/
synchronizing.py
1557 lines (1299 loc) · 64.3 KB
/
synchronizing.py
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
import asyncio
import collections
import functools
import json
import logging
import os
import re
import tempfile
import uuid
from collections import defaultdict
from gettext import gettext as _ # noqa:F401
from asgiref.sync import sync_to_async
from django.conf import settings
from django.core.exceptions import ObjectDoesNotExist
from django.core.files import File
from django.db import transaction
from django.db.models import Q
from aiohttp.client_exceptions import ClientResponseError
from aiohttp.web_exceptions import HTTPNotFound
import createrepo_c as cr
import libcomps
from pulpcore.plugin.util import get_domain
from pulpcore.plugin.models import (
Artifact,
ContentArtifact,
ProgressReport,
Remote,
PublishedArtifact,
PublishedMetadata,
)
from pulpcore.plugin.stages import (
ACSArtifactHandler,
ArtifactDownloader,
ArtifactSaver,
ContentSaver,
DeclarativeArtifact,
DeclarativeContent,
DeclarativeVersion,
RemoteArtifactSaver,
Stage,
QueryExistingArtifacts,
QueryExistingContents,
)
from pulp_rpm.app.advisory import hash_update_record
from pulp_rpm.app.constants import (
CHECKSUM_TYPES,
COMPS_REPODATA,
DIST_TREE_MAIN_REPO_PATH,
MODULAR_REPODATA,
PACKAGE_DB_REPODATA,
PACKAGE_REPODATA,
PULP_MODULE_ATTR,
SYNC_POLICIES,
UPDATE_REPODATA,
)
from pulp_rpm.app.models import (
Addon,
Checksum,
DistributionTree,
Image,
Variant,
Modulemd,
ModulemdDefaults,
ModulemdObsolete,
Package,
RepoMetadataFile,
PackageGroup,
PackageCategory,
PackageEnvironment,
PackageLangpacks,
RpmPublication,
RpmRemote,
RpmRepository,
UlnRemote,
UpdateCollection,
UpdateCollectionPackage,
UpdateRecord,
UpdateReference,
)
from pulp_rpm.app.modulemd import parse_modular
from pulp_rpm.app.comps import strdict_to_dict, dict_digest
from pulp_rpm.app.kickstart.treeinfo import PulpTreeInfo, TreeinfoData
from pulp_rpm.app.metadata_parsing import MetadataParser
from pulp_rpm.app.shared_utils import (
is_previous_version,
get_sha256,
urlpath_sanitize,
)
from pulp_rpm.app.rpm_version import RpmVersion
log = logging.getLogger(__name__)
# TODO: https://pulp.plan.io/issues/8687
# A global dictionary for storing data about the remote's metadata files, used for mirroring
# Indexed by repository.pk due to sub-repos.
metadata_files_for_mirroring = collections.defaultdict(dict)
# A global dictionary for storing data mapping pkgid to location_href for all packages, used
# for mirroring. Indexed by repository.pk due to sub-repos.
pkgid_to_location_href = collections.defaultdict(functools.partial(collections.defaultdict, set))
MIRROR_INCOMPATIBLE_REPO_ERR_MSG = (
"This repository uses features which are incompatible with 'mirror' sync. "
"Please sync without mirroring enabled."
)
def store_metadata_for_mirroring(repo, md_path, relative_path):
"""Used to store data about the downloaded metadata for mirror-publishing after the sync.
Args:
repo: Which repository the metadata is associated with
md_path: The path to the metadata file
relative_path: The relative path to the metadata file within the repository
"""
global metadata_files_for_mirroring
metadata_files_for_mirroring[str(repo.pk)][relative_path] = md_path
def store_package_for_mirroring(repo, pkgid, location_href):
"""Used to store data about the packages for mirror-publishing after the sync.
Args:
repo: Which repository the metadata is associated with
pkgid: The checksum of the package
location_href: The relative path to the package within the repository
"""
global pkgid_to_location_href
# this shouldn't really add the location_href to a list, really it ought to set the value
# but unfortunately some repositories have the same packages present in multiple places
# same pkgid, >1 different location_hrefs
pkgid_to_location_href[str(repo.pk)][pkgid].add(location_href)
def add_metadata_to_publication(publication, version, prefix=""):
"""Create a mirrored publication for the given repository version.
