_data(fname, nelems, False)
with igzip._IndexedGzipFile(fname, auto_build=False) as f:
# Seeking to 0 should work, but
# anywhere else should fail
f.seek(0)
for off in [1, 2, 20, 200]:
with pytest.raises(igzip.NotCoveredError):
f.seek(off)
# Reading from beginning should work
readval = read_element(f, 0, seek=False)
assert readval == 0
# but subsequent reads should fail
# (n.b. this might change in the future)
with pytest.raises(igzip.NotCoveredError):
readval = read_element(f, 1, seek=False)
# Seek should still fail even after read
with pytest.raises(igzip.NotCoveredError):
f.seek(8)
# But reading from beginning should still work
f.seek(0)
readval = read_element(f, 0, seek=False)
assert readval == 0
# But after building the index,
# seeking and reading should work
f.build_full_index()
for i in range(5):
element = np.random.randint(0, nelems, 1)[0]
readval = read_element(f, element)
assert readval == element
@pytest.mark.parametrize('drop', [False, True])
def test_read_all(testfile, nelems, use_mmap, drop):
if use_mmap:
pytest.skip('skipping test_read_all test as '
'it will require too much memory')
with igzip._IndexedGzipFile(filename=testfile, drop_handles=drop) as f:
data = f.read(nelems * 8)
data = np.ndarray(shape=nelems, dtype=np.uint64, buffer=data)
# Check that every value is valid
assert check_data_valid(data, 0)
def test_simple_read_with_null_padding():
fileobj = BytesIO()
with gzip.GzipFile(fileobj=fileobj, mode='wb') as f:
f.write(b"hello world")
fileobj.write(b"\0" * 100)
with igzip._IndexedGzipFile(fileobj=fileobj) as f:
assert f.read() == b"hello world"
f.seek(3)
assert f.read() == b"lo world"
f.seek(20)
assert f.read() == b""
def test_read_with_null_padding(testfile, nelems, use_mmap):
if use_mmap:
pytest.skip('skipping test_read_with_null_padding test '
'as it will require too much memory')
fileobj = BytesIO(open(testfile, "rb").read() + b"\0" * 100)
with igzip._IndexedGzipFile(fileobj=fileobj) as f:
data = f.read(nelems * 8)
# Read a bit further so we reach the zero-padded area.
# This line should not throw an exception.
f.read(1)
data = np.ndarray(shape=nelems, dtype=np.uint64, buffer=data)
# Check that every value is valid
assert check_data_valid(data, 0)
@pytest.mark.parametrize('drop', [False, True])
def test_read_beyond_end(concat, drop):
with tempdir() as tdir:
nelems = 65536
testfile = op.join(tdir, 'test.gz')
gen_test_data(testfile, nelems, concat)
with igzip._IndexedGzipFile(filename=testfile,
readall_buf_size=1024,
drop_handles=drop) as f:
# Try with a specific number of bytes
data1 = f.read(nelems * 8 + 10)
# And also with unspecified numbytes
f.seek(0)
data2 = f.read()
data1 = np.ndarray(shape=nelems, dtype=np.uint64, buffer=data1)
data2 = np.ndarray(shape=nelems, dtype=np.uint64, buffer=data2)
assert check_data_valid(data1, 0)
assert check_data_valid(data2, 0)
del f
f = None
os.remove(testfile)
def test_seek(concat):
with tempdir() as tdir:
nelems = 262144 # == 2MB
testfile = op.join(tdir, 'test.gz')
gen_test_data(testfile, nelems, concat)
results = []
with igzip._IndexedGzipFile(testfile, spacing=131072) as f:
results.append((f.read(8), 0))
f.seek(24, SEEK_SET)
results.append((f.read(8), 3))
f.seek(-16, SEEK_CUR)
results.append((f.read(8), 2))
f.seek(16, SEEK_CUR)
results.append((f.read(8), 5))
