diff --git a/swh/web/browse/snapshot_context.py b/swh/web/browse/snapshot_context.py
index 4395aeb2..81f89a81 100644
--- a/swh/web/browse/snapshot_context.py
+++ b/swh/web/browse/snapshot_context.py
@@ -1,1427 +1,1424 @@
# Copyright (C) 2018-2020 The Software Heritage developers
# See the AUTHORS file at the top-level directory of this distribution
# License: GNU Affero General Public License version 3, or any later version
# See top-level LICENSE file for more information
# Utility module for browsing the archive in a snapshot context.
from collections import defaultdict
-from copy import copy
from typing import Any, Dict, List, Optional, Tuple
from django.core.cache import cache
from django.shortcuts import render
from django.template.defaultfilters import filesizeformat
from django.utils.html import escape
from swh.model.identifiers import CONTENT, DIRECTORY, RELEASE, REVISION, SNAPSHOT, swhid
from swh.model.model import Snapshot
from swh.web.browse.utils import (
content_display_max_size,
format_log_entries,
gen_content_link,
gen_directory_link,
gen_release_link,
gen_revision_link,
gen_revision_log_link,
gen_revision_url,
gen_snapshot_link,
get_directory_entries,
get_readme_to_display,
prepare_content_for_display,
request_content,
)
from swh.web.common import archive, highlightjs
from swh.web.common.exc import BadInputExc, NotFoundExc
from swh.web.common.identifiers import get_swhids_info
from swh.web.common.origin_visits import get_origin_visit
from swh.web.common.typing import (
ContentMetadata,
DirectoryMetadata,
OriginInfo,
SnapshotBranchInfo,
SnapshotContext,
SnapshotReleaseInfo,
SWHObjectInfo,
)
from swh.web.common.utils import (
format_utc_iso_date,
gen_path_info,
reverse,
swh_object_icons,
)
from swh.web.config import get_config
_empty_snapshot_id = Snapshot(branches={}).id.hex()
def _get_branch(branches, branch_name, snapshot_id):
"""
Utility function to get a specific branch from a branches list.
Its purpose is to get the default HEAD branch as some software origin
(e.g those with svn type) does not have it. In that latter case, check
if there is a master branch instead and returns it.
"""
filtered_branches = [b for b in branches if b["name"] == branch_name]
if filtered_branches:
return filtered_branches[0]
elif branch_name == "HEAD":
filtered_branches = [b for b in branches if b["name"].endswith("master")]
if filtered_branches:
return filtered_branches[0]
elif branches:
return branches[0]
else:
# case where a large branches list has been truncated
snp = archive.lookup_snapshot(
snapshot_id,
branches_from=branch_name,
branches_count=1,
target_types=["revision", "alias"],
)
snp_branch, _ = process_snapshot_branches(snp)
if snp_branch and snp_branch[0]["name"] == branch_name:
branches.append(snp_branch[0])
return snp_branch[0]
def _get_release(releases, release_name, snapshot_id):
"""
Utility function to get a specific release from a releases list.
Returns None if the release can not be found in the list.
"""
filtered_releases = [r for r in releases if r["name"] == release_name]
if filtered_releases:
return filtered_releases[0]
else:
# case where a large branches list has been truncated
try:
# git origins have specific branches for releases
snp = archive.lookup_snapshot(
snapshot_id,
branches_from=f"refs/tags/{release_name}",
branches_count=1,
target_types=["release"],
)
except NotFoundExc:
snp = archive.lookup_snapshot(
snapshot_id,
branches_from=release_name,
branches_count=1,
target_types=["release"],
)
_, snp_release = process_snapshot_branches(snp)
if snp_release and snp_release[0]["name"] == release_name:
releases.append(snp_release[0])
return snp_release[0]
def _branch_not_found(
branch_type, branch, snapshot_id, snapshot_sizes, origin_info, timestamp, visit_id
):
"""
Utility function to raise an exception when a specified branch/release
can not be found.
"""
if branch_type == "branch":
branch_type = "Branch"
branch_type_plural = "branches"
target_type = "revision"
else:
branch_type = "Release"
branch_type_plural = "releases"
target_type = "release"
if snapshot_id and snapshot_sizes[target_type] == 0:
msg = "Snapshot with id %s has an empty list" " of %s!" % (
snapshot_id,
branch_type_plural,
)
elif snapshot_id:
msg = "%s %s for snapshot with id %s" " not found!" % (
branch_type,
branch,
snapshot_id,
)
elif visit_id and snapshot_sizes[target_type] == 0:
msg = (
"Origin with url %s"
" for visit with id %s has an empty list"
" of %s!" % (origin_info["url"], visit_id, branch_type_plural)
)
elif visit_id:
msg = (
"%s %s associated to visit with"
" id %s for origin with url %s"
" not found!" % (branch_type, branch, visit_id, origin_info["url"])
)
elif snapshot_sizes[target_type] == 0:
msg = (
"Origin with url %s"
" for visit with timestamp %s has an empty list"
" of %s!" % (origin_info["url"], timestamp, branch_type_plural)
)
else:
msg = (
"%s %s associated to visit with"
" timestamp %s for origin with "
"url %s not found!" % (branch_type, branch, timestamp, origin_info["url"])
)
raise NotFoundExc(escape(msg))
def process_snapshot_branches(
snapshot: Dict[str, Any]
) -> Tuple[List[SnapshotBranchInfo], List[SnapshotReleaseInfo]]:
"""
Process a dictionary describing snapshot branches: extract those
targeting revisions and releases, put them in two different lists,
then sort those lists in lexicographical order of the branches' names.
Args:
snapshot: A dict describing a snapshot as returned for instance by
:func:`swh.web.common.archive.lookup_snapshot`
Returns:
A tuple whose first member is the sorted list of branches
targeting revisions and second member the sorted list of branches
targeting releases
"""
snapshot_branches = snapshot["branches"]
branches: Dict[str, SnapshotBranchInfo] = {}
branch_aliases: Dict[str, str] = {}
releases: Dict[str, SnapshotReleaseInfo] = {}
revision_to_branch = defaultdict(set)
revision_to_release = defaultdict(set)
release_to_branch = defaultdict(set)
for branch_name, target in snapshot_branches.items():
if not target:
# FIXME: display branches with an unknown target anyway
continue
target_id = target["target"]
target_type = target["target_type"]
if target_type == "revision":
branches[branch_name] = SnapshotBranchInfo(
name=branch_name,
+ alias=False,
revision=target_id,
date=None,
directory=None,
message=None,
url=None,
)
revision_to_branch[target_id].add(branch_name)
elif target_type == "release":
release_to_branch[target_id].add(branch_name)
elif target_type == "alias":
branch_aliases[branch_name] = target_id
# FIXME: handle pointers to other object types
- def _add_release_info(branch, release):
+ def _add_release_info(branch, release, alias=False):
releases[branch] = SnapshotReleaseInfo(
name=release["name"],
+ alias=alias,
branch_name=branch,
date=format_utc_iso_date(release["date"]),
directory=None,
id=release["id"],
message=release["message"],
target_type=release["target_type"],
target=release["target"],
url=None,
)
- def _add_branch_info(branch, revision):
+ def _add_branch_info(branch, revision, alias=False):
branches[branch] = SnapshotBranchInfo(
name=branch,
+ alias=alias,
revision=revision["id"],
directory=revision["directory"],
date=format_utc_iso_date(revision["date"]),
message=revision["message"],
url=None,
)
releases_info = archive.lookup_release_multiple(release_to_branch.keys())
for release in releases_info:
if release is None:
continue
branches_to_update = release_to_branch[release["id"]]
for branch in branches_to_update:
_add_release_info(branch, release)
if release["target_type"] == "revision":
revision_to_release[release["target"]].update(branches_to_update)
revisions = archive.lookup_revision_multiple(
set(revision_to_branch.keys()) | set(revision_to_release.keys())
)
for revision in revisions:
if not revision:
continue
for branch in revision_to_branch[revision["id"]]:
_add_branch_info(branch, revision)
for release_id in revision_to_release[revision["id"]]:
releases[release_id]["directory"] = revision["directory"]
for branch_alias, branch_target in branch_aliases.items():
- if branch_target in branches:
- branches[branch_alias] = copy(branches[branch_target])
- else:
- snp = archive.lookup_snapshot(
- snapshot["id"], branches_from=branch_target, branches_count=1
- )
- if snp and branch_target in snp["branches"]:
+ resolved_alias = archive.lookup_snapshot_alias(snapshot["id"], branch_alias)
+ if resolved_alias is None:
+ continue
- if snp["branches"][branch_target] is None:
- continue
+ target_type = resolved_alias["target_type"]
+ target = resolved_alias["target"]
- target_type = snp["branches"][branch_target]["target_type"]
- target = snp["branches"][branch_target]["target"]
- if target_type == "revision":
- branches[branch_alias] = snp["branches"][branch_target]
- revision = archive.lookup_revision(target)
- _add_branch_info(branch_alias, revision)
- elif target_type == "release":
- release = archive.lookup_release(target)
- _add_release_info(branch_alias, release)
+ if target_type == "revision":
+ branches[branch_alias] = target
+ revision = archive.lookup_revision(target)
+ _add_branch_info(branch_alias, revision, alias=True)
+ elif target_type == "release":
+ release = archive.lookup_release(target)
+ _add_release_info(branch_alias, release, alias=True)
if branch_alias in branches:
branches[branch_alias]["name"] = branch_alias
ret_branches = list(sorted(branches.values(), key=lambda b: b["name"]))
ret_releases = list(sorted(releases.values(), key=lambda b: b["name"]))
return ret_branches, ret_releases
def get_snapshot_content(
snapshot_id: str,
) -> Tuple[List[SnapshotBranchInfo], List[SnapshotReleaseInfo]]:
"""Returns the lists of branches and releases
associated to a swh snapshot.
That list is put in cache in order to speedup the navigation
in the swh-web/browse ui.
.. warning:: At most 1000 branches contained in the snapshot
will be returned for performance reasons.
Args:
snapshot_id: hexadecimal representation of the snapshot identifier
Returns:
A tuple with two members. The first one is a list of dict describing
the snapshot branches. The second one is a list of dict describing the
snapshot releases.
Raises:
NotFoundExc if the snapshot does not exist
"""
cache_entry_id = "swh_snapshot_%s" % snapshot_id
cache_entry = cache.get(cache_entry_id)
if cache_entry:
return cache_entry["branches"], cache_entry["releases"]
branches: List[SnapshotBranchInfo] = []
releases: List[SnapshotReleaseInfo] = []
snapshot_content_max_size = get_config()["snapshot_content_max_size"]
if snapshot_id:
snapshot = archive.lookup_snapshot(
snapshot_id, branches_count=snapshot_content_max_size
)
branches, releases = process_snapshot_branches(snapshot)
cache.set(cache_entry_id, {"branches": branches, "releases": releases,})
return branches, releases
def get_origin_visit_snapshot(
origin_info: OriginInfo,
visit_ts: Optional[str] = None,
visit_id: Optional[int] = None,
snapshot_id: Optional[str] = None,
) -> Tuple[List[SnapshotBranchInfo], List[SnapshotReleaseInfo]]:
"""Returns the lists of branches and releases associated to an origin for
a given visit.
The visit is expressed by either:
* a snapshot identifier
* a timestamp, if no visit with that exact timestamp is found,
the closest one from the provided timestamp will be used.
If no visit parameter is provided, it returns the list of branches
found for the latest visit.
That list is put in cache in order to speedup the navigation
in the swh-web/browse ui.
.. warning:: At most 1000 branches contained in the snapshot
will be returned for performance reasons.
Args:
origin_info: a dict filled with origin information
visit_ts: an ISO 8601 datetime string to parse
visit_id: visit id for disambiguation in case several visits have
the same timestamp
snapshot_id: if provided, visit associated to the snapshot will be processed
Returns:
A tuple with two members. The first one is a list of dict describing
the origin branches for the given visit.
The second one is a list of dict describing the origin releases
for the given visit.
Raises:
NotFoundExc if the origin or its visit are not found
"""
visit_info = get_origin_visit(origin_info, visit_ts, visit_id, snapshot_id)
return get_snapshot_content(visit_info["snapshot"])
def get_snapshot_context(
snapshot_id: Optional[str] = None,
origin_url: Optional[str] = None,
timestamp: Optional[str] = None,
visit_id: Optional[int] = None,
branch_name: Optional[str] = None,
release_name: Optional[str] = None,
revision_id: Optional[str] = None,
path: Optional[str] = None,
browse_context: str = "directory",
) -> SnapshotContext:
"""
Utility function to compute relevant information when navigating
the archive in a snapshot context. The snapshot is either
referenced by its id or it will be retrieved from an origin visit.
Args:
snapshot_id: hexadecimal representation of a snapshot identifier
origin_url: an origin_url
timestamp: a datetime string for retrieving the closest
visit of the origin
visit_id: optional visit id for disambiguation in case
of several visits with the same timestamp
branch_name: optional branch name set when browsing the snapshot in
that scope (will default to "HEAD" if not provided)
release_name: optional release name set when browsing the snapshot in
that scope
revision_id: optional revision identifier set when browsing the snapshot in
that scope
path: optional path of the object currently browsed in the snapshot
browse_context: indicates which type of object is currently browsed
Returns:
A dict filled with snapshot context information.
Raises:
swh.web.common.exc.NotFoundExc: if no snapshot is found for the visit
of an origin.
"""
assert origin_url is not None or snapshot_id is not None
origin_info = None
visit_info = None
url_args = {}
query_params: Dict[str, Any] = {}
origin_visits_url = None
if origin_url:
if visit_id is not None:
query_params["visit_id"] = visit_id
elif snapshot_id is not None:
query_params["snapshot"] = snapshot_id
origin_info = archive.lookup_origin({"url": origin_url})
visit_info = get_origin_visit(origin_info, timestamp, visit_id, snapshot_id)
formatted_date = format_utc_iso_date(visit_info["date"])
visit_info["formatted_date"] = formatted_date
snapshot_id = visit_info["snapshot"]
if not snapshot_id:
raise NotFoundExc(
"No snapshot associated to the visit of origin "
"%s on %s" % (escape(origin_url), formatted_date)
)
# provided timestamp is not necessarily equals to the one
# of the retrieved visit, so get the exact one in order
# to use it in the urls generated below
if timestamp:
timestamp = visit_info["date"]
branches, releases = get_origin_visit_snapshot(
origin_info, timestamp, visit_id, snapshot_id
)
query_params["origin_url"] = origin_info["url"]
origin_visits_url = reverse(
"browse-origin-visits", query_params={"origin_url": origin_info["url"]}
)
if timestamp is not None:
query_params["timestamp"] = format_utc_iso_date(
timestamp, "%Y-%m-%dT%H:%M:%SZ"
)
visit_url = reverse("browse-origin-directory", query_params=query_params)
visit_info["url"] = visit_url
branches_url = reverse("browse-origin-branches", query_params=query_params)
releases_url = reverse("browse-origin-releases", query_params=query_params)
else:
assert snapshot_id is not None
branches, releases = get_snapshot_content(snapshot_id)
url_args = {"snapshot_id": snapshot_id}
branches_url = reverse("browse-snapshot-branches", url_args=url_args)
releases_url = reverse("browse-snapshot-releases", url_args=url_args)
releases = list(reversed(releases))
snapshot_sizes = archive.lookup_snapshot_sizes(snapshot_id)
- is_empty = sum(snapshot_sizes.values()) == 0
+ is_empty = (snapshot_sizes["release"] + snapshot_sizes["revision"]) == 0
swh_snp_id = swhid("snapshot", snapshot_id)
if visit_info:
timestamp = format_utc_iso_date(visit_info["date"])
if origin_info:
browse_view_name = f"browse-origin-{browse_context}"
else:
browse_view_name = f"browse-snapshot-{browse_context}"
release_id = None
root_directory = None
- snapshot_total_size = sum(snapshot_sizes.values())
+ snapshot_total_size = snapshot_sizes["release"] + snapshot_sizes["revision"]
if path is not None:
query_params["path"] = path
if snapshot_total_size and revision_id is not None:
revision = archive.lookup_revision(revision_id)
root_directory = revision["directory"]
branches.append(
SnapshotBranchInfo(
name=revision_id,
+ alias=False,
revision=revision_id,
directory=root_directory,
date=revision["date"],
message=revision["message"],
url=None,
)
)
branch_name = revision_id
query_params["revision"] = revision_id
elif snapshot_total_size and release_name:
release = _get_release(releases, release_name, snapshot_id)
if release is None:
_branch_not_found(
"release",
release_name,
snapshot_id,
snapshot_sizes,
origin_info,
timestamp,
visit_id,
)
else:
root_directory = release["directory"]
revision_id = release["target"]
release_id = release["id"]
query_params["release"] = release_name
elif snapshot_total_size:
if branch_name:
query_params["branch"] = branch_name
branch = _get_branch(branches, branch_name or "HEAD", snapshot_id)
if branch is None:
_branch_not_found(
"branch",
branch_name,
snapshot_id,
snapshot_sizes,
origin_info,
timestamp,
visit_id,
)
else:
branch_name = branch["name"]
revision_id = branch["revision"]
root_directory = branch["directory"]
for b in branches:
branch_query_params = dict(query_params)
branch_query_params.pop("release", None)
if b["name"] != b["revision"]:
branch_query_params.pop("revision", None)
branch_query_params["branch"] = b["name"]
b["url"] = reverse(
browse_view_name, url_args=url_args, query_params=branch_query_params
)
for r in releases:
release_query_params = dict(query_params)
release_query_params.pop("branch", None)
release_query_params.pop("revision", None)
release_query_params["release"] = r["name"]
r["url"] = reverse(
browse_view_name, url_args=url_args, query_params=release_query_params,
)
revision_info = None
if revision_id:
try:
revision_info = archive.lookup_revision(revision_id)
except NotFoundExc:
pass
else:
revision_info["date"] = format_utc_iso_date(revision_info["date"])
revision_info["committer_date"] = format_utc_iso_date(
revision_info["committer_date"]
)
if revision_info["message"]:
message_lines = revision_info["message"].split("\n")
revision_info["message_header"] = message_lines[0]
else:
revision_info["message_header"] = ""
snapshot_context = SnapshotContext(
branch=branch_name,
branches=branches,
branches_url=branches_url,
is_empty=is_empty,
origin_info=origin_info,
origin_visits_url=origin_visits_url,
release=release_name,
release_id=release_id,
query_params=query_params,
releases=releases,
releases_url=releases_url,
revision_id=revision_id,
revision_info=revision_info,
root_directory=root_directory,
snapshot_id=snapshot_id,
snapshot_sizes=snapshot_sizes,
snapshot_swhid=swh_snp_id,
url_args=url_args,
visit_info=visit_info,
)
if revision_info:
revision_info["revision_url"] = gen_revision_url(revision_id, snapshot_context)
return snapshot_context
def _build_breadcrumbs(snapshot_context: SnapshotContext, path: str):
origin_info = snapshot_context["origin_info"]
url_args = snapshot_context["url_args"]
query_params = dict(snapshot_context["query_params"])
root_directory = snapshot_context["root_directory"]
path_info = gen_path_info(path)
if origin_info:
browse_view_name = "browse-origin-directory"
else:
browse_view_name = "browse-snapshot-directory"
breadcrumbs = []
if root_directory:
query_params.pop("path", None)
breadcrumbs.append(
{
"name": root_directory[:7],
"url": reverse(
browse_view_name, url_args=url_args, query_params=query_params
),
}
)
for pi in path_info:
query_params["path"] = pi["path"]
breadcrumbs.append(
{
"name": pi["name"],
"url": reverse(
browse_view_name, url_args=url_args, query_params=query_params
),
}
)
return breadcrumbs
def _check_origin_url(snapshot_id, origin_url):
if snapshot_id is None and origin_url is None:
raise BadInputExc("An origin URL must be provided as query parameter.")
def browse_snapshot_directory(
request, snapshot_id=None, origin_url=None, timestamp=None, path=None
):
"""
Django view implementation for browsing a directory in a snapshot context.
"""
_check_origin_url(snapshot_id, origin_url)
snapshot_context = get_snapshot_context(
snapshot_id=snapshot_id,
origin_url=origin_url,
timestamp=timestamp,
visit_id=request.GET.get("visit_id"),
path=path,
browse_context="directory",
branch_name=request.GET.get("branch"),
release_name=request.GET.get("release"),
revision_id=request.GET.get("revision"),
)
root_directory = snapshot_context["root_directory"]
sha1_git = root_directory
if root_directory and path:
dir_info = archive.lookup_directory_with_path(root_directory, path)
sha1_git = dir_info["target"]
dirs = []
files = []
if sha1_git:
dirs, files = get_directory_entries(sha1_git)
origin_info = snapshot_context["origin_info"]
visit_info = snapshot_context["visit_info"]
url_args = snapshot_context["url_args"]
query_params = dict(snapshot_context["query_params"])
revision_id = snapshot_context["revision_id"]
snapshot_id = snapshot_context["snapshot_id"]
if origin_info:
browse_view_name = "browse-origin-directory"
else:
browse_view_name = "browse-snapshot-directory"
breadcrumbs = _build_breadcrumbs(snapshot_context, path)
path = "" if path is None else (path + "/")
for d in dirs:
if d["type"] == "rev":
d["url"] = reverse("browse-revision", url_args={"sha1_git": d["target"]})
else:
query_params["path"] = path + d["name"]
d["url"] = reverse(
browse_view_name, url_args=url_args, query_params=query_params
)
sum_file_sizes = 0
readmes = {}
if origin_info:
browse_view_name = "browse-origin-content"
else:
browse_view_name = "browse-snapshot-content"
for f in files:
query_params["path"] = path + f["name"]
f["url"] = reverse(
browse_view_name, url_args=url_args, query_params=query_params
)
if f["length"] is not None:
sum_file_sizes += f["length"]
f["length"] = filesizeformat(f["length"])
if f["name"].lower().startswith("readme"):
readmes[f["name"]] = f["checksums"]["sha1"]
readme_name, readme_url, readme_html = get_readme_to_display(readmes)
if origin_info:
browse_view_name = "browse-origin-log"
else:
browse_view_name = "browse-snapshot-log"
history_url = None
if snapshot_id != _empty_snapshot_id:
query_params.pop("path", None)
history_url = reverse(
browse_view_name, url_args=url_args, query_params=query_params
)
nb_files = None
nb_dirs = None
dir_path = None
if root_directory:
nb_files = len(files)
nb_dirs = len(dirs)
sum_file_sizes = filesizeformat(sum_file_sizes)
dir_path = "/" + path
browse_dir_link = gen_directory_link(sha1_git)
browse_rev_link = gen_revision_link(revision_id)
browse_snp_link = gen_snapshot_link(snapshot_id)
revision_found = True
if sha1_git is None and revision_id is not None:
try:
archive.lookup_revision(revision_id)
except NotFoundExc:
revision_found = False
swh_objects = [
SWHObjectInfo(object_type=DIRECTORY, object_id=sha1_git),
SWHObjectInfo(object_type=REVISION, object_id=revision_id),
SWHObjectInfo(object_type=SNAPSHOT, object_id=snapshot_id),
]
visit_date = None
visit_type = None
if visit_info:
visit_date = format_utc_iso_date(visit_info["date"])
visit_type = visit_info["type"]
release_id = snapshot_context["release_id"]
browse_rel_link = None
if release_id:
swh_objects.append(SWHObjectInfo(object_type=RELEASE, object_id=release_id))
browse_rel_link = gen_release_link(release_id)
dir_metadata = DirectoryMetadata(
object_type=DIRECTORY,
object_id=sha1_git,
directory=sha1_git,
directory_url=browse_dir_link,
nb_files=nb_files,
nb_dirs=nb_dirs,
sum_file_sizes=sum_file_sizes,
root_directory=root_directory,
path=dir_path,
revision=revision_id,
revision_found=revision_found,
revision_url=browse_rev_link,
release=release_id,
release_url=browse_rel_link,
snapshot=snapshot_id,
snapshot_url=browse_snp_link,
origin_url=origin_url,
visit_date=visit_date,
visit_type=visit_type,
)
vault_cooking = {
"directory_context": True,
"directory_id": sha1_git,
"revision_context": True,
"revision_id": revision_id,
}
swhids_info = get_swhids_info(swh_objects, snapshot_context, dir_metadata)
dir_path = "/".join([bc["name"] for bc in breadcrumbs]) + "/"
context_found = "snapshot: %s" % snapshot_context["snapshot_id"]
if origin_info:
context_found = "origin: %s" % origin_info["url"]
heading = "Directory - %s - %s - %s" % (
dir_path,
snapshot_context["branch"],
context_found,
)
top_right_link = None
if not snapshot_context["is_empty"]:
top_right_link = {
"url": history_url,
"icon": swh_object_icons["revisions history"],
"text": "History",
}
return render(
request,
"browse/directory.html",
{
"heading": heading,
"swh_object_name": "Directory",
"swh_object_metadata": dir_metadata,
"dirs": dirs,
"files": files,
"breadcrumbs": breadcrumbs if root_directory else [],
"top_right_link": top_right_link,
"readme_name": readme_name,
"readme_url": readme_url,
"readme_html": readme_html,
"snapshot_context": snapshot_context,
"vault_cooking": vault_cooking,
"show_actions": True,
"swhids_info": swhids_info,
},
)
def browse_snapshot_content(
request,
snapshot_id=None,
origin_url=None,
timestamp=None,
path=None,
selected_language=None,
):
"""
Django view implementation for browsing a content in a snapshot context.