Uses the `metadata_files` global data.
Args:
publication: The publication to add downloaded repo metadata to
version: The repository version the repo corresponds to
Kwargs:
prefix: Subdirectory underneath the root repository (if a sub-repo)
"""
repo_metadata_files = metadata_files_for_mirroring[str(version.repository.pk)]
for relative_path, metadata_file_path in repo_metadata_files.items():
with open(metadata_file_path, "rb") as metadata_fd:
PublishedMetadata.create_from_file(
file=File(metadata_fd),
relative_path=os.path.join(prefix, relative_path),
publication=publication,
)
published_artifacts = []
# Handle packages
pkg_data = ContentArtifact.objects.filter(
content__in=version.content, content__pulp_type=Package.get_pulp_type()
).values("pk", "content__rpm_package__pkgId")
for ca in pkg_data.iterator():
for relative_path in pkgid_to_location_href[str(version.repository.pk)][
ca["content__rpm_package__pkgId"]
]:
pa = PublishedArtifact(
content_artifact_id=ca["pk"],
relative_path=os.path.join(prefix, relative_path),
publication=publication,
)
published_artifacts.append(pa)
# Handle everything else
# TODO: this code is copied directly from publication, we should deduplicate it later
# (if possible)
is_treeinfo = Q(relative_path__in=["treeinfo", ".treeinfo"])
unpublishable_types = Q(
content__pulp_type__in=[
RepoMetadataFile.get_pulp_type(),
Modulemd.get_pulp_type(),
ModulemdDefaults.get_pulp_type(),
# already dealt with
Package.get_pulp_type(),
]
)
contentartifact_qs = (
ContentArtifact.objects.filter(content__in=version.content)
.exclude(unpublishable_types)
.exclude(is_treeinfo)
)
for content_artifact in contentartifact_qs.values("pk", "relative_path").iterator():
published_artifacts.append(
PublishedArtifact(
relative_path=content_artifact["relative_path"],
publication=publication,
content_artifact_id=content_artifact["pk"],
)
)
PublishedArtifact.objects.bulk_create(published_artifacts)
def get_repomd_file(remote, url):
"""
Check if repodata exists.
Args:
remote (RpmRemote or UlnRemote): An RpmRemote or UlnRemote to download with.
url (str): A remote repository URL
Returns:
pulpcore.plugin.download.DownloadResult: downloaded repomd.xml
"""
# URLs, esp mirrorlist URLs, can come into this method with parameters attached.
# This causes the urlpath_sanitize() below to return something like
# "http://path?param¶m/repodata/repomd.xml", which is **not** an expected/useful response.
# Make sure we're only looking for the repomd.xml file, no matter what weirdness comes
# in. See https://pulp.plan.io/issues/8981 for more details.
url = url.split("?")[0]
downloader = remote.get_downloader(url=urlpath_sanitize(url, "repodata/repomd.xml"))
return downloader.fetch()
def fetch_mirror(remote):
"""Fetch the first valid mirror from a list of all available mirrors from a mirror list feed.
URLs which are commented out or have any punctuations in front of them are being ignored.
"""
downloader = remote.get_downloader(url=remote.url.rstrip("/"), urlencode=False)
result = downloader.fetch()
url_pattern = re.compile(r"(^|^[\w\s=]+\s)((http(s)?)://.*)")
with open(result.path) as mirror_list_file:
for mirror in mirror_list_file:
match = re.match(url_pattern, mirror)
if not match:
continue
mirror_url = match.group(2)
try:
get_repomd_file(remote, mirror_url)
# just check if the metadata exists
return mirror_url
except Exception as exc:
log.warning(
"Url '{}' from mirrorlist was tried and failed with error: {}".format(
mirror_url, exc
)
)
continue
return None
def fetch_remote_url(remote, custom_url=None):
"""Fetch a single remote from which can be content synced."""
def normalize_url(url_to_normalize):
return url_to_normalize.rstrip("/") + "/"
url = custom_url or remote.url
try:
normalized_remote_url = normalize_url(url)
get_repomd_file(remote, normalized_remote_url)
# just check if the metadata exists
return normalized_remote_url
except ClientResponseError as exc:
# If 'custom_url' is passed it is a call from ACS refresh
# which doesn't support mirror lists.
if custom_url:
raise ValueError(
_(
"ACS remote for url '{}' raised an error '{}: {}'. "
"Please check your ACS remote configuration."