# SEEK_END only works when index is built
with pytest.raises(ValueError):
f.seek(-100, SEEK_END)
f.build_full_index()
f.seek(-800, SEEK_END)
results.append((f.read(8), 262044))
f.seek(-3200, SEEK_END)
results.append((f.read(8), 261744))
for data, expected in results:
val = np.frombuffer(data, dtype=np.uint64)
assert val == expected
@pytest.mark.parametrize('drop', [False, True])
def test_seek_and_read(testfile, nelems, niters, seed, drop):
with igzip._IndexedGzipFile(filename=testfile, drop_handles=drop) as f:
# Pick some random elements and make
# sure their values are all right
seekelems = np.random.randint(0, nelems, niters)
for i, testval in enumerate(seekelems):
readval = read_element(f, testval)
ft = f.tell()
assert ft == (testval + 1) * 8
assert readval == testval
@pytest.mark.parametrize('drop', [False, True])
def test_seek_and_tell(testfile, nelems, niters, seed, drop):
filesize = nelems * 8
with igzip._IndexedGzipFile(filename=testfile, drop_handles=drop) as f:
# Pick some random seek positions
# and make sure that seek and tell
# return their location correctly
seeklocs = np.random.randint(0, filesize, niters)
for seekloc in seeklocs:
st = f.seek(seekloc)
ft = f.tell()
assert ft == seekloc
assert st == seekloc
# Also test that seeking beyond
# EOF is clamped to EOF
eofseeks = [filesize,
filesize + 1,
filesize + 2,
filesize + 3,
filesize + 4,
filesize + 1000,
filesize * 1000]
for es in eofseeks:
assert f.seek(es) == filesize
assert f.tell() == filesize
def test_pread():
with tempdir() as td:
nelems = 1024
testfile = op.join(td, 'test.gz')
gen_test_data(testfile, nelems, False)
with igzip.IndexedGzipFile(testfile) as f:
for i in range(20):
off = np.random.randint(0, nelems, 1)[0]
data = f.pread(8, off * 8)
val = np.frombuffer(data, dtype=np.uint64)
assert val[0] == off
@pytest.mark.parametrize('drop', [False, True])
def test_readinto(drop):
lines = textwrap.dedent("""
line 1
line 2
this is line 3
line the fourth
here is the fifth line
""").strip().split('\n')
def line_offset(idx):
return sum([len(l) for l in lines[:idx]]) + idx
with tempdir() as td:
testfile = op.join(td, 'test.gz')
write_text_to_gzip_file(testfile, lines)
with igzip._IndexedGzipFile(filename=testfile, drop_handles=drop) as f:
# read first line into a byte array
buf = bytearray(len(lines[0]))
f.seek(0)
assert f.readinto(buf) == len(lines[0])
assert buf.decode() == lines[0]
# read first line into memoryvew
buf = memoryview(bytearray(len(lines[0])))
f.seek(0)
assert f.readinto(buf) == len(lines[0])
assert buf.tobytes().decode() == lines[0]
# read an arbitrary line
offset = line_offset(2)
buf = bytearray(len(lines[2]))
f.seek(offset)
assert f.readinto(buf) == len(lines[2])
assert buf.decode() == lines[2]
# read the end line, sans-newline
offset = line_offset(len(lines) - 1)
buf = bytearray(len(lines[-1]))
f.seek(offset)
assert f.readinto(buf) == len(lines[-1])
assert buf.decode() == lines[-1]
# read the end line, with newline
buf = bytearray(len(lines[-1]) + 1)
f.seek(offset)
assert f.readinto(buf) == len(lines[-1]) + 1
assert buf.decode() == lines[-1] + '\n'
# read the end line with a bigger buffer
buf = bytearray(len(lines[-1]) + 10)
f.seek(offset)
assert f.readinto(buf) == len(lines[-1]) + 1
assert buf.decode() == lines[-1] + '\n' + (b'\0' * 9).decode()
