def _get_visit_info_for_save_request(save_request):
visit_date = None
visit_status = None
time_now = datetime.now(tz=timezone.utc)
time_delta = time_now - save_request.request_date
# stop trying to find a visit date one month after save request submission
# as those requests to storage are expensive and associated loading task
# surely ended up with errors
if time_delta.days <= 30:
try:
origin = {"url": save_request.origin_url}
origin_info = archive.lookup_origin(origin)
origin_visits = get_origin_visits(origin_info)
visit_dates = [
parse_iso8601_date_to_utc(v["date"]) for v in origin_visits
]
i = bisect_right(visit_dates, save_request.request_date)
if i != len(visit_dates):
visit_date = visit_dates[i]
visit_status = origin_visits[i]["status"]
if origin_visits[i]["status"] not in ("full", "partial",
"not_found"):
visit_date = None
except Exception as exc:
sentry_sdk.capture_exception(exc)
return visit_date, visit_status
def _origin_visits_browse(request, origin_url):
if origin_url is None:
raise BadInputExc("An origin URL must be provided as query parameter.")
origin_info = archive.lookup_origin({"url": origin_url})
origin_visits = get_origin_visits(origin_info)
snapshot_context = get_snapshot_context(origin_url=origin_url)
for i, visit in enumerate(origin_visits):
url_date = format_utc_iso_date(visit["date"], "%Y-%m-%dT%H:%M:%SZ")
visit["formatted_date"] = format_utc_iso_date(visit["date"])
query_params = {"origin_url": origin_url, "timestamp": url_date}
if i < len(origin_visits) - 1:
if visit["date"] == origin_visits[i + 1]["date"]:
query_params = {"visit_id": visit["visit"]}
if i > 0:
if visit["date"] == origin_visits[i - 1]["date"]:
query_params = {"visit_id": visit["visit"]}
snapshot = visit["snapshot"] if visit["snapshot"] else ""
visit["url"] = reverse(
"browse-origin-directory",
query_params=query_params,
)
if not snapshot:
visit["snapshot"] = ""
visit["date"] = parse_iso8601_date_to_utc(visit["date"]).timestamp()
heading = "Origin visits - %s" % origin_url
return render(
request,
"browse/origin-visits.html",
{
"heading": heading,
"swh_object_name": "Visits",
"swh_object_metadata": origin_info,
"origin_visits": origin_visits,
"origin_info": origin_info,
"snapshot_context": snapshot_context,
"vault_cooking": None,
"show_actions": False,
},
)
def _stat_counters(request):
stat_counters = archive.stat_counters()
url = get_config()["history_counters_url"]
stat_counters_history = {}
try:
response = requests.get(url, timeout=5)
stat_counters_history = json.loads(response.text)
for d, object_counts in _stat_counters_backfill.items():
# convert date to javascript timestamp (in ms)
timestamp = int(parse_iso8601_date_to_utc(d).timestamp()) * 1000
for object_type, object_count in object_counts.items():
stat_counters_history[object_type].append(
[timestamp, object_count])
except Exception as exc:
sentry_sdk.capture_exception(exc)
counters = {
"stat_counters": stat_counters,
"stat_counters_history": stat_counters_history,
}
return JsonResponse(counters)
def get_origin_visit(
origin_info: OriginInfo,
visit_ts: Optional[str] = None,
visit_id: Optional[int] = None,
snapshot_id: Optional[str] = None,
) -> OriginVisitInfo:
"""Function that returns information about a visit for a given origin.
If a timestamp is provided, the closest visit from that
timestamp is returned.
If a snapshot identifier is provided, the first visit with that snapshot
is returned.
If no search hints are provided, return the most recent full visit with
a valid snapshot or the most recent partial visit with a valid snapshot
otherwise.
Args:
origin_info: a dict filled with origin information
visit_ts: an ISO 8601 datetime string to parse
snapshot_id: a snapshot identifier
Returns:
A dict containing the visit info.
Raises:
swh.web.common.exc.NotFoundExc: if no visit can be found
"""
# returns the latest full visit with a valid snapshot
visit = archive.lookup_origin_visit_latest(origin_info["url"],
allowed_statuses=["full"],
require_snapshot=True)
if not visit:
# or the latest partial visit with a valid snapshot otherwise
visit = archive.lookup_origin_visit_latest(
origin_info["url"],
allowed_statuses=["partial"],
require_snapshot=True)
if not visit_ts and not visit_id and not snapshot_id:
if visit:
return visit
else:
raise NotFoundExc(
f"No valid visit for origin with url {origin_info['url']} found!"
