diff --git a/swh/web/browse/utils.py b/swh/web/browse/utils.py
index 0962f604..dcbe2d1b 100644
--- a/swh/web/browse/utils.py
+++ b/swh/web/browse/utils.py
@@ -1,1111 +1,1139 @@
# Copyright (C) 2017-2019 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 base64
import magic
import pypandoc
import stat
import textwrap
from collections import defaultdict
from threading import Lock
from django.core.cache import cache
from django.utils.safestring import mark_safe
from django.utils.html import escape
from swh.model.identifiers import persistent_identifier
from swh.web.common import highlightjs, service
from swh.web.common.exc import NotFoundExc, http_status_code_message
from swh.web.common.origin_visits import get_origin_visit
from swh.web.common.utils import (
reverse, format_utc_iso_date, get_swh_persistent_id,
swh_object_icons
)
from swh.web.config import get_config
def get_directory_entries(sha1_git):
"""Function that retrieves the content of a directory
from the archive.
The directories entries are first sorted in lexicographical order.
Sub-directories and regular files are then extracted.
Args:
sha1_git: sha1_git identifier of the directory
Returns:
A tuple whose first member corresponds to the sub-directories list
and second member the regular files list
Raises:
NotFoundExc if the directory is not found
"""
cache_entry_id = 'directory_entries_%s' % sha1_git
cache_entry = cache.get(cache_entry_id)
if cache_entry:
return cache_entry
entries = list(service.lookup_directory(sha1_git))
for e in entries:
e['perms'] = stat.filemode(e['perms'])
if e['type'] == 'rev':
# modify dir entry name to explicitly show it points
# to a revision
e['name'] = '%s @ %s' % (e['name'], e['target'][:7])
dirs = [e for e in entries if e['type'] in ('dir', 'rev')]
files = [e for e in entries if e['type'] == 'file']
dirs = sorted(dirs, key=lambda d: d['name'])
files = sorted(files, key=lambda f: f['name'])
cache.set(cache_entry_id, (dirs, files))
return dirs, files
_lock = Lock()
def get_mimetype_and_encoding_for_content(content):
"""Function that returns the mime type and the encoding associated to
a content buffer using the magic module under the hood.
Args:
content (bytes): a content buffer
Returns:
A tuple (mimetype, encoding), for instance ('text/plain', 'us-ascii'),
associated to the provided content.
"""
# https://pypi.org/project/python-magic/
# packaged as python3-magic in debian buster
if hasattr(magic, 'from_buffer'):
m = magic.Magic(mime=True, mime_encoding=True)
mime_encoding = m.from_buffer(content)
mime_type, encoding = mime_encoding.split(';')
encoding = encoding.replace(' charset=', '')
# https://pypi.org/project/file-magic/
# packaged as python3-magic in debian stretch
else:
# TODO: Remove that code when production environment is upgraded
# to debian buster
# calls to the file-magic API are not thread-safe so they must
# be protected with a Lock to guarantee they will succeed
_lock.acquire()
magic_result = magic.detect_from_content(content)
_lock.release()
mime_type = magic_result.mime_type
encoding = magic_result.encoding
return mime_type, encoding
# maximum authorized content size in bytes for HTML display
# with code highlighting
content_display_max_size = get_config()['content_display_max_size']
snapshot_content_max_size = get_config()['snapshot_content_max_size']
def _re_encode_content(mimetype, encoding, content_data):
# encode textual content to utf-8 if needed
if mimetype.startswith('text/'):
# probably a malformed UTF-8 content, re-encode it
# by replacing invalid chars with a substitution one
if encoding == 'unknown-8bit':
content_data = content_data.decode('utf-8', 'replace')\
.encode('utf-8')
elif encoding not in ['utf-8', 'binary']:
content_data = content_data.decode(encoding, 'replace')\
.encode('utf-8')
elif mimetype.startswith('application/octet-stream'):
# file may detect a text content as binary
# so try to decode it for display
encodings = ['us-ascii']
encodings += ['iso-8859-%s' % i for i in range(1, 17)]
for encoding in encodings:
try:
content_data = content_data.decode(encoding)\
.encode('utf-8')
except Exception:
pass
else:
# ensure display in content view
mimetype = 'text/plain'
break
return mimetype, content_data
def request_content(query_string, max_size=content_display_max_size,
raise_if_unavailable=True, re_encode=True):
"""Function that retrieves a content from the archive.
Raw bytes content is first retrieved, then the content mime type.
If the mime type is not stored in the archive, it will be computed
using Python magic module.
Args:
query_string: a string of the form "[ALGO_HASH:]HASH" where
optional ALGO_HASH can be either ``sha1``, ``sha1_git``,
``sha256``, or ``blake2s256`` (default to ``sha1``) and HASH
the hexadecimal representation of the hash value
max_size: the maximum size for a content to retrieve (default to 1MB,
no size limit if None)
Returns:
A tuple whose first member corresponds to the content raw bytes
and second member the content mime type
Raises:
NotFoundExc if the content is not found
"""
content_data = service.lookup_content(query_string)
filetype = None
language = None
license = None
# requests to the indexer db may fail so properly handle
# those cases in order to avoid content display errors
try:
filetype = service.lookup_content_filetype(query_string)
language = service.lookup_content_language(query_string)
license = service.lookup_content_license(query_string)
except Exception:
pass
mimetype = 'unknown'
encoding = 'unknown'
if filetype:
mimetype = filetype['mimetype']
encoding = filetype['encoding']
# workaround when encountering corrupted data due to implicit
# conversion from bytea to text in the indexer db (see T818)
# TODO: Remove that code when all data have been correctly converted
if mimetype.startswith('\\'):
filetype = None
content_data['error_code'] = 200
content_data['error_message'] = ''
content_data['error_description'] = ''
if not max_size or content_data['length'] < max_size:
try:
content_raw = service.lookup_content_raw(query_string)
except Exception as e:
if raise_if_unavailable:
raise e
else:
content_data['raw_data'] = None
content_data['error_code'] = 404
content_data['error_description'] = \
'The bytes of the content are currently not available in the archive.' # noqa
content_data['error_message'] = \
http_status_code_message[content_data['error_code']]
else:
content_data['raw_data'] = content_raw['data']
if not filetype:
mimetype, encoding = \
get_mimetype_and_encoding_for_content(content_data['raw_data']) # noqa
if re_encode:
mimetype, raw_data = _re_encode_content(
mimetype, encoding, content_data['raw_data'])
content_data['raw_data'] = raw_data
else:
content_data['raw_data'] = None
content_data['mimetype'] = mimetype
content_data['encoding'] = encoding
if language:
content_data['language'] = language['lang']
else:
content_data['language'] = 'not detected'
if license:
content_data['licenses'] = ', '.join(license['facts'][0]['licenses'])
else:
content_data['licenses'] = 'not detected'
return content_data
_browsers_supported_image_mimes = set(['image/gif', 'image/png',
'image/jpeg', 'image/bmp',
'image/webp', 'image/svg',
'image/svg+xml'])
def prepare_content_for_display(content_data, mime_type, path):
"""Function that prepares a content for HTML display.