# start at EOF, and try to read something
filelen = sum([len(l) for l in lines]) + len(lines)
f.seek(filelen)
buf = bytearray([99 for i in range(len(buf))])
assert f.readinto(buf) == 0
assert all([b == chr(99) for b in buf.decode()])
@pytest.mark.parametrize('drop', [False, True])
def test_readline(drop):
lines = textwrap.dedent("""
this is
some text
split across
several lines
how creative
""").strip().split('\n')
with tempdir() as td:
fname = op.join(td, 'test.gz')
write_text_to_gzip_file(fname, lines)
with igzip._IndexedGzipFile(fname, drop_handles=drop) as f:
seekpos = 0
for line in lines:
assert f.readline() == (line + '\n').encode()
seekpos += len(line) + 1
assert f.tell() == seekpos
# Should return empty string after EOF
assert f.readline() == b''
f.seek(0)
assert f.readline(0) == b''
@pytest.mark.parametrize('drop', [False, True])
def test_readline_sizelimit(drop):
lines = ['line one', 'line two']
with tempdir() as td:
fname = op.join(td, 'test.gz')
write_text_to_gzip_file(fname, lines)
with igzip._IndexedGzipFile(fname, drop_handles=drop) as f:
# limit to one character before the end of the first line
l = f.readline(len(lines[0]) - 1)
assert l == (lines[0][:-1]).encode()
# limit to the last character of the first line
f.seek(0)
l = f.readline(len(lines[0]) - 1)
assert l == (lines[0][:-1]).encode()
# limit to the newline at the end of the first line
f.seek(0)
l = f.readline(len(lines[0]) + 1)
assert l == (lines[0] + '\n').encode()
# limit to the first character after the first line
f.seek(0)
l = f.readline(len(lines[0]) + 2)
assert l == (lines[0] + '\n').encode()
@pytest.mark.parametrize('drop', [False, True])
def test_readlines(drop):
lines = textwrap.dedent("""
this is
some more text
split across
several lines
super imaginative
test data
""").strip().split('\n')
with tempdir() as td:
fname = op.join(td, 'test.gz')
write_text_to_gzip_file(fname, lines)
with igzip._IndexedGzipFile(fname, drop_handles=drop) as f:
gotlines = f.readlines()
assert len(lines) == len(gotlines)
for expl, gotl in zip(lines, gotlines):
assert (expl + '\n').encode() == gotl
assert f.read() == b''
@pytest.mark.parametrize('drop', [False, True])
def test_readlines_sizelimit(drop):
lines = ['line one', 'line two']
data = '\n'.join(lines) + '\n'
with tempdir() as td:
fname = op.join(td, 'test.gz')
write_text_to_gzip_file(fname, lines)
limits = range(len(data) + 2)
with igzip._IndexedGzipFile(fname, drop_handles=drop) as f:
for lim in limits:
f.seek(0)
gotlines = f.readlines(lim)
# Expect the first line
if lim < len(lines[0]) + 1:
assert len(gotlines) == 1
assert gotlines[0] == (lines[0] + '\n').encode()
# Expect both lines
else:
assert len(gotlines) == 2
assert gotlines[0] == (lines[0] + '\n').encode()
assert gotlines[1] == (lines[1] + '\n').encode()
@pytest.mark.parametrize('drop', [False, True])
def test_iter(drop):
lines = textwrap.dedent("""
this is
even more text
that is split
across several lines
the creativity
involved in generating
this test data is
unparalleled
""").strip().split('\n')
with tempdir() as td:
fname = op.join(td, 'test.gz')
write_text_to_gzip_file(fname, lines)
with igzip._IndexedGzipFile(fname, drop_handles=drop) as f:
for i, gotline in enumerate(f):
assert (lines[i] + '\n').encode() == gotline
with pytest.raises(StopIteration):
next(f)