"""
_check_origin_url(snapshot_id, origin_url)
if path is None:
raise BadInputExc("The path of a content must be given as query parameter.")
snapshot_context = get_snapshot_context(
snapshot_id=snapshot_id,
origin_url=origin_url,
timestamp=timestamp,
visit_id=request.GET.get("visit_id"),
path=path,
browse_context="content",
branch_name=request.GET.get("branch"),
release_name=request.GET.get("release"),
revision_id=request.GET.get("revision"),
)
root_directory = snapshot_context["root_directory"]
sha1_git = None
query_string = None
content_data = {}
directory_id = None
split_path = path.split("/")
filename = split_path[-1]
filepath = path[: -len(filename)]
if root_directory:
content_info = archive.lookup_directory_with_path(root_directory, path)
sha1_git = content_info["target"]
query_string = "sha1_git:" + sha1_git
content_data = request_content(query_string, raise_if_unavailable=False)
if filepath:
dir_info = archive.lookup_directory_with_path(root_directory, filepath)
directory_id = dir_info["target"]
else:
directory_id = root_directory
revision_id = snapshot_context["revision_id"]
origin_info = snapshot_context["origin_info"]
visit_info = snapshot_context["visit_info"]
snapshot_id = snapshot_context["snapshot_id"]
if content_data.get("raw_data") is not None:
content_display_data = prepare_content_for_display(
content_data["raw_data"], content_data["mimetype"], path
)
content_data.update(content_display_data)
# Override language with user-selected language
if selected_language is not None:
content_data["language"] = selected_language
available_languages = None
if content_data.get("mimetype") is not None and "text/" in content_data["mimetype"]:
available_languages = highlightjs.get_supported_languages()
breadcrumbs = _build_breadcrumbs(snapshot_context, filepath)
breadcrumbs.append({"name": filename, "url": None})
browse_content_link = gen_content_link(sha1_git)
content_raw_url = None
if query_string:
content_raw_url = reverse(
"browse-content-raw",
url_args={"query_string": query_string},
query_params={"filename": filename},
)
browse_rev_link = gen_revision_link(revision_id)
browse_dir_link = gen_directory_link(directory_id)
content_checksums = content_data.get("checksums", {})
swh_objects = [
SWHObjectInfo(object_type=CONTENT, object_id=content_checksums.get("sha1_git")),
SWHObjectInfo(object_type=DIRECTORY, object_id=directory_id),
SWHObjectInfo(object_type=REVISION, object_id=revision_id),
SWHObjectInfo(object_type=SNAPSHOT, object_id=snapshot_id),
]
visit_date = None
visit_type = None
if visit_info:
visit_date = format_utc_iso_date(visit_info["date"])
visit_type = visit_info["type"]
release_id = snapshot_context["release_id"]
browse_rel_link = None
if release_id:
swh_objects.append(SWHObjectInfo(object_type=RELEASE, object_id=release_id))
browse_rel_link = gen_release_link(release_id)
content_metadata = ContentMetadata(
object_type=CONTENT,
object_id=content_checksums.get("sha1_git"),
sha1=content_checksums.get("sha1"),
sha1_git=content_checksums.get("sha1_git"),
sha256=content_checksums.get("sha256"),
blake2s256=content_checksums.get("blake2s256"),
content_url=browse_content_link,
mimetype=content_data.get("mimetype"),
encoding=content_data.get("encoding"),
size=filesizeformat(content_data.get("length", 0)),
language=content_data.get("language"),
root_directory=root_directory,
path=f"/{filepath}",
filename=filename,
directory=directory_id,
directory_url=browse_dir_link,
revision=revision_id,
revision_url=browse_rev_link,
release=release_id,
release_url=browse_rel_link,
snapshot=snapshot_id,
snapshot_url=gen_snapshot_link(snapshot_id),
origin_url=origin_url,
visit_date=visit_date,
visit_type=visit_type,
)
swhids_info = get_swhids_info(swh_objects, snapshot_context, content_metadata)
content_path = "/".join([bc["name"] for bc in breadcrumbs])
context_found = "snapshot: %s" % snapshot_context["snapshot_id"]
if origin_info:
context_found = "origin: %s" % origin_info["url"]
heading = "Content - %s - %s - %s" % (
content_path,
snapshot_context["branch"],
context_found,
)
top_right_link = None
if not snapshot_context["is_empty"]:
top_right_link = {
"url": content_raw_url,
"icon": swh_object_icons["content"],
"text": "Raw File",
}
return render(
request,
"browse/content.html",
{
"heading": heading,
"swh_object_name": "Content",
"swh_object_metadata": content_metadata,
"content": content_data.get("content_data"),
"content_size": content_data.get("length"),
"max_content_size": content_display_max_size,
"filename": filename,
"encoding": content_data.get("encoding"),
"mimetype": content_data.get("mimetype"),
"language": content_data.get("language"),
"available_languages": available_languages,
"breadcrumbs": breadcrumbs if root_directory else [],
"top_right_link": top_right_link,
"snapshot_context": snapshot_context,
"vault_cooking": None,
"show_actions": True,
"swhids_info": swhids_info,
"error_code": content_data.get("error_code"),
"error_message": content_data.get("error_message"),
"error_description": content_data.get("error_description"),
},
status=content_data.get("error_code", 200),
)
PER_PAGE = 100
def browse_snapshot_log(request, snapshot_id=None, origin_url=None, timestamp=None):
"""
Django view implementation for browsing a revision history in a
snapshot context.
"""
_check_origin_url(snapshot_id, origin_url)
snapshot_context = get_snapshot_context(
snapshot_id=snapshot_id,
origin_url=origin_url,
timestamp=timestamp,
visit_id=request.GET.get("visit_id"),
browse_context="log",
branch_name=request.GET.get("branch"),
release_name=request.GET.get("release"),
revision_id=request.GET.get("revision"),
)
revision_id = snapshot_context["revision_id"]
per_page = int(request.GET.get("per_page", PER_PAGE))
offset = int(request.GET.get("offset", 0))
revs_ordering = request.GET.get("revs_ordering", "committer_date")
session_key = "rev_%s_log_ordering_%s" % (revision_id, revs_ordering)
rev_log_session = request.session.get(session_key, None)
rev_log = []
revs_walker_state = None
if rev_log_session:
rev_log = rev_log_session["rev_log"]
revs_walker_state = rev_log_session["revs_walker_state"]
if len(rev_log) < offset + per_page:
revs_walker = archive.get_revisions_walker(
revs_ordering,
revision_id,
max_revs=offset + per_page + 1,
state=revs_walker_state,
)
rev_log += [rev["id"] for rev in revs_walker]
revs_walker_state = revs_walker.export_state()
revs = rev_log[offset : offset + per_page]
revision_log = archive.lookup_revision_multiple(revs)
request.session[session_key] = {
"rev_log": rev_log,
"revs_walker_state": revs_walker_state,
}
origin_info = snapshot_context["origin_info"]
visit_info = snapshot_context["visit_info"]
url_args = snapshot_context["url_args"]
query_params = snapshot_context["query_params"]
snapshot_id = snapshot_context["snapshot_id"]
query_params["per_page"] = per_page
revs_ordering = request.GET.get("revs_ordering", "")
query_params["revs_ordering"] = revs_ordering or None
if origin_info:
browse_view_name = "browse-origin-log"
else:
browse_view_name = "browse-snapshot-log"
prev_log_url = None
if len(rev_log) > offset + per_page:
query_params["offset"] = offset + per_page
prev_log_url = reverse(
browse_view_name, url_args=url_args, query_params=query_params
)
next_log_url = None
if offset != 0:
query_params["offset"] = offset - per_page
next_log_url = reverse(
browse_view_name, url_args=url_args, query_params=query_params
)
revision_log_data = format_log_entries(revision_log, per_page, snapshot_context)
browse_rev_link = gen_revision_link(revision_id)
browse_log_link = gen_revision_log_link(revision_id)
browse_snp_link = gen_snapshot_link(snapshot_id)
revision_metadata = {
"context-independent revision": browse_rev_link,
"context-independent revision history": browse_log_link,
"context-independent snapshot": browse_snp_link,
"snapshot": snapshot_id,
}
if origin_info:
revision_metadata["origin url"] = origin_info["url"]
revision_metadata["origin visit date"] = format_utc_iso_date(visit_info["date"])
revision_metadata["origin visit type"] = visit_info["type"]
swh_objects = [
SWHObjectInfo(object_type=REVISION, object_id=revision_id),
SWHObjectInfo(object_type=SNAPSHOT, object_id=snapshot_id),
]
release_id = snapshot_context["release_id"]
if release_id:
swh_objects.append(SWHObjectInfo(object_type=RELEASE, object_id=release_id))
browse_rel_link = gen_release_link(release_id)
revision_metadata["release"] = release_id
revision_metadata["context-independent release"] = browse_rel_link
swhids_info = get_swhids_info(swh_objects, snapshot_context)
context_found = "snapshot: %s" % snapshot_context["snapshot_id"]
if origin_info:
context_found = "origin: %s" % origin_info["url"]
heading = "Revision history - %s - %s" % (snapshot_context["branch"], context_found)
return render(
request,
"browse/revision-log.html",
{
"heading": heading,
"swh_object_name": "Revisions history",
"swh_object_metadata": revision_metadata,
"revision_log": revision_log_data,
"revs_ordering": revs_ordering,
"next_log_url": next_log_url,
"prev_log_url": prev_log_url,
"breadcrumbs": None,
"top_right_link": None,
"snapshot_context": snapshot_context,
"vault_cooking": None,
"show_actions": True,
"swhids_info": swhids_info,
},
)
def browse_snapshot_branches(
request, snapshot_id=None, origin_url=None, timestamp=None
):
"""
Django view implementation for browsing a list of branches in a snapshot
context.
"""
_check_origin_url(snapshot_id, origin_url)
snapshot_context = get_snapshot_context(
snapshot_id=snapshot_id,
origin_url=origin_url,
timestamp=timestamp,
visit_id=request.GET.get("visit_id"),
)
branches_bc = request.GET.get("branches_breadcrumbs", "")
branches_bc = branches_bc.split(",") if branches_bc else []
branches_from = branches_bc[-1] if branches_bc else ""
origin_info = snapshot_context["origin_info"]
url_args = snapshot_context["url_args"]
query_params = snapshot_context["query_params"]
if origin_info:
browse_view_name = "browse-origin-directory"
else:
browse_view_name = "browse-snapshot-directory"
snapshot = archive.lookup_snapshot(
snapshot_context["snapshot_id"],
branches_from,
PER_PAGE + 1,
target_types=["revision", "alias"],
)
displayed_branches, _ = process_snapshot_branches(snapshot)
for branch in displayed_branches:
rev_query_params = {}
if origin_info:
rev_query_params["origin_url"] = origin_info["url"]
revision_url = reverse(
"browse-revision",
url_args={"sha1_git": branch["revision"]},
query_params=query_params,
)
query_params["branch"] = branch["name"]
directory_url = reverse(
browse_view_name, url_args=url_args, query_params=query_params
)
del query_params["branch"]
branch["revision_url"] = revision_url
branch["directory_url"] = directory_url
if origin_info:
browse_view_name = "browse-origin-branches"
else:
browse_view_name = "browse-snapshot-branches"
prev_branches_url = None
next_branches_url = None
if branches_bc:
query_params_prev = dict(query_params)
query_params_prev["branches_breadcrumbs"] = ",".join(branches_bc[:-1])
prev_branches_url = reverse(
browse_view_name, url_args=url_args, query_params=query_params_prev
)
elif branches_from:
prev_branches_url = reverse(
browse_view_name, url_args=url_args, query_params=query_params
)
if snapshot["next_branch"] is not None:
query_params_next = dict(query_params)
next_branch = displayed_branches[-1]["name"]
del displayed_branches[-1]
branches_bc.append(next_branch)
query_params_next["branches_breadcrumbs"] = ",".join(branches_bc)
next_branches_url = reverse(
browse_view_name, url_args=url_args, query_params=query_params_next
)
heading = "Branches - "
if origin_info:
heading += "origin: %s" % origin_info["url"]
else:
heading += "snapshot: %s" % snapshot_id
return render(
request,
"browse/branches.html",
{
"heading": heading,
"swh_object_name": "Branches",
"swh_object_metadata": {},
"top_right_link": None,
"displayed_branches": displayed_branches,
"prev_branches_url": prev_branches_url,
"next_branches_url": next_branches_url,
"snapshot_context": snapshot_context,
},
)
def browse_snapshot_releases(
request, snapshot_id=None, origin_url=None, timestamp=None
):
"""
Django view implementation for browsing a list of releases in a snapshot
context.
"""
_check_origin_url(snapshot_id, origin_url)
snapshot_context = get_snapshot_context(
snapshot_id=snapshot_id,
origin_url=origin_url,
timestamp=timestamp,
visit_id=request.GET.get("visit_id"),
)
rel_bc = request.GET.get("releases_breadcrumbs", "")
rel_bc = rel_bc.split(",") if rel_bc else []
rel_from = rel_bc[-1] if rel_bc else ""
origin_info = snapshot_context["origin_info"]
url_args = snapshot_context["url_args"]
query_params = snapshot_context["query_params"]
snapshot = archive.lookup_snapshot(
snapshot_context["snapshot_id"],
rel_from,
PER_PAGE + 1,
target_types=["release", "alias"],
)
_, displayed_releases = process_snapshot_branches(snapshot)
for release in displayed_releases:
query_params_tgt = {"snapshot": snapshot_id}
if origin_info:
query_params_tgt["origin_url"] = origin_info["url"]
release_url = reverse(
"browse-release",
url_args={"sha1_git": release["id"]},
query_params=query_params_tgt,
)
target_url = ""
if release["target_type"] == "revision":
target_url = reverse(
"browse-revision",
url_args={"sha1_git": release["target"]},
query_params=query_params_tgt,
)
elif release["target_type"] == "directory":
target_url = reverse(
"browse-directory",
url_args={"sha1_git": release["target"]},
query_params=query_params_tgt,
)
elif release["target_type"] == "content":
target_url = reverse(
"browse-content",
url_args={"query_string": release["target"]},
query_params=query_params_tgt,
)
elif release["target_type"] == "release":
target_url = reverse(
"browse-release",
url_args={"sha1_git": release["target"]},
query_params=query_params_tgt,
)
release["release_url"] = release_url
release["target_url"] = target_url
if origin_info:
browse_view_name = "browse-origin-releases"
else:
browse_view_name = "browse-snapshot-releases"
prev_releases_url = None
next_releases_url = None
if rel_bc:
query_params_prev = dict(query_params)
query_params_prev["releases_breadcrumbs"] = ",".join(rel_bc[:-1])
prev_releases_url = reverse(
browse_view_name, url_args=url_args, query_params=query_params_prev
)
elif rel_from:
prev_releases_url = reverse(
browse_view_name, url_args=url_args, query_params=query_params
)
if snapshot["next_branch"] is not None:
query_params_next = dict(query_params)
next_rel = displayed_releases[-1]["branch_name"]
del displayed_releases[-1]
rel_bc.append(next_rel)
query_params_next["releases_breadcrumbs"] = ",".join(rel_bc)
next_releases_url = reverse(
browse_view_name, url_args=url_args, query_params=query_params_next
)
heading = "Releases - "
if origin_info:
heading += "origin: %s" % origin_info["url"]
else:
heading += "snapshot: %s" % snapshot_id
return render(
request,
"browse/releases.html",
{
"heading": heading,
"top_panel_visible": False,
"top_panel_collapsible": False,
"swh_object_name": "Releases",
"swh_object_metadata": {},
"top_right_link": None,
"displayed_releases": displayed_releases,
"prev_releases_url": prev_releases_url,
"next_releases_url": next_releases_url,
"snapshot_context": snapshot_context,
"vault_cooking": None,
"show_actions": False,
},
)
diff --git a/swh/web/common/archive.py b/swh/web/common/archive.py
index 6dc8d524..ad86c60c 100644
--- a/swh/web/common/archive.py
+++ b/swh/web/common/archive.py
@@ -1,1363 +1,1364 @@
# Copyright (C) 2015-2020 The Software Heritage developers
# See the AUTHORS file at the top-level directory of this distribution
# License: GNU Affero General Public License version 3, or any later version
# See top-level LICENSE file for more information
from collections import defaultdict
import itertools
import os
import re
from typing import Any, Dict, Iterable, Iterator, List, Optional, Set, Tuple, Union
from swh.model import hashutil
from swh.model.identifiers import CONTENT, DIRECTORY, RELEASE, REVISION, SNAPSHOT
from swh.model.model import OriginVisit, Revision
from swh.storage.algos import diff, revisions_walker
from swh.storage.algos.origin import origin_get_latest_visit_status
-from swh.storage.algos.snapshot import snapshot_get_latest
+from swh.storage.algos.snapshot import snapshot_get_latest, snapshot_resolve_alias
from swh.vault.exc import NotFoundExc as VaultNotFoundExc
from swh.web import config
from swh.web.common import converters, query
from swh.web.common.exc import BadInputExc, NotFoundExc
from swh.web.common.typing import (
OriginInfo,
OriginMetadataInfo,
OriginVisitInfo,
PagedResult,
)
search = config.search()
storage = config.storage()
vault = config.vault()
idx_storage = config.indexer_storage()
MAX_LIMIT = 50 # Top limit the users can ask for
def _first_element(lst):
"""Returns the first element in the provided list or None
if it is empty or None"""
return next(iter(lst or []), None)
def lookup_multiple_hashes(hashes):
"""Lookup the passed hashes in a single DB connection, using batch
processing.
Args:
An array of {filename: X, sha1: Y}, string X, hex sha1 string Y.
Returns:
The same array with elements updated with elem['found'] = true if
the hash is present in storage, elem['found'] = false if not.
"""
hashlist = [hashutil.hash_to_bytes(elem["sha1"]) for elem in hashes]
content_missing = storage.content_missing_per_sha1(hashlist)
missing = [hashutil.hash_to_hex(x) for x in content_missing]
for x in hashes:
x.update({"found": True})
for h in hashes:
if h["sha1"] in missing:
h["found"] = False
return hashes
def lookup_expression(expression, last_sha1, per_page):
"""Lookup expression in raw content.
Args:
expression (str): An expression to lookup through raw indexed
content
last_sha1 (str): Last sha1 seen
per_page (int): Number of results per page
Yields:
ctags whose content match the expression
"""
limit = min(per_page, MAX_LIMIT)
ctags = idx_storage.content_ctags_search(
expression, last_sha1=last_sha1, limit=limit
)
for ctag in ctags:
ctag = converters.from_swh(ctag, hashess={"id"})
ctag["sha1"] = ctag["id"]
ctag.pop("id")
yield ctag
def lookup_hash(q: str) -> Dict[str, Any]:
"""Check if the storage contains a given content checksum and return it if found.
Args:
q: query string of the form <hash_algo:hash>
Returns:
Dict with key found containing the hash info if the
hash is present, None if not.
"""
algo, hash_ = query.parse_hash(q)
found = _first_element(storage.content_find({algo: hash_}))
if found:
content = converters.from_content(found.to_dict())
else:
content = None
return {"found": content, "algo": algo}
def search_hash(q: str) -> Dict[str, bool]:
"""Search storage for a given content checksum.
Args:
q: query string of the form <hash_algo:hash>
Returns:
Dict with key found to True or False, according to
whether the checksum is present or not
"""
algo, hash_ = query.parse_hash(q)
found = _first_element(storage.content_find({algo: hash_}))
return {"found": found is not None}
def _lookup_content_sha1(q: str) -> Optional[bytes]:
"""Given a possible input, query for the content's sha1.
Args:
q: query string of the form <hash_algo:hash>
Returns:
binary sha1 if found or None
"""
algo, hash_ = query.parse_hash(q)
if algo != "sha1":
hashes = _first_element(storage.content_find({algo: hash_}))
if not hashes:
return None
return hashes.sha1
return hash_
def lookup_content_ctags(q):
"""Return ctags information from a specified content.
Args:
q: query string of the form <hash_algo:hash>
Yields:
ctags information (dict) list if the content is found.
"""
sha1 = _lookup_content_sha1(q)
if not sha1:
return None
ctags = list(idx_storage.content_ctags_get([sha1]))
if not ctags:
return None
for ctag in ctags:
yield converters.from_swh(ctag, hashess={"id"})
def lookup_content_filetype(q):
"""Return filetype information from a specified content.
Args:
q: query string of the form <hash_algo:hash>
Yields:
filetype information (dict) list if the content is found.
"""
sha1 = _lookup_content_sha1(q)
if not sha1:
return None
filetype = _first_element(list(idx_storage.content_mimetype_get([sha1])))
if not filetype:
return None
return converters.from_filetype(filetype.to_dict())
def lookup_content_language(q):
"""Always returns None.
This used to return language information from a specified content,
but this is currently disabled.
Args:
q: query string of the form <hash_algo:hash>
Yields:
language information (dict) list if the content is found.
"""
return None
def lookup_content_license(q):
"""Return license information from a specified content.
Args:
q: query string of the form <hash_algo:hash>
Yields:
license information (dict) list if the content is found.
"""
sha1 = _lookup_content_sha1(q)
if not sha1:
return None
licenses = list(idx_storage.content_fossology_license_get([sha1]))
if not licenses:
return None
license_dicts = [license.to_dict() for license in licenses]
for license_dict in license_dicts:
del license_dict["id"]
lic = {
"id": sha1,
"facts": license_dicts,
}
return converters.from_swh(lic, hashess={"id"})
def lookup_origin(origin: OriginInfo) -> OriginInfo:
"""Return information about the origin matching dict origin.
Args:
origin: origin's dict with 'url' key
Returns:
origin information as dict.
"""
origin_urls = [origin["url"]]
if origin["url"]:
# handle case when user provided an origin url with a trailing
# slash while the url in storage does not have it (e.g. GitHub)
if origin["url"].endswith("/"):
origin_urls.append(origin["url"][:-1])
# handle case when user provided an origin url without a trailing
# slash while the url in storage have it (e.g. Debian source package)
else:
origin_urls.append(f"{origin['url']}/")
origins = [o for o in storage.origin_get(origin_urls) if o is not None]
if not origins:
msg = "Origin with url %s not found!" % origin["url"]
raise NotFoundExc(msg)
return converters.from_origin(origins[0].to_dict())
def lookup_origins(
page_token: Optional[str], limit: int = 100
) -> PagedResult[OriginInfo]:
"""Get list of archived software origins in a paginated way.
Origins are sorted by id before returning them
Args:
origin_from (int): The minimum id of the origins to return
origin_count (int): The maximum number of origins to return
Returns:
Page of OriginInfo
"""
page = storage.origin_list(page_token=page_token, limit=limit)
return PagedResult(
[converters.from_origin(o.to_dict()) for o in page.results],
next_page_token=page.next_page_token,
)
def search_origin(
url_pattern: str,
limit: int = 50,
with_visit: bool = False,
page_token: Optional[str] = None,
) -> Tuple[List[OriginInfo], Optional[str]]:
"""Search for origins whose urls contain a provided string pattern
or match a provided regular expression.
Args:
url_pattern: the string pattern to search for in origin urls
limit: the maximum number of found origins to return
page_token: opaque string used to get the next results of a search
Returns:
list of origin information as dict.
"""
if page_token:
assert isinstance(page_token, str)
if search:
page_result = search.origin_search(
url_pattern=url_pattern,
page_token=page_token,
with_visit=with_visit,
limit=limit,
)
origins = [converters.from_origin(ori_dict) for ori_dict in page_result.results]
else:
# Fallback to swh-storage if swh-search is not configured
search_words = [re.escape(word) for word in url_pattern.split()]
if len(search_words) >= 7:
url_pattern = ".*".join(search_words)
else:
pattern_parts = []
for permut in itertools.permutations(search_words):
pattern_parts.append(".*".join(permut))
url_pattern = "|".join(pattern_parts)
page_result = storage.origin_search(
url_pattern,
page_token=page_token,
with_visit=with_visit,
limit=limit,
regexp=True,
)
origins = [converters.from_origin(ori.to_dict()) for ori in page_result.results]
return (origins, page_result.next_page_token)
def search_origin_metadata(
fulltext: str, limit: int = 50
) -> Iterable[OriginMetadataInfo]:
"""Search for origins whose metadata match a provided string pattern.
Args:
fulltext: the string pattern to search for in origin metadata
limit: the maximum number of found origins to return
Returns:
Iterable of origin metadata information for existing origins
"""
matches = idx_storage.origin_intrinsic_metadata_search_fulltext(
conjunction=[fulltext], limit=limit
)
matches = [match.to_dict() for match in matches]
results = []
origins = storage.origin_get([match["id"] for match in matches])
for origin, match in zip(origins, matches):
if not origin:
continue
match["from_revision"] = hashutil.hash_to_hex(match["from_revision"])
del match["id"]
results.append(OriginMetadataInfo(url=origin.url, metadata=match))
return results
def lookup_origin_intrinsic_metadata(origin_url: str) -> Dict[str, Any]:
"""Return intrinsic metadata for origin whose origin matches given
origin.
Args:
origin_url: origin url
Raises:
NotFoundExc when the origin is not found
Returns:
origin metadata.
"""
origins = [origin_url]
origin_info = storage.origin_get(origins)[0]
if not origin_info:
raise NotFoundExc(f"Origin with url {origin_url} not found!")
match = _first_element(idx_storage.origin_intrinsic_metadata_get(origins))
result = {}
if match:
result = match.metadata
return result
def _to_sha1_bin(sha1_hex):
_, sha1_git_bin = query.parse_hash_with_algorithms_or_throws(
sha1_hex, ["sha1"], "Only sha1_git is supported." # HACK: sha1_git really
)
return sha1_git_bin
def _check_directory_exists(sha1_git, sha1_git_bin):
if len(list(storage.directory_missing([sha1_git_bin]))):
raise NotFoundExc("Directory with sha1_git %s not found" % sha1_git)
def lookup_directory(sha1_git):
"""Return information about the directory with id sha1_git.
Args:
sha1_git as string
Returns:
directory information as dict.
"""
empty_dir_sha1 = "4b825dc642cb6eb9a060e54bf8d69288fbee4904"
if sha1_git == empty_dir_sha1:
return []
sha1_git_bin = _to_sha1_bin(sha1_git)
_check_directory_exists(sha1_git, sha1_git_bin)
directory_entries = storage.directory_ls(sha1_git_bin)
return map(converters.from_directory_entry, directory_entries)
def lookup_directory_with_path(sha1_git, path_string):
"""Return directory information for entry with path path_string w.r.t.
root directory pointed by directory_sha1_git
Args:
- directory_sha1_git: sha1_git corresponding to the directory
to which we append paths to (hopefully) find the entry
- the relative path to the entry starting from the directory pointed by
directory_sha1_git
Raises:
NotFoundExc if the directory entry is not found
"""
sha1_git_bin = _to_sha1_bin(sha1_git)
_check_directory_exists(sha1_git, sha1_git_bin)
paths = path_string.strip(os.path.sep).split(os.path.sep)
queried_dir = storage.directory_entry_get_by_path(
sha1_git_bin, list(map(lambda p: p.encode("utf-8"), paths))
)
if not queried_dir:
raise NotFoundExc(
("Directory entry with path %s from %s not found") % (path_string, sha1_git)
)
return converters.from_directory_entry(queried_dir)
def lookup_release(release_sha1_git: str) -> Dict[str, Any]:
"""Return information about the release with sha1 release_sha1_git.