).format(custom_url, exc.status, exc.message)
)
log.info(
_("Attempting to resolve a true url from potential mirrolist url '{}'").format(url)
)
remote_url = fetch_mirror(remote)
if remote_url:
log.info(
_("Using url '{}' from mirrorlist in place of the provided url {}").format(
remote_url, url
)
)
return normalize_url(remote_url)
if exc.status == 404:
raise ValueError(_("An invalid remote URL was provided: {}").format(url))
raise exc
def should_optimize_sync(sync_details, last_sync_details):
"""
Check whether the sync should be optimized by comparing its parameters with the previous sync.
Args:
sync_details (dict): A collection of details about the current sync configuration.
last_sync_details (dict): A collection of details about the previous sync configuration.
Returns:
bool: True, if sync is optimized; False, otherwise.
"""
might_download_content = (
last_sync_details.get("download_policy") != "immediate"
and sync_details["download_policy"] == "immediate"
)
might_create_publication = (
last_sync_details.get("sync_policy") != SYNC_POLICIES.MIRROR_COMPLETE
and sync_details["sync_policy"] == SYNC_POLICIES.MIRROR_COMPLETE
)
if might_download_content or might_create_publication:
return False
url_has_changed = last_sync_details.get("url") != sync_details["url"]
retain_package_versions_has_changed = (
last_sync_details.get("retain_package_versions") != sync_details["retain_package_versions"]
)
repository_has_been_modified = (
last_sync_details.get("most_recent_version") != sync_details["most_recent_version"]
)
if url_has_changed or repository_has_been_modified or retain_package_versions_has_changed:
return False
old_revision = is_previous_version(sync_details["revision"], last_sync_details.get("revision"))
same_revision = last_sync_details.get("revision") == sync_details["revision"]
same_repomd_checksum = (
last_sync_details.get("repomd_checksum") == sync_details["repomd_checksum"]
)
same_treeinfo_checksum = (
last_sync_details.get("treeinfo_checksum") == sync_details["treeinfo_checksum"]
)
if not old_revision or not (same_revision and same_repomd_checksum and same_treeinfo_checksum):
return False
return True
def synchronize(remote_pk, repository_pk, sync_policy, skip_types, optimize, url=None):
"""
Sync content from the remote repository.
Create a new version of the repository that is synchronized with the remote.
If sync_policy=mirror_complete, a publication will be created with a copy of the original
metadata. This comes with some limitations, namely:
* SRPMs and other types listed in "skip_types" will *not* be skipped.
* If the repository uses the xml:base / location_base feature, then the sync will fail.
This feature is incompatible with the intentions of most Pulp users, because the metadata
will tell clients to look for files at some source outside of the Pulp-hosted repo.
* If the repository uses Delta RPMs, the sync will fail, because Pulp does not support them,
and cannot change the repository metadata to remove them.
* If the repository metadata is invalid in specific ways, the sync will fail. These repos
may "work" with Yum / DNF but the results could be unexpected. Pulp is less tolerant of
metadata with these ambiguities.
Args:
remote_pk (str): The remote PK.
repository_pk (str): The repository PK.
sync_policy (str): How to perform the sync.
skip_types (list): List of content to skip.
optimize(bool): Optimize mode.
url(str): Custom URL to use instead of Remote's URL
Raises:
ValueError: If the remote does not specify a url to sync.