)
# no need to fetch all visits list and search in it if the latest
# visit matches some criteria
if visit and (visit["snapshot"] == snapshot_id
or visit["visit"] == visit_id):
return visit
visits = get_origin_visits(origin_info)
if not visits:
raise NotFoundExc(
f"No visits associated to origin with url {origin_info['url']}!")
if snapshot_id:
visits = [v for v in visits if v["snapshot"] == snapshot_id]
if len(visits) == 0:
raise NotFoundExc(
("Visit for snapshot with id %s for origin with"
" url %s not found!" % (snapshot_id, origin_info["url"])))
return visits[0]
if visit_id:
visits = [v for v in visits if v["visit"] == int(visit_id)]
if len(visits) == 0:
raise NotFoundExc(
("Visit with id %s for origin with"
" url %s not found!" % (visit_id, origin_info["url"])))
return visits[0]
if visit_ts:
target_visit_ts = math.floor(
parse_iso8601_date_to_utc(visit_ts).timestamp())
# Find the visit with date closest to the target (in absolute value)
(abs_time_delta, visit_idx) = min(
((math.floor(parse_iso8601_date_to_utc(
visit["date"]).timestamp()), i)
for (i, visit) in enumerate(visits)),
key=lambda ts_and_i: abs(ts_and_i[0] - target_visit_ts),
)
if visit_idx is not None:
visit = visits[visit_idx]
# If multiple visits have the same date, select the one with
# the largest id.
while (visit_idx < len(visits) - 1
and visit["date"] == visits[visit_idx + 1]["date"]):
visit_idx = visit_idx + 1
visit = visits[visit_idx]
return visit
else:
raise NotFoundExc(
("Visit with timestamp %s for origin with "
"url %s not found!" % (visit_ts, origin_info["url"])))
return visits[-1]
def _visit_sort_key(visit):
ts = parse_iso8601_date_to_utc(visit["date"]).timestamp()
return ts + (float(visit["visit"]) / 10e3)
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 ({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 ({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"])
assert_not_contains(resp, "swh-metadata-popover")
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 ({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 ({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"])
assert_not_contains(resp, "swh-metadata-popover")
def test_revision_log_browse(client, archive_data, revision):
per_page = 10
revision_log = archive_data.revision_log(revision)
revision_log_sorted = sorted(
revision_log,
key=lambda rev: -parse_iso8601_date_to_utc(rev["committer_date"]).
timestamp(),
)
url = reverse(
"browse-revision-log",
url_args={"sha1_git": revision},
query_params={"per_page": per_page},
)
next_page_url = reverse(
"browse-revision-log",
url_args={"sha1_git": revision},
query_params={
"offset": per_page,
"per_page": per_page,
},
)
nb_log_entries = per_page
if len(revision_log_sorted) < per_page:
nb_log_entries = len(revision_log_sorted)
resp = check_html_get_response(client,
url,
status_code=200,
template_used="browse/revision-log.html")
assert_contains(resp,
'<tr class="swh-revision-log-entry',
count=nb_log_entries)
assert_contains(resp, '<a class="page-link">Newer</a>')
if len(revision_log_sorted) > per_page:
assert_contains(
resp,
'<a class="page-link" href="%s">Older</a>' % escape(next_page_url),
)
for log in revision_log_sorted[:per_page]:
revision_url = reverse("browse-revision",
url_args={"sha1_git": log["id"]})
assert_contains(resp, log["id"][:7])
assert_contains(resp, log["author"]["name"])
assert_contains(resp, format_utc_iso_date(log["date"]))
assert_contains(resp, escape(log["message"]))
assert_contains(resp, format_utc_iso_date(log["committer_date"]))
assert_contains(resp, revision_url)
if len(revision_log_sorted) <= per_page:
return
resp = check_html_get_response(client,
next_page_url,
status_code=200,
template_used="browse/revision-log.html")
prev_page_url = reverse(
"browse-revision-log",
url_args={"sha1_git": revision},
query_params={
"offset": 0,
"per_page": per_page
},
)
next_page_url = reverse(
"browse-revision-log",
url_args={"sha1_git": revision},
query_params={
"offset": 2 * per_page,
"per_page": per_page
},
)
nb_log_entries = len(revision_log_sorted) - per_page
if nb_log_entries > per_page:
nb_log_entries = per_page
assert_contains(resp,
'<tr class="swh-revision-log-entry',
count=nb_log_entries)
assert_contains(
resp,
'<a class="page-link" href="%s">Newer</a>' % escape(prev_page_url))
if len(revision_log_sorted) > 2 * per_page:
assert_contains(
resp,
'<a class="page-link" href="%s">Older</a>' % escape(next_page_url),
)
if len(revision_log_sorted) <= 2 * per_page:
return
resp = check_html_get_response(client,
next_page_url,
status_code=200,
template_used="browse/revision-log.html")
prev_page_url = reverse(
"browse-revision-log",
url_args={"sha1_git": revision},
query_params={
"offset": per_page,
"per_page": per_page
},
)
next_page_url = reverse(
"browse-revision-log",
url_args={"sha1_git": revision},
query_params={
"offset": 3 * per_page,
"per_page": per_page
},
)
nb_log_entries = len(revision_log_sorted) - 2 * per_page
if nb_log_entries > per_page:
nb_log_entries = per_page
assert_contains(resp,
'<tr class="swh-revision-log-entry',
count=nb_log_entries)
assert_contains(
resp,
'<a class="page-link" href="%s">Newer</a>' % escape(prev_page_url))
if len(revision_log_sorted) > 3 * per_page:
assert_contains(
resp,
'<a class="page-link" href="%s">Older</a>' % escape(next_page_url),
)
def test_parse_iso8601_date_to_utc_ko(invalid_iso8601_timestamp):
with pytest.raises(BadInputExc):
utils.parse_iso8601_date_to_utc(invalid_iso8601_timestamp)
def test_parse_iso8601_date_to_utc_ok(input_timestamp, output_date):
assert utils.parse_iso8601_date_to_utc(input_timestamp) == output_date