Args:
release_sha1_git: The release's sha1 as hexadecimal
Returns:
Release information as dict.
Raises:
ValueError if the identifier provided is not of sha1 nature.
"""
sha1_git_bin = _to_sha1_bin(release_sha1_git)
release = _first_element(storage.release_get([sha1_git_bin]))
if not release:
raise NotFoundExc(f"Release with sha1_git {release_sha1_git} not found.")
return converters.from_release(release)
def lookup_release_multiple(sha1_git_list) -> Iterator[Optional[Dict[str, Any]]]:
"""Return information about the releases identified with
their sha1_git identifiers.
Args:
sha1_git_list: A list of release sha1_git identifiers
Returns:
Iterator of Release metadata information as dict.
Raises:
ValueError if the identifier provided is not of sha1 nature.
"""
sha1_bin_list = [_to_sha1_bin(sha1_git) for sha1_git in sha1_git_list]
releases = storage.release_get(sha1_bin_list)
for r in releases:
if r is not None:
yield converters.from_release(r)
else:
yield None
def lookup_revision(rev_sha1_git) -> Dict[str, Any]:
"""Return information about the revision with sha1 revision_sha1_git.
Args:
revision_sha1_git: The revision's sha1 as hexadecimal
Returns:
Revision information as dict.
Raises:
ValueError if the identifier provided is not of sha1 nature.
NotFoundExc if there is no revision with the provided sha1_git.
"""
sha1_git_bin = _to_sha1_bin(rev_sha1_git)
revision = storage.revision_get([sha1_git_bin])[0]
if not revision:
raise NotFoundExc(f"Revision with sha1_git {rev_sha1_git} not found.")
return converters.from_revision(revision)
def lookup_revision_multiple(sha1_git_list) -> Iterator[Optional[Dict[str, Any]]]:
"""Return information about the revisions identified with
their sha1_git identifiers.
Args:
sha1_git_list: A list of revision sha1_git identifiers
Yields:
revision information as dict if the revision exists, None otherwise.
Raises:
ValueError if the identifier provided is not of sha1 nature.
"""
sha1_bin_list = [_to_sha1_bin(sha1_git) for sha1_git in sha1_git_list]
revisions = storage.revision_get(sha1_bin_list)
for revision in revisions:
if revision is not None:
yield converters.from_revision(revision)
else:
yield None
def lookup_revision_message(rev_sha1_git) -> Dict[str, bytes]:
"""Return the raw message of the revision with sha1 revision_sha1_git.
Args:
revision_sha1_git: The revision's sha1 as hexadecimal
Returns:
Decoded revision message as dict {'message': <the_message>}
Raises:
ValueError if the identifier provided is not of sha1 nature.
NotFoundExc if the revision is not found, or if it has no message
"""
sha1_git_bin = _to_sha1_bin(rev_sha1_git)
revision = storage.revision_get([sha1_git_bin])[0]
if not revision:
raise NotFoundExc(f"Revision with sha1_git {rev_sha1_git} not found.")
if not revision.message:
raise NotFoundExc(f"No message for revision with sha1_git {rev_sha1_git}.")
return {"message": revision.message}
def _lookup_revision_id_by(origin, branch_name, timestamp):
def _get_snapshot_branch(snapshot, branch_name):
snapshot = lookup_snapshot(
visit["snapshot"], branches_from=branch_name, branches_count=10
)
branch = None
if branch_name in snapshot["branches"]:
branch = snapshot["branches"][branch_name]
return branch
if isinstance(origin, int):
origin = {"id": origin}
elif isinstance(origin, str):
origin = {"url": origin}
else:
raise TypeError('"origin" must be an int or a string.')
from swh.web.common.origin_visits import get_origin_visit
visit = get_origin_visit(origin, visit_ts=timestamp)
branch = _get_snapshot_branch(visit["snapshot"], branch_name)
rev_id = None
if branch and branch["target_type"] == "revision":
rev_id = branch["target"]
elif branch and branch["target_type"] == "alias":
branch = _get_snapshot_branch(visit["snapshot"], branch["target"])
if branch and branch["target_type"] == "revision":
rev_id = branch["target"]
if not rev_id:
raise NotFoundExc(
"Revision for origin %s and branch %s not found."
% (origin.get("url"), branch_name)
)
return rev_id
def lookup_revision_by(origin, branch_name="HEAD", timestamp=None):
"""Lookup revision by origin, snapshot branch name and visit timestamp.
If branch_name is not provided, lookup using 'HEAD' as default.
If timestamp is not provided, use the most recent.
Args:
origin (Union[int,str]): origin of the revision
branch_name (str): snapshot branch name
timestamp (str/int): origin visit time frame
Returns:
dict: The revision matching the criterions
Raises:
NotFoundExc if no revision corresponds to the criterion
"""
rev_id = _lookup_revision_id_by(origin, branch_name, timestamp)
return lookup_revision(rev_id)
def lookup_revision_log(rev_sha1_git, limit):
"""Lookup revision log by revision id.
Args:
rev_sha1_git (str): The revision's sha1 as hexadecimal
limit (int): the maximum number of revisions returned
Returns:
list: Revision log as list of revision dicts
Raises:
ValueError: if the identifier provided is not of sha1 nature.
swh.web.common.exc.NotFoundExc: if there is no revision with the
provided sha1_git.
"""
lookup_revision(rev_sha1_git)
sha1_git_bin = _to_sha1_bin(rev_sha1_git)
revision_entries = storage.revision_log([sha1_git_bin], limit)
return map(converters.from_revision, revision_entries)
def lookup_revision_log_by(origin, branch_name, timestamp, limit):
"""Lookup revision by origin, snapshot branch name and visit timestamp.
Args:
origin (Union[int,str]): origin of the revision
branch_name (str): snapshot branch
timestamp (str/int): origin visit time frame
limit (int): the maximum number of revisions returned
Returns:
list: Revision log as list of revision dicts
Raises:
swh.web.common.exc.NotFoundExc: if no revision corresponds to the
criterion
"""
rev_id = _lookup_revision_id_by(origin, branch_name, timestamp)
return lookup_revision_log(rev_id, limit)
def lookup_revision_with_context_by(
origin, branch_name, timestamp, sha1_git, limit=100
):
"""Return information about revision sha1_git, limited to the
sub-graph of all transitive parents of sha1_git_root.
sha1_git_root being resolved through the lookup of a revision by origin,
branch_name and ts.
In other words, sha1_git is an ancestor of sha1_git_root.
Args:
- origin: origin of the revision.
- branch_name: revision's branch.
- timestamp: revision's time frame.
- sha1_git: one of sha1_git_root's ancestors.
- limit: limit the lookup to 100 revisions back.
Returns:
Pair of (root_revision, revision).
Information on sha1_git if it is an ancestor of sha1_git_root
including children leading to sha1_git_root
Raises:
- BadInputExc in case of unknown algo_hash or bad hash.
- NotFoundExc if either revision is not found or if sha1_git is not an
ancestor of sha1_git_root.
"""
rev_root_id = _lookup_revision_id_by(origin, branch_name, timestamp)
rev_root_id_bin = hashutil.hash_to_bytes(rev_root_id)
rev_root = storage.revision_get([rev_root_id_bin])[0]
return (
converters.from_revision(rev_root) if rev_root else None,
lookup_revision_with_context(rev_root, sha1_git, limit),
)
def lookup_revision_with_context(
sha1_git_root: Union[str, Dict[str, Any], Revision], sha1_git: str, limit: int = 100
) -> Dict[str, Any]:
"""Return information about revision sha1_git, limited to the
sub-graph of all transitive parents of sha1_git_root.
In other words, sha1_git is an ancestor of sha1_git_root.
Args:
sha1_git_root: latest revision. The type is either a sha1 (as an hex
string) or a non converted dict.
sha1_git: one of sha1_git_root's ancestors
limit: limit the lookup to 100 revisions back
Returns:
Information on sha1_git if it is an ancestor of sha1_git_root
including children leading to sha1_git_root
Raises:
BadInputExc in case of unknown algo_hash or bad hash
NotFoundExc if either revision is not found or if sha1_git is not an
ancestor of sha1_git_root
"""
sha1_git_bin = _to_sha1_bin(sha1_git)
revision = storage.revision_get([sha1_git_bin])[0]
if not revision:
raise NotFoundExc(f"Revision {sha1_git} not found")
if isinstance(sha1_git_root, str):
sha1_git_root_bin = _to_sha1_bin(sha1_git_root)
revision_root = storage.revision_get([sha1_git_root_bin])[0]
if not revision_root:
raise NotFoundExc(f"Revision root {sha1_git_root} not found")
elif isinstance(sha1_git_root, Revision):
sha1_git_root_bin = sha1_git_root.id
else:
sha1_git_root_bin = sha1_git_root["id"]
revision_log = storage.revision_log([sha1_git_root_bin], limit)
parents: Dict[str, List[str]] = {}
children = defaultdict(list)
for rev in revision_log:
rev_id = rev["id"]
parents[rev_id] = []
for parent_id in rev["parents"]:
parents[rev_id].append(parent_id)
children[parent_id].append(rev_id)
if revision.id not in parents:
raise NotFoundExc(f"Revision {sha1_git} is not an ancestor of {sha1_git_root}")
revision_d = revision.to_dict()
revision_d["children"] = children[revision.id]
return converters.from_revision(revision_d)
def lookup_directory_with_revision(sha1_git, dir_path=None, with_data=False):
"""Return information on directory pointed by revision with sha1_git.
If dir_path is not provided, display top level directory.
Otherwise, display the directory pointed by dir_path (if it exists).
Args:
sha1_git: revision's hash.
dir_path: optional directory pointed to by that revision.
with_data: boolean that indicates to retrieve the raw data if the path
resolves to a content. Default to False (for the api)
Returns:
Information on the directory pointed to by that revision.
Raises:
BadInputExc in case of unknown algo_hash or bad hash.
NotFoundExc either if the revision is not found or the path referenced
does not exist.
NotImplementedError in case of dir_path exists but do not reference a
type 'dir' or 'file'.
"""
sha1_git_bin = _to_sha1_bin(sha1_git)
revision = storage.revision_get([sha1_git_bin])[0]
if not revision:
raise NotFoundExc(f"Revision {sha1_git} not found")
dir_sha1_git_bin = revision.directory
if dir_path:
paths = dir_path.strip(os.path.sep).split(os.path.sep)
entity = storage.directory_entry_get_by_path(
dir_sha1_git_bin, list(map(lambda p: p.encode("utf-8"), paths))
)
if not entity:
raise NotFoundExc(
"Directory or File '%s' pointed to by revision %s not found"
% (dir_path, sha1_git)
)
else:
entity = {"type": "dir", "target": dir_sha1_git_bin}
if entity["type"] == "dir":
directory_entries = storage.directory_ls(entity["target"]) or []
return {
"type": "dir",
"path": "." if not dir_path else dir_path,
"revision": sha1_git,
"content": list(map(converters.from_directory_entry, directory_entries)),
}
elif entity["type"] == "file": # content
content = _first_element(storage.content_find({"sha1_git": entity["target"]}))
if not content:
raise NotFoundExc(f"Content not found for revision {sha1_git}")
content_d = content.to_dict()
if with_data:
data = storage.content_get_data(content.sha1)
if data:
content_d["data"] = data
return {
"type": "file",
"path": "." if not dir_path else dir_path,
"revision": sha1_git,
"content": converters.from_content(content_d),
}
elif entity["type"] == "rev": # revision
revision = storage.revision_get([entity["target"]])[0]
return {
"type": "rev",
"path": "." if not dir_path else dir_path,
"revision": sha1_git,
"content": converters.from_revision(revision) if revision else None,
}
else:
raise NotImplementedError("Entity of type %s not implemented." % entity["type"])
def lookup_content(q: str) -> Dict[str, Any]:
"""Lookup the content designed by q.
Args:
q: The release's sha1 as hexadecimal
Raises:
NotFoundExc if the requested content is not found
"""
algo, hash_ = query.parse_hash(q)
c = _first_element(storage.content_find({algo: hash_}))
if not c:
hhex = hashutil.hash_to_hex(hash_)
raise NotFoundExc(f"Content with {algo} checksum equals to {hhex} not found!")
return converters.from_content(c.to_dict())
def lookup_content_raw(q: str) -> Dict[str, Any]:
"""Lookup the content defined by q.
Args:
q: query string of the form <hash_algo:hash>
Returns:
dict with 'sha1' and 'data' keys.
data representing its raw data decoded.
Raises:
NotFoundExc if the requested content is not found or
if the content bytes are not available in the storage
"""
c = lookup_content(q)
content_sha1_bytes = hashutil.hash_to_bytes(c["checksums"]["sha1"])
content_data = storage.content_get_data(content_sha1_bytes)
if content_data is None:
algo, hash_ = query.parse_hash(q)
raise NotFoundExc(
f"Bytes of content with {algo} checksum equals "
f"to {hashutil.hash_to_hex(hash_)} are not available!"
)
return converters.from_content({"sha1": content_sha1_bytes, "data": content_data})
def stat_counters():
"""Return the stat counters for Software Heritage
Returns:
A dict mapping textual labels to integer values.
"""
return storage.stat_counters()
def _lookup_origin_visits(
origin_url: str, last_visit: Optional[int] = None, limit: int = 10
) -> Iterator[OriginVisit]:
"""Yields the origin origins' visits.
Args:
origin_url (str): origin to list visits for
last_visit (int): last visit to lookup from
limit (int): Number of elements max to display
Yields:
OriginVisit for that origin
"""
limit = min(limit, MAX_LIMIT)
page_token: Optional[str]
if last_visit is not None:
page_token = str(last_visit)
else:
page_token = None
visit_page = storage.origin_visit_get(
origin_url, page_token=page_token, limit=limit
)
yield from visit_page.results
def lookup_origin_visits(
origin: str, last_visit: Optional[int] = None, per_page: int = 10
) -> Iterator[OriginVisitInfo]:
"""Yields the origin origins' visits.
Args:
origin: origin to list visits for
Yields:
Dictionaries of origin_visit for that origin
"""
for visit in _lookup_origin_visits(origin, last_visit=last_visit, limit=per_page):
visit_status = storage.origin_visit_status_get_latest(origin, visit.visit)
yield converters.from_origin_visit(
{**visit_status.to_dict(), "type": visit.type}
)
def lookup_origin_visit_latest(
origin_url: str,
require_snapshot: bool = False,
type: Optional[str] = None,
allowed_statuses: Optional[List[str]] = None,
) -> Optional[OriginVisitInfo]:
"""Return the origin's latest visit
Args:
origin_url: origin to list visits for
type: Optional visit type to filter on (e.g git, tar, dsc, svn,
hg, npm, pypi, ...)
allowed_statuses: list of visit statuses considered
to find the latest visit. For instance,
``allowed_statuses=['full']`` will only consider visits that
have successfully run to completion.
require_snapshot: filter out origins without a snapshot
Returns:
The origin visit info as dict if found
"""
visit_and_status = origin_get_latest_visit_status(
storage,
origin_url,
type=type,
allowed_statuses=allowed_statuses,
require_snapshot=require_snapshot,
)
return (
converters.from_origin_visit(
{**visit_and_status[0].to_dict(), **visit_and_status[1].to_dict()}
)
if visit_and_status
else None
)
def lookup_origin_visit(origin_url: str, visit_id: int) -> OriginVisitInfo:
"""Return information about visit visit_id with origin origin.
Args:
origin: origin concerned by the visit
visit_id: the visit identifier to lookup
Yields:
The dict origin_visit concerned
"""
visit = storage.origin_visit_get_by(origin_url, visit_id)
visit_status = storage.origin_visit_status_get_latest(origin_url, visit_id)
if not visit:
raise NotFoundExc(
f"Origin {origin_url} or its visit with id {visit_id} not found!"
)
return converters.from_origin_visit({**visit_status.to_dict(), "type": visit.type})
-def lookup_snapshot_sizes(snapshot_id):
+def lookup_snapshot_sizes(snapshot_id: str) -> Dict[str, int]:
"""Count the number of branches in the snapshot with the given id
Args:
snapshot_id (str): sha1 identifier of the snapshot
Returns:
dict: A dict whose keys are the target types of branches and
values their corresponding amount
"""
snapshot_id_bin = _to_sha1_bin(snapshot_id)
- snapshot_sizes = storage.snapshot_count_branches(snapshot_id_bin)
- if "revision" not in snapshot_sizes:
- snapshot_sizes["revision"] = 0
- if "release" not in snapshot_sizes:
- snapshot_sizes["release"] = 0
- # adjust revision / release count for display if aliases are defined
- if "alias" in snapshot_sizes:
- aliases = lookup_snapshot(
- snapshot_id, branches_count=snapshot_sizes["alias"], target_types=["alias"]
- )
- for alias in aliases["branches"].values():
- try:
- for target_type in ("revision", "release"):
- snapshot = lookup_snapshot(
- snapshot_id,
- branches_from=alias["target"],
- branches_count=1,
- target_types=[target_type],
- )
- if snapshot and alias["target"] in snapshot["branches"]:
- snapshot_sizes[target_type] += 1
- except NotFoundExc:
- # aliased revision or release is missing in the snapshot
- pass
- del snapshot_sizes["alias"]
- # remove possible None key returned by snapshot_count_branches
- # when null branches are present in the snapshot
- snapshot_sizes.pop(None, None)
+ snapshot_sizes = dict.fromkeys(("alias", "release", "revision"), 0)
+ branch_counts = storage.snapshot_count_branches(snapshot_id_bin)
+ # remove possible None key returned by snapshot_count_branches
+ # when null branches are present in the snapshot
+ branch_counts.pop(None, None)
+ snapshot_sizes.update(branch_counts)
return snapshot_sizes
def lookup_snapshot(
snapshot_id, branches_from="", branches_count=1000, target_types=None
):
"""Return information about a snapshot, aka the list of named
branches found during a specific visit of an origin.
Args:
snapshot_id (str): sha1 identifier of the snapshot
branches_from (str): optional parameter used to skip branches
whose name is lesser than it before returning them
branches_count (int): optional parameter used to restrain
the amount of returned branches
target_types (list): optional parameter used to filter the
target types of branch to return (possible values that can be
contained in that list are `'content', 'directory',
'revision', 'release', 'snapshot', 'alias'`)
Returns:
A dict filled with the snapshot content.
"""
snapshot_id_bin = _to_sha1_bin(snapshot_id)
partial_branches = storage.snapshot_get_branches(
snapshot_id_bin, branches_from.encode(), branches_count, target_types
)
if not partial_branches:
raise NotFoundExc(f"Snapshot with id {snapshot_id} not found!")
return converters.from_partial_branches(partial_branches)
def lookup_latest_origin_snapshot(
origin: str, allowed_statuses: List[str] = None
) -> Optional[Dict[str, Any]]:
"""Return information about the latest snapshot of an origin.
.. warning:: At most 1000 branches contained in the snapshot
will be returned for performance reasons.
Args:
origin: URL or integer identifier of the origin
allowed_statuses: list of visit statuses considered
to find the latest snapshot for the visit. For instance,
``allowed_statuses=['full']`` will only consider visits that
have successfully run to completion.
Returns:
A dict filled with the snapshot content.
"""
snp = snapshot_get_latest(
storage, origin, allowed_statuses=allowed_statuses, branches_count=1000
)
return converters.from_snapshot(snp.to_dict()) if snp is not None else None
def lookup_snapshot_branch_name_from_tip_revision(
snapshot_id: str, revision_id: str
) -> Optional[str]:
"""Check if a revision corresponds to the tip of a snapshot branch
Args:
snapshot_id: hexadecimal representation of a snapshot id
revision_id: hexadecimal representation of a revision id
Returns:
The name of the first found branch or None otherwise
"""
per_page = 10000
branches_from = ""
snapshot: Dict[str, Any] = {"branches": {}}
branches = []
while not branches_from or len(snapshot["branches"]) == per_page + 1:
snapshot = lookup_snapshot(
snapshot_id,
target_types=[REVISION],
branches_from=branches_from,
branches_count=per_page + 1,
)
branches += [
{"name": k, "target": v["target"]} for k, v in snapshot["branches"].items()
]
branches_from = branches[-1]["name"]
for branch in branches:
if branch["target"] == revision_id:
return branch["name"]
return None
+def lookup_snapshot_alias(
+ snapshot_id: str, alias_name: str
+) -> Optional[Dict[str, Any]]:
+ """Try to resolve a branch alias in a snapshot.
+
+ Args:
+ snapshot_id: hexadecimal representation of a snapshot id
+ alias_name: name of the branch alias to resolve
+
+ Returns:
+ Target branch information or None if the alias does not exist
+ or target a dangling branch.
+ """
+ resolved_alias = snapshot_resolve_alias(
+ storage, _to_sha1_bin(snapshot_id), alias_name.encode()
+ )
+ return (
+ converters.from_swh(resolved_alias.to_dict(), hashess={"target"})
+ if resolved_alias is not None
+ else None
+ )
+
+
def lookup_revision_through(revision, limit=100):
"""Retrieve a revision from the criterion stored in revision dictionary.
Args:
revision: Dictionary of criterion to lookup the revision with.
Here are the supported combination of possible values:
- origin_url, branch_name, ts, sha1_git
- origin_url, branch_name, ts
- sha1_git_root, sha1_git
- sha1_git
Returns:
None if the revision is not found or the actual revision.
"""
if (
"origin_url" in revision
and "branch_name" in revision
and "ts" in revision
and "sha1_git" in revision
):
return lookup_revision_with_context_by(
revision["origin_url"],
revision["branch_name"],
revision["ts"],
revision["sha1_git"],
limit,
)
if "origin_url" in revision and "branch_name" in revision and "ts" in revision:
return lookup_revision_by(
revision["origin_url"], revision["branch_name"], revision["ts"]
)
if "sha1_git_root" in revision and "sha1_git" in revision:
return lookup_revision_with_context(
revision["sha1_git_root"], revision["sha1_git"], limit
)
if "sha1_git" in revision:
return lookup_revision(revision["sha1_git"])
# this should not happen
raise NotImplementedError("Should not happen!")
def lookup_directory_through_revision(revision, path=None, limit=100, with_data=False):
"""Retrieve the directory information from the revision.
Args:
revision: dictionary of criterion representing a revision to lookup
path: directory's path to lookup.
limit: optional query parameter to limit the revisions log (default to
100). For now, note that this limit could impede the transitivity
conclusion about sha1_git not being an ancestor of.
with_data: indicate to retrieve the content's raw data if path resolves
to a content.
Returns:
The directory pointing to by the revision criterions at path.
"""
rev = lookup_revision_through(revision, limit)
if not rev:
raise NotFoundExc("Revision with criterion %s not found!" % revision)
return (rev["id"], lookup_directory_with_revision(rev["id"], path, with_data))
def _vault_request(vault_fn, *args, **kwargs):
try:
return vault_fn(*args, **kwargs)
except VaultNotFoundExc:
return None
def vault_cook(obj_type, obj_id, email=None):
"""Cook a vault bundle.
"""
return _vault_request(vault.cook, obj_type, obj_id, email=email)
def vault_fetch(obj_type, obj_id):
"""Fetch a vault bundle.
"""
return _vault_request(vault.fetch, obj_type, obj_id)
def vault_progress(obj_type, obj_id):
"""Get the current progress of a vault bundle.
"""
return _vault_request(vault.progress, obj_type, obj_id)
def diff_revision(rev_id):
"""Get the list of file changes (insertion / deletion / modification /
renaming) for a particular revision.
"""
rev_sha1_git_bin = _to_sha1_bin(rev_id)
changes = diff.diff_revision(storage, rev_sha1_git_bin, track_renaming=True)
for change in changes:
change["from"] = converters.from_directory_entry(change["from"])
change["to"] = converters.from_directory_entry(change["to"])
if change["from_path"]:
change["from_path"] = change["from_path"].decode("utf-8")
if change["to_path"]:
change["to_path"] = change["to_path"].decode("utf-8")
return changes
class _RevisionsWalkerProxy(object):
"""
Proxy class wrapping a revisions walker iterator from
swh-storage and performing needed conversions.
"""
def __init__(self, rev_walker_type, rev_start, *args, **kwargs):
rev_start_bin = hashutil.hash_to_bytes(rev_start)
self.revisions_walker = revisions_walker.get_revisions_walker(
rev_walker_type, storage, rev_start_bin, *args, **kwargs
)
def export_state(self):
return self.revisions_walker.export_state()
def __next__(self):
return converters.from_revision(next(self.revisions_walker))
def __iter__(self):
return self
def get_revisions_walker(rev_walker_type, rev_start, *args, **kwargs):
"""
Utility function to instantiate a revisions walker of a given type,
see :mod:`swh.storage.algos.revisions_walker`.
Args:
rev_walker_type (str): the type of revisions walker to return,
possible values are: ``committer_date``, ``dfs``, ``dfs_post``,
``bfs`` and ``path``
rev_start (str): hexadecimal representation of a revision identifier
args (list): position arguments to pass to the revisions walker
constructor
kwargs (dict): keyword arguments to pass to the revisions walker
constructor
"""
# first check if the provided revision is valid
lookup_revision(rev_start)
return _RevisionsWalkerProxy(rev_walker_type, rev_start, *args, **kwargs)
def lookup_object(object_type: str, object_id: str) -> Dict[str, Any]:
"""
Utility function for looking up an object in the archive by its type
and id.
Args:
object_type (str): the type of object to lookup, either *content*,
*directory*, *release*, *revision* or *snapshot*
object_id (str): the *sha1_git* checksum identifier in hexadecimal
form of the object to lookup
Returns:
Dict[str, Any]: A dictionary describing the object or a list of
dictionary for the directory object type.
Raises:
swh.web.common.exc.NotFoundExc: if the object could not be found in
the archive
BadInputExc: if the object identifier is invalid
"""
if object_type == CONTENT:
return lookup_content(f"sha1_git:{object_id}")
elif object_type == DIRECTORY:
return {"id": object_id, "content": list(lookup_directory(object_id))}
elif object_type == RELEASE:
return lookup_release(object_id)
elif object_type == REVISION:
return lookup_revision(object_id)
elif object_type == SNAPSHOT:
return lookup_snapshot(object_id)
raise BadInputExc(
(
"Invalid swh object type! Valid types are "
f"{CONTENT}, {DIRECTORY}, {RELEASE} "
f"{REVISION} or {SNAPSHOT}."