The function tries to associate a programming language to a
content in order to perform syntax highlighting client-side
using highlightjs. The language is determined using either
the content filename or its mime type.
If the mime type corresponds to an image format supported
by web browsers, the content will be encoded in base64
for displaying the image.
Args:
content_data (bytes): raw bytes of the content
mime_type (string): mime type of the content
path (string): path of the content including filename
Returns:
A dict containing the content bytes (possibly different from the one
provided as parameter if it is an image) under the key 'content_data
and the corresponding highlightjs language class under the
key 'language'.
"""
language = highlightjs.get_hljs_language_from_filename(path)
if not language:
language = highlightjs.get_hljs_language_from_mime_type(mime_type)
if not language:
language = 'nohighlight'
elif mime_type.startswith('application/'):
mime_type = mime_type.replace('application/', 'text/')
if mime_type.startswith('image/'):
if mime_type in _browsers_supported_image_mimes:
content_data = base64.b64encode(content_data)
content_data = content_data.decode('utf-8')
else:
content_data = None
if mime_type.startswith('image/svg'):
mime_type = 'image/svg+xml'
return {'content_data': content_data,
'language': language,
'mimetype': mime_type}
def process_snapshot_branches(snapshot):
"""
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_branches (dict): A dict describing the branches of a snapshot
as returned for instance by
:func:`swh.web.common.service.lookup_snapshot`
Returns:
tuple: 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 = {}
branch_aliases = {}
releases = {}
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] = {
'name': branch_name,
'revision': target_id,
}
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 _enrich_release_branch(branch, release):
releases[branch] = {
'name': release['name'],
'branch_name': branch,
'date': format_utc_iso_date(release['date']),
'id': release['id'],
'message': release['message'],
'target_type': release['target_type'],
'target': release['target'],
}
def _enrich_revision_branch(branch, revision):
branches[branch].update({
'revision': revision['id'],
'directory': revision['directory'],
'date': format_utc_iso_date(revision['date']),
'message': revision['message']
})
releases_info = service.lookup_release_multiple(
release_to_branch.keys()
)
for release in releases_info:
branches_to_update = release_to_branch[release['id']]
for branch in branches_to_update:
_enrich_release_branch(branch, release)
if release['target_type'] == 'revision':
revision_to_release[release['target']].update(
branches_to_update
)
revisions = service.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']]:
_enrich_revision_branch(branch, revision)
for release in revision_to_release[revision['id']]:
releases[release]['directory'] = revision['directory']
for branch_alias, branch_target in branch_aliases.items():
if branch_target in branches:
branches[branch_alias] = dict(branches[branch_target])
else:
snp = service.lookup_snapshot(snapshot['id'],
branches_from=branch_target,
branches_count=1)
if snp and branch_target in snp['branches']:
if snp['branches'][branch_target] is None:
continue
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 = service.lookup_revision(target)
_enrich_revision_branch(branch_alias, revision)
elif target_type == 'release':
release = service.lookup_release(target)
_enrich_release_branch(branch_alias, release)
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):
"""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 (str): 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 = []
releases = []
if snapshot_id:
snapshot = service.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, visit_ts=None, visit_id=None,
snapshot_id=None):
"""Returns the lists of branches and releases
associated to a swh origin for a given visit.
The visit is expressed by a timestamp. In the latter case,
the closest visit 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 (dict): a dict filled with origin information
(id, url, type)
visit_ts (int or str): an ISO date string or Unix timestamp to parse
visit_id (int): optional visit id for disambiguation in case
several visits have the same timestamp
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 gen_link(url, link_text=None, link_attrs=None):
"""
Utility function for generating an HTML link to insert
in Django templates.
Args:
url (str): an url
link_text (str): optional text for the produced link,
if not provided the url will be used
link_attrs (dict): optional attributes (e.g. class)
to add to the link
Returns:
An HTML link in the form 'link_text'
"""
attrs = ' '
if link_attrs:
for k, v in link_attrs.items():
attrs += '%s="%s" ' % (k, v)
if not link_text:
link_text = url
link = '%s' \
% (attrs, escape(url), escape(link_text))
return mark_safe(link)
def _snapshot_context_query_params(snapshot_context):
query_params = None
if snapshot_context and snapshot_context['origin_info']:
origin_info = snapshot_context['origin_info']
query_params = {'origin': origin_info['url']}
if 'timestamp' in snapshot_context['url_args']:
query_params['timestamp'] = \
snapshot_context['url_args']['timestamp']
if 'visit_id' in snapshot_context['query_params']:
query_params['visit_id'] = \
snapshot_context['query_params']['visit_id']
elif snapshot_context:
query_params = {'snapshot_id': snapshot_context['snapshot_id']}
return query_params
def gen_revision_url(revision_id, snapshot_context=None):
"""
Utility function for generating an url to a revision.
Args:
revision_id (str): a revision id
snapshot_context (dict): if provided, generate snapshot-dependent
browsing url
Returns:
str: The url to browse the revision
"""
query_params = _snapshot_context_query_params(snapshot_context)
return reverse('browse-revision',
url_args={'sha1_git': revision_id},
query_params=query_params)
def gen_revision_link(revision_id, shorten_id=False, snapshot_context=None,
link_text='Browse',
link_attrs={'class': 'btn btn-default btn-sm',
'role': 'button'}):
"""
Utility function for generating a link to a revision HTML view
to insert in Django templates.
Args:
revision_id (str): a revision id
shorten_id (boolean): whether to shorten the revision id to 7
characters for the link text
snapshot_context (dict): if provided, generate snapshot-dependent
browsing link
link_text (str): optional text for the generated link
(the revision id will be used by default)
link_attrs (dict): optional attributes (e.g. class)
to add to the link
Returns:
str: An HTML link in the form 'revision_id'
"""
if not revision_id:
return None
revision_url = gen_revision_url(revision_id, snapshot_context)
if shorten_id:
return gen_link(revision_url, revision_id[:7], link_attrs)
else:
if not link_text:
link_text = revision_id
return gen_link(revision_url, link_text, link_attrs)
def gen_directory_link(sha1_git, snapshot_context=None, link_text='Browse',
link_attrs={'class': 'btn btn-default btn-sm',
'role': 'button'}):
"""
Utility function for generating a link to a directory HTML view
to insert in Django templates.