"""
try:
remote = RpmRemote.objects.get(pk=remote_pk)
except ObjectDoesNotExist:
remote = UlnRemote.objects.get(pk=remote_pk)
repository = RpmRepository.objects.get(pk=repository_pk)
if not remote.url and not url:
raise ValueError(_("A remote must have a url specified to synchronize."))
log.info(_("Synchronizing: repository={r} remote={p}").format(r=repository.name, p=remote.name))
deferred_download = remote.policy != Remote.IMMEDIATE # Interpret download policy
skip_treeinfo = "treeinfo" in skip_types
def get_treeinfo_data(remote, remote_url):
"""Get Treeinfo data from remote."""
treeinfo_serialized = {}
if skip_treeinfo:
return treeinfo_serialized
namespaces = [".treeinfo", "treeinfo"]
for namespace in namespaces:
downloader = remote.get_downloader(
url=urlpath_sanitize(remote_url, namespace),
silence_errors_for_response_status_codes={403, 404},
)
try:
result = downloader.fetch()
except FileNotFoundError:
continue
treeinfo = PulpTreeInfo()
treeinfo.load(f=result.path)
sha256 = result.artifact_attributes["sha256"]
treeinfo_data = TreeinfoData(treeinfo.parsed_sections())
# get the data we need before changing the original
treeinfo_serialized = treeinfo_data.to_dict(hash=sha256, filename=namespace)
# rewrite the treeinfo file such that the variant repository and package location
# is a relative subtree
treeinfo.rewrite_subrepo_paths(treeinfo_data)
# TODO: better way to do this?
main_variant = treeinfo.original_parser._sections.get("general", {}).get(
"variant", None
)
treeinfo_file = tempfile.NamedTemporaryFile(dir=".", delete=False)
treeinfo.dump(treeinfo_file.name, main_variant=main_variant)
store_metadata_for_mirroring(repository, treeinfo_file.name, namespace)
break
return treeinfo_serialized
def get_sync_details(remote, url, sync_policy, repository):
version = repository.latest_version()
with tempfile.TemporaryDirectory(dir="."):
result = get_repomd_file(remote, url)
repomd_path = result.path
repomd = cr.Repomd(repomd_path)
repomd_checksum = get_sha256(repomd_path)
treeinfo_file_data = get_treeinfo_data(remote, url)
treeinfo_checksum = treeinfo_file_data.get("hash", "")
return {
"url": remote.url, # use the original remote url so that mirrorlists are optimizable
"download_policy": remote.policy,
"sync_policy": sync_policy,
"most_recent_version": version.number,
"revision": repomd.revision,
"repomd_checksum": repomd_checksum,
"treeinfo_checksum": treeinfo_checksum,
"retain_package_versions": repository.retain_package_versions,
}
mirror = sync_policy.startswith("mirror")
mirror_metadata = sync_policy == SYNC_POLICIES.MIRROR_COMPLETE
repo_sync_config = {}
# this is the "directory" of the repo within the target repo location - for the primary
# repo, they are the same
PRIMARY_REPO = ""
def is_subrepo(directory):
return directory != PRIMARY_REPO
with tempfile.TemporaryDirectory(dir="."):
remote_url = fetch_remote_url(remote, url)
# Find and set up to deal with any subtrees
treeinfo = get_treeinfo_data(remote, remote_url)
if treeinfo:
treeinfo["repositories"] = {}
for repodata in set(treeinfo["download"]["repodatas"]):
if repodata == DIST_TREE_MAIN_REPO_PATH:
treeinfo["repositories"].update({repodata: None})
continue
name = f"{repodata}-{treeinfo['hash']}-{repository.pulp_id}"
sub_repo, created = RpmRepository.objects.get_or_create(name=name, user_hidden=True)
if created:
sub_repo.save()
directory = treeinfo["repo_map"][repodata]
treeinfo["repositories"].update({directory: str(sub_repo.pk)})
path = f"{repodata}/"
new_url = urlpath_sanitize(remote_url, path)
try:
subrepo_sync_details = get_sync_details(remote, new_url, sync_policy, sub_repo)
except ClientResponseError as exc:
if is_subrepo(directory) and exc.status == 404:
log.warning("Unable to sync sub-repo '{}' from treeinfo.".format(directory))
continue
raise exc
repo_sync_config[directory] = {
"should_skip": should_optimize_sync(
subrepo_sync_details, sub_repo.last_sync_details
),
"sync_details": subrepo_sync_details,
"url": new_url,
"repo": sub_repo,
}
# Set up to deal with the primary repository
sync_details = get_sync_details(remote, remote_url, sync_policy, repository)
repo_sync_config[PRIMARY_REPO] = {
"should_skip": should_optimize_sync(sync_details, repository.last_sync_details),
"sync_details": sync_details,
"url": remote_url,
"repo": repository,
}
# If all repos are exactly the same, we should skip all further processing, even in
# metadata-mirror mode
if optimize and all([config["should_skip"] for config in repo_sync_config.values()]):
with ProgressReport(
message="Skipping Sync (no change from previous sync)", code="sync.was_skipped"
) as pb:
pb.done = len(repo_sync_config)
pb.total = len(repo_sync_config)