)
)
def lookup_missing_hashes(grouped_swhids: Dict[str, List[bytes]]) -> Set[str]:
"""Lookup missing Software Heritage persistent identifier hash, using
batch processing.
Args:
A dictionary with:
keys: object types
values: object hashes
Returns:
A set(hexadecimal) of the hashes not found in the storage
"""
missing_hashes = []
for obj_type, obj_ids in grouped_swhids.items():
if obj_type == CONTENT:
missing_hashes.append(storage.content_missing_per_sha1_git(obj_ids))
elif obj_type == DIRECTORY:
missing_hashes.append(storage.directory_missing(obj_ids))
elif obj_type == REVISION:
missing_hashes.append(storage.revision_missing(obj_ids))
elif obj_type == RELEASE:
missing_hashes.append(storage.directory_missing(obj_ids))
elif obj_type == SNAPSHOT:
missing_hashes.append(storage.directory_missing(obj_ids))
missing = set(
map(lambda x: hashutil.hash_to_hex(x), itertools.chain(*missing_hashes))
)
return missing
def lookup_origins_by_sha1s(sha1s: List[str]) -> Iterator[Optional[OriginInfo]]:
"""Lookup origins from the sha1 hash values of their URLs.
Args:
sha1s: list of sha1s hexadecimal representation
Yields:
origin information as dict
"""
sha1s_bytes = [hashutil.hash_to_bytes(sha1) for sha1 in sha1s]
origins = storage.origin_get_by_sha1(sha1s_bytes)
for origin in origins:
yield converters.from_origin(origin)
diff --git a/swh/web/common/typing.py b/swh/web/common/typing.py
index 0cf59f0e..5e26c9c0 100644
--- a/swh/web/common/typing.py
+++ b/swh/web/common/typing.py
@@ -1,223 +1,227 @@
# Copyright (C) 2020 The Software Heritage developers
# See the AUTHORS file at the top-level directory of this distribution
# License: GNU Affero General Public License version 3, or any later version
# See top-level LICENSE file for more information
from typing import Any, Dict, List, Optional, TypeVar, Union
from typing_extensions import TypedDict
from django.http import QueryDict
from swh.core.api.classes import PagedResult as CorePagedResult
QueryParameters = Union[Dict[str, Any], QueryDict]
class OriginInfo(TypedDict):
url: str
"""URL of the origin"""
class OriginMetadataInfo(TypedDict):
url: str
"""URL of the origin"""
metadata: Dict[str, Any]
"""Origin metadata associated to the origin"""
class OriginVisitInfo(TypedDict):
date: str
"""date of the visit in iso format"""
formatted_date: str
"""formatted date of the visit"""
metadata: Dict[str, Any]
"""metadata associated to the visit"""
origin: str
"""visited origin URL"""
snapshot: str
"""snapshot identifier computed during the visit"""
status: str
"""status of the visit ("ongoing", "full" or "partial") """
type: str
"""visit type (git, hg, debian, ...)"""
url: str
"""URL to browse the snapshot"""
visit: int
"""visit identifier"""
class SnapshotBranchInfo(TypedDict):
date: Optional[str]
""""author date of branch heading revision"""
directory: Optional[str]
"""directory associated to branch heading revision"""
message: Optional[str]
"""message of branch heading revision"""
name: str
"""branch name"""
+ alias: bool
+ """define if the branch is an alias"""
revision: str
"""branch heading revision"""
url: Optional[str]
"""optional browse URL (content, directory, ...) scoped to branch"""
class SnapshotReleaseInfo(TypedDict):
branch_name: str
"""branch name associated to release in snapshot"""
date: str
"""release date"""
directory: Optional[str]
"""optional directory associatd to the release"""
id: str
"""release identifier"""
message: str
"""release message"""
name: str
"""release name"""
+ alias: bool
+ """define if the branch is an alias"""
target: str
"""release target"""
target_type: str
"""release target_type"""
url: Optional[str]
"""optional browse URL (content, directory, ...) scoped to release"""
class SnapshotContext(TypedDict):
branch: Optional[str]
"""optional branch name set when browsing snapshot in that scope"""
branches: List[SnapshotBranchInfo]
"""list of snapshot branches (possibly truncated)"""
branches_url: str
"""snapshot branches list browse URL"""
is_empty: bool
"""indicates if the snapshot is empty"""
origin_info: Optional[OriginInfo]
"""optional origin info associated to the snapshot"""
origin_visits_url: Optional[str]
"""optional origin visits URL"""
query_params: QueryParameters
"""common query parameters when browsing snapshot content"""
release: Optional[str]
"""optional release name set when browsing snapshot in that scope"""
release_id: Optional[str]
"""optional release identifier set when browsing snapshot in that scope"""
releases: List[SnapshotReleaseInfo]
"""list of snapshot releases (possibly truncated)"""
releases_url: str
"""snapshot releases list browse URL"""
revision_id: Optional[str]
"""optional revision identifier set when browsing snapshot in that scope"""
revision_info: Optional[Dict[str, Any]]
"""optional revision info set when browsing snapshot in that scope"""
root_directory: Optional[str]
"""optional root directory identifier set when browsing snapshot content"""
snapshot_id: str
"""snapshot identifier"""
snapshot_sizes: Dict[str, int]
"""snapshot sizes grouped by branch target type"""
snapshot_swhid: str
"""snapshot SWHID"""
url_args: Dict[str, Any]
"""common URL arguments when browsing snapshot content"""
visit_info: Optional[OriginVisitInfo]
"""optional origin visit info associated to the snapshot"""
class SWHObjectInfo(TypedDict):
object_type: str
object_id: str
class SWHIDContext(TypedDict, total=False):
origin: str
anchor: str
visit: str
path: str
lines: str
class SWHIDInfo(SWHObjectInfo):
swhid: str
swhid_url: str
context: SWHIDContext
swhid_with_context: Optional[str]
swhid_with_context_url: Optional[str]
class SWHObjectInfoMetadata(TypedDict, total=False):
origin_url: Optional[str]
visit_date: Optional[str]
visit_type: Optional[str]
directory_url: Optional[str]
revision_url: Optional[str]
release_url: Optional[str]
snapshot_url: Optional[str]
class ContentMetadata(SWHObjectInfo, SWHObjectInfoMetadata):
sha1: str
sha1_git: str
sha256: str
blake2s256: str
content_url: str
mimetype: str
encoding: str
size: str
language: str
path: Optional[str]
filename: Optional[str]
directory: Optional[str]
root_directory: Optional[str]
revision: Optional[str]
release: Optional[str]
snapshot: Optional[str]
class DirectoryMetadata(SWHObjectInfo, SWHObjectInfoMetadata):
directory: str
nb_files: int
nb_dirs: int
sum_file_sizes: str
root_directory: Optional[str]
path: str
revision: Optional[str]
revision_found: Optional[bool]
release: Optional[str]
snapshot: Optional[str]
class ReleaseMetadata(SWHObjectInfo, SWHObjectInfoMetadata):
release: str
author: str
author_url: str
date: str
name: str
synthetic: bool
target: str
target_type: str
target_url: str
snapshot: Optional[str]
class RevisionMetadata(SWHObjectInfo, SWHObjectInfoMetadata):
revision: str
author: str
author_url: str
committer: str
committer_url: str
date: str
committer_date: str
directory: str
merge: bool
metadata: str
parents: List[str]
synthetic: bool
type: str
snapshot: Optional[str]
TResult = TypeVar("TResult")
PagedResult = CorePagedResult[TResult, str]
diff --git a/swh/web/tests/browse/test_snapshot_context.py b/swh/web/tests/browse/test_snapshot_context.py
index 24f17682..c1eac880 100644
--- a/swh/web/tests/browse/test_snapshot_context.py
+++ b/swh/web/tests/browse/test_snapshot_context.py
@@ -1,398 +1,401 @@
# Copyright (C) 2020 The Software Heritage developers
# See the AUTHORS file at the top-level directory of this distribution
# License: GNU Affero General Public License version 3, or any later version
# See top-level LICENSE file for more information
import random
from hypothesis import given
from swh.web.browse.snapshot_context import (
_get_release,
get_origin_visit_snapshot,
get_snapshot_content,
get_snapshot_context,
)
from swh.web.browse.utils import gen_revision_url
from swh.web.common.identifiers import gen_swhid
from swh.web.common.origin_visits import get_origin_visit, get_origin_visits
from swh.web.common.typing import (
SnapshotBranchInfo,
SnapshotContext,
SnapshotReleaseInfo,
)
from swh.web.common.utils import format_utc_iso_date, reverse
from swh.web.tests.strategies import (
origin_with_multiple_visits,
origin_with_releases,
snapshot,
)
@given(origin_with_multiple_visits())
def test_get_origin_visit_snapshot_simple(archive_data, origin):
visits = archive_data.origin_visit_get(origin["url"])
for visit in visits:
snapshot = archive_data.snapshot_get(visit["snapshot"])
branches = []
releases = []
- def _process_branch_data(branch, branch_data):
+ def _process_branch_data(branch, branch_data, alias=False):
if branch_data["target_type"] == "revision":
rev_data = archive_data.revision_get(branch_data["target"])
branches.append(
SnapshotBranchInfo(
name=branch,
+ alias=alias,
revision=branch_data["target"],
directory=rev_data["directory"],
date=format_utc_iso_date(rev_data["date"]),
message=rev_data["message"],
url=None,
)
)
elif branch_data["target_type"] == "release":
rel_data = archive_data.release_get(branch_data["target"])
rev_data = archive_data.revision_get(rel_data["target"])
releases.append(
SnapshotReleaseInfo(
name=rel_data["name"],
+ alias=alias,
branch_name=branch,
date=format_utc_iso_date(rel_data["date"]),
id=rel_data["id"],
message=rel_data["message"],
target_type=rel_data["target_type"],
target=rel_data["target"],
directory=rev_data["directory"],
url=None,
)
)
for branch in sorted(snapshot["branches"].keys()):
branch_data = snapshot["branches"][branch]
if branch_data["target_type"] == "alias":
target_data = snapshot["branches"][branch_data["target"]]
- _process_branch_data(branch, target_data)
+ _process_branch_data(branch, target_data, alias=True)
else:
_process_branch_data(branch, branch_data)
assert branches and releases, "Incomplete test data."
origin_visit_branches = get_origin_visit_snapshot(
origin, visit_id=visit["visit"]
)
assert origin_visit_branches == (branches, releases)
@given(snapshot())
def test_get_snapshot_context_no_origin(archive_data, snapshot):
for browse_context, kwargs in (
("content", {"snapshot_id": snapshot, "path": "/some/path"}),
("directory", {"snapshot_id": snapshot}),
("log", {"snapshot_id": snapshot}),
):
url_args = {"snapshot_id": snapshot}
query_params = dict(kwargs)
query_params.pop("snapshot_id")
snapshot_context = get_snapshot_context(**kwargs, browse_context=browse_context)
branches, releases = get_snapshot_content(snapshot)
releases = list(reversed(releases))
revision_id = None
root_directory = None
for branch in branches:
if branch["name"] == "HEAD":
revision_id = branch["revision"]
root_directory = branch["directory"]
branch["url"] = reverse(
f"browse-snapshot-{browse_context}",
url_args=url_args,
query_params={"branch": branch["name"], **query_params},
)
for release in releases:
release["url"] = reverse(
f"browse-snapshot-{browse_context}",
url_args=url_args,
query_params={"release": release["name"], **query_params},
)
branches_url = reverse("browse-snapshot-branches", url_args=url_args)
releases_url = reverse("browse-snapshot-releases", url_args=url_args)
is_empty = not branches and not releases
snapshot_swhid = gen_swhid("snapshot", snapshot)
- snapshot_sizes = {"revision": len(branches), "release": len(releases)}
+ snapshot_sizes = archive_data.snapshot_count_branches(snapshot)
expected = SnapshotContext(
branch="HEAD",
branches=branches,
branches_url=branches_url,
is_empty=is_empty,
origin_info=None,
origin_visits_url=None,
release=None,
release_id=None,
query_params=query_params,
releases=releases,
releases_url=releases_url,
revision_id=revision_id,
revision_info=_get_revision_info(archive_data, revision_id),
root_directory=root_directory,
snapshot_id=snapshot,
snapshot_sizes=snapshot_sizes,
snapshot_swhid=snapshot_swhid,
url_args=url_args,
visit_info=None,
)
if revision_id:
expected["revision_info"]["revision_url"] = gen_revision_url(
revision_id, snapshot_context
)
assert snapshot_context == expected
_check_branch_release_revision_parameters(
archive_data, expected, browse_context, kwargs, branches, releases
)
@given(origin_with_multiple_visits())
def test_get_snapshot_context_with_origin(archive_data, origin):
origin_visits = get_origin_visits(origin)
timestamp = format_utc_iso_date(origin_visits[0]["date"], "%Y-%m-%dT%H:%M:%SZ")
visit_id = origin_visits[1]["visit"]
for browse_context, kwargs in (
("content", {"origin_url": origin["url"], "path": "/some/path"}),
("directory", {"origin_url": origin["url"]}),
("log", {"origin_url": origin["url"]}),
("directory", {"origin_url": origin["url"], "timestamp": timestamp,},),
("directory", {"origin_url": origin["url"], "visit_id": visit_id,},),
):
visit_id = kwargs["visit_id"] if "visit_id" in kwargs else None
visit_ts = kwargs["timestamp"] if "timestamp" in kwargs else None
visit_info = get_origin_visit(
{"url": kwargs["origin_url"]}, visit_ts=visit_ts, visit_id=visit_id
)
snapshot = visit_info["snapshot"]
snapshot_context = get_snapshot_context(**kwargs, browse_context=browse_context)
query_params = dict(kwargs)
branches, releases = get_snapshot_content(snapshot)
releases = list(reversed(releases))
revision_id = None
root_directory = None
for branch in branches:
if branch["name"] == "HEAD":
revision_id = branch["revision"]
root_directory = branch["directory"]
branch["url"] = reverse(
f"browse-origin-{browse_context}",
query_params={"branch": branch["name"], **query_params},
)
for release in releases:
release["url"] = reverse(
f"browse-origin-{browse_context}",
query_params={"release": release["name"], **query_params},
)
query_params.pop("path", None)
branches_url = reverse("browse-origin-branches", query_params=query_params)
releases_url = reverse("browse-origin-releases", query_params=query_params)
origin_visits_url = reverse(
"browse-origin-visits", query_params={"origin_url": kwargs["origin_url"]}
)
is_empty = not branches and not releases
snapshot_swhid = gen_swhid("snapshot", snapshot)
- snapshot_sizes = {"revision": len(branches), "release": len(releases)}
+ snapshot_sizes = archive_data.snapshot_count_branches(snapshot)
visit_info["url"] = reverse(
"browse-origin-directory", query_params=query_params
)
visit_info["formatted_date"] = format_utc_iso_date(visit_info["date"])
if "path" in kwargs:
query_params["path"] = kwargs["path"]
expected = SnapshotContext(
branch="HEAD",
branches=branches,
branches_url=branches_url,
is_empty=is_empty,
origin_info={"url": origin["url"]},
origin_visits_url=origin_visits_url,
release=None,
release_id=None,
query_params=query_params,
releases=releases,
releases_url=releases_url,
revision_id=revision_id,
revision_info=_get_revision_info(archive_data, revision_id),
root_directory=root_directory,
snapshot_id=snapshot,
snapshot_sizes=snapshot_sizes,
snapshot_swhid=snapshot_swhid,
url_args={},
visit_info=visit_info,
)
if revision_id:
expected["revision_info"]["revision_url"] = gen_revision_url(
revision_id, snapshot_context
)
assert snapshot_context == expected
_check_branch_release_revision_parameters(
archive_data, expected, browse_context, kwargs, branches, releases
)
def _check_branch_release_revision_parameters(
archive_data, base_expected_context, browse_context, kwargs, branches, releases,
):
branch = random.choice(branches)
snapshot_context = get_snapshot_context(
**kwargs, browse_context=browse_context, branch_name=branch["name"]
)
url_args = dict(kwargs)
url_args.pop("path", None)
url_args.pop("timestamp", None)
url_args.pop("visit_id", None)
url_args.pop("origin_url", None)
query_params = dict(kwargs)
query_params.pop("snapshot_id", None)
expected_branch = dict(base_expected_context)
expected_branch["branch"] = branch["name"]
expected_branch["revision_id"] = branch["revision"]
expected_branch["revision_info"] = _get_revision_info(
archive_data, branch["revision"]
)
expected_branch["root_directory"] = branch["directory"]
expected_branch["query_params"] = {"branch": branch["name"], **query_params}
expected_branch["revision_info"]["revision_url"] = gen_revision_url(
branch["revision"], expected_branch
)
assert snapshot_context == expected_branch
if releases:
release = random.choice(releases)
snapshot_context = get_snapshot_context(
**kwargs, browse_context=browse_context, release_name=release["name"]
)
expected_release = dict(base_expected_context)
expected_release["branch"] = None
expected_release["release"] = release["name"]
expected_release["release_id"] = release["id"]
if release["target_type"] == "revision":
expected_release["revision_id"] = release["target"]
expected_release["revision_info"] = _get_revision_info(
archive_data, release["target"]
)
expected_release["root_directory"] = release["directory"]
expected_release["query_params"] = {"release": release["name"], **query_params}
expected_release["revision_info"]["revision_url"] = gen_revision_url(
release["target"], expected_release
)
assert snapshot_context == expected_release
revision_log = archive_data.revision_log(branch["revision"])
revision = revision_log[-1]
snapshot_context = get_snapshot_context(
**kwargs, browse_context=browse_context, revision_id=revision["id"]
)
if "origin_url" in kwargs:
view_name = f"browse-origin-{browse_context}"
else:
view_name = f"browse-snapshot-{browse_context}"
kwargs.pop("visit_id", None)
revision_browse_url = reverse(
view_name,
url_args=url_args,
query_params={"revision": revision["id"], **query_params},
)
branches.append(
SnapshotBranchInfo(
name=revision["id"],
+ alias=False,
revision=revision["id"],
directory=revision["directory"],
date=revision["date"],
message=revision["message"],
url=revision_browse_url,
)
)
expected_revision = dict(base_expected_context)
expected_revision["branch"] = revision["id"]
expected_revision["branches"] = branches
expected_revision["revision_id"] = revision["id"]
expected_revision["revision_info"] = _get_revision_info(
archive_data, revision["id"]
)
expected_revision["root_directory"] = revision["directory"]
expected_revision["query_params"] = {"revision": revision["id"], **query_params}
expected_revision["revision_info"]["revision_url"] = gen_revision_url(
revision["id"], expected_revision
)
assert snapshot_context == expected_revision
@given(origin_with_releases())
def test_get_release_large_snapshot(archive_data, origin):
snapshot = archive_data.snapshot_get_latest(origin["url"])
release_id = random.choice(
[
v["target"]
for v in snapshot["branches"].values()
if v["target_type"] == "release"
]
)
release_data = archive_data.release_get(release_id)
# simulate large snapshot processing by providing releases parameter
# as an empty list
release = _get_release(
releases=[], release_name=release_data["name"], snapshot_id=snapshot["id"]
)
assert release_data["name"] == release["name"]
assert release_data["id"] == release["id"]
def _get_revision_info(archive_data, revision_id):
revision_info = None
if revision_id:
revision_info = archive_data.revision_get(revision_id)
revision_info["message_header"] = revision_info["message"].split("\n")[0]
revision_info["date"] = format_utc_iso_date(revision_info["date"])
revision_info["committer_date"] = format_utc_iso_date(
revision_info["committer_date"]
)
return revision_info
diff --git a/swh/web/tests/browse/views/test_content.py b/swh/web/tests/browse/views/test_content.py
index bb0dfde1..3b022e07 100644
--- a/swh/web/tests/browse/views/test_content.py
+++ b/swh/web/tests/browse/views/test_content.py
@@ -1,588 +1,596 @@
# Copyright (C) 2017-2020 The Software Heritage developers
# See the AUTHORS file at the top-level directory of this distribution
# License: GNU Affero General Public License version 3, or any later version
# See top-level LICENSE file for more information
import random
from hypothesis import given
from django.utils.html import escape
from swh.model.identifiers import CONTENT, DIRECTORY, RELEASE, REVISION, SNAPSHOT
from swh.web.browse.snapshot_context import process_snapshot_branches
from swh.web.browse.utils import (
_re_encode_content,
get_mimetype_and_encoding_for_content,
prepare_content_for_display,
)
from swh.web.common.exc import NotFoundExc
from swh.web.common.identifiers import gen_swhid
from swh.web.common.utils import gen_path_info, reverse
from swh.web.tests.django_asserts import assert_contains, assert_not_contains
from swh.web.tests.strategies import (
content,
content_image_type,
content_text,
content_text_no_highlight,
content_text_non_utf8,
content_unsupported_image_type_rendering,
content_utf8_detected_as_binary,
invalid_sha1,
origin_with_multiple_visits,
unknown_content,
)
from swh.web.tests.utils import check_html_get_response, check_http_get_response
@given(content_text())
def test_content_view_text(client, archive_data, content):
sha1_git = content["sha1_git"]
url = reverse(
"browse-content",
url_args={"query_string": content["sha1"]},
query_params={"path": content["path"]},
)
url_raw = reverse("browse-content-raw", url_args={"query_string": content["sha1"]})
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/content.html"
)
content_display = _process_content_for_display(archive_data, content)
mimetype = content_display["mimetype"]
if mimetype.startswith("text/"):
assert_contains(resp, '<code class="%s">' % content_display["language"])
assert_contains(resp, escape(content_display["content_data"]))
assert_contains(resp, url_raw)
swh_cnt_id = gen_swhid(CONTENT, sha1_git)
swh_cnt_id_url = reverse("browse-swhid", url_args={"swhid": swh_cnt_id})
assert_contains(resp, swh_cnt_id)
assert_contains(resp, swh_cnt_id_url)
@given(content_text_no_highlight())
def test_content_view_text_no_highlight(client, archive_data, content):
sha1_git = content["sha1_git"]
url = reverse("browse-content", url_args={"query_string": content["sha1"]})
url_raw = reverse("browse-content-raw", url_args={"query_string": content["sha1"]})
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/content.html"
)
content_display = _process_content_for_display(archive_data, content)
assert_contains(resp, '<code class="nohighlight">')
assert_contains(resp, escape(content_display["content_data"]))
assert_contains(resp, url_raw)
swh_cnt_id = gen_swhid(CONTENT, sha1_git)
swh_cnt_id_url = reverse("browse-swhid", url_args={"swhid": swh_cnt_id})
assert_contains(resp, swh_cnt_id)
assert_contains(resp, swh_cnt_id_url)
@given(content_text_non_utf8())
def test_content_view_no_utf8_text(client, archive_data, content):
sha1_git = content["sha1_git"]
url = reverse("browse-content", url_args={"query_string": content["sha1"]})
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/content.html"
)
content_display = _process_content_for_display(archive_data, content)
swh_cnt_id = gen_swhid(CONTENT, sha1_git)
swh_cnt_id_url = reverse("browse-swhid", url_args={"swhid": swh_cnt_id})
assert_contains(resp, swh_cnt_id_url)
assert_contains(resp, escape(content_display["content_data"]))
@given(content_image_type())
def test_content_view_image(client, archive_data, content):
url = reverse("browse-content", url_args={"query_string": content["sha1"]})
url_raw = reverse("browse-content-raw", url_args={"query_string": content["sha1"]})
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/content.html"
)
content_display = _process_content_for_display(archive_data, content)
mimetype = content_display["mimetype"]
content_data = content_display["content_data"]
assert_contains(resp, '<img src="data:%s;base64,%s"/>' % (mimetype, content_data))
assert_contains(resp, url_raw)
@given(content_unsupported_image_type_rendering())
def test_content_view_image_no_rendering(client, archive_data, content):
url = reverse("browse-content", url_args={"query_string": content["sha1"]})
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/content.html"
)
mimetype = content["mimetype"]
encoding = content["encoding"]
assert_contains(
resp,
(
f"Content with mime type {mimetype} and encoding {encoding} "
"cannot be displayed."