Args:
sha1_git (str): directory identifier
link_text (str): optional text for the generated link
(the directory id will be used by default)
link_attrs (dict): optional attributes (e.g. class)
to add to the link
Returns:
An HTML link in the form 'link_text'
"""
if not sha1_git:
return None
query_params = _snapshot_context_query_params(snapshot_context)
directory_url = reverse('browse-directory',
url_args={'sha1_git': sha1_git},
query_params=query_params)
if not link_text:
link_text = sha1_git
return gen_link(directory_url, link_text, link_attrs)
def gen_snapshot_link(snapshot_id, snapshot_context=None, link_text='Browse',
link_attrs={'class': 'btn btn-default btn-sm',
'role': 'button'}):
"""
Utility function for generating a link to a snapshot HTML view
to insert in Django templates.
Args:
snapshot_id (str): snapshot identifier
link_text (str): optional text for the generated link
(the snapshot id will be used by default)
link_attrs (dict): optional attributes (e.g. class)
to add to the link
Returns:
An HTML link in the form 'link_text'
"""
query_params = _snapshot_context_query_params(snapshot_context)
snapshot_url = reverse('browse-snapshot',
url_args={'snapshot_id': snapshot_id},
query_params=query_params)
if not link_text:
link_text = snapshot_id
return gen_link(snapshot_url, link_text, link_attrs)
def gen_content_link(sha1_git, snapshot_context=None, link_text='Browse',
link_attrs={'class': 'btn btn-default btn-sm',
'role': 'button'}):
"""
Utility function for generating a link to a content HTML view
to insert in Django templates.
Args:
sha1_git (str): content identifier
link_text (str): optional text for the generated link
(the content sha1_git will be used by default)
link_attrs (dict): optional attributes (e.g. class)
to add to the link
Returns:
An HTML link in the form 'link_text'
"""
if not sha1_git:
return None
query_params = _snapshot_context_query_params(snapshot_context)
content_url = reverse('browse-content',
url_args={'query_string': 'sha1_git:' + sha1_git},
query_params=query_params)
if not link_text:
link_text = sha1_git
return gen_link(content_url, link_text, link_attrs)
def get_revision_log_url(revision_id, snapshot_context=None):
"""
Utility function for getting the URL for a revision log HTML view
(possibly in the context of an origin).
Args:
revision_id (str): revision identifier the history heads to
snapshot_context (dict): if provided, generate snapshot-dependent
browsing link
Returns:
The revision log view URL
"""
query_params = {'revision': revision_id}
if snapshot_context and snapshot_context['origin_info']:
origin_info = snapshot_context['origin_info']
url_args = {'origin_url': origin_info['url']}
if 'timestamp' in snapshot_context['url_args']:
url_args['timestamp'] = \
snapshot_context['url_args']['timestamp']
if 'visit_id' in snapshot_context['query_params']:
query_params['visit_id'] = \
snapshot_context['query_params']['visit_id']
revision_log_url = reverse('browse-origin-log',
url_args=url_args,
query_params=query_params)
elif snapshot_context:
url_args = {'snapshot_id': snapshot_context['snapshot_id']}
revision_log_url = reverse('browse-snapshot-log',
url_args=url_args,
query_params=query_params)
else:
revision_log_url = reverse('browse-revision-log',
url_args={'sha1_git': revision_id})
return revision_log_url
def gen_revision_log_link(revision_id, snapshot_context=None,
link_text='Browse',
link_attrs={'class': 'btn btn-default btn-sm',
'role': 'button'}):
"""
Utility function for generating a link to a revision log HTML view
(possibly in the context of an origin) to insert in Django templates.
Args:
revision_id (str): revision identifier the history heads to
snapshot_context (dict): if provided, generate snapshot-dependent
browsing link
link_text (str): optional text to use for the generated link
(the revision id will be used by default)
link_attrs (dict): optional attributes (e.g. class)
to add to the link
Returns:
An HTML link in the form
'link_text'
"""
if not revision_id:
return None
revision_log_url = get_revision_log_url(revision_id, snapshot_context)
if not link_text:
link_text = revision_id
return gen_link(revision_log_url, link_text, link_attrs)
+def gen_person_mail_link(person, link_text=None):
+ """
+ Utility function for generating a mail link to a person to insert
+ in Django templates.
+
+ Args:
+ person (dict): dictionary containing person data
+ (*name*, *email*, *fullname*)
+ link_text (str): optional text to use for the generated mail link
+ (the person name will be used by default)
+
+ Returns:
+ str: A mail link to the person or the person name if no email is
+ present in person data
+ """
+ person_name = person['name'] or person['fullname']
+ if link_text is None:
+ link_text = person_name
+ person_email = person['email'] if person['email'] else None
+ if person_email is None and '@' in person_name and ' ' not in person_name:
+ person_email = person_name
+ if person_email:
+ return gen_link(url='mailto:%s' % person_email,
+ link_text=link_text)
+ else:
+ return person_name
+
+
def gen_release_link(sha1_git, snapshot_context=None, link_text='Browse',
link_attrs={'class': 'btn btn-default btn-sm',
'role': 'button'}):
"""
Utility function for generating a link to a release HTML view
to insert in Django templates.
Args:
sha1_git (str): release identifier
link_text (str): optional text for the generated link
(the release id will be used by default)
link_attrs (dict): optional attributes (e.g. class)
to add to the link
Returns:
An HTML link in the form 'link_text'
"""
query_params = _snapshot_context_query_params(snapshot_context)
release_url = reverse('browse-release',
url_args={'sha1_git': sha1_git},
query_params=query_params)
if not link_text:
link_text = sha1_git
return gen_link(release_url, link_text, link_attrs)
def format_log_entries(revision_log, per_page, snapshot_context=None):
"""
Utility functions that process raw revision log data for HTML display.
Its purpose is to:
* add links to relevant browse views
* format date in human readable format
* truncate the message log
Args:
revision_log (list): raw revision log as returned by the swh-web api
per_page (int): number of log entries per page
snapshot_context (dict): if provided, generate snapshot-dependent
browsing link
"""
revision_log_data = []
for i, rev in enumerate(revision_log):
if i == per_page:
break
author_name = 'None'
author_fullname = 'None'
committer_fullname = 'None'
if rev['author']:
- author_name = rev['author']['name'] or rev['author']['fullname']
+ author_name = gen_person_mail_link(rev['author'])
author_fullname = rev['author']['fullname']
if rev['committer']:
committer_fullname = rev['committer']['fullname']
author_date = format_utc_iso_date(rev['date'])
committer_date = format_utc_iso_date(rev['committer_date'])
tooltip = 'revision %s\n' % rev['id']
tooltip += 'author: %s\n' % author_fullname
tooltip += 'author date: %s\n' % author_date
tooltip += 'committer: %s\n' % committer_fullname
tooltip += 'committer date: %s\n\n' % committer_date
if rev['message']:
tooltip += textwrap.indent(rev['message'], ' '*4)
revision_log_data.append({
'author': author_name,
'id': rev['id'][:7],
'message': rev['message'],
'date': author_date,
'commit_date': committer_date,
'url': gen_revision_url(rev['id'], snapshot_context),
'tooltip': tooltip
})
return revision_log_data
# list of origin types that can be found in the swh archive
# TODO: retrieve it dynamically in an efficient way instead
# of hardcoding it
_swh_origin_types = ['git', 'svn', 'deb', 'hg', 'ftp', 'deposit',
'pypi', 'npm']
def get_origin_info(origin_url, origin_type=None):
"""
Get info about a software origin.