return
skipped_syncs = 0
repo_sync_results = {}
# If some repos need to be synced and others do not, we go through them all
# items() returns in insertion-order - make sure PRIMARY is the LAST thing we process
# here, or autopublish will fail to find any subrepo-content.
for directory, repo_config in repo_sync_config.items():
repo = repo_config["repo"]
# If metadata_mirroring is enabled we cannot skip any syncs, because the generated
# publication needs to contain exactly the same metadata at the same paths.
if not mirror_metadata and optimize and repo_config["should_skip"]:
skipped_syncs += 1
repo_sync_results[directory] = repo.latest_version()
continue
stage = RpmFirstStage(
remote,
repo,
deferred_download,
mirror_metadata,
skip_types=skip_types,
new_url=repo_config["url"],
treeinfo=(treeinfo if not is_subrepo(directory) else None),
namespace=directory,
)
dv = RpmDeclarativeVersion(first_stage=stage, repository=repo, mirror=mirror)
repo_version = dv.create() or repo.latest_version()
repo_config["sync_details"]["most_recent_version"] = repo_version.number
repo.last_sync_details = repo_config["sync_details"]
repo.save()
repo_sync_results[directory] = repo_version
if skipped_syncs:
with ProgressReport(
message="Skipping Sync (no change from previous sync)", code="sync.was_skipped"
) as pb:
pb.done = skipped_syncs
pb.total = len(repo_sync_config)
if mirror_metadata:
with RpmPublication.create(
repo_sync_results[PRIMARY_REPO], pass_through=False
) as publication:
gpgcheck = repository.repo_config.get("gpgcheck", 0)
has_repomd_signature = (
"repodata/repomd.xml.asc" in metadata_files_for_mirroring[str(repository.pk)].keys()
)
repo_gpgcheck = has_repomd_signature and repository.repo_config.get("repo_gpgcheck", 0)
publication.checksum_type = CHECKSUM_TYPES.UNKNOWN
publication.package_checksum_type = CHECKSUM_TYPES.UNKNOWN
publication.metadata_checksum_type = CHECKSUM_TYPES.UNKNOWN
publication.repo_config = {
"repo_gpgcheck": int(repo_gpgcheck),
"gpgcheck": int(gpgcheck),
}
for path, repo_version in repo_sync_results.items():
add_metadata_to_publication(publication, repo_version, prefix=path)
return repo_sync_results[PRIMARY_REPO]
class RpmDeclarativeVersion(DeclarativeVersion):
"""
Subclassed Declarative version creates a custom pipeline for RPM sync.
"""
def __init__(self, *args, **kwargs):
"""
Adding support for ACS.
Adding it here, because we call RpmDeclarativeVersion multiple times in sync.
"""
kwargs["acs"] = True
super().__init__(*args, **kwargs)
def pipeline_stages(self, new_version):
"""
Build a list of stages feeding into the ContentUnitAssociation stage.
This defines the "architecture" of the entire sync.
Args:
new_version (:class:`~pulpcore.plugin.models.RepositoryVersion`): The
new repository version that is going to be built.