),
)
@given(content_text())
def test_content_view_text_with_path(client, archive_data, content):
path = content["path"]
url = reverse(
"browse-content",
url_args={"query_string": content["sha1"]},
query_params={"path": path},
)
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/content.html"
)
assert_contains(resp, '<nav class="bread-crumbs')
content_display = _process_content_for_display(archive_data, content)
mimetype = content_display["mimetype"]
if mimetype.startswith("text/"):
hljs_language = content["hljs_language"]
assert_contains(resp, '<code class="%s">' % hljs_language)
assert_contains(resp, escape(content_display["content_data"]))
split_path = path.split("/")
root_dir_sha1 = split_path[0]
filename = split_path[-1]
path = path.replace(root_dir_sha1 + "/", "").replace(filename, "")
swhid_context = {
"anchor": gen_swhid(DIRECTORY, root_dir_sha1),
"path": f"/{path}{filename}",
}
swh_cnt_id = gen_swhid(CONTENT, content["sha1_git"], metadata=swhid_context)
swh_cnt_id_url = reverse("browse-swhid", url_args={"swhid": swh_cnt_id})
assert_contains(resp, swh_cnt_id)
assert_contains(resp, swh_cnt_id_url)
path_info = gen_path_info(path)
root_dir_url = reverse("browse-directory", url_args={"sha1_git": root_dir_sha1})
assert_contains(resp, '<li class="swh-path">', count=len(path_info) + 1)
assert_contains(
resp, '<a href="' + root_dir_url + '">' + root_dir_sha1[:7] + "</a>"
)
for p in path_info:
dir_url = reverse(
"browse-directory",
url_args={"sha1_git": root_dir_sha1},
query_params={"path": p["path"]},
)
assert_contains(resp, '<a href="' + dir_url + '">' + p["name"] + "</a>")
assert_contains(resp, "<li>" + filename + "</li>")
url_raw = reverse(
"browse-content-raw",
url_args={"query_string": content["sha1"]},
query_params={"filename": filename},
)
assert_contains(resp, url_raw)
url = reverse(
"browse-content",
url_args={"query_string": content["sha1"]},
query_params={"path": filename},
)
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/content.html"
)
assert_not_contains(resp, '<nav class="bread-crumbs')
invalid_path = "%s/foo/bar/baz" % root_dir_sha1
url = reverse(
"browse-content",
url_args={"query_string": content["sha1"]},
query_params={"path": invalid_path},
)
resp = check_html_get_response(
client, url, status_code=404, template_used="error.html"
)
@given(content_text())
def test_content_raw_text(client, archive_data, content):
url = reverse("browse-content-raw", url_args={"query_string": content["sha1"]})
resp = check_http_get_response(
client, url, status_code=200, content_type="text/plain"
)
content_data = archive_data.content_get_data(content["sha1"])["data"]
assert resp["Content-Type"] == "text/plain"
assert resp["Content-disposition"] == ("filename=%s_%s" % ("sha1", content["sha1"]))
assert resp.content == content_data
filename = content["path"].split("/")[-1]
url = reverse(
"browse-content-raw",
url_args={"query_string": content["sha1"]},
query_params={"filename": filename},
)
resp = check_http_get_response(
client, url, status_code=200, content_type="text/plain"
)
assert resp["Content-Type"] == "text/plain"
assert resp["Content-disposition"] == "filename=%s" % filename
assert resp.content == content_data
@given(content_text_non_utf8())
def test_content_raw_no_utf8_text(client, content):
url = reverse("browse-content-raw", url_args={"query_string": content["sha1"]})
resp = check_http_get_response(
client, url, status_code=200, content_type="text/plain"
)
_, encoding = get_mimetype_and_encoding_for_content(resp.content)
assert encoding == content["encoding"]
@given(content_image_type())
def test_content_raw_bin(client, archive_data, content):
url = reverse("browse-content-raw", url_args={"query_string": content["sha1"]})
resp = check_http_get_response(
client, url, status_code=200, content_type="application/octet-stream"
)
filename = content["path"].split("/")[-1]
content_data = archive_data.content_get_data(content["sha1"])["data"]
assert resp["Content-Type"] == "application/octet-stream"
assert resp["Content-disposition"] == "attachment; filename=%s_%s" % (
"sha1",
content["sha1"],
)
assert resp.content == content_data
url = reverse(
"browse-content-raw",
url_args={"query_string": content["sha1"]},
query_params={"filename": filename},
)
resp = check_http_get_response(
client, url, status_code=200, content_type="application/octet-stream"
)
assert resp["Content-Type"] == "application/octet-stream"
assert resp["Content-disposition"] == "attachment; filename=%s" % filename
assert resp.content == content_data
@given(invalid_sha1(), unknown_content())
def test_content_request_errors(client, invalid_sha1, unknown_content):
url = reverse("browse-content", url_args={"query_string": invalid_sha1})
check_html_get_response(client, url, status_code=400, template_used="error.html")
url = reverse("browse-content", url_args={"query_string": unknown_content["sha1"]})
check_html_get_response(client, url, status_code=404, template_used="error.html")
@given(content())
def test_content_bytes_missing(client, archive_data, mocker, content):
mock_archive = mocker.patch("swh.web.browse.utils.archive")
content_data = archive_data.content_get(content["sha1"])
mock_archive.lookup_content.return_value = content_data
mock_archive.lookup_content_filetype.side_effect = Exception()
mock_archive.lookup_content_raw.side_effect = NotFoundExc(
"Content bytes not available!"
)
url = reverse("browse-content", url_args={"query_string": content["sha1"]})
check_html_get_response(
client, url, status_code=404, template_used="browse/content.html"
)
def test_content_too_large(client, mocker):
mock_request_content = mocker.patch("swh.web.browse.views.content.request_content")
stub_content_too_large_data = {
"checksums": {
"sha1": "8624bcdae55baeef00cd11d5dfcfa60f68710a02",
"sha1_git": "94a9ed024d3859793618152ea559a168bbcbb5e2",
"sha256": (
"8ceb4b9ee5adedde47b31e975c1d90c73ad27b6b16" "5a1dcd80c7c545eb65b903"
),
"blake2s256": (
"38702b7168c7785bfe748b51b45d9856070ba90" "f9dc6d90f2ea75d4356411ffe"
),
},
"length": 30000000,
"raw_data": None,
"mimetype": "text/plain",
"encoding": "us-ascii",
"language": "not detected",
"licenses": "GPL",
"error_code": 200,
"error_message": "",
"error_description": "",
}
content_sha1 = stub_content_too_large_data["checksums"]["sha1"]
mock_request_content.return_value = stub_content_too_large_data
url = reverse("browse-content", url_args={"query_string": content_sha1})
url_raw = reverse("browse-content-raw", url_args={"query_string": content_sha1})
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/content.html"
)
assert_contains(resp, "Content is too large to be displayed")
assert_contains(resp, url_raw)
@given(content())
def test_content_uppercase(client, content):
url = reverse(
"browse-content-uppercase-checksum",
url_args={"query_string": content["sha1"].upper()},
)
resp = check_html_get_response(client, url, status_code=302)
redirect_url = reverse("browse-content", url_args={"query_string": content["sha1"]})
assert resp["location"] == redirect_url
@given(content_utf8_detected_as_binary())
def test_content_utf8_detected_as_binary_display(client, archive_data, content):
url = reverse("browse-content", url_args={"query_string": content["sha1"]})
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/content.html"
)
content_display = _process_content_for_display(archive_data, content)
assert_contains(resp, escape(content_display["content_data"]))
@given(origin_with_multiple_visits())
def test_content_origin_snapshot_branch_browse(client, archive_data, origin):
visits = archive_data.origin_visit_get(origin["url"])
visit = random.choice(visits)
snapshot = archive_data.snapshot_get(visit["snapshot"])
+ snapshot_sizes = archive_data.snapshot_count_branches(visit["snapshot"])
branches, releases = process_snapshot_branches(snapshot)
branch_info = random.choice(branches)
directory = archive_data.revision_get(branch_info["revision"])["directory"]
directory_content = archive_data.directory_ls(directory)
directory_file = random.choice(
[e for e in directory_content if e["type"] == "file"]
)
url = reverse(
"browse-content",
url_args={"query_string": directory_file["checksums"]["sha1"]},
query_params={
"origin_url": origin["url"],
"snapshot": snapshot["id"],
"branch": branch_info["name"],
"path": directory_file["name"],
},
)
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/content.html"
)
- _check_origin_snapshot_related_html(resp, origin, snapshot, branches, releases)
+ _check_origin_snapshot_related_html(
+ resp, origin, snapshot, snapshot_sizes, branches, releases
+ )
assert_contains(resp, directory_file["name"])
assert_contains(resp, f"Branch: <strong>{branch_info['name']}</strong>")
cnt_swhid = gen_swhid(
CONTENT,
directory_file["checksums"]["sha1_git"],
metadata={
"origin": origin["url"],
"visit": gen_swhid(SNAPSHOT, snapshot),
"anchor": gen_swhid(REVISION, branch_info["revision"]),
"path": f"/{directory_file['name']}",
},
)
assert_contains(resp, cnt_swhid)
dir_swhid = gen_swhid(
DIRECTORY,
directory,
metadata={
"origin": origin["url"],
"visit": gen_swhid(SNAPSHOT, snapshot),
"anchor": gen_swhid(REVISION, branch_info["revision"]),
},
)
assert_contains(resp, dir_swhid)
rev_swhid = gen_swhid(
REVISION,
branch_info["revision"],
metadata={"origin": origin["url"], "visit": gen_swhid(SNAPSHOT, snapshot),},
)
assert_contains(resp, rev_swhid)
snp_swhid = gen_swhid(SNAPSHOT, snapshot, metadata={"origin": origin["url"],},)
assert_contains(resp, snp_swhid)
@given(origin_with_multiple_visits())
def test_content_origin_snapshot_release_browse(client, archive_data, origin):
visits = archive_data.origin_visit_get(origin["url"])
visit = random.choice(visits)
snapshot = archive_data.snapshot_get(visit["snapshot"])
+ snapshot_sizes = archive_data.snapshot_count_branches(visit["snapshot"])
branches, releases = process_snapshot_branches(snapshot)
release_info = random.choice(releases)
directory_content = archive_data.directory_ls(release_info["directory"])
directory_file = random.choice(
[e for e in directory_content if e["type"] == "file"]
)
url = reverse(
"browse-content",
url_args={"query_string": directory_file["checksums"]["sha1"]},
query_params={
"origin_url": origin["url"],
"snapshot": snapshot["id"],
"release": release_info["name"],
"path": directory_file["name"],
},
)
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/content.html"
)
- _check_origin_snapshot_related_html(resp, origin, snapshot, branches, releases)
+ _check_origin_snapshot_related_html(
+ resp, origin, snapshot, snapshot_sizes, branches, releases
+ )
assert_contains(resp, directory_file["name"])
assert_contains(resp, f"Release: <strong>{release_info['name']}</strong>")
cnt_swhid = gen_swhid(
CONTENT,
directory_file["checksums"]["sha1_git"],
metadata={
"origin": origin["url"],
"visit": gen_swhid(SNAPSHOT, snapshot),
"anchor": gen_swhid(RELEASE, release_info["id"]),
"path": f"/{directory_file['name']}",
},
)
assert_contains(resp, cnt_swhid)
dir_swhid = gen_swhid(
DIRECTORY,
release_info["directory"],
metadata={
"origin": origin["url"],
"visit": gen_swhid(SNAPSHOT, snapshot),
"anchor": gen_swhid(RELEASE, release_info["id"]),
},
)
assert_contains(resp, dir_swhid)
rev_swhid = gen_swhid(
REVISION,
release_info["target"],
metadata={"origin": origin["url"], "visit": gen_swhid(SNAPSHOT, snapshot),},
)
assert_contains(resp, rev_swhid)
rel_swhid = gen_swhid(
RELEASE,
release_info["id"],
metadata={"origin": origin["url"], "visit": gen_swhid(SNAPSHOT, snapshot),},
)
assert_contains(resp, rel_swhid)
snp_swhid = gen_swhid(SNAPSHOT, snapshot, metadata={"origin": origin["url"],},)
assert_contains(resp, snp_swhid)
-def _check_origin_snapshot_related_html(resp, origin, snapshot, branches, releases):
+def _check_origin_snapshot_related_html(
+ resp, origin, snapshot, snapshot_sizes, branches, releases
+):
browse_origin_url = reverse(
"browse-origin", query_params={"origin_url": origin["url"]}
)
assert_contains(resp, f'href="{browse_origin_url}"')
origin_branches_url = reverse(
"browse-origin-branches",
query_params={"origin_url": origin["url"], "snapshot": snapshot["id"]},
)
assert_contains(resp, f'href="{escape(origin_branches_url)}"')
- assert_contains(resp, f"Branches ({len(branches)})")
+ assert_contains(resp, f"Branches ({snapshot_sizes['revision']})")
origin_releases_url = reverse(
"browse-origin-releases",
query_params={"origin_url": origin["url"], "snapshot": snapshot["id"]},
)
assert_contains(resp, f'href="{escape(origin_releases_url)}"')
- assert_contains(resp, f"Releases ({len(releases)})")
+ assert_contains(resp, f"Releases ({snapshot_sizes['release']})")
assert_contains(resp, '<li class="swh-branch">', count=len(branches))
assert_contains(resp, '<li class="swh-release">', count=len(releases))
def _process_content_for_display(archive_data, content):
content_data = archive_data.content_get_data(content["sha1"])
mime_type, encoding = get_mimetype_and_encoding_for_content(content_data["data"])
mime_type, encoding, content_data = _re_encode_content(
mime_type, encoding, content_data["data"]
)
content_display = prepare_content_for_display(
content_data, mime_type, content["path"]
)
assert type(content_display["content_data"]) == str
return content_display
diff --git a/swh/web/tests/browse/views/test_directory.py b/swh/web/tests/browse/views/test_directory.py
index 62b7f789..fda45098 100644
--- a/swh/web/tests/browse/views/test_directory.py
+++ b/swh/web/tests/browse/views/test_directory.py
@@ -1,429 +1,437 @@
# Copyright (C) 2017-2020 The Software Heritage developers
# See the AUTHORS file at the top-level directory of this distribution
# License: GNU Affero General Public License version 3, or any later version
# See top-level LICENSE file for more information
import random
from hypothesis import given
from django.utils.html import escape
from swh.model.from_disk import DentryPerms
from swh.model.hashutil import hash_to_bytes, hash_to_hex
from swh.model.identifiers import DIRECTORY, RELEASE, REVISION, SNAPSHOT
from swh.model.model import (
Directory,
DirectoryEntry,
Origin,
OriginVisit,
OriginVisitStatus,
Revision,
RevisionType,
Snapshot,
SnapshotBranch,
TargetType,
TimestampWithTimezone,
)
from swh.storage.utils import now
from swh.web.browse.snapshot_context import process_snapshot_branches
from swh.web.common.identifiers import gen_swhid
from swh.web.common.utils import gen_path_info, reverse
from swh.web.tests.django_asserts import assert_contains
from swh.web.tests.strategies import (
directory,
directory_with_subdirs,
empty_directory,
invalid_sha1,
new_person,
new_swh_date,
origin_with_multiple_visits,
unknown_directory,
)
from swh.web.tests.utils import check_html_get_response
@given(directory())
def test_root_directory_view(client, archive_data, directory):
_directory_view_checks(client, directory, archive_data.directory_ls(directory))
@given(directory_with_subdirs())
def test_sub_directory_view(client, archive_data, directory):
dir_content = archive_data.directory_ls(directory)
subdir = random.choice([e for e in dir_content if e["type"] == "dir"])
subdir_content = archive_data.directory_ls(subdir["target"])
_directory_view_checks(client, directory, subdir_content, subdir["name"])
@given(empty_directory(), new_person(), new_swh_date())
def test_sub_directory_view_origin_context(
client, archive_data, empty_directory, person, date
):
origin_url = "test_sub_directory_view_origin_context"
subdir = Directory(
entries=(
DirectoryEntry(
name=b"foo",
type="dir",
target=hash_to_bytes(empty_directory),
perms=DentryPerms.directory,
),
DirectoryEntry(
name=b"bar",
type="dir",
target=hash_to_bytes(empty_directory),
perms=DentryPerms.directory,
),
)
)
parentdir = Directory(
entries=(
DirectoryEntry(
name=b"baz", type="dir", target=subdir.id, perms=DentryPerms.directory,
),
)
)
archive_data.directory_add([subdir, parentdir])
revision = Revision(
directory=parentdir.id,
author=person,
committer=person,
message=b"commit message",
date=TimestampWithTimezone.from_datetime(date),
committer_date=TimestampWithTimezone.from_datetime(date),
synthetic=False,
type=RevisionType.GIT,
)
archive_data.revision_add([revision])
snapshot = Snapshot(
branches={
b"HEAD": SnapshotBranch(
target="refs/head/master".encode(), target_type=TargetType.ALIAS,
),
b"refs/head/master": SnapshotBranch(
target=revision.id, target_type=TargetType.REVISION,
),
}
)
archive_data.snapshot_add([snapshot])
archive_data.origin_add([Origin(url=origin_url)])
date = now()
visit = OriginVisit(origin=origin_url, date=date, type="git")
visit = archive_data.origin_visit_add([visit])[0]
visit_status = OriginVisitStatus(
origin=origin_url,
visit=visit.visit,
date=date,
status="full",
snapshot=snapshot.id,
)
archive_data.origin_visit_status_add([visit_status])
dir_content = archive_data.directory_ls(hash_to_hex(parentdir.id))
subdir = dir_content[0]
subdir_content = archive_data.directory_ls(subdir["target"])
_directory_view_checks(
client,
hash_to_hex(parentdir.id),
subdir_content,
subdir["name"],
origin_url,
hash_to_hex(snapshot.id),
hash_to_hex(revision.id),
)
@given(invalid_sha1(), unknown_directory())
def test_directory_request_errors(client, invalid_sha1, unknown_directory):
dir_url = reverse("browse-directory", url_args={"sha1_git": invalid_sha1})
check_html_get_response(
client, dir_url, status_code=400, template_used="error.html"
)
dir_url = reverse("browse-directory", url_args={"sha1_git": unknown_directory})
check_html_get_response(
client, dir_url, status_code=404, template_used="error.html"
)
@given(directory())
def test_directory_uppercase(client, directory):
url = reverse(
"browse-directory-uppercase-checksum", url_args={"sha1_git": directory.upper()}
)
resp = check_html_get_response(client, url, status_code=302)
redirect_url = reverse("browse-directory", url_args={"sha1_git": directory})
assert resp["location"] == redirect_url
@given(directory())
def test_permalink_box_context(client, tests_data, directory):
origin_url = random.choice(tests_data["origins"])["url"]
url = reverse(
"browse-directory",
url_args={"sha1_git": directory},
query_params={"origin_url": origin_url},
)
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/directory.html"
)
assert_contains(resp, 'id="swhid-context-option-directory"')
@given(origin_with_multiple_visits())
def test_directory_origin_snapshot_branch_browse(client, archive_data, origin):
visits = archive_data.origin_visit_get(origin["url"])
visit = random.choice(visits)
snapshot = archive_data.snapshot_get(visit["snapshot"])
+ snapshot_sizes = archive_data.snapshot_count_branches(visit["snapshot"])
branches, releases = process_snapshot_branches(snapshot)
branch_info = random.choice(branches)
directory = archive_data.revision_get(branch_info["revision"])["directory"]
directory_content = archive_data.directory_ls(directory)
directory_subdir = random.choice(
[e for e in directory_content if e["type"] == "dir"]
)
url = reverse(
"browse-directory",
url_args={"sha1_git": directory},
query_params={
"origin_url": origin["url"],
"snapshot": snapshot["id"],
"branch": branch_info["name"],
"path": directory_subdir["name"],
},
)
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/directory.html"
)
- _check_origin_snapshot_related_html(resp, origin, snapshot, branches, releases)
+ _check_origin_snapshot_related_html(
+ resp, origin, snapshot, snapshot_sizes, branches, releases
+ )
assert_contains(resp, directory_subdir["name"])
assert_contains(resp, f"Branch: <strong>{branch_info['name']}</strong>")
dir_swhid = gen_swhid(
DIRECTORY,
directory_subdir["target"],
metadata={
"origin": origin["url"],
"visit": gen_swhid(SNAPSHOT, snapshot),
"anchor": gen_swhid(REVISION, branch_info["revision"]),
"path": "/",
},
)
assert_contains(resp, dir_swhid)
rev_swhid = gen_swhid(
REVISION,
branch_info["revision"],
metadata={"origin": origin["url"], "visit": gen_swhid(SNAPSHOT, snapshot),},
)
assert_contains(resp, rev_swhid)
snp_swhid = gen_swhid(SNAPSHOT, snapshot, metadata={"origin": origin["url"],},)
assert_contains(resp, snp_swhid)
@given(origin_with_multiple_visits())
def test_content_origin_snapshot_release_browse(client, archive_data, origin):
visits = archive_data.origin_visit_get(origin["url"])
visit = random.choice(visits)
snapshot = archive_data.snapshot_get(visit["snapshot"])
+ snapshot_sizes = archive_data.snapshot_count_branches(visit["snapshot"])
branches, releases = process_snapshot_branches(snapshot)
release_info = random.choice(releases)
directory = release_info["directory"]
directory_content = archive_data.directory_ls(directory)
directory_subdir = random.choice(
[e for e in directory_content if e["type"] == "dir"]
)
url = reverse(
"browse-directory",
url_args={"sha1_git": directory},
query_params={
"origin_url": origin["url"],
"snapshot": snapshot["id"],
"release": release_info["name"],
"path": directory_subdir["name"],
},
)
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/directory.html"
)
- _check_origin_snapshot_related_html(resp, origin, snapshot, branches, releases)
+ _check_origin_snapshot_related_html(
+ resp, origin, snapshot, snapshot_sizes, branches, releases
+ )
assert_contains(resp, directory_subdir["name"])
assert_contains(resp, f"Release: <strong>{release_info['name']}</strong>")
dir_swhid = gen_swhid(
DIRECTORY,
directory_subdir["target"],
metadata={
"origin": origin["url"],
"visit": gen_swhid(SNAPSHOT, snapshot),
"anchor": gen_swhid(RELEASE, release_info["id"]),
"path": "/",
},
)
assert_contains(resp, dir_swhid)
rev_swhid = gen_swhid(
REVISION,
release_info["target"],
metadata={"origin": origin["url"], "visit": gen_swhid(SNAPSHOT, snapshot),},
)
assert_contains(resp, rev_swhid)
rel_swhid = gen_swhid(
RELEASE,
release_info["id"],
metadata={"origin": origin["url"], "visit": gen_swhid(SNAPSHOT, snapshot),},
)
assert_contains(resp, rel_swhid)
snp_swhid = gen_swhid(SNAPSHOT, snapshot, metadata={"origin": origin["url"],},)
assert_contains(resp, snp_swhid)
-def _check_origin_snapshot_related_html(resp, origin, snapshot, branches, releases):
+def _check_origin_snapshot_related_html(
+ resp, origin, snapshot, snapshot_sizes, branches, releases
+):
browse_origin_url = reverse(
"browse-origin", query_params={"origin_url": origin["url"]}
)
assert_contains(resp, f'href="{browse_origin_url}"')
origin_branches_url = reverse(
"browse-origin-branches",
query_params={"origin_url": origin["url"], "snapshot": snapshot["id"]},
)
assert_contains(resp, f'href="{escape(origin_branches_url)}"')
- assert_contains(resp, f"Branches ({len(branches)})")
+ assert_contains(resp, f"Branches ({snapshot_sizes['revision']})")
origin_releases_url = reverse(
"browse-origin-releases",
query_params={"origin_url": origin["url"], "snapshot": snapshot["id"]},
)
assert_contains(resp, f'href="{escape(origin_releases_url)}"')
- assert_contains(resp, f"Releases ({len(releases)})")
+ assert_contains(resp, f"Releases ({snapshot_sizes['release']})")
assert_contains(resp, '<li class="swh-branch">', count=len(branches))
assert_contains(resp, '<li class="swh-release">', count=len(releases))
def _directory_view_checks(
client,
root_directory_sha1,
directory_entries,
path=None,
origin_url=None,
snapshot_id=None,
revision_id=None,
):
dirs = [e for e in directory_entries if e["type"] in ("dir", "rev")]
files = [e for e in directory_entries if e["type"] == "file"]
url_args = {"sha1_git": root_directory_sha1}
query_params = {"origin_url": origin_url, "snapshot": snapshot_id}