Its main purpose is to automatically find an origin type
when it is not provided as parameter.
Args:
origin_url (str): complete url of a software origin
origin_type (str): optional origin type
Returns:
A dict with the following entries:
* type: the origin type
* url: the origin url
* id: the internal id of the origin
"""
if origin_type:
return service.lookup_origin({'type': origin_type,
'url': origin_url})
else:
for origin_type in _swh_origin_types:
try:
origin_info = service.lookup_origin({'type': origin_type,
'url': origin_url})
return origin_info
except Exception:
pass
raise NotFoundExc('Origin with url %s not found!' % escape(origin_url))
def get_snapshot_context(snapshot_id=None, origin_type=None, origin_url=None,
timestamp=None, visit_id=None):
"""
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 (str): hexadecimal representation of a snapshot identifier,
all other parameters will be ignored if it is provided
origin_type (str): the origin type (git, svn, deposit, ...)
origin_url (str): the origin_url
(e.g. https://github.com/(user)/(repo)/)
timestamp (str): a datetime string for retrieving the closest
visit of the origin
visit_id (int): optional visit id for disambiguation in case
of several visits with the same timestamp
Returns:
A dict with the following entries:
* origin_info: dict containing origin information
* visit_info: dict containing visit information
* branches: the list of branches for the origin found
during the visit
* releases: the list of releases for the origin found
during the visit
* origin_browse_url: the url to browse the origin
* origin_branches_url: the url to browse the origin branches
* origin_releases_url': the url to browse the origin releases
* origin_visit_url: the url to browse the snapshot of the origin
found during the visit
* url_args: dict containing url arguments to use when browsing in
the context of the origin and its visit
Raises:
NotFoundExc: if no snapshot is found for the visit of an origin.
"""
origin_info = None
visit_info = None
url_args = None
query_params = {}
branches = []
releases = []
browse_url = None
visit_url = None
branches_url = None
releases_url = None
swh_type = 'snapshot'
if origin_url:
swh_type = 'origin'
origin_info = get_origin_info(origin_url, origin_type)
visit_info = get_origin_visit(origin_info, timestamp, visit_id,
snapshot_id)
fmt_date = format_utc_iso_date(visit_info['date'])
visit_info['fmt_date'] = fmt_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), fmt_date))
# provided timestamp is not necessarily equals to the one
# of the retrieved visit, so get the exact one in order
# 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)
url_args = {'origin_type': origin_type,
'origin_url': origin_info['url']}
query_params = {'visit_id': visit_id}
browse_url = reverse('browse-origin-visits',
url_args=url_args)
if timestamp:
url_args['timestamp'] = format_utc_iso_date(timestamp,
'%Y-%m-%dT%H:%M:%S')
visit_url = reverse('browse-origin-directory',
url_args=url_args,
query_params=query_params)
visit_info['url'] = visit_url
branches_url = reverse('browse-origin-branches',
url_args=url_args,
query_params=query_params)
releases_url = reverse('browse-origin-releases',
url_args=url_args,
query_params=query_params)
elif snapshot_id:
branches, releases = get_snapshot_content(snapshot_id)
url_args = {'snapshot_id': snapshot_id}
browse_url = reverse('browse-snapshot',
url_args=url_args)
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_size = service.lookup_snapshot_size(snapshot_id)
is_empty = sum(snapshot_size.values()) == 0
swh_snp_id = persistent_identifier('snapshot', snapshot_id)
return {
'swh_type': swh_type,
'swh_object_id': swh_snp_id,
'snapshot_id': snapshot_id,
'snapshot_size': snapshot_size,
'is_empty': is_empty,
'origin_info': origin_info,
# keep track if the origin type was provided as url argument
'origin_type': origin_type,
'visit_info': visit_info,
'branches': branches,
'releases': releases,
'branch': None,
'release': None,
'browse_url': browse_url,
'branches_url': branches_url,
'releases_url': releases_url,
'url_args': url_args,
'query_params': query_params
}
# list of common readme names ordered by preference
# (lower indices have higher priority)
_common_readme_names = [
"readme.markdown",
"readme.md",
"readme.rst",
"readme.txt",
"readme"
]
def get_readme_to_display(readmes):
"""
Process a list of readme files found in a directory
in order to find the adequate one to display.
Args:
readmes: a list of dict where keys are readme file names and values
are readme sha1s
Returns:
A tuple (readme_name, readme_sha1)
"""
readme_name = None
readme_url = None
readme_sha1 = None
readme_html = None
lc_readmes = {k.lower(): {'orig_name': k, 'sha1': v}
for k, v in readmes.items()}
# look for readme names according to the preference order
# defined by the _common_readme_names list
for common_readme_name in _common_readme_names:
if common_readme_name in lc_readmes:
readme_name = lc_readmes[common_readme_name]['orig_name']
readme_sha1 = lc_readmes[common_readme_name]['sha1']
readme_url = reverse('browse-content-raw',
url_args={'query_string': readme_sha1},
query_params={'re_encode': 'true'})
break
# otherwise pick the first readme like file if any
if not readme_name and len(readmes.items()) > 0:
readme_name = next(iter(readmes))
readme_sha1 = readmes[readme_name]
readme_url = reverse('browse-content-raw',
url_args={'query_string': readme_sha1},
query_params={'re_encode': 'true'})
# convert rst README to html server side as there is
# no viable solution to perform that task client side
if readme_name and readme_name.endswith('.rst'):
cache_entry_id = 'readme_%s' % readme_sha1
cache_entry = cache.get(cache_entry_id)
if cache_entry:
readme_html = cache_entry
else:
try:
rst_doc = request_content(readme_sha1)
readme_html = pypandoc.convert_text(rst_doc['raw_data'],
'html', format='rst')
cache.set(cache_entry_id, readme_html)
except Exception:
readme_html = 'Readme bytes are not available'
return readme_name, readme_url, readme_html
def get_swh_persistent_ids(swh_objects, snapshot_context=None):
"""
Returns a list of dict containing info related to persistent
identifiers of swh objects.
