diff --git a/swh/web/ui/controller.py b/swh/web/ui/controller.py index 8ffa879f..0ad983d6 100644 --- a/swh/web/ui/controller.py +++ b/swh/web/ui/controller.py @@ -1,333 +1,362 @@ # Copyright (C) 2015 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 logging from flask import redirect, render_template, url_for, jsonify, request from flask import make_response from swh.core.hashutil import ALGORITHMS from swh.web.ui.main import app -from swh.web.ui import service, query +from swh.web.ui import service from swh.web.ui.decorators import jsonp hash_filter_keys = ALGORITHMS @app.route('/') def main(): """Main application view. At the moment, redirect to the content search view. """ return redirect(url_for('info')) @app.route('/info') def info(): """A simple api to define what the server is all about. """ logging.info('Dev SWH UI') return 'Dev SWH UI' @app.route('/search') def search(): """Search for hashes in swh-storage. """ q = request.args.get('q', '') env = {'q': q, 'message': '', 'found': None} try: if q: env['found'] = service.lookup_hash(q) except ValueError: env['message'] = 'Error: invalid query string' return render_template('search.html', **env) @app.route('/browse/revision/') def revision(sha1_git): """Show commit information. Args: sha1_git: the revision's sha1 Returns: Revision information """ return render_template('revision.html', sha1_git=sha1_git) @app.route('/browse/directory/') def directory(sha1_git): """Show directory information. Args: sha1_git: the directory's sha1 Returns: Directory information """ return render_template('directory.html', sha1_git=sha1_git) @app.route('/browse/directory//') def directory_at_path(sha1_git, p): """Show directory information for the sha1_git at path. Args: sha1_git: the directory's sha1 path: file or directory pointed to Returns: Directory information at sha1_git + path """ return render_template('directory.html', sha1_git=sha1_git, path=p) +def _origin_seen(hash, data): + """Given an origin, compute a message string with the right information. + + Args: + origin: a dictionary with keys: + - origin: a dictionary with type and url keys + - occurrence: a dictionary with a validity range + + Returns: + message as a string + + """ + if data is None: + return 'Content with hash %s is unknown as of now.' % hash + + origin_type = data['origin_type'] + origin_url = data['origin_url'] + revision = data['revision'] + branch = data['branch'] + path = data['path'] + + return """The content with hash %s has been seen on origin with type '%s' +at url '%s'. The revision was identified at '%s' on branch '%s'. +The file's path referenced was '%s'.""" % (hash, + origin_type, + origin_url, + revision, + branch, + path) + + @app.route('/browse/content/:') def content(hash, sha): """Show content information. Args: hash: hash according to HASH_ALGO, where HASH_ALGO is one of: sha1, sha1_git, sha256. This means that several different URLs (at least one per HASH_ALGO) will point to the same content sha: the sha with 'hash' format Returns: The content's information at sha1_git """ # Checks user input if hash not in hash_filter_keys: return make_response( 'Bad request, sha must be one of sha1, sha1_git, sha256', 400) - h = query.categorize_hash(sha) - if h == {}: - return make_response( - 'Bad request, %s is not of type %s' % (sha, hash), - 400) - - if hash == 'sha256' and not h.get(hash): - return make_response( - 'Bad request, %s is not of type sha256' % (sha,), - 400) - - if hash != 'sha256' and not h.get('sha1') and not h.get('sha1_git'): - return make_response( - 'Bad request, %s is not of type sha1 or sha1_git' % (sha,), - 400) - - message = service.lookup_hash_origin(h) + q = "%s:%s" % (hash, sha) + found = service.lookup_hash(q) + if not found: + message = "Hash %s was not found." % hash + else: + origin = service.lookup_hash_origin(q) + message = _origin_seen(hash, origin) return render_template('content.html', hash=hash, sha=sha, message=message) @app.route('/browse/release/') def release(sha1_git): """Show release's