url = reverse(
"browse-directory",
url_args=url_args,
query_params={**query_params, "path": path},
)
root_dir_url = reverse(
"browse-directory", url_args=url_args, query_params=query_params,
)
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/directory.html"
)
assert_contains(
resp, '<a href="' + root_dir_url + '">' + root_directory_sha1[:7] + "</a>",
)
assert_contains(resp, '<td class="swh-directory">', count=len(dirs))
assert_contains(resp, '<td class="swh-content">', count=len(files))
for d in dirs:
if d["type"] == "rev":
dir_url = reverse("browse-revision", url_args={"sha1_git": d["target"]})
else:
dir_path = d["name"]
if path:
dir_path = "%s/%s" % (path, d["name"])
dir_url = reverse(
"browse-directory",
url_args={"sha1_git": root_directory_sha1},
query_params={**query_params, "path": dir_path},
)
assert_contains(resp, dir_url)
for f in files:
file_path = "%s/%s" % (root_directory_sha1, f["name"])
if path:
file_path = "%s/%s/%s" % (root_directory_sha1, path, f["name"])
query_string = "sha1_git:" + f["target"]
file_url = reverse(
"browse-content",
url_args={"query_string": query_string},
query_params={**query_params, "path": file_path},
)
assert_contains(resp, file_url)
path_info = gen_path_info(path)
assert_contains(resp, '<li class="swh-path">', count=len(path_info) + 1)
assert_contains(
resp, '<a href="%s">%s</a>' % (root_dir_url, root_directory_sha1[:7])
)
for p in path_info:
dir_url = reverse(
"browse-directory",
url_args={"sha1_git": root_directory_sha1},
query_params={**query_params, "path": p["path"]},
)
assert_contains(resp, '<a href="%s">%s</a>' % (dir_url, p["name"]))
assert_contains(resp, "vault-cook-directory")
swh_dir_id = gen_swhid(DIRECTORY, directory_entries[0]["dir_id"])
swh_dir_id_url = reverse("browse-swhid", url_args={"swhid": swh_dir_id})
swhid_context = {}
if origin_url:
swhid_context["origin"] = origin_url
if snapshot_id:
swhid_context["visit"] = gen_swhid(SNAPSHOT, snapshot_id)
if root_directory_sha1 != directory_entries[0]["dir_id"]:
swhid_context["anchor"] = gen_swhid(DIRECTORY, root_directory_sha1)
if root_directory_sha1 != directory_entries[0]["dir_id"]:
swhid_context["anchor"] = gen_swhid(DIRECTORY, root_directory_sha1)
if revision_id:
swhid_context["anchor"] = gen_swhid(REVISION, revision_id)
swhid_context["path"] = f"/{path}/" if path else None
swh_dir_id = gen_swhid(
DIRECTORY, directory_entries[0]["dir_id"], metadata=swhid_context
)
swh_dir_id_url = reverse("browse-swhid", url_args={"swhid": swh_dir_id})
assert_contains(resp, swh_dir_id)
assert_contains(resp, swh_dir_id_url)
diff --git a/swh/web/tests/browse/views/test_origin.py b/swh/web/tests/browse/views/test_origin.py
index 0551750a..4863f173 100644
--- a/swh/web/tests/browse/views/test_origin.py
+++ b/swh/web/tests/browse/views/test_origin.py
@@ -1,1249 +1,1276 @@
# Copyright (C) 2017-2020 The Software Heritage developers
# See the AUTHORS file at the top-level directory of this distribution
# License: GNU Affero General Public License version 3, or any later version
# See top-level LICENSE file for more information
import random
import re
import string
from hypothesis import given
from django.utils.html import escape
from swh.model.hashutil import hash_to_bytes
from swh.model.identifiers import CONTENT, DIRECTORY, RELEASE, REVISION, SNAPSHOT
from swh.model.model import (
OriginVisit,
OriginVisitStatus,
Snapshot,
SnapshotBranch,
TargetType,
)
from swh.storage.utils import now
from swh.web.browse.snapshot_context import process_snapshot_branches
from swh.web.common.exc import NotFoundExc
from swh.web.common.identifiers import gen_swhid
from swh.web.common.utils import (
format_utc_iso_date,
gen_path_info,
parse_iso8601_date_to_utc,
reverse,
)
from swh.web.tests.data import get_content, random_sha1
from swh.web.tests.django_asserts import assert_contains
from swh.web.tests.strategies import (
new_origin,
new_snapshot,
origin,
origin_with_multiple_visits,
origin_with_releases,
)
from swh.web.tests.strategies import release as existing_release
from swh.web.tests.strategies import revisions, unknown_revision, visit_dates
from swh.web.tests.utils import check_html_get_response
@given(origin_with_multiple_visits())
def test_origin_visits_browse(client, archive_data, origin):
url = reverse("browse-origin-visits", query_params={"origin_url": origin["url"]})
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/origin-visits.html"
)
visits = archive_data.origin_visit_get(origin["url"])
for v in visits:
vdate = format_utc_iso_date(v["date"], "%Y-%m-%dT%H:%M:%SZ")
browse_dir_url = reverse(
"browse-origin-directory",
query_params={"origin_url": origin["url"], "timestamp": vdate},
)
assert_contains(resp, browse_dir_url)
_check_origin_link(resp, origin["url"])
@given(origin_with_multiple_visits())
def test_origin_content_view(client, archive_data, origin):
origin_visits = archive_data.origin_visit_get(origin["url"])
def _get_archive_data(visit_idx):
snapshot = archive_data.snapshot_get(origin_visits[visit_idx]["snapshot"])
head_rev_id = archive_data.snapshot_get_head(snapshot)
head_rev = archive_data.revision_get(head_rev_id)
dir_content = archive_data.directory_ls(head_rev["directory"])
dir_files = [e for e in dir_content if e["type"] == "file"]
dir_file = random.choice(dir_files)
branches, releases = process_snapshot_branches(snapshot)
return {
"branches": branches,
"releases": releases,
"root_dir_sha1": head_rev["directory"],
"content": get_content(dir_file["checksums"]["sha1"]),
"visit": origin_visits[visit_idx],
+ "snapshot_sizes": archive_data.snapshot_count_branches(snapshot["id"]),
}
tdata = _get_archive_data(-1)
_origin_content_view_test_helper(
client,
archive_data,
origin,
origin_visits[-1],
+ tdata["snapshot_sizes"],
tdata["branches"],
tdata["releases"],
tdata["root_dir_sha1"],
tdata["content"],
)
_origin_content_view_test_helper(
client,
archive_data,
origin,
origin_visits[-1],
+ tdata["snapshot_sizes"],
tdata["branches"],
tdata["releases"],
tdata["root_dir_sha1"],
tdata["content"],
timestamp=tdata["visit"]["date"],
)
_origin_content_view_test_helper(
client,
archive_data,
origin,
origin_visits[-1],
+ tdata["snapshot_sizes"],
tdata["branches"],
tdata["releases"],
tdata["root_dir_sha1"],
tdata["content"],
snapshot_id=tdata["visit"]["snapshot"],
)
tdata = _get_archive_data(0)
_origin_content_view_test_helper(
client,
archive_data,
origin,
origin_visits[0],
+ tdata["snapshot_sizes"],
tdata["branches"],
tdata["releases"],
tdata["root_dir_sha1"],
tdata["content"],
visit_id=tdata["visit"]["visit"],
)
_origin_content_view_test_helper(
client,
archive_data,
origin,
origin_visits[0],
+ tdata["snapshot_sizes"],
tdata["branches"],
tdata["releases"],
tdata["root_dir_sha1"],
tdata["content"],
snapshot_id=tdata["visit"]["snapshot"],
)
@given(origin())
def test_origin_root_directory_view(client, archive_data, origin):
origin_visits = archive_data.origin_visit_get(origin["url"])
visit = origin_visits[-1]
snapshot = archive_data.snapshot_get(visit["snapshot"])
+ snapshot_sizes = archive_data.snapshot_count_branches(snapshot["id"])
head_rev_id = archive_data.snapshot_get_head(snapshot)
head_rev = archive_data.revision_get(head_rev_id)
root_dir_sha1 = head_rev["directory"]
dir_content = archive_data.directory_ls(root_dir_sha1)
branches, releases = process_snapshot_branches(snapshot)
_origin_directory_view_test_helper(
client,
archive_data,
origin,
visit,
+ snapshot_sizes,
branches,
releases,
root_dir_sha1,
dir_content,
)
_origin_directory_view_test_helper(
client,
archive_data,
origin,
visit,
+ snapshot_sizes,
branches,
releases,
root_dir_sha1,
dir_content,
visit_id=visit["visit"],
)
_origin_directory_view_test_helper(
client,
archive_data,
origin,
visit,
+ snapshot_sizes,
branches,
releases,
root_dir_sha1,
dir_content,
timestamp=visit["date"],
)
_origin_directory_view_test_helper(
client,
archive_data,
origin,
visit,
+ snapshot_sizes,
branches,
releases,
root_dir_sha1,
dir_content,
snapshot_id=visit["snapshot"],
)
_origin_directory_view_test_helper(
client,
archive_data,
origin,
visit,
+ snapshot_sizes,
branches,
releases,
root_dir_sha1,
dir_content,
)
_origin_directory_view_test_helper(
client,
archive_data,
origin,
visit,
+ snapshot_sizes,
branches,
releases,
root_dir_sha1,
dir_content,
visit_id=visit["visit"],
)
_origin_directory_view_test_helper(
client,
archive_data,
origin,
visit,
+ snapshot_sizes,
branches,
releases,
root_dir_sha1,
dir_content,
timestamp=visit["date"],
)
_origin_directory_view_test_helper(
client,
archive_data,
origin,
visit,
+ snapshot_sizes,
branches,
releases,
root_dir_sha1,
dir_content,
snapshot_id=visit["snapshot"],
)
@given(origin())
def test_origin_sub_directory_view(client, archive_data, origin):
origin_visits = archive_data.origin_visit_get(origin["url"])
visit = origin_visits[-1]
snapshot = archive_data.snapshot_get(visit["snapshot"])
+ snapshot_sizes = archive_data.snapshot_count_branches(snapshot["id"])
head_rev_id = archive_data.snapshot_get_head(snapshot)
head_rev = archive_data.revision_get(head_rev_id)
root_dir_sha1 = head_rev["directory"]
subdirs = [
e for e in archive_data.directory_ls(root_dir_sha1) if e["type"] == "dir"
]
branches, releases = process_snapshot_branches(snapshot)
if len(subdirs) == 0:
return
subdir = random.choice(subdirs)
subdir_content = archive_data.directory_ls(subdir["target"])
subdir_path = subdir["name"]
_origin_directory_view_test_helper(
client,
archive_data,
origin,
visit,
+ snapshot_sizes,
branches,
releases,
root_dir_sha1,
subdir_content,
path=subdir_path,
)
_origin_directory_view_test_helper(
client,
archive_data,
origin,
visit,
+ snapshot_sizes,
branches,
releases,
root_dir_sha1,
subdir_content,
path=subdir_path,
visit_id=visit["visit"],
)
_origin_directory_view_test_helper(
client,
archive_data,
origin,
visit,
+ snapshot_sizes,
branches,
releases,
root_dir_sha1,
subdir_content,
path=subdir_path,
timestamp=visit["date"],
)
_origin_directory_view_test_helper(
client,
archive_data,
origin,
visit,
+ snapshot_sizes,
branches,
releases,
root_dir_sha1,
subdir_content,
path=subdir_path,
snapshot_id=visit["snapshot"],
)
_origin_directory_view_test_helper(
client,
archive_data,
origin,
visit,
+ snapshot_sizes,
branches,
releases,
root_dir_sha1,
subdir_content,
path=subdir_path,
)
_origin_directory_view_test_helper(
client,
archive_data,
origin,
visit,
+ snapshot_sizes,
branches,
releases,
root_dir_sha1,
subdir_content,
path=subdir_path,
visit_id=visit["visit"],
)
_origin_directory_view_test_helper(
client,
archive_data,
origin,
visit,
+ snapshot_sizes,
branches,
releases,
root_dir_sha1,
subdir_content,
path=subdir_path,
timestamp=visit["date"],
)
_origin_directory_view_test_helper(
client,
archive_data,
origin,
visit,
+ snapshot_sizes,
branches,
releases,
root_dir_sha1,
subdir_content,
path=subdir_path,
snapshot_id=visit["snapshot"],
)
@given(origin())
def test_origin_branches(client, archive_data, origin):
origin_visits = archive_data.origin_visit_get(origin["url"])
visit = origin_visits[-1]
snapshot = archive_data.snapshot_get(visit["snapshot"])
+ snapshot_sizes = archive_data.snapshot_count_branches(snapshot["id"])
snapshot_content = process_snapshot_branches(snapshot)
- _origin_branches_test_helper(client, origin, snapshot_content)
+ _origin_branches_test_helper(client, origin, snapshot_content, snapshot_sizes)
_origin_branches_test_helper(
- client, origin, snapshot_content, snapshot_id=visit["snapshot"]
+ client, origin, snapshot_content, snapshot_sizes, snapshot_id=visit["snapshot"]
)
@given(origin())
def test_origin_releases(client, archive_data, origin):
origin_visits = archive_data.origin_visit_get(origin["url"])
visit = origin_visits[-1]
snapshot = archive_data.snapshot_get(visit["snapshot"])
+ snapshot_sizes = archive_data.snapshot_count_branches(snapshot["id"])
snapshot_content = process_snapshot_branches(snapshot)
- _origin_releases_test_helper(client, origin, snapshot_content)
+ _origin_releases_test_helper(client, origin, snapshot_content, snapshot_sizes)
_origin_releases_test_helper(
- client, origin, snapshot_content, snapshot_id=visit["snapshot"]
+ client, origin, snapshot_content, snapshot_sizes, snapshot_id=visit["snapshot"]
)
@given(
new_origin(),
new_snapshot(min_size=4, max_size=4),
visit_dates(),
revisions(min_size=3, max_size=3),
)
def test_origin_snapshot_null_branch(
client, archive_data, new_origin, new_snapshot, visit_dates, revisions
):
snp_dict = new_snapshot.to_dict()
archive_data.origin_add([new_origin])
for i, branch in enumerate(snp_dict["branches"].keys()):
if i == 0:
snp_dict["branches"][branch] = None
else:
snp_dict["branches"][branch] = {
"target_type": "revision",
"target": hash_to_bytes(revisions[i - 1]),
}
archive_data.snapshot_add([Snapshot.from_dict(snp_dict)])
visit = archive_data.origin_visit_add(
[OriginVisit(origin=new_origin.url, date=visit_dates[0], type="git",)]
)[0]
visit_status = OriginVisitStatus(
origin=new_origin.url,
visit=visit.visit,
date=now(),
status="partial",
snapshot=snp_dict["id"],
)
archive_data.origin_visit_status_add([visit_status])
url = reverse(
"browse-origin-directory", query_params={"origin_url": new_origin.url}
)
check_html_get_response(
client, url, status_code=200, template_used="browse/directory.html"
)
@given(
new_origin(),
new_snapshot(min_size=4, max_size=4),
visit_dates(),
revisions(min_size=4, max_size=4),
)
def test_origin_snapshot_invalid_branch(
client, archive_data, new_origin, new_snapshot, visit_dates, revisions
):
snp_dict = new_snapshot.to_dict()
archive_data.origin_add([new_origin])
for i, branch in enumerate(snp_dict["branches"].keys()):
snp_dict["branches"][branch] = {
"target_type": "revision",
"target": hash_to_bytes(revisions[i]),
}
archive_data.snapshot_add([Snapshot.from_dict(snp_dict)])
visit = archive_data.origin_visit_add(
[OriginVisit(origin=new_origin.url, date=visit_dates[0], type="git",)]
)[0]
visit_status = OriginVisitStatus(
origin=new_origin.url,
visit=visit.visit,
date=now(),
status="full",
snapshot=snp_dict["id"],
)
archive_data.origin_visit_status_add([visit_status])
url = reverse(
"browse-origin-directory",
query_params={"origin_url": new_origin.url, "branch": "invalid_branch"},
)
check_html_get_response(client, url, status_code=404, template_used="error.html")
@given(new_origin())
def test_browse_visits_origin_not_found(client, new_origin):
url = reverse("browse-origin-visits", query_params={"origin_url": new_origin.url})
resp = check_html_get_response(
client, url, status_code=404, template_used="error.html"
)
assert_contains(
resp, f"Origin with url {new_origin.url} not found", status_code=404
)
@given(origin())
def test_browse_origin_directory_no_visit(client, mocker, origin):
mock_get_origin_visits = mocker.patch(
"swh.web.common.origin_visits.get_origin_visits"
)
mock_get_origin_visits.return_value = []
mock_archive = mocker.patch("swh.web.common.origin_visits.archive")
mock_archive.lookup_origin_visit_latest.return_value = None
url = reverse("browse-origin-directory", query_params={"origin_url": origin["url"]})
resp = check_html_get_response(
client, url, status_code=404, template_used="error.html"
)
assert_contains(resp, "No valid visit", status_code=404)
assert not mock_get_origin_visits.called
@given(origin())
def test_browse_origin_directory_unknown_visit(client, mocker, origin):
mock_get_origin_visits = mocker.patch(
"swh.web.common.origin_visits.get_origin_visits"
)
mock_get_origin_visits.return_value = [{"visit": 1}]
url = reverse(
"browse-origin-directory",
query_params={"origin_url": origin["url"], "visit_id": 2},
)
resp = check_html_get_response(
client, url, status_code=404, template_used="error.html"
)
assert re.search("Visit.*not found", resp.content.decode("utf-8"))
assert mock_get_origin_visits.called
@given(origin())
def test_browse_origin_directory_not_found(client, origin):
url = reverse(
"browse-origin-directory",
query_params={"origin_url": origin["url"], "path": "/invalid/dir/path/"},
)
resp = check_html_get_response(
client, url, status_code=404, template_used="error.html"
)
assert re.search("Directory.*not found", resp.content.decode("utf-8"))
@given(origin())
def test_browse_origin_content_no_visit(client, mocker, origin):
mock_get_origin_visits = mocker.patch(
"swh.web.common.origin_visits.get_origin_visits"
)
mock_get_origin_visits.return_value = []
mock_archive = mocker.patch("swh.web.common.origin_visits.archive")
mock_archive.lookup_origin_visit_latest.return_value = None
url = reverse(
"browse-origin-content",
query_params={"origin_url": origin["url"], "path": "foo"},
)
resp = check_html_get_response(
client, url, status_code=404, template_used="error.html"
)
assert_contains(resp, "No valid visit", status_code=404)
assert not mock_get_origin_visits.called
@given(origin())
def test_browse_origin_content_unknown_visit(client, mocker, origin):
mock_get_origin_visits = mocker.patch(
"swh.web.common.origin_visits.get_origin_visits"
)
mock_get_origin_visits.return_value = [{"visit": 1}]
url = reverse(
"browse-origin-content",
query_params={"origin_url": origin["url"], "path": "foo", "visit_id": 2},
)
resp = check_html_get_response(
client, url, status_code=404, template_used="error.html"
)
assert re.search("Visit.*not found", resp.content.decode("utf-8"))
assert mock_get_origin_visits.called
@given(origin())
def test_browse_origin_content_directory_empty_snapshot(client, mocker, origin):
mock_snapshot_archive = mocker.patch("swh.web.browse.snapshot_context.archive")
mock_get_origin_visit_snapshot = mocker.patch(
"swh.web.browse.snapshot_context.get_origin_visit_snapshot"
)
mock_get_origin_visit_snapshot.return_value = ([], [])
mock_snapshot_archive.lookup_origin.return_value = origin
mock_snapshot_archive.lookup_snapshot_sizes.return_value = {
"revision": 0,
"release": 0,
}
for browse_context in ("content", "directory"):
url = reverse(
f"browse-origin-{browse_context}",
query_params={"origin_url": origin["url"], "path": "baz"},
)
resp = check_html_get_response(
client, url, status_code=200, template_used=f"browse/{browse_context}.html"
)
assert re.search("snapshot.*is empty", resp.content.decode("utf-8"))
assert mock_get_origin_visit_snapshot.called
assert mock_snapshot_archive.lookup_origin.called
assert mock_snapshot_archive.lookup_snapshot_sizes.called
@given(origin())
def test_browse_origin_content_not_found(client, origin):
url = reverse(
"browse-origin-content",
query_params={"origin_url": origin["url"], "path": "/invalid/file/path"},
)
resp = check_html_get_response(
client, url, status_code=404, template_used="error.html"
)
assert re.search("Directory entry.*not found", resp.content.decode("utf-8"))
@given(origin())
def test_browse_directory_snapshot_not_found(client, mocker, origin):
mock_get_snapshot_context = mocker.patch(
"swh.web.browse.snapshot_context.get_snapshot_context"
)
mock_get_snapshot_context.side_effect = NotFoundExc("Snapshot not found")
url = reverse("browse-origin-directory", query_params={"origin_url": origin["url"]})
resp = check_html_get_response(
client, url, status_code=404, template_used="error.html"
)
assert_contains(resp, "Snapshot not found", status_code=404)
assert mock_get_snapshot_context.called
@given(origin())
def test_origin_empty_snapshot(client, mocker, origin):
mock_archive = mocker.patch("swh.web.browse.snapshot_context.archive")
mock_get_origin_visit_snapshot = mocker.patch(
"swh.web.browse.snapshot_context.get_origin_visit_snapshot"
)
mock_get_origin_visit_snapshot.return_value = ([], [])
mock_archive.lookup_snapshot_sizes.return_value = {
"revision": 0,
"release": 0,
}
mock_archive.lookup_origin.return_value = origin
url = reverse("browse-origin-directory", query_params={"origin_url": origin["url"]})
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/directory.html"
)
resp_content = resp.content.decode("utf-8")
assert re.search("snapshot.*is empty", resp_content)
assert not re.search("swh-tr-link", resp_content)
assert mock_get_origin_visit_snapshot.called
assert mock_archive.lookup_snapshot_sizes.called
@given(new_origin())
def test_origin_empty_snapshot_null_revision(client, archive_data, new_origin):
snapshot = Snapshot(
branches={
b"HEAD": SnapshotBranch(
target="refs/head/master".encode(), target_type=TargetType.ALIAS,
),
b"refs/head/master": None,
}
)
archive_data.origin_add([new_origin])
archive_data.snapshot_add([snapshot])
visit = archive_data.origin_visit_add(
[OriginVisit(origin=new_origin.url, date=now(), type="git",)]
)[0]
visit_status = OriginVisitStatus(
origin=new_origin.url,
visit=visit.visit,
date=now(),
status="partial",
snapshot=snapshot.id,
)
archive_data.origin_visit_status_add([visit_status])
url = reverse(
"browse-origin-directory", query_params={"origin_url": new_origin.url},
)
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/directory.html"
)
resp_content = resp.content.decode("utf-8")
assert re.search("snapshot.*is empty", resp_content)
assert not re.search("swh-tr-link", resp_content)
@given(origin_with_releases())
def test_origin_release_browse(client, archive_data, origin):
snapshot = archive_data.snapshot_get_latest(origin["url"])
release = [
b for b in snapshot["branches"].values() if b["target_type"] == "release"
][-1]
release_data = archive_data.release_get(release["target"])
revision_data = archive_data.revision_get(release_data["target"])
url = reverse(
"browse-origin-directory",
query_params={"origin_url": origin["url"], "release": release_data["name"]},
)
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/directory.html"
)
assert_contains(resp, release_data["name"])
assert_contains(resp, release["target"])
swhid_context = {
"origin": origin["url"],
"visit": gen_swhid(SNAPSHOT, snapshot["id"]),
"anchor": gen_swhid(RELEASE, release_data["id"]),
}
swh_dir_id = gen_swhid(
DIRECTORY, revision_data["directory"], metadata=swhid_context
)
swh_dir_id_url = reverse("browse-swhid", url_args={"swhid": swh_dir_id})
assert_contains(resp, swh_dir_id)
assert_contains(resp, swh_dir_id_url)
@given(origin_with_releases())
def test_origin_release_browse_not_found(client, origin):
invalid_release_name = "swh-foo-bar"
url = reverse(
"browse-origin-directory",
query_params={"origin_url": origin["url"], "release": invalid_release_name},
)
resp = check_html_get_response(
client, url, status_code=404, template_used="error.html"
)
assert re.search(
f"Release {invalid_release_name}.*not found", resp.content.decode("utf-8")
)
@given(new_origin(), unknown_revision())
def test_origin_browse_directory_branch_with_non_resolvable_revision(
client, archive_data, new_origin, unknown_revision
):
branch_name = "master"
snapshot = Snapshot(
branches={
branch_name.encode(): SnapshotBranch(
target=hash_to_bytes(unknown_revision), target_type=TargetType.REVISION,
)
}
)
archive_data.origin_add([new_origin])
archive_data.snapshot_add([snapshot])
visit = archive_data.origin_visit_add(
[OriginVisit(origin=new_origin.url, date=now(), type="git",)]
)[0]
visit_status = OriginVisitStatus(
origin=new_origin.url,
visit=visit.visit,
date=now(),
status="partial",
snapshot=snapshot.id,
)
archive_data.origin_visit_status_add([visit_status])
url = reverse(
"browse-origin-directory",
query_params={"origin_url": new_origin.url, "branch": branch_name},
)
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/directory.html"
)
assert_contains(
resp, f"Revision {unknown_revision } could not be found in the archive."