@pytest.mark.slow_test
def test_get_index_seek_points():
with tempdir() as td:
fname = op.join(td, 'test.gz')
spacing = 1048576
# make a test file
data = np.arange(spacing, dtype=np.uint64)
with gzip.open(fname, 'wb') as f:
f.write(data.tobytes())
# check points before and after index creation
with igzip._IndexedGzipFile(fname, spacing=spacing) as f:
assert not list(f.seek_points())
f.build_full_index()
expected_number_of_seek_points = 1 + int(data.nbytes / spacing)
seek_points = list(f.seek_points())
assert len(seek_points) == expected_number_of_seek_points
# check monotonic growth
uncmp_offsets = [point[0] for point in seek_points]
assert sorted(uncmp_offsets) == uncmp_offsets
def test_import_export_index():
with tempdir() as td:
fname = op.join(td, 'test.gz')
idxfname = op.join(td, 'test.gzidx')
# make a test file
data = np.arange(524288, dtype=np.uint64)
with gzip.open(fname, 'wb') as f:
f.write(data.tobytes())
# generate an index file
with igzip._IndexedGzipFile(fname, spacing=131072) as f:
f.build_full_index()
points = list(f.seek_points())
f.export_index(idxfname)
# Check that index file works via __init__
with igzip._IndexedGzipFile(fname, index_file=idxfname) as f:
f.seek(65535 * 8)
val = np.frombuffer(f.read(8), dtype=np.uint64)
assert val[0] == 65535
assert points == list(f.seek_points())
# Check that index file works via import_index
with igzip._IndexedGzipFile(fname) as f:
f.import_index(idxfname)
f.seek(65535 * 8)
val = np.frombuffer(f.read(8), dtype=np.uint64)
assert val[0] == 65535
assert points == list(f.seek_points())
def test_import_export_index_open_file():
with tempdir() as td:
fname = op.join(td, 'test.gz')
idxfname = op.join(td, 'test.gzidx')
# make a test file
data = np.arange(524288, dtype=np.uint64)
with gzip.open(fname, 'wb') as f:
f.write(data.tobytes())
# generate an index file from open file handle
with igzip._IndexedGzipFile(fname, spacing=131072) as f:
f.build_full_index()
points = list(f.seek_points())
# Should raise if wrong permissions
with pytest.raises(ValueError):
touch(idxfname)
with open(idxfname, 'rb') as idxf:
f.export_index(fileobj=idxf)
with open(idxfname, 'wb') as idxf:
f.export_index(fileobj=idxf)
# Check that we can read it back
with igzip._IndexedGzipFile(fname) as f:
# Should raise if wrong permissions
# (append, so existing contents are
# not overwritten)
with pytest.raises(ValueError):
with open(idxfname, 'ab') as idxf:
f.import_index(fileobj=idxf)
with open(idxfname, 'rb') as idxf:
f.import_index(fileobj=idxf)
f.seek(65535 * 8)
val = np.frombuffer(f.read(8), dtype=np.uint64)
assert val[0] == 65535
assert points == list(f.seek_points())
# Test exporting to / importing from a file-like object
idxf = BytesIO()
with igzip._IndexedGzipFile(fname, spacing=131072) as f:
f.build_full_index()
f.export_index(fileobj=idxf)
points = list(f.seek_points())
idxf.seek(0)
with igzip._IndexedGzipFile(fname) as f:
f.import_index(fileobj=idxf)
f.seek(65535 * 8)
val = np.frombuffer(f.read(8), dtype=np.uint64)
assert val[0] == 65535
assert points == list(f.seek_points())
def test_build_index_from_unseekable():
with tempdir() as td:
fname = op.join(td, 'test.gz')
idxfname = op.join(td, 'test.gzidx')
# make a test file
data = np.arange(524288, dtype=np.uint64)
with gzip.open(fname, 'wb') as f:
f.write(data.tobytes())