Returns:
list: List of :class:`~pulpcore.plugin.stages.Stage` instances
"""
pipeline = [
self.first_stage,
QueryExistingArtifacts(),
]
if self.acs:
pipeline.append(ACSArtifactHandler())
pipeline.extend(
[
ArtifactDownloader(),
ArtifactSaver(),
QueryExistingContents(),
RpmContentSaver(),
RpmInterrelateContent(),
RemoteArtifactSaver(fix_mismatched_remote_artifacts=True),
]
)
return pipeline
class RpmFirstStage(Stage):
"""
First stage of the Asyncio Stage Pipeline.
Create a :class:`~pulpcore.plugin.stages.DeclarativeContent` object for each content unit
that should exist in the new :class:`~pulpcore.plugin.models.RepositoryVersion`.
"""
def __init__(
self,
remote,
repository,
deferred_download,
mirror_metadata,
skip_types=None,
new_url=None,
treeinfo=None,
namespace="",
):
"""
The first stage of a pulp_rpm sync pipeline.
Args:
remote (RpmRemote or UlnRemote): The remote data to be used when syncing
repository (RpmRepository): The repository to be compared when optimizing sync
deferred_download (bool): If True the downloading will not happen now. If False, it will
happen immediately.
mirror_metadata (bool): Influences which metadata files are downloaded and what
is done with them.
Keyword Args:
skip_types (list): List of content to skip
new_url(str): URL to replace remote url
treeinfo(dict): Treeinfo data
namespace(str): Path where this repo is located relative to some parent repo.
"""
super().__init__()
self.remote = remote
self.repository = repository
self.deferred_download = deferred_download
self.mirror_metadata = mirror_metadata
# How many directories deep this repo is nested within another repo (if at all).
# Backwards relative paths that are shallower than this depth are permitted (in mirror
# mode), to accomodate sub-repos which re-use packages from the parent repo.
self.namespace_depth = 0 if not namespace else len(namespace.strip("/").split("/"))
self.treeinfo = treeinfo
self.skip_types = [] if skip_types is None else skip_types
self.remote_url = new_url or self.remote.url
self.nevra_to_module = defaultdict(dict)
self.pkgname_to_groups = defaultdict(list)
def is_illegal_relative_path(self, path):
"""Whether a relative path points outside the repository being synced."""
return path.count("../") > self.namespace_depth
@staticmethod
async def parse_updateinfo(updateinfo_xml_path):
"""
Parse updateinfo.xml to extact update info.
Args:
updateinfo_xml_path: a path to a downloaded updateinfo.xml
Returns:
:obj:`list` of :obj:`createrepo_c.UpdateRecord`: parsed update records
"""
uinfo = cr.UpdateInfo()
# TODO: handle parsing errors/warnings, warningcb callback can be used
cr.xml_parse_updateinfo(updateinfo_xml_path, uinfo)
return uinfo.updates
async def run(self):
"""Build `DeclarativeContent` from the repodata."""