)
@given(origin())
def test_origin_content_no_path(client, origin):
url = reverse("browse-origin-content", query_params={"origin_url": origin["url"]})
resp = check_html_get_response(
client, url, status_code=400, template_used="error.html"
)
assert_contains(
resp, "The path of a content must be given as query parameter.", status_code=400
)
def test_origin_views_no_url_query_parameter(client):
for browse_context in (
"content",
"directory",
"log",
"branches",
"releases",
"visits",
):
url = reverse(f"browse-origin-{browse_context}")
resp = check_html_get_response(
client, url, status_code=400, template_used="error.html"
)
assert_contains(
resp, "An origin URL must be provided as query parameter.", status_code=400
)
def _origin_content_view_test_helper(
client,
archive_data,
origin_info,
origin_visit,
+ snapshot_sizes,
origin_branches,
origin_releases,
root_dir_sha1,
content,
visit_id=None,
timestamp=None,
snapshot_id=None,
):
content_path = "/".join(content["path"].split("/")[1:])
if not visit_id and not snapshot_id:
visit_id = origin_visit["visit"]
query_params = {"origin_url": origin_info["url"], "path": content_path}
if timestamp:
query_params["timestamp"] = timestamp
if visit_id:
query_params["visit_id"] = visit_id
elif snapshot_id:
query_params["snapshot"] = snapshot_id
url = reverse("browse-origin-content", query_params=query_params)
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/content.html"
)
assert type(content["data"]) == str
assert_contains(resp, '<code class="%s">' % content["hljs_language"])
assert_contains(resp, escape(content["data"]))
split_path = content_path.split("/")
filename = split_path[-1]
path = content_path.replace(filename, "")[:-1]
path_info = gen_path_info(path)
del query_params["path"]
if timestamp:
query_params["timestamp"] = format_utc_iso_date(
parse_iso8601_date_to_utc(timestamp).isoformat(), "%Y-%m-%dT%H:%M:%SZ"
)
root_dir_url = reverse("browse-origin-directory", query_params=query_params)
assert_contains(resp, '<li class="swh-path">', count=len(path_info) + 1)
assert_contains(resp, '<a href="%s">%s</a>' % (root_dir_url, root_dir_sha1[:7]))
for p in path_info:
query_params["path"] = p["path"]
dir_url = reverse("browse-origin-directory", query_params=query_params)
assert_contains(resp, '<a href="%s">%s</a>' % (dir_url, p["name"]))
assert_contains(resp, "<li>%s</li>" % filename)
query_string = "sha1_git:" + content["sha1_git"]
url_raw = reverse(
"browse-content-raw",
url_args={"query_string": query_string},
query_params={"filename": filename},
)
assert_contains(resp, url_raw)
if "path" in query_params:
del query_params["path"]
origin_branches_url = reverse("browse-origin-branches", query_params=query_params)
assert_contains(resp, f'href="{escape(origin_branches_url)}"')
- assert_contains(resp, f"Branches ({len(origin_branches)})")
+ assert_contains(resp, f"Branches ({snapshot_sizes['revision']})")
origin_releases_url = reverse("browse-origin-releases", query_params=query_params)
assert_contains(resp, f'href="{escape(origin_releases_url)}">')
- assert_contains(resp, f"Releases ({len(origin_releases)})")
+ assert_contains(resp, f"Releases ({snapshot_sizes['release']})")
assert_contains(resp, '<li class="swh-branch">', count=len(origin_branches))
query_params["path"] = content_path
for branch in origin_branches:
root_dir_branch_url = reverse(
"browse-origin-content",
query_params={"branch": branch["name"], **query_params},
)
assert_contains(resp, '<a href="%s">' % root_dir_branch_url)
assert_contains(resp, '<li class="swh-release">', count=len(origin_releases))
query_params["branch"] = None
for release in origin_releases:
root_dir_release_url = reverse(
"browse-origin-content",
query_params={"release": release["name"], **query_params},
)
assert_contains(resp, '<a href="%s">' % root_dir_release_url)
url = reverse("browse-origin-content", query_params=query_params)
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/content.html"
)
snapshot = archive_data.snapshot_get(origin_visit["snapshot"])
head_rev_id = archive_data.snapshot_get_head(snapshot)
swhid_context = {
"origin": origin_info["url"],
"visit": gen_swhid(SNAPSHOT, snapshot["id"]),
"anchor": gen_swhid(REVISION, head_rev_id),
"path": f"/{content_path}",
}
swh_cnt_id = gen_swhid(CONTENT, content["sha1_git"], metadata=swhid_context)
swh_cnt_id_url = reverse("browse-swhid", url_args={"swhid": swh_cnt_id})
assert_contains(resp, swh_cnt_id)
assert_contains(resp, swh_cnt_id_url)
assert_contains(resp, "swh-take-new-snapshot")
_check_origin_link(resp, origin_info["url"])
def _origin_directory_view_test_helper(
client,
archive_data,
origin_info,
origin_visit,
+ snapshot_sizes,
origin_branches,
origin_releases,
root_directory_sha1,
directory_entries,
visit_id=None,
timestamp=None,
snapshot_id=None,
path=None,
):
dirs = [e for e in directory_entries if e["type"] in ("dir", "rev")]
files = [e for e in directory_entries if e["type"] == "file"]
if not visit_id and not snapshot_id:
visit_id = origin_visit["visit"]
query_params = {"origin_url": origin_info["url"]}
if timestamp:
query_params["timestamp"] = timestamp
elif visit_id:
query_params["visit_id"] = visit_id
else:
query_params["snapshot"] = snapshot_id
if path:
query_params["path"] = path
url = reverse("browse-origin-directory", query_params=query_params)
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/directory.html"
)
assert_contains(resp, '<td class="swh-directory">', count=len(dirs))
assert_contains(resp, '<td class="swh-content">', count=len(files))
if timestamp:
query_params["timestamp"] = format_utc_iso_date(
parse_iso8601_date_to_utc(timestamp).isoformat(), "%Y-%m-%dT%H:%M:%SZ"
)
for d in dirs:
if d["type"] == "rev":
dir_url = reverse("browse-revision", url_args={"sha1_git": d["target"]})
else:
dir_path = d["name"]
if path:
dir_path = "%s/%s" % (path, d["name"])
query_params["path"] = dir_path
dir_url = reverse("browse-origin-directory", query_params=query_params,)
assert_contains(resp, dir_url)
for f in files:
file_path = f["name"]
if path:
file_path = "%s/%s" % (path, f["name"])
query_params["path"] = file_path
file_url = reverse("browse-origin-content", query_params=query_params)
assert_contains(resp, file_url)
if "path" in query_params:
del query_params["path"]
root_dir_branch_url = reverse("browse-origin-directory", query_params=query_params)
nb_bc_paths = 1
if path:
nb_bc_paths = len(path.split("/")) + 1
assert_contains(resp, '<li class="swh-path">', count=nb_bc_paths)
assert_contains(
resp, '<a href="%s">%s</a>' % (root_dir_branch_url, root_directory_sha1[:7])
)
origin_branches_url = reverse("browse-origin-branches", query_params=query_params)
assert_contains(resp, f'href="{escape(origin_branches_url)}"')
- assert_contains(resp, f"Branches ({len(origin_branches)})")
+ assert_contains(resp, f"Branches ({snapshot_sizes['revision']})")
origin_releases_url = reverse("browse-origin-releases", query_params=query_params)
nb_releases = len(origin_releases)
if nb_releases > 0:
assert_contains(resp, f'href="{escape(origin_releases_url)}"')
- assert_contains(resp, f"Releases ({nb_releases})")
+ assert_contains(resp, f"Releases ({snapshot_sizes['release']})")
if path:
query_params["path"] = path
assert_contains(resp, '<li class="swh-branch">', count=len(origin_branches))
for branch in origin_branches:
query_params["branch"] = branch["name"]
root_dir_branch_url = reverse(
"browse-origin-directory", query_params=query_params
)
assert_contains(resp, '<a href="%s">' % root_dir_branch_url)
assert_contains(resp, '<li class="swh-release">', count=len(origin_releases))
query_params["branch"] = None
for release in origin_releases:
query_params["release"] = release["name"]
root_dir_release_url = reverse(
"browse-origin-directory", query_params=query_params
)
assert_contains(resp, 'href="%s"' % root_dir_release_url)
assert_contains(resp, "vault-cook-directory")
assert_contains(resp, "vault-cook-revision")
snapshot = archive_data.snapshot_get(origin_visit["snapshot"])
head_rev_id = archive_data.snapshot_get_head(snapshot)
swhid_context = {
"origin": origin_info["url"],
"visit": gen_swhid(SNAPSHOT, snapshot["id"]),
"anchor": gen_swhid(REVISION, head_rev_id),
"path": f"/{path}" if path else None,
}
swh_dir_id = gen_swhid(
DIRECTORY, directory_entries[0]["dir_id"], metadata=swhid_context
)
swh_dir_id_url = reverse("browse-swhid", url_args={"swhid": swh_dir_id})
assert_contains(resp, swh_dir_id)
assert_contains(resp, swh_dir_id_url)
assert_contains(resp, "swh-take-new-snapshot")
_check_origin_link(resp, origin_info["url"])
def _origin_branches_test_helper(
- client, origin_info, origin_snapshot, snapshot_id=None
+ client, origin_info, origin_snapshot, snapshot_sizes, snapshot_id=None
):
query_params = {"origin_url": origin_info["url"], "snapshot": snapshot_id}
url = reverse("browse-origin-branches", query_params=query_params)
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/branches.html"
)
origin_branches = origin_snapshot[0]
origin_releases = origin_snapshot[1]
origin_branches_url = reverse("browse-origin-branches", query_params=query_params)
assert_contains(resp, f'href="{escape(origin_branches_url)}"')
- assert_contains(resp, f"Branches ({len(origin_branches)})")
+ assert_contains(resp, f"Branches ({snapshot_sizes['revision']})")
origin_releases_url = reverse("browse-origin-releases", query_params=query_params)
nb_releases = len(origin_releases)
if nb_releases > 0:
assert_contains(resp, f'href="{escape(origin_releases_url)}">')
- assert_contains(resp, f"Releases ({nb_releases})")
+ assert_contains(resp, f"Releases ({snapshot_sizes['release']})")
assert_contains(resp, '<tr class="swh-branch-entry', count=len(origin_branches))
for branch in origin_branches:
browse_branch_url = reverse(
"browse-origin-directory",
query_params={"branch": branch["name"], **query_params},
)
assert_contains(resp, '<a href="%s">' % escape(browse_branch_url))
browse_revision_url = reverse(
"browse-revision",
url_args={"sha1_git": branch["revision"]},
query_params=query_params,
)
assert_contains(resp, '<a href="%s">' % escape(browse_revision_url))
_check_origin_link(resp, origin_info["url"])
def _origin_releases_test_helper(
- client, origin_info, origin_snapshot, snapshot_id=None
+ client, origin_info, origin_snapshot, snapshot_sizes, snapshot_id=None
):
query_params = {"origin_url": origin_info["url"], "snapshot": snapshot_id}
url = reverse("browse-origin-releases", query_params=query_params)
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/releases.html"
)
- origin_branches = origin_snapshot[0]
origin_releases = origin_snapshot[1]
origin_branches_url = reverse("browse-origin-branches", query_params=query_params)
assert_contains(resp, f'href="{escape(origin_branches_url)}"')
- assert_contains(resp, f"Branches ({len(origin_branches)})")
+ assert_contains(resp, f"Branches ({snapshot_sizes['revision']})")
origin_releases_url = reverse("browse-origin-releases", query_params=query_params)
nb_releases = len(origin_releases)
if nb_releases > 0:
assert_contains(resp, f'href="{escape(origin_releases_url)}"')
- assert_contains(resp, f"Releases ({nb_releases})")
+ assert_contains(resp, f"Releases ({snapshot_sizes['release']}")
assert_contains(resp, '<tr class="swh-release-entry', count=nb_releases)
for release in origin_releases:
browse_release_url = reverse(
"browse-release",
url_args={"sha1_git": release["id"]},
query_params=query_params,
)
browse_revision_url = reverse(
"browse-revision",
url_args={"sha1_git": release["target"]},
query_params=query_params,
)
assert_contains(resp, '<a href="%s">' % escape(browse_release_url))
assert_contains(resp, '<a href="%s">' % escape(browse_revision_url))
_check_origin_link(resp, origin_info["url"])
@given(
new_origin(), visit_dates(), revisions(min_size=10, max_size=10), existing_release()
)
def test_origin_branches_pagination_with_alias(
client, archive_data, mocker, new_origin, visit_dates, revisions, existing_release
):
"""
When a snapshot contains a branch or a release alias, pagination links
in the branches / releases view should be displayed.
"""
mocker.patch("swh.web.browse.snapshot_context.PER_PAGE", len(revisions) / 2)
snp_dict = {"branches": {}, "id": hash_to_bytes(random_sha1())}
for i in range(len(revisions)):
branch = "".join(random.choices(string.ascii_lowercase, k=8))
snp_dict["branches"][branch.encode()] = {
"target_type": "revision",
"target": hash_to_bytes(revisions[i]),
}
release = "".join(random.choices(string.ascii_lowercase, k=8))
snp_dict["branches"][b"RELEASE_ALIAS"] = {
"target_type": "alias",
"target": release.encode(),
}
snp_dict["branches"][release.encode()] = {
"target_type": "release",
"target": hash_to_bytes(existing_release),
}
archive_data.origin_add([new_origin])
archive_data.snapshot_add([Snapshot.from_dict(snp_dict)])
visit = archive_data.origin_visit_add(
[OriginVisit(origin=new_origin.url, date=visit_dates[0], type="git",)]
)[0]
visit_status = OriginVisitStatus(
origin=new_origin.url,
visit=visit.visit,
date=now(),
status="full",
snapshot=snp_dict["id"],
)
archive_data.origin_visit_status_add([visit_status])
url = reverse("browse-origin-branches", query_params={"origin_url": new_origin.url})
resp = check_html_get_response(
client, url, status_code=200, template_used="browse/branches.html"
)
assert_contains(resp, '<ul class="pagination')
def _check_origin_link(resp, origin_url):
browse_origin_url = reverse(
"browse-origin", query_params={"origin_url": origin_url}
)
assert_contains(resp, f'href="{browse_origin_url}"')
diff --git a/swh/web/tests/common/test_archive.py b/swh/web/tests/common/test_archive.py
index 8f47e054..1755cb8a 100644
--- a/swh/web/tests/common/test_archive.py
+++ b/swh/web/tests/common/test_archive.py
@@ -1,1015 +1,1040 @@
# Copyright (C) 2015-2020 The Software Heritage developers
# See the AUTHORS file at the top-level directory of this distribution
# License: GNU Affero General Public License version 3, or any later version
# See top-level LICENSE file for more information
from collections import defaultdict
import hashlib
import itertools
import random
from hypothesis import given
import pytest
from swh.model.from_disk import DentryPerms
from swh.model.hashutil import hash_to_bytes, hash_to_hex
from swh.model.identifiers import CONTENT, DIRECTORY, RELEASE, REVISION, SNAPSHOT
from swh.model.model import Directory, DirectoryEntry, Origin, OriginVisit, Revision
from swh.web.common import archive
from swh.web.common.exc import BadInputExc, NotFoundExc
from swh.web.common.typing import OriginInfo
from swh.web.tests.conftest import ctags_json_missing, fossology_missing
from swh.web.tests.data import random_content, random_sha1
from swh.web.tests.strategies import (
ancestor_revisions,
content,
contents,
contents_with_ctags,
directory,
empty_content,
empty_directory,
invalid_sha1,
new_origin,
new_revision,
non_ancestor_revisions,
origin,
release,
releases,
revision,
revision_with_submodules,
revisions,
sha256,
snapshot,
unknown_content,
unknown_contents,
unknown_directory,
unknown_release,
unknown_revision,
unknown_snapshot,
visit_dates,
)
@given(contents())
def test_lookup_multiple_hashes_all_present(contents):
input_data = []
expected_output = []
for cnt in contents:
input_data.append({"sha1": cnt["sha1"]})
expected_output.append({"sha1": cnt["sha1"], "found": True})
assert archive.lookup_multiple_hashes(input_data) == expected_output
@given(contents(), unknown_contents())
def test_lookup_multiple_hashes_some_missing(contents, unknown_contents):
input_contents = list(itertools.chain(contents, unknown_contents))
random.shuffle(input_contents)
input_data = []
expected_output = []
for cnt in input_contents:
input_data.append({"sha1": cnt["sha1"]})
expected_output.append({"sha1": cnt["sha1"], "found": cnt in contents})
assert archive.lookup_multiple_hashes(input_data) == expected_output
def test_lookup_hash_does_not_exist():
unknown_content_ = random_content()
actual_lookup = archive.lookup_hash("sha1_git:%s" % unknown_content_["sha1_git"])
assert actual_lookup == {"found": None, "algo": "sha1_git"}
@given(content())
def test_lookup_hash_exist(archive_data, content):
actual_lookup = archive.lookup_hash("sha1:%s" % content["sha1"])
content_metadata = archive_data.content_get(content["sha1"])
assert {"found": content_metadata, "algo": "sha1"} == actual_lookup
def test_search_hash_does_not_exist():
unknown_content_ = random_content()
actual_lookup = archive.search_hash("sha1_git:%s" % unknown_content_["sha1_git"])
assert {"found": False} == actual_lookup
@given(content())
def test_search_hash_exist(content):
actual_lookup = archive.search_hash("sha1:%s" % content["sha1"])
assert {"found": True} == actual_lookup
@pytest.mark.skipif(
ctags_json_missing, reason="requires ctags with json output support"
)
@given(contents_with_ctags())
def test_lookup_content_ctags(indexer_data, contents_with_ctags):
content_sha1 = random.choice(contents_with_ctags["sha1s"])
indexer_data.content_add_ctags(content_sha1)
actual_ctags = list(archive.lookup_content_ctags("sha1:%s" % content_sha1))
expected_data = list(indexer_data.content_get_ctags(content_sha1))
for ctag in expected_data:
ctag["id"] = content_sha1
assert actual_ctags == expected_data
def test_lookup_content_ctags_no_hash():
unknown_content_ = random_content()
actual_ctags = list(
archive.lookup_content_ctags("sha1:%s" % unknown_content_["sha1"])
)
assert actual_ctags == []
@given(content())
def test_lookup_content_filetype(indexer_data, content):
indexer_data.content_add_mimetype(content["sha1"])
actual_filetype = archive.lookup_content_filetype(content["sha1"])
expected_filetype = indexer_data.content_get_mimetype(content["sha1"])
assert actual_filetype == expected_filetype
@given(contents_with_ctags())
def test_lookup_expression(indexer_data, contents_with_ctags):
per_page = 10
expected_ctags = []
for content_sha1 in contents_with_ctags["sha1s"]:
if len(expected_ctags) == per_page:
break
indexer_data.content_add_ctags(content_sha1)
for ctag in indexer_data.content_get_ctags(content_sha1):
if len(expected_ctags) == per_page:
break
if ctag["name"] == contents_with_ctags["symbol_name"]:
del ctag["id"]
ctag["sha1"] = content_sha1
expected_ctags.append(ctag)
actual_ctags = list(
archive.lookup_expression(
contents_with_ctags["symbol_name"], last_sha1=None, per_page=10
)
)
assert actual_ctags == expected_ctags
def test_lookup_expression_no_result():
expected_ctags = []
actual_ctags = list(
archive.lookup_expression("barfoo", last_sha1=None, per_page=10)
)
assert actual_ctags == expected_ctags
@pytest.mark.skipif(fossology_missing, reason="requires fossology-nomossa installed")
@given(content())
def test_lookup_content_license(indexer_data, content):
indexer_data.content_add_license(content["sha1"])
actual_license = archive.lookup_content_license(content["sha1"])
expected_license = indexer_data.content_get_license(content["sha1"])
assert actual_license == expected_license
def test_stat_counters(archive_data):
actual_stats = archive.stat_counters()
assert actual_stats == archive_data.stat_counters()
@given(new_origin(), visit_dates())
def test_lookup_origin_visits(archive_data, new_origin, visit_dates):
archive_data.origin_add([new_origin])
archive_data.origin_visit_add(
[OriginVisit(origin=new_origin.url, date=ts, type="git",) for ts in visit_dates]
)
actual_origin_visits = list(
archive.lookup_origin_visits(new_origin.url, per_page=100)
)
expected_visits = archive_data.origin_visit_get(new_origin.url)
for expected_visit in expected_visits:
expected_visit["origin"] = new_origin.url
assert actual_origin_visits == expected_visits
@given(new_origin(), visit_dates())
def test_lookup_origin_visit(archive_data, new_origin, visit_dates):
archive_data.origin_add([new_origin])
visits = archive_data.origin_visit_add(
[OriginVisit(origin=new_origin.url, date=ts, type="git",) for ts in visit_dates]
)
visit = random.choice(visits).visit
actual_origin_visit = archive.lookup_origin_visit(new_origin.url, visit)
expected_visit = dict(archive_data.origin_visit_get_by(new_origin.url, visit))
assert actual_origin_visit == expected_visit
@given(new_origin())
def test_lookup_origin(archive_data, new_origin):
archive_data.origin_add([new_origin])
actual_origin = archive.lookup_origin({"url": new_origin.url})
expected_origin = archive_data.origin_get([new_origin.url])[0]
assert actual_origin == expected_origin
@given(invalid_sha1())
def test_lookup_release_ko_id_checksum_not_a_sha1(invalid_sha1):
with pytest.raises(BadInputExc) as e:
archive.lookup_release(invalid_sha1)
assert e.match("Invalid checksum")
@given(sha256())
def test_lookup_release_ko_id_checksum_too_long(sha256):
with pytest.raises(BadInputExc) as e:
archive.lookup_release(sha256)
assert e.match("Only sha1_git is supported.")
@given(releases())
def test_lookup_release_multiple(archive_data, releases):
actual_releases = list(archive.lookup_release_multiple(releases))
expected_releases = []
for release_id in releases:
release_info = archive_data.release_get(release_id)
expected_releases.append(release_info)
assert actual_releases == expected_releases
def test_lookup_release_multiple_none_found():
unknown_releases_ = [random_sha1(), random_sha1(), random_sha1()]
actual_releases = list(archive.lookup_release_multiple(unknown_releases_))
assert actual_releases == [None] * len(unknown_releases_)
@given(directory())
def test_lookup_directory_with_path_not_found(directory):
path = "some/invalid/path/here"
with pytest.raises(NotFoundExc) as e:
archive.lookup_directory_with_path(directory, path)
assert e.match("Directory entry with path %s from %s not found" % (path, directory))
@given(directory())
def test_lookup_directory_with_path_found(archive_data, directory):
directory_content = archive_data.directory_ls(directory)
directory_entry = random.choice(directory_content)
path = directory_entry["name"]
actual_result = archive.lookup_directory_with_path(directory, path)
assert actual_result == directory_entry
@given(release())
def test_lookup_release(archive_data, release):
actual_release = archive.lookup_release(release)
assert actual_release == archive_data.release_get(release)
@given(revision(), invalid_sha1(), sha256())
def test_lookup_revision_with_context_ko_not_a_sha1(revision, invalid_sha1, sha256):
sha1_git_root = revision
sha1_git = invalid_sha1
with pytest.raises(BadInputExc) as e:
archive.lookup_revision_with_context(sha1_git_root, sha1_git)
assert e.match("Invalid checksum query string")
sha1_git = sha256
with pytest.raises(BadInputExc) as e:
archive.lookup_revision_with_context(sha1_git_root, sha1_git)
assert e.match("Only sha1_git is supported")
@given(revision(), unknown_revision())
def test_lookup_revision_with_context_ko_sha1_git_does_not_exist(
revision, unknown_revision
):
sha1_git_root = revision
sha1_git = unknown_revision
with pytest.raises(NotFoundExc) as e:
archive.lookup_revision_with_context(sha1_git_root, sha1_git)
assert e.match("Revision %s not found" % sha1_git)
@given(revision(), unknown_revision())
def test_lookup_revision_with_context_ko_root_sha1_git_does_not_exist(
revision, unknown_revision
):
sha1_git_root = unknown_revision
sha1_git = revision
with pytest.raises(NotFoundExc) as e:
archive.lookup_revision_with_context(sha1_git_root, sha1_git)
assert e.match("Revision root %s not found" % sha1_git_root)
@given(ancestor_revisions())
def test_lookup_revision_with_context(archive_data, ancestor_revisions):
sha1_git = ancestor_revisions["sha1_git"]
root_sha1_git = ancestor_revisions["sha1_git_root"]
for sha1_git_root in (root_sha1_git, {"id": hash_to_bytes(root_sha1_git)}):
actual_revision = archive.lookup_revision_with_context(sha1_git_root, sha1_git)
children = []
for rev in archive_data.revision_log(root_sha1_git):
for p_rev in rev["parents"]:
p_rev_hex = hash_to_hex(p_rev)
if p_rev_hex == sha1_git:
children.append(rev["id"])
expected_revision = archive_data.revision_get(sha1_git)
expected_revision["children"] = children
assert actual_revision == expected_revision
@given(non_ancestor_revisions())
def test_lookup_revision_with_context_ko(non_ancestor_revisions):
sha1_git = non_ancestor_revisions["sha1_git"]
root_sha1_git = non_ancestor_revisions["sha1_git_root"]
with pytest.raises(NotFoundExc) as e:
archive.lookup_revision_with_context(root_sha1_git, sha1_git)
assert e.match("Revision %s is not an ancestor of %s" % (sha1_git, root_sha1_git))
def test_lookup_directory_with_revision_not_found():
unknown_revision_ = random_sha1()
with pytest.raises(NotFoundExc) as e:
archive.lookup_directory_with_revision(unknown_revision_)
assert e.match("Revision %s not found" % unknown_revision_)
@given(new_revision())
def test_lookup_directory_with_revision_unknown_content(archive_data, new_revision):
unknown_content_ = random_content()
dir_path = "README.md"
# A directory that points to unknown content
dir = Directory(
entries=(
DirectoryEntry(
name=bytes(dir_path.encode("utf-8")),
type="file",
target=hash_to_bytes(unknown_content_["sha1_git"]),
perms=DentryPerms.content,
),
)
)
# Create a revision that points to a directory
# Which points to unknown content
new_revision = new_revision.to_dict()
new_revision["directory"] = dir.id
del new_revision["id"]
new_revision = Revision.from_dict(new_revision)
# Add the directory and revision in mem
archive_data.directory_add([dir])
archive_data.revision_add([new_revision])
new_revision_id = hash_to_hex(new_revision.id)
with pytest.raises(NotFoundExc) as e:
archive.lookup_directory_with_revision(new_revision_id, dir_path)
assert e.match("Content not found for revision %s" % new_revision_id)
@given(revision())
def test_lookup_directory_with_revision_ko_path_to_nowhere(revision):
invalid_path = "path/to/something/unknown"
with pytest.raises(NotFoundExc) as e:
archive.lookup_directory_with_revision(revision, invalid_path)
assert e.match("Directory or File")
assert e.match(invalid_path)
assert e.match("revision %s" % revision)
assert e.match("not found")
@given(revision_with_submodules())
def test_lookup_directory_with_revision_submodules(
archive_data, revision_with_submodules
):
rev_sha1_git = revision_with_submodules["rev_sha1_git"]
rev_dir_path = revision_with_submodules["rev_dir_rev_path"]
actual_data = archive.lookup_directory_with_revision(rev_sha1_git, rev_dir_path)
revision = archive_data.revision_get(revision_with_submodules["rev_sha1_git"])
directory = archive_data.directory_ls(revision["directory"])
rev_entry = next(e for e in directory if e["name"] == rev_dir_path)
expected_data = {
"content": archive_data.revision_get(rev_entry["target"]),
"path": rev_dir_path,
"revision": rev_sha1_git,
"type": "rev",
}
assert actual_data == expected_data
@given(revision())
def test_lookup_directory_with_revision_without_path(archive_data, revision):
actual_directory_entries = archive.lookup_directory_with_revision(revision)
revision_data = archive_data.revision_get(revision)
expected_directory_entries = archive_data.directory_ls(revision_data["directory"])
assert actual_directory_entries["type"] == "dir"
assert actual_directory_entries["content"] == expected_directory_entries
@given(revision())
def test_lookup_directory_with_revision_with_path(archive_data, revision):
rev_data = archive_data.revision_get(revision)
dir_entries = [
e
for e in archive_data.directory_ls(rev_data["directory"])
if e["type"] in ("file", "dir")
]
expected_dir_entry = random.choice(dir_entries)
actual_dir_entry = archive.lookup_directory_with_revision(
revision, expected_dir_entry["name"]
)
assert actual_dir_entry["type"] == expected_dir_entry["type"]
assert actual_dir_entry["revision"] == revision
assert actual_dir_entry["path"] == expected_dir_entry["name"]
if actual_dir_entry["type"] == "file":
del actual_dir_entry["content"]["checksums"]["blake2s256"]
for key in ("checksums", "status", "length"):
assert actual_dir_entry["content"][key] == expected_dir_entry[key]
else:
sub_dir_entries = archive_data.directory_ls(expected_dir_entry["target"])
assert actual_dir_entry["content"] == sub_dir_entries
@given(revision())
def test_lookup_directory_with_revision_with_path_to_file_and_data(
archive_data, revision
):
rev_data = archive_data.revision_get(revision)
dir_entries = [
e
for e in archive_data.directory_ls(rev_data["directory"])
if e["type"] == "file"
]
expected_dir_entry = random.choice(dir_entries)
expected_data = archive_data.content_get_data(
expected_dir_entry["checksums"]["sha1"]
)
actual_dir_entry = archive.lookup_directory_with_revision(
revision, expected_dir_entry["name"], with_data=True
)
assert actual_dir_entry["type"] == expected_dir_entry["type"]
assert actual_dir_entry["revision"] == revision
assert actual_dir_entry["path"] == expected_dir_entry["name"]
del actual_dir_entry["content"]["checksums"]["blake2s256"]
for key in ("checksums", "status", "length"):
assert actual_dir_entry["content"][key] == expected_dir_entry[key]
assert actual_dir_entry["content"]["data"] == expected_data["data"]
@given(revision())
def test_lookup_revision(archive_data, revision):
actual_revision = archive.lookup_revision(revision)
assert actual_revision == archive_data.revision_get(revision)
@given(new_revision())
def test_lookup_revision_invalid_msg(archive_data, new_revision):
new_revision = new_revision.to_dict()
new_revision["message"] = b"elegant fix for bug \xff"
archive_data.revision_add([Revision.from_dict(new_revision)])
revision = archive.lookup_revision(hash_to_hex(new_revision["id"]))
assert revision["message"] == "elegant fix for bug \\xff"
assert revision["decoding_failures"] == ["message"]
@given(new_revision())
def test_lookup_revision_msg_ok(archive_data, new_revision):
archive_data.revision_add([new_revision])
revision_message = archive.lookup_revision_message(hash_to_hex(new_revision.id))
assert revision_message == {"message": new_revision.message}
def test_lookup_revision_msg_no_rev():
unknown_revision_ = random_sha1()
with pytest.raises(NotFoundExc) as e:
archive.lookup_revision_message(unknown_revision_)
assert e.match("Revision with sha1_git %s not found." % unknown_revision_)
@given(revisions())
def test_lookup_revision_multiple(archive_data, revisions):
actual_revisions = list(archive.lookup_revision_multiple(revisions))
expected_revisions = []
for rev in revisions:
expected_revisions.append(archive_data.revision_get(rev))
assert actual_revisions == expected_revisions
def test_lookup_revision_multiple_none_found():
unknown_revisions_ = [random_sha1(), random_sha1(), random_sha1()]
actual_revisions = list(archive.lookup_revision_multiple(unknown_revisions_))
assert actual_revisions == [None] * len(unknown_revisions_)
@given(revision())
def test_lookup_revision_log(archive_data, revision):
actual_revision_log = list(archive.lookup_revision_log(revision, limit=25))
expected_revision_log = archive_data.revision_log(revision, limit=25)
assert actual_revision_log == expected_revision_log
def _get_origin_branches(archive_data, origin):
origin_visit = archive_data.origin_visit_get(origin["url"])[-1]
snapshot = archive_data.snapshot_get(origin_visit["snapshot"])
branches = {
k: v
for (k, v) in snapshot["branches"].items()
if v["target_type"] == "revision"
}
return branches
@given(origin())
def test_lookup_revision_log_by(archive_data, origin):
branches = _get_origin_branches(archive_data, origin)
branch_name = random.choice(list(branches.keys()))
actual_log = list(
archive.lookup_revision_log_by(origin["url"], branch_name, None, limit=25)
)
expected_log = archive_data.revision_log(branches[branch_name]["target"], limit=25)
assert actual_log == expected_log
@given(origin())
def test_lookup_revision_log_by_notfound(origin):
with pytest.raises(NotFoundExc):
archive.lookup_revision_log_by(
origin["url"], "unknown_branch_name", None, limit=100
)
def test_lookup_content_raw_not_found():
unknown_content_ = random_content()
with pytest.raises(NotFoundExc) as e:
archive.lookup_content_raw("sha1:" + unknown_content_["sha1"])
assert e.match(
"Content with %s checksum equals to %s not found!"