# Test creating the index when file is unseekable,
# then using the index when file is seekable.
with open(fname, 'rb') as f:
b = f.read()
fileobj = BytesIO(b)
def new_seek(*args, **kwargs):
raise OSError()
def new_tell(*args, **kwargs):
raise OSError()
old_seek = fileobj.seek
old_tell = fileobj.tell
fileobj.seekable = lambda: False
fileobj.seek = new_seek
fileobj.tell = new_tell
# generate an index file
with igzip._IndexedGzipFile(fileobj, spacing=131072) as f:
f.build_full_index()
f.export_index(idxfname)
points = list(f.seek_points())
fileobj.seek = old_seek
fileobj.tell = old_tell
fileobj.seekable = lambda: True
fileobj.seek(0)
# Check that index file works via __init__
with igzip._IndexedGzipFile(fileobj, index_file=idxfname) as f:
f.seek(65535 * 8)
val = np.frombuffer(f.read(8), dtype=np.uint64)
assert val[0] == 65535
assert points == list(f.seek_points())
def test_wrapper_class():
with tempdir() as td:
fname = op.join(td, 'test.gz')
idxfname = op.join(td, 'test.gzidx')
data = np.arange(65536, dtype=np.uint64)
with gzip.open(fname, 'wb') as f:
f.write(data.tobytes())
with igzip.IndexedGzipFile(fname, drop_handles=False) as f:
assert f.fileno() == f.fileobj().fileno()
assert not f.drop_handles
f.build_full_index()
f.export_index(idxfname)
f.import_index(idxfname)
def gcd(num):
if num <= 3:
return 1
candidates = list(range(int(np.ceil(np.sqrt(num))), 2, -2))
candidates.extend((2, 1))
for divisor in candidates:
if num % divisor == 0:
return divisor
return 1
def test_size_multiple_of_readbuf():
fname = 'test.gz'
with tempdir():
while True:
data = np.random.randint(1, 1000, 100000, dtype=np.uint32)
with gzip.open(fname, 'wb') as f:
f.write(data.tobytes())
del f
f = None
# we need a file size that is divisible
# by the minimum readbuf size
fsize = op.getsize(fname)
if gcd(fsize) >= 128:
break
# readbuf size == file size
bufsz = fsize
with igzip.IndexedGzipFile(fname, readbuf_size=bufsz) as f:
assert f.seek(fsize) == fsize
del f
f = None
with igzip.IndexedGzipFile(fname, readbuf_size=bufsz) as f:
read = np.ndarray(shape=100000, dtype=np.uint32, buffer=f.read())
assert np.all(read == data)
del f
f = None
# Use a buf size that is a divisor of the file size
bufsz = gcd(fsize)
with igzip.IndexedGzipFile(fname, readbuf_size=bufsz) as f:
assert f.seek(fsize) == fsize
del f
f = None
with igzip.IndexedGzipFile(fname, readbuf_size=bufsz) as f:
read = np.ndarray(shape=100000, dtype=np.uint32, buffer=f.read())
assert np.all(read == data)
del f
f = None
@pytest.mark.slow_test
def test_picklable():
# default behaviour is for drop_handles=True,
# which means that an IndexedGzipFile object
# should be picklable/serialisable
fname = 'test.gz'
with tempdir():
data = np.random.randint(1, 1000, (10000, 10000), dtype=np.uint32)
with open(fname+'.bin', 'wb') as f:
f.write(data.tobytes())
compress(fname+'.bin', fname)
del f
gzf = igzip.IndexedGzipFile(fname)
first50MB = gzf.read(1048576 * 50)
gzf.seek(gzf.tell())
pickled = pickle.dumps(gzf)
second50MB = gzf.read(1048576 * 50)
gzf.seek(gzf.tell())