Args:
swh_objects (list): a list of dict with the following keys:
* type: swh object type
(content/directory/release/revision/snapshot)
* id: swh object id
snapshot_context (dict): optional parameter describing the snapshot in
which the object has been found
Returns:
list: a list of dict with the following keys:
* object_type: the swh object type
(content/directory/release/revision/snapshot)
* object_icon: the swh object icon to use in HTML views
* swh_id: the computed swh object persistent identifier
* swh_id_url: the url resolving the persistent identifier
* show_options: boolean indicating if the persistent id options
must be displayed in persistent ids HTML view
"""
swh_ids = []
for swh_object in swh_objects:
if not swh_object['id']:
continue
swh_id = get_swh_persistent_id(swh_object['type'], swh_object['id'])
show_options = swh_object['type'] == 'content' or \
(snapshot_context and snapshot_context['origin_info'] is not None)
object_icon = swh_object_icons[swh_object['type']]
swh_ids.append({
'object_type': swh_object['type'],
'object_icon': object_icon,
'swh_id': swh_id,
'swh_id_url': reverse('browse-swh-id',
url_args={'swh_id': swh_id}),
'show_options': show_options
})
return swh_ids
diff --git a/swh/web/browse/views/release.py b/swh/web/browse/views/release.py
index 9dba53e1..d3fba27c 100644
--- a/swh/web/browse/views/release.py
+++ b/swh/web/browse/views/release.py
@@ -1,192 +1,189 @@
# Copyright (C) 2017-2019 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 django.shortcuts import render
from swh.web.common import service
from swh.web.common.utils import (
reverse, format_utc_iso_date
)
from swh.web.common.exc import NotFoundExc, handle_view_exception
from swh.web.browse.browseurls import browse_route
from swh.web.browse.utils import (
gen_revision_link, get_snapshot_context, gen_link,
gen_snapshot_link, get_swh_persistent_ids, gen_directory_link,
- gen_content_link, gen_release_link
+ gen_content_link, gen_release_link, gen_person_mail_link
)
@browse_route(r'release/(?P[0-9a-f]+)/',
view_name='browse-release',
checksum_args=['sha1_git'])
def release_browse(request, sha1_git):
"""
Django view that produces an HTML display of a release
identified by its id.
The url that points to it is :http:get:`/browse/release/(sha1_git)/`.
"""
try:
release = service.lookup_release(sha1_git)
snapshot_context = None
origin_info = None
snapshot_id = request.GET.get('snapshot_id', None)
origin_type = request.GET.get('origin_type', None)
origin_url = request.GET.get('origin_url', None)
if not origin_url:
origin_url = request.GET.get('origin', None)
timestamp = request.GET.get('timestamp', None)
visit_id = request.GET.get('visit_id', None)
if origin_url:
try:
snapshot_context = \
get_snapshot_context(snapshot_id, origin_type,
origin_url, timestamp,
visit_id)
except Exception:
raw_rel_url = reverse('browse-release',
url_args={'sha1_git': sha1_git})
error_message = \
('The Software Heritage archive has a release '
'with the hash you provided but the origin '
'mentioned in your request appears broken: %s. '
'Please check the URL and try again.\n\n'
'Nevertheless, you can still browse the release '
'without origin information: %s'
% (gen_link(origin_url), gen_link(raw_rel_url)))
raise NotFoundExc(error_message)
origin_info = snapshot_context['origin_info']
elif snapshot_id:
snapshot_context = get_snapshot_context(snapshot_id)
except Exception as exc:
return handle_view_exception(request, exc)
release_data = {}
- author_name = 'None'
release_data['author'] = 'None'
if release['author']:
- author_name = release['author']['name'] or \
- release['author']['fullname']
- release_data['author'] = author_name,
+ release_data['author'] = gen_person_mail_link(release['author'])
release_data['date'] = format_utc_iso_date(release['date'])
release_data['release'] = sha1_git
release_data['name'] = release['name']
release_data['synthetic'] = release['synthetic']
release_data['target'] = release['target']
release_data['target type'] = release['target_type']
if snapshot_context:
if release['target_type'] == 'revision':
release_data['context-independent target'] = \
gen_revision_link(release['target'])
elif release['target_type'] == 'content':
release_data['context-independent target'] = \
gen_content_link(release['target'])
elif release['target_type'] == 'directory':
release_data['context-independent target'] = \
gen_directory_link(release['target'])
elif release['target_type'] == 'release':
release_data['context-independent target'] = \
gen_release_link(release['target'])
release_note_lines = []
if release['message']:
release_note_lines = release['message'].split('\n')
vault_cooking = None
target_link = None
if release['target_type'] == 'revision':
target_link = gen_revision_link(release['target'],
snapshot_context=snapshot_context,
link_text=None, link_attrs=None)
try:
revision = service.lookup_revision(release['target'])
vault_cooking = {
'directory_context': True,
'directory_id': revision['directory'],
'revision_context': True,
'revision_id': release['target']
}
except Exception:
pass
elif release['target_type'] == 'directory':
target_link = gen_directory_link(release['target'],
snapshot_context=snapshot_context,
link_text=None, link_attrs=None)
try:
revision = service.lookup_directory(release['target'])
vault_cooking = {
'directory_context': True,
'directory_id': revision['directory'],
'revision_context': False,
'revision_id': None
}
except Exception:
pass
elif release['target_type'] == 'content':
target_link = gen_content_link(release['target'],
snapshot_context=snapshot_context,
link_text=None, link_attrs=None)
elif release['target_type'] == 'release':
target_link = gen_release_link(release['target'],
snapshot_context=snapshot_context,
link_text=None, link_attrs=None)
release['target_link'] = target_link
if snapshot_context:
release_data['snapshot'] = snapshot_context['snapshot_id']
if origin_info:
release_data['context-independent release'] = \
gen_release_link(release['id'])
release_data['origin type'] = origin_info['type']
release_data['origin url'] = gen_link(origin_info['url'],
origin_info['url'])
browse_snapshot_link = \
gen_snapshot_link(snapshot_context['snapshot_id'])
release_data['context-independent snapshot'] = browse_snapshot_link
swh_objects = [{'type': 'release',
'id': sha1_git}]
if snapshot_context:
snapshot_id = snapshot_context['snapshot_id']
if snapshot_id:
swh_objects.append({'type': 'snapshot',
'id': snapshot_id})
swh_ids = get_swh_persistent_ids(swh_objects, snapshot_context)
note_header = 'None'
if len(release_note_lines) > 0:
note_header = release_note_lines[0]
release['note_header'] = note_header
release['note_body'] = '\n'.join(release_note_lines[1:])
heading = 'Release - %s' % release['name']
if snapshot_context:
context_found = 'snapshot: %s' % snapshot_context['snapshot_id']
if origin_info:
context_found = 'origin: %s' % origin_info['url']
heading += ' - %s' % context_found
return render(request, 'browse/release.html',
{'heading': heading,
'swh_object_id': swh_ids[0]['swh_id'],
'swh_object_name': 'Release',
'swh_object_metadata': release_data,
'release': release,
'snapshot_context': snapshot_context,
'show_actions_menu': True,
'breadcrumbs': None,
'vault_cooking': vault_cooking,
'top_right_link': None,
'swh_ids': swh_ids})
diff --git a/swh/web/browse/views/revision.py b/swh/web/browse/views/revision.py
index a6125b81..b2f86a35 100644
--- a/swh/web/browse/views/revision.py
+++ b/swh/web/browse/views/revision.py
@@ -1,534 +1,531 @@
# Copyright (C) 2017-2019 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 hashlib
import json
import textwrap
from django.http import HttpResponse
from django.shortcuts import render
from django.template.defaultfilters import filesizeformat
from django.utils.html import escape
from django.utils.safestring import mark_safe
from swh.model.identifiers import persistent_identifier
from swh.web.common import service
from swh.web.common.utils import (
reverse, format_utc_iso_date, gen_path_info, swh_object_icons
)
from swh.web.common.exc import NotFoundExc, handle_view_exception
from swh.web.browse.browseurls import browse_route
from swh.web.browse.utils import (
gen_link, gen_revision_link, gen_revision_url,
get_snapshot_context, get_revision_log_url, get_directory_entries,
gen_directory_link, request_content, prepare_content_for_display,
content_display_max_size, gen_snapshot_link, get_readme_to_display,
- get_swh_persistent_ids, format_log_entries
+ get_swh_persistent_ids, format_log_entries, gen_person_mail_link
)
def _gen_content_url(revision, query_string, path, snapshot_context):
if snapshot_context:
url_args = snapshot_context['url_args']
url_args['path'] = path
query_params = snapshot_context['query_params']
query_params['revision'] = revision['id']
content_url = reverse('browse-origin-content',
url_args=url_args,
query_params=query_params)
else:
content_path = '%s/%s' % (revision['directory'], path)
content_url = reverse('browse-content',
url_args={'query_string': query_string},
query_params={'path': content_path})
return content_url
def _gen_diff_link(idx, diff_anchor, link_text):
if idx < _max_displayed_file_diffs:
return gen_link(diff_anchor, link_text)
else:
return link_text
# TODO: put in conf
_max_displayed_file_diffs = 1000
def _gen_revision_changes_list(revision, changes, snapshot_context):
"""
Returns a HTML string describing the file changes
introduced in a revision.
As this string will be displayed in the browse revision view,
links to adequate file diffs are also generated.
Args:
revision (str): hexadecimal representation of a revision identifier
changes (list): list of file changes in the revision
snapshot_context (dict): optional origin context used to reverse
the content urls
Returns:
A string to insert in a revision HTML view.
"""
changes_msg = []
for i, change in enumerate(changes):
hasher = hashlib.sha1()
from_query_string = ''
to_query_string = ''
diff_id = 'diff-'
if change['from']:
from_query_string = 'sha1_git:' + change['from']['target']
diff_id += change['from']['target'] + '-' + change['from_path']
diff_id += '-'
if change['to']:
to_query_string = 'sha1_git:' + change['to']['target']
diff_id += change['to']['target'] + change['to_path']
change['path'] = change['to_path'] or change['from_path']
url_args = {'from_query_string': from_query_string,
'to_query_string': to_query_string}
query_params = {'path': change['path']}
change['diff_url'] = reverse('diff-contents',
url_args=url_args,
query_params=query_params)
hasher.update(diff_id.encode('utf-8'))
diff_id = hasher.hexdigest()
change['id'] = diff_id
panel_diff_link = '#panel_' + diff_id
if change['type'] == 'modify':
change['content_url'] = \
_gen_content_url(revision, to_query_string,
change['to_path'], snapshot_context)
changes_msg.append('modified: %s' %
_gen_diff_link(i, panel_diff_link,
change['to_path']))
elif change['type'] == 'insert':
change['content_url'] = \
_gen_content_url(revision, to_query_string,
change['to_path'], snapshot_context)
changes_msg.append('new file: %s' %
_gen_diff_link(i, panel_diff_link,
change['to_path']))
elif change['type'] == 'delete':
parent = service.lookup_revision(revision['parents'][0])
change['content_url'] = \
_gen_content_url(parent,
from_query_string,
change['from_path'], snapshot_context)
changes_msg.append('deleted: %s' %
_gen_diff_link(i, panel_diff_link,
change['from_path']))
elif change['type'] == 'rename':
change['content_url'] = \
_gen_content_url(revision, to_query_string,
change['to_path'], snapshot_context)
link_text = change['from_path'] + ' → ' + change['to_path']
changes_msg.append('renamed: %s' %
_gen_diff_link(i, panel_diff_link, link_text))
if not changes:
changes_msg.append('No changes')
return mark_safe('\n'.join(changes_msg))
@browse_route(r'revision/(?P[0-9a-f]+)/diff/',
view_name='diff-revision',
checksum_args=['sha1_git'])
def _revision_diff(request, sha1_git):
"""
Browse internal endpoint to compute revision diff
"""
try:
revision = service.lookup_revision(sha1_git)
snapshot_context = None
origin_type = request.GET.get('origin_type', None)
origin_url = request.GET.get('origin_url', None)
if not origin_url:
origin_url = request.GET.get('origin', None)
timestamp = request.GET.get('timestamp', None)
visit_id = request.GET.get('visit_id', None)
if origin_url:
snapshot_context = get_snapshot_context(None, origin_type,
origin_url,
timestamp, visit_id)
except Exception as exc:
return handle_view_exception(request, exc)
changes = service.diff_revision(sha1_git)
changes_msg = _gen_revision_changes_list(revision, changes,
snapshot_context)
diff_data = {
'total_nb_changes': len(changes),
'changes': changes[:_max_displayed_file_diffs],
'changes_msg': changes_msg
}
diff_data_json = json.dumps(diff_data, separators=(',', ': '))
return HttpResponse(diff_data_json, content_type='application/json')
NB_LOG_ENTRIES = 100
@browse_route(r'revision/(?P[0-9a-f]+)/log/',
view_name='browse-revision-log',
checksum_args=['sha1_git'])
def revision_log_browse(request, sha1_git):
"""
Django view that produces an HTML display of the history
log for a revision identified by its id.