with tempfile.TemporaryDirectory(dir="."):
progress_data = dict(
message="Downloading Metadata Files", code="sync.downloading.metadata"
)
async with ProgressReport(**progress_data) as metadata_pb:
# download repomd.xml
downloader = self.remote.get_downloader(
url=urlpath_sanitize(self.remote_url, "repodata/repomd.xml")
)
result = await downloader.run()
store_metadata_for_mirroring(self.repository, result.path, "repodata/repomd.xml")
await metadata_pb.aincrement()
repomd_path = result.path
repomd = cr.Repomd(repomd_path)
if repomd.warnings:
for warn_type, warn_msg in repomd.warnings:
log.warn(warn_msg)
msg = (
"Problems encountered parsing repomd.xml - proxy used: {}, url: {}".format(
self.remote.proxy_url,
result.url,
)
)
log.warn(msg)
checksum_types = {}
repomd_downloaders = {}
repomd_files = {}
types_to_download = (
set(PACKAGE_REPODATA)
| set(UPDATE_REPODATA)
| set(COMPS_REPODATA)
| set(MODULAR_REPODATA)
)
async def run_repomdrecord_download(name, location_href, downloader):
result = await downloader.run()
return name, location_href, result
for record in repomd.records:
record_checksum_type = getattr(CHECKSUM_TYPES, record.checksum_type.upper())
checksum_types[record.type] = record_checksum_type
record.checksum_type = record_checksum_type
if self.mirror_metadata:
uses_base_url = record.location_base
illegal_relative_path = self.is_illegal_relative_path(record.location_href)
if (
uses_base_url
or illegal_relative_path
or record.type in RepoMetadataFile.UNSUPPORTED_METADATA
):
raise ValueError(MIRROR_INCOMPATIBLE_REPO_ERR_MSG)
if not self.mirror_metadata and record.type not in types_to_download:
continue
base_url = record.location_base or self.remote_url
downloader = self.remote.get_downloader(
url=urlpath_sanitize(base_url, record.location_href),
expected_size=record.size,
expected_digests={record_checksum_type: record.checksum},
)
repomd_downloaders[record.type] = asyncio.ensure_future(
run_repomdrecord_download(record.type, record.location_href, downloader)
)
self.repository.original_checksum_types = checksum_types
try:
for future in asyncio.as_completed(list(repomd_downloaders.values())):
name, location_href, result = await future
store_metadata_for_mirroring(self.repository, result.path, location_href)
repomd_files[name] = result
await metadata_pb.aincrement()
except ClientResponseError as exc:
raise HTTPNotFound(reason=_("File not found: {}".format(exc.request_info.url)))
except FileNotFoundError:
raise
if self.mirror_metadata:
# optional signature and key files for repomd metadata
for file_href in ["repodata/repomd.xml.asc", "repodata/repomd.xml.key"]:
try:
downloader = self.remote.get_downloader(
url=urlpath_sanitize(self.remote_url, file_href),
silence_errors_for_response_status_codes={403, 404},
)
result = await downloader.run()
store_metadata_for_mirroring(self.repository, result.path, file_href)
await metadata_pb.aincrement()
except (ClientResponseError, FileNotFoundError):
pass
# extra files to copy, e.g. EULA, LICENSE
try:
downloader = self.remote.get_downloader(
url=urlpath_sanitize(self.remote_url, "extra_files.json"),
silence_errors_for_response_status_codes={403, 404},
)
result = await downloader.run()
store_metadata_for_mirroring(
self.repository, result.path, "extra_files.json"
)
await metadata_pb.aincrement()
except (ClientResponseError, FileNotFoundError):
pass
else:
try:
with open(result.path, "r") as f:
extra_files = json.loads(f.read())
for data in extra_files["data"]:
filtered_checksums = {
digest: value
for digest, value in data["checksums"].items()
if digest in settings.ALLOWED_CONTENT_CHECKSUMS
}
downloader = self.remote.get_downloader(
url=urlpath_sanitize(self.remote_url, data["file"]),
expected_size=data["size"],
expected_digests=filtered_checksums,
)
result = await downloader.run()
store_metadata_for_mirroring(
self.repository, result.path, data["file"]
)
await metadata_pb.aincrement()
except ClientResponseError as exc:
raise HTTPNotFound(
reason=_("File not found: {}".format(exc.request_info.url))
)
except FileNotFoundError:
raise
await self.parse_repository_metadata(repomd, repomd_files)
async def parse_distribution_tree(self):
"""Parse content from the file treeinfo if present."""
if self.treeinfo:
d_artifacts = [
DeclarativeArtifact(
artifact=Artifact(),
url=urlpath_sanitize(self.remote_url, self.treeinfo["filename"]),
relative_path=".treeinfo",
remote=self.remote,
deferred_download=False,
)
]
for path, checksum in self.treeinfo["download"]["images"].items():
artifact = Artifact(**checksum)
da = DeclarativeArtifact(
artifact=artifact,
url=urlpath_sanitize(self.remote_url, path),
relative_path=path,
remote=self.remote,
deferred_download=self.deferred_download,
)
d_artifacts.append(da)
tree_digest = f'{self.treeinfo["hash"]}-{self.repository.pulp_id}'
self.treeinfo["distribution_tree"]["digest"] = tree_digest
distribution_tree = DistributionTree(**self.treeinfo["distribution_tree"])
dc = DeclarativeContent(content=distribution_tree, d_artifacts=d_artifacts)
dc.extra_data = self.treeinfo
await self.put(dc)
async def parse_repository_metadata(self, repomd, metadata_results):
"""Parse repository metadata."""