gzf.close()
del gzf
gzf = pickle.loads(pickled)
assert gzf.tell() == 1048576 * 50
assert gzf.read(1048576 * 50) == second50MB
gzf.seek(0)
assert gzf.read(1048576 * 50) == first50MB
gzf.close()
del gzf
# if drop_handles=False, no pickle
with tempdir():
data = np.random.randint(1, 1000, 50000, dtype=np.uint32)
with gzip.open(fname, 'wb') as f:
f.write(data.tobytes())
del f
gzf = igzip.IndexedGzipFile(fname, drop_handles=False)
with pytest.raises(pickle.PicklingError):
pickled = pickle.dumps(gzf)
gzf.close()
del gzf
def test_copyable():
fname = 'test.gz'
with tempdir():
data = np.random.randint(1, 1000, (10000, 10000), dtype=np.uint32)
with open(fname+'.bin', 'wb') as f:
f.write(data.tobytes())
compress(fname+'.bin', fname)
del f
gzf = igzip.IndexedGzipFile(fname)
gzf_copy = cp.deepcopy(gzf)
first50MB = gzf.read(1048576 * 50)
gzf.seek(gzf.tell())
gzf_copy2 = cp.deepcopy(gzf)
second50MB = gzf.read(1048576 * 50)
gzf.seek(gzf.tell())
gzf.close()
del gzf
assert gzf_copy.tell() == 0
assert gzf_copy2.tell() == 1048576 * 50
assert gzf_copy.read(1048576 * 50) == first50MB
assert gzf_copy2.read(1048576 * 50) == second50MB
gzf_copy2.seek(0)
assert gzf_copy2.read(1048576 * 50) == first50MB
gzf_copy.close()
gzf_copy2.close()
del gzf_copy
del gzf_copy2
with tempdir():
data = np.random.randint(1, 1000, 50000, dtype=np.uint32)
with gzip.open(fname, 'wb') as f:
f.write(data.tobytes())
del f
# if drop_handles=False, no copy
gzf = igzip.IndexedGzipFile(fname, drop_handles=False)
with pytest.raises(pickle.PicklingError):
gzf_copy = cp.deepcopy(gzf)
gzf.close()
del gzf
# If passed an open filehandle, no copy
with open(fname, 'rb') as fobj:
gzf = igzip.IndexedGzipFile(fileobj=fobj)
with pytest.raises(pickle.PicklingError):
gzf_copy = cp.deepcopy(gzf)
gzf.close()
del gzf
del fobj
def _mpfunc(gzf, size, offset):
gzf.seek(offset)
bytes = gzf.read(size)
val = np.ndarray(int(size / 4), np.uint32, buffer=bytes)
gzf.close()
del gzf
return val.sum()
@pytest.mark.slowtest
def test_multiproc_serialise():
fname = 'test.gz'
with tempdir():
data = np.arange(10000000, dtype=np.uint32)
with gzip.open(fname, 'wb') as f:
f.write(data.tobytes())
del f
gzf = igzip.IndexedGzipFile(fname)
size = int(len(data) / 16)
offsets = np.arange(0, len(data), size)
func = ft.partial(_mpfunc, gzf, size * 4)
pool = mp.Pool(8)
results = pool.map(func, offsets * 4)
pool.close()
pool.join()
gzf.close()
del gzf
del pool
expected = [data[off:off+size].sum() for off in offsets]
assert results == expected
@pytest.mark.slow_test
def test_32bit_overflow(niters, seed):
with tempdir():
block = 2 ** 24 # 128MB
nelems = block * 48 # 6GB
data = np.ones(block, dtype=np.uint64).tobytes()
with gzip.open('test.gz', 'wb') as f:
for i in range(48):
print('Generated to {}...'.format(block * i))
f.write(data)
with igzip._IndexedGzipFile(filename='test.gz') as f:
seekelems = np.random.randint(0, nelems, niters)
for i, testval in enumerate(seekelems):
readval = read_element(f, testval)
ft = f.tell()
assert ft == int(testval + 1) * 8
assert readval == 1
|