The url that points to it is :http:get:`/browse/revision/(sha1_git)/log/`
"""
try:
per_page = int(request.GET.get('per_page', NB_LOG_ENTRIES))
offset = int(request.GET.get('offset', 0))
revs_ordering = request.GET.get('revs_ordering', 'committer_date')
session_key = 'rev_%s_log_ordering_%s' % (sha1_git, 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 = \
service.get_revisions_walker(revs_ordering, sha1_git,
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 = service.lookup_revision_multiple(revs)
request.session[session_key] = {
'rev_log': rev_log,
'revs_walker_state': revs_walker_state
}
except Exception as exc:
return handle_view_exception(request, exc)
revs_ordering = request.GET.get('revs_ordering', '')
prev_log_url = None
if len(rev_log) > offset + per_page:
prev_log_url = reverse('browse-revision-log',
url_args={'sha1_git': sha1_git},
query_params={'per_page': per_page,
'offset': offset + per_page,
'revs_ordering': revs_ordering})
next_log_url = None
if offset != 0:
next_log_url = reverse('browse-revision-log',
url_args={'sha1_git': sha1_git},
query_params={'per_page': per_page,
'offset': offset - per_page,
'revs_ordering': revs_ordering})
revision_log_data = format_log_entries(revision_log, per_page)
swh_rev_id = persistent_identifier('revision', sha1_git)
return render(request, 'browse/revision-log.html',
{'heading': 'Revision history',
'swh_object_id': swh_rev_id,
'swh_object_name': 'Revisions history',
'swh_object_metadata': None,
'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': None,
'vault_cooking': None,
'show_actions_menu': True,
'swh_ids': None})
@browse_route(r'revision/(?P[0-9a-f]+)/',
r'revision/(?P[0-9a-f]+)/(?P.+)/',
view_name='browse-revision',
checksum_args=['sha1_git'])
def revision_browse(request, sha1_git, extra_path=None):
"""
Django view that produces an HTML display of a revision
identified by its id.
The url that points to it is :http:get:`/browse/revision/(sha1_git)/`.
"""
try:
revision = service.lookup_revision(sha1_git)
origin_info = None
snapshot_context = None
origin_type = request.GET.get('origin_type', None)
origin_url = request.GET.get('origin_url', None)
if not origin_url:
origin_url = request.GET.get('origin', None)
timestamp = request.GET.get('timestamp', None)
visit_id = request.GET.get('visit_id', None)
snapshot_id = request.GET.get('snapshot_id', None)
path = request.GET.get('path', None)
dir_id = None
dirs, files = None, None
content_data = None
if origin_url:
try:
snapshot_context = get_snapshot_context(None, origin_type,
origin_url,
timestamp, visit_id)
except Exception:
raw_rev_url = reverse('browse-revision',
url_args={'sha1_git': sha1_git})
error_message = \
('The Software Heritage archive has a revision '
'with the hash you provided but the origin '
'mentioned in your request appears broken: %s. '
'Please check the URL and try again.\n\n'
'Nevertheless, you can still browse the revision '
'without origin information: %s'
% (gen_link(origin_url), gen_link(raw_rev_url)))
raise NotFoundExc(error_message)
origin_info = snapshot_context['origin_info']
snapshot_id = snapshot_context['snapshot_id']
elif snapshot_id:
snapshot_context = get_snapshot_context(snapshot_id)
if path:
file_info = \
service.lookup_directory_with_path(revision['directory'], path)
if file_info['type'] == 'dir':
dir_id = file_info['target']
else:
query_string = 'sha1_git:' + file_info['target']
content_data = request_content(query_string,
raise_if_unavailable=False)
else:
dir_id = revision['directory']
if dir_id:
path = '' if path is None else (path + '/')
dirs, files = get_directory_entries(dir_id)
except Exception as exc:
return handle_view_exception(request, exc)
revision_data = {}
- author_name = 'None'
revision_data['author'] = 'None'
if revision['author']:
- author_name = revision['author']['name'] or \
- revision['author']['fullname']
- revision_data['author'] = author_name,
+ author_link = gen_person_mail_link(revision['author'])
+ revision_data['author'] = author_link
revision_data['committer'] = 'None'
if revision['committer']:
- committer_name = revision['committer']['name'] or \
- revision['committer']['fullname']
- revision_data['committer'] = committer_name
+ committer_link = gen_person_mail_link(revision['committer'])
+ revision_data['committer'] = committer_link
revision_data['committer date'] = \
format_utc_iso_date(revision['committer_date'])
revision_data['date'] = format_utc_iso_date(revision['date'])
revision_data['directory'] = revision['directory']
if snapshot_context:
revision_data['snapshot'] = snapshot_id
browse_snapshot_link = \
gen_snapshot_link(snapshot_id)
revision_data['context-independent snapshot'] = browse_snapshot_link
revision_data['context-independent directory'] = \
gen_directory_link(revision['directory'])
revision_data['revision'] = sha1_git
revision_data['merge'] = revision['merge']
revision_data['metadata'] = escape(json.dumps(revision['metadata'],
sort_keys=True,
indent=4, separators=(',', ': ')))
if origin_info:
revision_data['origin type'] = origin_info['type']
revision_data['origin url'] = gen_link(origin_info['url'],
origin_info['url'])
revision_data['context-independent revision'] = \
gen_revision_link(sha1_git)
parents = ''
for p in revision['parents']:
parent_link = gen_revision_link(p, link_text=None, link_attrs=None,
snapshot_context=snapshot_context)
parents += parent_link + '
'
revision_data['parents'] = mark_safe(parents)
revision_data['synthetic'] = revision['synthetic']
revision_data['type'] = revision['type']
message_lines = ['None']
if revision['message']:
message_lines = revision['message'].split('\n')
parents = []
for p in revision['parents']:
parent_url = gen_revision_url(p, snapshot_context)
parents.append({'id': p, 'url': parent_url})
path_info = gen_path_info(path)
query_params = {'snapshot_id': snapshot_id,
'origin_type': origin_type,
'origin': origin_url,
'timestamp': timestamp,
'visit_id': visit_id}
breadcrumbs = []
breadcrumbs.append({'name': revision['directory'][:7],
'url': reverse('browse-revision',
url_args={'sha1_git': sha1_git},
query_params=query_params)})
for pi in path_info:
query_params['path'] = pi['path']
breadcrumbs.append({'name': pi['name'],
'url': reverse('browse-revision',
url_args={'sha1_git': sha1_git},
query_params=query_params)})
vault_cooking = {
'directory_context': False,
'directory_id': None,
'revision_context': True,
'revision_id': sha1_git
}
swh_objects = [{'type': 'revision',
'id': sha1_git}]
content = None