% ("sha1", unknown_content_["sha1"])
)
@given(content())
def test_lookup_content_raw(archive_data, content):
actual_content = archive.lookup_content_raw("sha256:%s" % content["sha256"])
expected_content = archive_data.content_get_data(content["sha1"])
assert actual_content == expected_content
@given(empty_content())
def test_lookup_empty_content_raw(archive_data, empty_content):
content_raw = archive.lookup_content_raw(f"sha1_git:{empty_content['sha1_git']}")
assert content_raw["data"] == b""
def test_lookup_content_not_found():
unknown_content_ = random_content()
with pytest.raises(NotFoundExc) as e:
archive.lookup_content("sha1:%s" % unknown_content_["sha1"])
assert e.match(
"Content with %s checksum equals to %s not found!"
% ("sha1", unknown_content_["sha1"])
)
@given(content())
def test_lookup_content_with_sha1(archive_data, content):
actual_content = archive.lookup_content(f"sha1:{content['sha1']}")
expected_content = archive_data.content_get(content["sha1"])
assert actual_content == expected_content
@given(content())
def test_lookup_content_with_sha256(archive_data, content):
actual_content = archive.lookup_content(f"sha256:{content['sha256']}")
expected_content = archive_data.content_get(content["sha1"])
assert actual_content == expected_content
def test_lookup_directory_bad_checksum():
with pytest.raises(BadInputExc):
archive.lookup_directory("directory_id")
def test_lookup_directory_not_found():
unknown_directory_ = random_sha1()
with pytest.raises(NotFoundExc) as e:
archive.lookup_directory(unknown_directory_)
assert e.match("Directory with sha1_git %s not found" % unknown_directory_)
@given(directory())
def test_lookup_directory(archive_data, directory):
actual_directory_ls = list(archive.lookup_directory(directory))
expected_directory_ls = archive_data.directory_ls(directory)
assert actual_directory_ls == expected_directory_ls
@given(empty_directory())
def test_lookup_directory_empty(empty_directory):
actual_directory_ls = list(archive.lookup_directory(empty_directory))
assert actual_directory_ls == []
@given(origin())
def test_lookup_revision_by_nothing_found(origin):
with pytest.raises(NotFoundExc):
archive.lookup_revision_by(origin["url"], "invalid-branch-name")
@given(origin())
def test_lookup_revision_by(archive_data, origin):
branches = _get_origin_branches(archive_data, origin)
branch_name = random.choice(list(branches.keys()))
actual_revision = archive.lookup_revision_by(origin["url"], branch_name)
expected_revision = archive_data.revision_get(branches[branch_name]["target"])
assert actual_revision == expected_revision
@given(origin(), revision())
def test_lookup_revision_with_context_by_ko(origin, revision):
with pytest.raises(NotFoundExc):
archive.lookup_revision_with_context_by(
origin["url"], "invalid-branch-name", None, revision
)
@given(origin())
def test_lookup_revision_with_context_by(archive_data, origin):
branches = _get_origin_branches(archive_data, origin)
branch_name = random.choice(list(branches.keys()))
root_rev = branches[branch_name]["target"]
root_rev_log = archive_data.revision_log(root_rev)
children = defaultdict(list)
for rev in root_rev_log:
for rev_p in rev["parents"]:
children[rev_p].append(rev["id"])
rev = root_rev_log[-1]["id"]
actual_root_rev, actual_rev = archive.lookup_revision_with_context_by(
origin["url"], branch_name, None, rev
)
expected_root_rev = archive_data.revision_get(root_rev)
expected_rev = archive_data.revision_get(rev)
expected_rev["children"] = children[rev]
assert actual_root_rev == expected_root_rev
assert actual_rev == expected_rev
def test_lookup_revision_through_ko_not_implemented():
with pytest.raises(NotImplementedError):
archive.lookup_revision_through({"something-unknown": 10})
@given(origin())
def test_lookup_revision_through_with_context_by(archive_data, origin):
branches = _get_origin_branches(archive_data, origin)
branch_name = random.choice(list(branches.keys()))
root_rev = branches[branch_name]["target"]
root_rev_log = archive_data.revision_log(root_rev)
rev = root_rev_log[-1]["id"]
assert archive.lookup_revision_through(
{
"origin_url": origin["url"],
"branch_name": branch_name,
"ts": None,
"sha1_git": rev,
}
) == archive.lookup_revision_with_context_by(origin["url"], branch_name, None, rev)
@given(origin())
def test_lookup_revision_through_with_revision_by(archive_data, origin):
branches = _get_origin_branches(archive_data, origin)
branch_name = random.choice(list(branches.keys()))
assert archive.lookup_revision_through(
{"origin_url": origin["url"], "branch_name": branch_name, "ts": None,}
) == archive.lookup_revision_by(origin["url"], branch_name, None)
@given(ancestor_revisions())
def test_lookup_revision_through_with_context(ancestor_revisions):
sha1_git = ancestor_revisions["sha1_git"]
sha1_git_root = ancestor_revisions["sha1_git_root"]
assert archive.lookup_revision_through(
{"sha1_git_root": sha1_git_root, "sha1_git": sha1_git,}
) == archive.lookup_revision_with_context(sha1_git_root, sha1_git)
@given(revision())
def test_lookup_revision_through_with_revision(revision):
assert archive.lookup_revision_through(
{"sha1_git": revision}
) == archive.lookup_revision(revision)
@given(revision())
def test_lookup_directory_through_revision_ko_not_found(revision):
with pytest.raises(NotFoundExc):
archive.lookup_directory_through_revision(
{"sha1_git": revision}, "some/invalid/path"
)
@given(revision())
def test_lookup_directory_through_revision_ok(archive_data, revision):
rev_data = archive_data.revision_get(revision)
dir_entries = [
e
for e in archive_data.directory_ls(rev_data["directory"])
if e["type"] == "file"
]
dir_entry = random.choice(dir_entries)
assert archive.lookup_directory_through_revision(
{"sha1_git": revision}, dir_entry["name"]
) == (revision, archive.lookup_directory_with_revision(revision, dir_entry["name"]))
@given(revision())
def test_lookup_directory_through_revision_ok_with_data(archive_data, revision):
rev_data = archive_data.revision_get(revision)
dir_entries = [
e
for e in archive_data.directory_ls(rev_data["directory"])
if e["type"] == "file"
]
dir_entry = random.choice(dir_entries)
assert archive.lookup_directory_through_revision(
{"sha1_git": revision}, dir_entry["name"], with_data=True
) == (
revision,
archive.lookup_directory_with_revision(
revision, dir_entry["name"], with_data=True
),
)
@given(content(), directory(), release(), revision(), snapshot())
def test_lookup_known_objects(
archive_data, content, directory, release, revision, snapshot
):
expected = archive_data.content_find(content)
assert archive.lookup_object(CONTENT, content["sha1_git"]) == expected
expected = archive_data.directory_get(directory)
assert archive.lookup_object(DIRECTORY, directory) == expected
expected = archive_data.release_get(release)
assert archive.lookup_object(RELEASE, release) == expected
expected = archive_data.revision_get(revision)
assert archive.lookup_object(REVISION, revision) == expected
expected = {**archive_data.snapshot_get(snapshot), "next_branch": None}
assert archive.lookup_object(SNAPSHOT, snapshot) == expected
@given(
unknown_content(),
unknown_directory(),
unknown_release(),
unknown_revision(),
unknown_snapshot(),
)
def test_lookup_unknown_objects(
unknown_content,
unknown_directory,
unknown_release,
unknown_revision,
unknown_snapshot,
):
with pytest.raises(NotFoundExc) as e:
archive.lookup_object(CONTENT, unknown_content["sha1_git"])
assert e.match(r"Content.*not found")
with pytest.raises(NotFoundExc) as e:
archive.lookup_object(DIRECTORY, unknown_directory)
assert e.match(r"Directory.*not found")
with pytest.raises(NotFoundExc) as e:
archive.lookup_object(RELEASE, unknown_release)
assert e.match(r"Release.*not found")
with pytest.raises(NotFoundExc) as e:
archive.lookup_object(REVISION, unknown_revision)
assert e.match(r"Revision.*not found")
with pytest.raises(NotFoundExc) as e:
archive.lookup_object(SNAPSHOT, unknown_snapshot)
assert e.match(r"Snapshot.*not found")
@given(invalid_sha1())
def test_lookup_invalid_objects(invalid_sha1):
with pytest.raises(BadInputExc) as e:
archive.lookup_object("foo", invalid_sha1)
assert e.match("Invalid swh object type")
with pytest.raises(BadInputExc) as e:
archive.lookup_object(CONTENT, invalid_sha1)
assert e.match("Invalid hash")
with pytest.raises(BadInputExc) as e:
archive.lookup_object(DIRECTORY, invalid_sha1)
assert e.match("Invalid checksum")
with pytest.raises(BadInputExc) as e:
archive.lookup_object(RELEASE, invalid_sha1)
assert e.match("Invalid checksum")
with pytest.raises(BadInputExc) as e:
archive.lookup_object(REVISION, invalid_sha1)
assert e.match("Invalid checksum")
with pytest.raises(BadInputExc) as e:
archive.lookup_object(SNAPSHOT, invalid_sha1)
assert e.match("Invalid checksum")
def test_lookup_missing_hashes_non_present():
missing_cnt = random_sha1()
missing_dir = random_sha1()
missing_rev = random_sha1()
missing_rel = random_sha1()
missing_snp = random_sha1()
grouped_swhids = {
CONTENT: [hash_to_bytes(missing_cnt)],
DIRECTORY: [hash_to_bytes(missing_dir)],
REVISION: [hash_to_bytes(missing_rev)],
RELEASE: [hash_to_bytes(missing_rel)],
SNAPSHOT: [hash_to_bytes(missing_snp)],
}
actual_result = archive.lookup_missing_hashes(grouped_swhids)
assert actual_result == {
missing_cnt,
missing_dir,
missing_rev,
missing_rel,
missing_snp,
}
@given(content(), directory())
def test_lookup_missing_hashes_some_present(archive_data, content, directory):
missing_rev = random_sha1()
missing_rel = random_sha1()
missing_snp = random_sha1()
grouped_swhids = {
CONTENT: [hash_to_bytes(content["sha1_git"])],
DIRECTORY: [hash_to_bytes(directory)],
REVISION: [hash_to_bytes(missing_rev)],
RELEASE: [hash_to_bytes(missing_rel)],
SNAPSHOT: [hash_to_bytes(missing_snp)],
}
actual_result = archive.lookup_missing_hashes(grouped_swhids)
assert actual_result == {missing_rev, missing_rel, missing_snp}
@given(origin())
def test_lookup_origin_extra_trailing_slash(origin):
origin_info = archive.lookup_origin({"url": f"{origin['url']}/"})
assert origin_info["url"] == origin["url"]
def test_lookup_origin_missing_trailing_slash(archive_data):
deb_origin = Origin(url="http://snapshot.debian.org/package/r-base/")
archive_data.origin_add([deb_origin])
origin_info = archive.lookup_origin({"url": deb_origin.url[:-1]})
assert origin_info["url"] == deb_origin.url
@given(snapshot())
def test_lookup_snapshot_branch_name_from_tip_revision(archive_data, snapshot_id):
snapshot = archive_data.snapshot_get(snapshot_id)
branches = [
{"name": k, "revision": v["target"]}
for k, v in snapshot["branches"].items()
if v["target_type"] == "revision"
]
branch_info = random.choice(branches)
possible_results = [
b["name"] for b in branches if b["revision"] == branch_info["revision"]
]
assert (
archive.lookup_snapshot_branch_name_from_tip_revision(
snapshot_id, branch_info["revision"]
)
in possible_results
)
@given(origin(), new_origin())
def test_lookup_origins_get_by_sha1s(origin, unknown_origin):
hasher = hashlib.sha1()
hasher.update(origin["url"].encode("ascii"))
origin_info = OriginInfo(url=origin["url"])
origin_sha1 = hasher.hexdigest()
hasher = hashlib.sha1()
hasher.update(unknown_origin.url.encode("ascii"))
unknown_origin_sha1 = hasher.hexdigest()
origins = list(archive.lookup_origins_by_sha1s([origin_sha1]))
assert origins == [origin_info]
origins = list(archive.lookup_origins_by_sha1s([origin_sha1, origin_sha1]))
assert origins == [origin_info, origin_info]
origins = list(archive.lookup_origins_by_sha1s([origin_sha1, unknown_origin_sha1]))
assert origins == [origin_info, None]
+
+
+@given(snapshot())
+def test_lookup_snapshot_sizes(archive_data, snapshot):
+ branches = archive_data.snapshot_get(snapshot)["branches"]
+
+ expected_sizes = {
+ "alias": 0,
+ "release": 0,
+ "revision": 0,
+ }
+
+ for branch_name, branch_info in branches.items():
+ if branch_info is not None:
+ expected_sizes[branch_info["target_type"]] += 1
+
+ assert archive.lookup_snapshot_sizes(snapshot) == expected_sizes
+
+
+@given(snapshot())
+def test_lookup_snapshot_alias(snapshot):
+ resolved_alias = archive.lookup_snapshot_alias(snapshot, "HEAD")
+ assert resolved_alias is not None
+ assert resolved_alias["target_type"] == "revision"
+ assert resolved_alias["target"] is not None
diff --git a/swh/web/tests/conftest.py b/swh/web/tests/conftest.py
index 964d94dc..ba905557 100644
--- a/swh/web/tests/conftest.py
+++ b/swh/web/tests/conftest.py
@@ -1,354 +1,360 @@
# Copyright (C) 2018-2020 The Software Heritage developers
# See the AUTHORS file at the top-level directory of this distribution
# License: GNU Affero General Public License version 3, or any later version
# See top-level LICENSE file for more information
import json
import os
import shutil
from subprocess import PIPE, run
import sys
from typing import Any, Dict, List, Optional
from hypothesis import HealthCheck, settings
import pytest
from django.core.cache import cache
from rest_framework.test import APIClient, APIRequestFactory
from swh.model.hashutil import ALGORITHMS, hash_to_bytes
from swh.storage.algos.origin import origin_get_latest_visit_status
from swh.storage.algos.snapshot import snapshot_get_all_branches, snapshot_get_latest
from swh.web.common import converters
from swh.web.common.typing import OriginVisitInfo
from swh.web.tests.data import get_tests_data, override_storages
# Used to skip some tests
ctags_json_missing = (
shutil.which("ctags") is None
or b"+json" not in run(["ctags", "--version"], stdout=PIPE).stdout
)
fossology_missing = shutil.which("nomossa") is None
# Register some hypothesis profiles
settings.register_profile("default", settings())
settings.register_profile(
"swh-web",
settings(
deadline=None,
suppress_health_check=[HealthCheck.too_slow, HealthCheck.filter_too_much],
),
)
settings.register_profile(
"swh-web-fast",
settings(
deadline=None,
max_examples=1,
suppress_health_check=[HealthCheck.too_slow, HealthCheck.filter_too_much],
),
)
def pytest_configure(config):
# Use fast hypothesis profile by default if none has been
# explicitly specified in pytest option
if config.getoption("--hypothesis-profile") is None:
settings.load_profile("swh-web-fast")
# Small hack in order to be able to run the unit tests
# without static assets generated by webpack.
# Those assets are not really needed for the Python tests
# but the django templates will fail to load due to missing
# generated file webpack-stats.json describing the js and css
# files to include.
# So generate a dummy webpack-stats.json file to overcome
# that issue.
test_dir = os.path.dirname(__file__)
# location of the static folder when running tests through tox
static_dir = os.path.join(sys.prefix, "share/swh/web/static")
if not os.path.exists(static_dir):
# location of the static folder when running tests locally with pytest
static_dir = os.path.join(test_dir, "../../../static")
webpack_stats = os.path.join(static_dir, "webpack-stats.json")
if os.path.exists(webpack_stats):
return
bundles_dir = os.path.join(test_dir, "../assets/src/bundles")
_, dirs, _ = next(os.walk(bundles_dir))
mock_webpack_stats = {"status": "done", "publicPath": "/static", "chunks": {}}
for bundle in dirs:
asset = "js/%s.js" % bundle
mock_webpack_stats["chunks"][bundle] = [
{
"name": asset,
"publicPath": "/static/%s" % asset,
"path": os.path.join(static_dir, asset),
}
]
with open(webpack_stats, "w") as outfile:
json.dump(mock_webpack_stats, outfile)
# Clear Django cache before each test
@pytest.fixture(autouse=True)
def django_cache_cleared():
cache.clear()
# Alias rf fixture from pytest-django
@pytest.fixture
def request_factory(rf):
return rf
# Fixture to get test client from Django REST Framework
@pytest.fixture(scope="module")
def api_client():
return APIClient()
# Fixture to get API request factory from Django REST Framework
@pytest.fixture(scope="module")
def api_request_factory():
return APIRequestFactory()
# Initialize tests data
@pytest.fixture(scope="session", autouse=True)
def tests_data():
data = get_tests_data(reset=True)
# Update swh-web configuration to use the in-memory storages
# instantiated in the tests.data module
override_storages(data["storage"], data["idx_storage"], data["search"])
return data
# Fixture to manipulate data from a sample archive used in the tests
@pytest.fixture(scope="session")
def archive_data(tests_data):
return _ArchiveData(tests_data)
# Fixture to manipulate indexer data from a sample archive used in the tests
@pytest.fixture(scope="session")
def indexer_data(tests_data):
return _IndexerData(tests_data)
# Custom data directory for requests_mock
@pytest.fixture
def datadir():
return os.path.join(os.path.abspath(os.path.dirname(__file__)), "resources")
class _ArchiveData:
"""
Helper class to manage data from a sample test archive.
It is initialized with a reference to an in-memory storage
containing raw tests data.
It is basically a proxy to Storage interface but it overrides some methods
to retrieve those tests data in a json serializable format in order to ease
tests implementation.
"""
def __init__(self, tests_data):
self.storage = tests_data["storage"]
def __getattr__(self, key):
if key == "storage":
raise AttributeError(key)
# Forward calls to non overridden Storage methods to wrapped
# storage instance
return getattr(self.storage, key)
def content_find(self, content: Dict[str, Any]) -> Dict[str, Any]:
cnt_ids_bytes = {
algo_hash: hash_to_bytes(content[algo_hash])
for algo_hash in ALGORITHMS
if content.get(algo_hash)
}
cnt = self.storage.content_find(cnt_ids_bytes)
return converters.from_content(cnt[0].to_dict()) if cnt else cnt
def content_get(self, cnt_id: str) -> Dict[str, Any]:
cnt_id_bytes = hash_to_bytes(cnt_id)
content = self.storage.content_get([cnt_id_bytes])[0]
if content:
content_d = content.to_dict()
content_d.pop("ctime", None)
else:
content_d = None
return converters.from_swh(
content_d, hashess={"sha1", "sha1_git", "sha256", "blake2s256"}
)
def content_get_data(self, cnt_id: str) -> Optional[Dict[str, Any]]:
cnt_id_bytes = hash_to_bytes(cnt_id)
cnt_data = self.storage.content_get_data(cnt_id_bytes)
if cnt_data is None:
return None
return converters.from_content({"data": cnt_data, "sha1": cnt_id_bytes})
def directory_get(self, dir_id):
return {"id": dir_id, "content": self.directory_ls(dir_id)}
def directory_ls(self, dir_id):
cnt_id_bytes = hash_to_bytes(dir_id)
dir_content = map(
converters.from_directory_entry, self.storage.directory_ls(cnt_id_bytes)
)
return list(dir_content)
def release_get(self, rel_id: str) -> Optional[Dict[str, Any]]:
rel_id_bytes = hash_to_bytes(rel_id)
rel_data = self.storage.release_get([rel_id_bytes])[0]
return converters.from_release(rel_data) if rel_data else None
def revision_get(self, rev_id: str) -> Optional[Dict[str, Any]]:
rev_id_bytes = hash_to_bytes(rev_id)
rev_data = self.storage.revision_get([rev_id_bytes])[0]
return converters.from_revision(rev_data) if rev_data else None
def revision_log(self, rev_id, limit=None):
rev_id_bytes = hash_to_bytes(rev_id)
return list(
map(
converters.from_revision,
self.storage.revision_log([rev_id_bytes], limit=limit),
)
)
def snapshot_get_latest(self, origin_url):
snp = snapshot_get_latest(self.storage, origin_url)
return converters.from_snapshot(snp.to_dict())
def origin_get(self, origin_urls):
origins = self.storage.origin_get(origin_urls)
return [converters.from_origin(o.to_dict()) for o in origins]
def origin_visit_get(self, origin_url):
next_page_token = None
visits = []
while True:
visit_page = self.storage.origin_visit_get(
origin_url, page_token=next_page_token
)
next_page_token = visit_page.next_page_token
for visit in visit_page.results:
visit_status = self.storage.origin_visit_status_get_latest(
origin_url, visit.visit
)
visits.append(
converters.from_origin_visit(
{**visit_status.to_dict(), "type": visit.type}
)
)
if not next_page_token:
break
return visits
def origin_visit_get_by(self, origin_url: str, visit_id: int) -> OriginVisitInfo:
visit = self.storage.origin_visit_get_by(origin_url, visit_id)
assert visit is not None
visit_status = self.storage.origin_visit_status_get_latest(origin_url, visit_id)
assert visit_status is not None
return converters.from_origin_visit(
{**visit_status.to_dict(), "type": visit.type}
)
def origin_visit_status_get_latest(
self,
origin_url,
type: Optional[str] = None,
allowed_statuses: Optional[List[str]] = None,
require_snapshot: bool = False,
):
visit_and_status = origin_get_latest_visit_status(
self.storage,
origin_url,
type=type,
allowed_statuses=allowed_statuses,
require_snapshot=require_snapshot,
)
return (
converters.from_origin_visit(
{**visit_and_status[0].to_dict(), **visit_and_status[1].to_dict()}
)
if visit_and_status
else None
)
def snapshot_get(self, snapshot_id):
snp = snapshot_get_all_branches(self.storage, hash_to_bytes(snapshot_id))
return converters.from_snapshot(snp.to_dict())
def snapshot_get_branches(
self, snapshot_id, branches_from="", branches_count=1000, target_types=None
):
partial_branches = self.storage.snapshot_get_branches(
hash_to_bytes(snapshot_id),
branches_from.encode(),
branches_count,
target_types,
)
return converters.from_partial_branches(partial_branches)
def snapshot_get_head(self, snapshot):
if snapshot["branches"]["HEAD"]["target_type"] == "alias":
target = snapshot["branches"]["HEAD"]["target"]
head = snapshot["branches"][target]["target"]
else:
head = snapshot["branches"]["HEAD"]["target"]
return head
+ def snapshot_count_branches(self, snapshot_id):
+ counts = dict.fromkeys(("alias", "release", "revision"), 0)
+ counts.update(self.storage.snapshot_count_branches(hash_to_bytes(snapshot_id)))
+ counts.pop(None, None)
+ return counts
+
class _IndexerData:
"""
Helper class to manage indexer tests data
It is initialized with a reference to an in-memory indexer storage
containing raw tests data.
It also defines class methods to retrieve those tests data in
a json serializable format in order to ease tests implementation.
"""
def __init__(self, tests_data):
self.idx_storage = tests_data["idx_storage"]
self.mimetype_indexer = tests_data["mimetype_indexer"]
self.license_indexer = tests_data["license_indexer"]
self.ctags_indexer = tests_data["ctags_indexer"]
def content_add_mimetype(self, cnt_id):
self.mimetype_indexer.run([hash_to_bytes(cnt_id)], "update-dups")
def content_get_mimetype(self, cnt_id):
mimetype = self.idx_storage.content_mimetype_get([hash_to_bytes(cnt_id)])[
0
].to_dict()
return converters.from_filetype(mimetype)
def content_add_license(self, cnt_id):
self.license_indexer.run([hash_to_bytes(cnt_id)], "update-dups")
def content_get_license(self, cnt_id):
cnt_id_bytes = hash_to_bytes(cnt_id)
licenses = self.idx_storage.content_fossology_license_get([cnt_id_bytes])
for license in licenses:
yield converters.from_swh(license.to_dict(), hashess={"id"})
def content_add_ctags(self, cnt_id):
self.ctags_indexer.run([hash_to_bytes(cnt_id)], "update-dups")
def content_get_ctags(self, cnt_id):
cnt_id_bytes = hash_to_bytes(cnt_id)
ctags = self.idx_storage.content_ctags_get([cnt_id_bytes])
for ctag in ctags:
yield converters.from_swh(ctag, hashess={"id"})