information. Args: sha1_git: sha1_git for this particular release Returns: Release's information """ return 'Release information at %s' % sha1_git @app.route('/browse/person/') def person(id): """Show Person's information at id. Args: id: person's unique identifier Returns: Person's information """ return 'Person information at %s' % id @app.route('/browse/origin/') def origin(id): """Show origin's information at id. Args: id: origin's unique identifier Returns: Origin's information """ return 'Origin information at %s' % id @app.route('/browse/project/') def project(id): """Show project's information at id. Args: id: project's unique identifier Returns: Project's information """ return 'Project information at %s' % id @app.route('/browse/organization/') def organization(id): """Show organization's information at id. Args: id: organization's unique identifier Returns: Organization's information """ return 'Organization information at %s' % id @app.route('/browse/directory//' '+|/' '|/') def directory_at_origin(timestamp, origin_type, origin_url, branch, path): """Show directory information at timestamp, origin-type, origin-url, branch and path. Those parameters are separated by the `|` terminator. Args: timestamp: the timestamp to look for. can be latest or some iso8601 date format. (TODO: decide the time matching policy.) origin_type: origin's type origin_url: origin's url (can contain `/`) branch: branch name which can contain `/` path: path to directory or file Returns: Directory information at the given parameters. """ return 'Directory at (%s, %s, %s, %s, %s)' % (timestamp, origin_type, origin_url, branch, path) @app.route('/browse/revision//' '+|/') def revision_at_origin_and_branch(timestamp, origin_type, origin_url, branch): """Show revision information at timestamp, origin, and branch. Those parameters are separated by the `|` terminator. Args: timestamp: the timestamp to look for. can be latest or some iso8601 date format. (TODO: decide the time matching policy.) origin_type: origin's type origin_url: origin's url (can contain `/`) branch: branch name which can contain / Returns: Revision information at the given parameters. """ return 'Revision at (ts=%s, type=%s, url=%s, branch=%s)' % (timestamp, origin_type, origin_url, branch) @app.route('/browse/revision//' '+|') def revision_at_origin(timestamp, origin_type, origin_url): """Show revision information at timestamp, origin, and branch. Those parameters are separated by the `|` terminator. Args: timestamp: the timestamp to look for. can be latest or iso8601 date format. (TODO: decide the time matching policy.) origin_type: origin's type origin_url: origin's url (can contain `/`) Returns: Revision information at the given parameters. """ return 'Revision at (timestamp=%s, type=%s, url=%s)' % (timestamp, origin_type, origin_url) @app.route('/api/1/stat/counters') @jsonp def api_stats(): """Return statistics as a JSON object""" return jsonify(service.stat_counters()) @app.route('/api/1/search//') @jsonp def api_search(q): """Return search results as a JSON object""" return jsonify({'query': q, 'found': service.lookup_hash(q)}) +@app.route('/api/1/browse//') +@jsonp +def api_browse(q): + """Return search results as a JSON object""" + return jsonify({'query': q, + 'origin': service.lookup_hash_origin(q)}) + + def run(conf): """Run the api's server. Args: conf is a dictionary of keywords: - 'db_url' the db url's access (through psycopg2 format) - 'content_storage_dir' revisions/directories/contents storage on disk - 'host' to override the default 127.0.0.1 to open or not the server to the world - 'port' to override the default of 5000 (from the underlying layer: flask) - 'debug' activate the verbose logs - 'secret_key' the flask secret key Returns: Never Raises: ? """ print("""SWH Web UI available at http://%s:%s/ debug: %s""" % (conf['host'], conf.get('port', None), conf['debug'])) app.secret_key = conf['secret_key'] app.config.update({'conf': conf}) app.run(host=conf['host'], port=conf.get('port', None), debug=conf['debug']) diff --git a/swh/web/ui/converters.py b/swh/web/ui/converters.py new file mode 100644 index 00000000..abf1d964 --- /dev/null +++ b/swh/web/ui/converters.py @@ -0,0 +1,22 @@ +# Copyright (C) 2015 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 swh.core import hashutil + + +def from_origin(origin): + """Convert from an swh origin to an origin dictionary. + + """ + new_origin = {} + for key, value in origin.items(): + if key == 'revision': + new_origin[key] = hashutil.hash_to_hex(value) + elif key == 'path': + new_origin[key] = value.decode('utf-8') + else: + new_origin[key] = value + + return new_origin diff --git a/swh/web/ui/service.py b/swh/web/ui/service.py index 9ae49c75..0a4bccab 100644 --- a/swh/web/ui/service.py +++ b/swh/web/ui/service.py @@ -1,77 +1,43 @@ # Copyright (C) 2015 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 swh.web.ui import main -from swh.web.ui import query +from swh.web.ui import converters, main, query def lookup_hash(q): """Checks if the storage contains a given content checksum Args: query string Returns: True or False, according to whether the checksum is present or not """ (algo, hash) = query.parse_hash(q) return main.storage().content_exist({algo: hash}) -def _origin_seen(hash, data): - """Given an origin, compute a message string with the right information. +def lookup_hash_origin(q): + """Return information about the checksum contained in the query q. - Args: - origin: a dictionary with keys: - - origin: a dictionary with type and url keys - - occurrence: a dictionary with a validity range - - Returns: - message as a string - - """ - if data is None: - return 'Content with hash %s is unknown as of now.' % hash - - origin_type = data['origin_type'] - origin_url = data['origin_url'] - revision = data['revision'] - branch = data['branch'] - path = data['path'] - - return """The content with hash %s has been seen on origin with type '%s' -at url '%s'. The revision was identified at '%s' on branch '%s'. -The file's path referenced was '%s'.""" % (hash, - origin_type, - origin_url, - revision, - branch, - path) - - -def lookup_hash_origin(hash): - """Given a hash, return the origin of such content if any is found. - - Args: - hash: key/value dictionary + Args: query string Returns: - The origin for such hash if it's found. + True or False, according to whether the checksum is present or not - Raises: - OSError (no route to host), etc... Network issues in general """ - data = main.storage().content_find_occurrence(hash) - return _origin_seen(hash, data) + algo, h = query.parse_hash(q) + origin = main.storage().content_find_occurrence({algo: h}) + return converters.from_origin(origin) def stat_counters(): """Return the stat counters for Software Heritage Returns: A dict mapping textual labels to integer values. """ return main.storage().stat_counters() diff --git a/swh/web/ui/tests/test_converters.py b/swh/web/ui/tests/test_converters.py new file mode 100644 index 00000000..91773567 --- /dev/null +++ b/swh/web/ui/tests/test_converters.py @@ -0,0 +1,38 @@ +# Copyright (C) 2015 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 unittest + +from nose.tools import istest + +from swh.web.ui import converters + + +class ConvertersTestCase(unittest.TestCase): + + @istest + def from_origin(self): + # given + origin_input = { + 'origin_type': 'ftp', + 'origin_url': 'rsync://ftp.gnu.org/gnu/octave', + 'branch': 'octave-3.4.0.tar.gz', + 'revision': b'\xb0L\xaf\x10\xe9SQ`\xd9\x0e\x87KE\xaaBm\xe7b\xf1\x9f', # noqa + 'path': b'octave-3.4.0/doc/interpreter/octave.html/doc_002dS_005fISREG.html' # noqa + } + + expected_origin = { + 'origin_type': 'ftp', + 'origin_url': 'rsync://ftp.gnu.org/gnu/octave', + 'branch': 'octave-3.4.0.tar.gz', + 'revision': 'b04caf10e9535160d90e874b45aa426de762f19f', + 'path': 'octave-3.4.0/doc/interpreter/octave.html/doc_002dS_005fISREG.html' # noqa + } + + # when + actual_origin = converters.from_origin(origin_input) + + # then + self.assertEqual(actual_origin, expected_origin)