needed_metadata = set(PACKAGE_REPODATA) - set(metadata_results.keys())
if needed_metadata:
raise FileNotFoundError(
_("XML file(s): {filenames} not found").format(filenames=", ".join(needed_metadata))
)
await self.parse_distribution_tree()
# modularity-parsing MUST COME BEFORE package-parsing!
# The only way to know if a package is 'modular' in a repo, is to
# know that it is referenced in modulemd.
modulemd_dcs = []
modulemd_result = metadata_results.get("modules", None)
modulemd_list = []
if modulemd_result:
# Need to check modules compression here because modules are parsed before
# all other metadata. And check for compression type of metadata few lines
# bellow only skip them if unsupported. If we cannot parse modulemd, package
# can't be flagged as 'modular' thus broken repository!
if modulemd_result.url.endswith("zck"):
raise TypeError(_("Modular data compressed with ZCK is not supported."))
(modulemd_dcs, modulemd_list) = await self.parse_modules_metadata(modulemd_result)
# **Now** we can successfully parse package-metadata
await self.parse_packages(
metadata_results["primary"],
metadata_results["filelists"],
metadata_results["other"],
modulemd_list=modulemd_list,
)
groups_list = []
comps_result = metadata_results.get("group", None)
if comps_result:
groups_list = await self.parse_packages_components(comps_result)
updateinfo_result = metadata_results.get("updateinfo", None)
if updateinfo_result:
await self.parse_advisories(updateinfo_result)
# now send modules and groups down the pipeline since all relations have been set up
for modulemd_dc in modulemd_dcs:
await self.put(modulemd_dc)
for group_dc in groups_list:
await self.put(group_dc)
record_types = (
set(PACKAGE_REPODATA)
| set(PACKAGE_DB_REPODATA)
| set(UPDATE_REPODATA)
| set(COMPS_REPODATA)
| set(MODULAR_REPODATA)
)
for record in repomd.records:
should_skip = False
if record.type not in record_types:
for suffix in ["_zck", "_gz", "_xz"]:
if suffix in record.type:
should_skip = True
if record.type in RepoMetadataFile.UNSUPPORTED_METADATA:
should_skip = True
if should_skip:
continue
sanitized_checksum_type = getattr(CHECKSUM_TYPES, record.checksum_type.upper())
file_data = {sanitized_checksum_type: record.checksum, "size": record.size}
da = DeclarativeArtifact(
artifact=Artifact(**file_data),
url=urlpath_sanitize(self.remote_url, record.location_href),
relative_path=record.location_href,
remote=self.remote,
deferred_download=False,
)
repo_metadata_file = RepoMetadataFile(
data_type=record.type,
checksum_type=sanitized_checksum_type,
checksum=record.checksum,
relative_path=record.location_href,
)
dc = DeclarativeContent(content=repo_metadata_file, d_artifacts=[da])
await self.put(dc)
async def parse_modules_metadata(self, modulemd_result):
"""
Parse modules' metadata which define what packages are built for specific releases.
Args:
modulemd_result(pulpcore.download.base.DownloadResult): downloaded modulemd file
"""
modulemd_all, defaults_all, obsoletes_all = parse_modular(modulemd_result.path)
modulemd_dcs = []
# Parsing modules happens all at one time, and from here on no useful work happens.
# So just report that it finished this stage.
modulemd_pb_data = {"message": "Parsed Modulemd", "code": "sync.parsing.modulemds"}
async with ProgressReport(**modulemd_pb_data) as modulemd_pb:
modulemd_total = len(modulemd_all)
modulemd_pb.total = modulemd_total
modulemd_pb.done = modulemd_total
for modulemd in modulemd_all:
modulemd_content = Modulemd(**modulemd)