content_size = None
mimetype = None
language = None
readme_name = None
readme_url = None
readme_html = None
readmes = {}
error_code = 200
error_message = ''
error_description = ''
if content_data:
breadcrumbs[-1]['url'] = None
content_size = content_data['length']
mimetype = content_data['mimetype']
if content_data['raw_data']:
content_display_data = prepare_content_for_display(
content_data['raw_data'], content_data['mimetype'], path)
content = content_display_data['content_data']
language = content_display_data['language']
mimetype = content_display_data['mimetype']
query_params = {}
if path:
filename = path_info[-1]['name']
query_params['filename'] = path_info[-1]['name']
revision_data['filename'] = filename
top_right_link = {
'url': reverse('browse-content-raw',
url_args={'query_string': query_string},
query_params=query_params),
'icon': swh_object_icons['content'],
'text': 'Raw File'
}
swh_objects.append({'type': 'content',
'id': file_info['target']})
error_code = content_data['error_code']
error_message = content_data['error_message']
error_description = content_data['error_description']
else:
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-revision',
url_args={'sha1_git': sha1_git},
query_params=query_params)
for f in files:
query_params['path'] = path + f['name']
f['url'] = reverse('browse-revision',
url_args={'sha1_git': sha1_git},
query_params=query_params)
if f['length'] is not None:
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)
top_right_link = {
'url': get_revision_log_url(sha1_git, snapshot_context),
'icon': swh_object_icons['revisions history'],
'text': 'History'
}
vault_cooking['directory_context'] = True
vault_cooking['directory_id'] = dir_id
swh_objects.append({'type': 'directory',
'id': dir_id})
diff_revision_url = reverse('diff-revision',
url_args={'sha1_git': sha1_git},
query_params={'origin_type': origin_type,
'origin': origin_url,
'timestamp': timestamp,
'visit_id': visit_id})
if snapshot_id:
swh_objects.append({'type': 'snapshot',
'id': snapshot_id})
swh_ids = get_swh_persistent_ids(swh_objects, snapshot_context)
heading = 'Revision - %s - %s' %\
(sha1_git[:7], textwrap.shorten(message_lines[0], width=70))
if snapshot_context:
context_found = 'snapshot: %s' % snapshot_context['snapshot_id']
if origin_info:
context_found = 'origin: %s' % origin_info['url']
heading += ' - %s' % context_found
return render(request, 'browse/revision.html',
{'heading': heading,
'swh_object_id': swh_ids[0]['swh_id'],
'swh_object_name': 'Revision',
'swh_object_metadata': revision_data,
'message_header': message_lines[0],
'message_body': '\n'.join(message_lines[1:]),
'parents': parents,
'snapshot_context': snapshot_context,
'dirs': dirs,
'files': files,
'content': content,
'content_size': content_size,
'max_content_size': content_display_max_size,
'mimetype': mimetype,
'language': language,
'readme_name': readme_name,
'readme_url': readme_url,
'readme_html': readme_html,
'breadcrumbs': breadcrumbs,
'top_right_link': top_right_link,
'vault_cooking': vault_cooking,
'diff_revision_url': diff_revision_url,
'show_actions_menu': True,
'swh_ids': swh_ids,
'error_code': error_code,
'error_message': error_message,
'error_description': error_description},
status=error_code)
diff --git a/swh/web/tests/browse/test_utils.py b/swh/web/tests/browse/test_utils.py
index fa626a2a..2858eaf8 100644
--- a/swh/web/tests/browse/test_utils.py
+++ b/swh/web/tests/browse/test_utils.py
@@ -1,80 +1,123 @@
# Copyright (C) 2017-2019 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 hypothesis import given
from swh.web.browse import utils
from swh.web.common.utils import reverse, format_utc_iso_date
from swh.web.tests.strategies import origin_with_multiple_visits
from swh.web.tests.testcase import WebTestCase
class SwhBrowseUtilsTestCase(WebTestCase):
def test_get_mimetype_and_encoding_for_content(self):
text = b'Hello world!'
self.assertEqual(utils.get_mimetype_and_encoding_for_content(text),
('text/plain', 'us-ascii'))
@given(origin_with_multiple_visits())
def test_get_origin_visit_snapshot_simple(self, origin):
visits = self.origin_visit_get(origin['url'])
for visit in visits:
snapshot = self.snapshot_get(visit['snapshot'])
branches = []
releases = []
for branch in sorted(snapshot['branches'].keys()):
branch_data = snapshot['branches'][branch]
if branch_data['target_type'] == 'revision':
rev_data = self.revision_get(branch_data['target'])
branches.append({
'name': branch,
'revision': branch_data['target'],
'directory': rev_data['directory'],
'date': format_utc_iso_date(rev_data['date']),
'message': rev_data['message']
})
elif branch_data['target_type'] == 'release':
rel_data = self.release_get(branch_data['target'])
rev_data = self.revision_get(rel_data['target'])
releases.append({
'name': rel_data['name'],
'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']
})
assert branches and releases, 'Incomplete test data.'
origin_visit_branches = utils.get_origin_visit_snapshot(
origin, visit_id=visit['visit'])
self.assertEqual(origin_visit_branches, (branches, releases))
def test_gen_link(self):
self.assertEqual(
utils.gen_link('https://www.softwareheritage.org/', 'swh'),
'swh')
def test_gen_revision_link(self):
revision_id = '28a0bc4120d38a394499382ba21d6965a67a3703'
revision_url = reverse('browse-revision',
url_args={'sha1_git': revision_id})
self.assertEqual(utils.gen_revision_link(revision_id, link_text=None,
link_attrs=None),
'%s' % (revision_url, revision_id))
self.assertEqual(
utils.gen_revision_link(revision_id, shorten_id=True,
link_attrs=None),
'%s' % (revision_url, revision_id[:7]))
+
+ def test_gen_person_mail_link(self):
+ person_full = {
+ 'name': 'John Doe',
+ 'email': 'john.doe@swh.org',
+ 'fullname': 'John Doe '
+ }
+
+ self.assertEqual(
+ utils.gen_person_mail_link(person_full),
+ '%s' % (person_full['email'],
+ person_full['name'])
+ )
+
+ link_text = 'Mail'
+ self.assertEqual(
+ utils.gen_person_mail_link(person_full, link_text=link_text),
+ '%s' % (person_full['email'],
+ link_text)
+ )
+
+ person_partial_email = {
+ 'name': None,
+ 'email': None,
+ 'fullname': 'john.doe@swh.org'
+ }
+
+ self.assertEqual(
+ utils.gen_person_mail_link(person_partial_email),
+ '%s' % (person_partial_email['fullname'],
+ person_partial_email['fullname'])
+ )
+
+ person_partial = {
+ 'name': None,
+ 'email': None,
+ 'fullname': 'John Doe '
+ }
+
+ self.assertEqual(
+ utils.gen_person_mail_link(person_partial),
+ person_partial['fullname']
+ )