diff --git a/requirements-swh.txt b/requirements-swh.txt
index 2fefb9d7..139763a0 100644
--- a/requirements-swh.txt
+++ b/requirements-swh.txt
@@ -1,6 +1,7 @@
 swh.model >= 0.0.32
 swh.storage >= 0.0.145
 swh.vault >= 0.0.23
 swh.indexer >= 0.0.120
 swh.scheduler >= 0.0.31
 swh.core >= 0.0.81
+swh.search >= 0.0.3
diff --git a/swh/web/common/service.py b/swh/web/common/service.py
index b48fc196..5bf22366 100644
--- a/swh/web/common/service.py
+++ b/swh/web/common/service.py
@@ -1,1164 +1,1173 @@
 # Copyright (C) 2015-2019  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 itertools
 import os
 import re
 
 from collections import defaultdict
 from typing import Any, Dict
 
 from swh.model import hashutil
 
 from swh.storage.algos import diff, revisions_walker
 
 from swh.model.identifiers import (
     CONTENT, DIRECTORY, RELEASE, REVISION, SNAPSHOT
 )
 from swh.web.common import converters
 from swh.web.common import query
 from swh.web.common.exc import BadInputExc, NotFoundExc
 from swh.web.common.origin_visits import get_origin_visit
 from swh.web import config
 
+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(l):
     """Returns the first element in the provided list or None
     if it is empty or None"""
     return next(iter(l 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):
     """Checks if the storage contains a given content checksum
 
     Args: 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}))
     return {'found': converters.from_content(found),
             'algo': algo}
 
 
 def search_hash(q):
     """Checks if the storage contains a given content checksum
 
     Args: 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):
     """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)
 
 
 def lookup_content_language(q):
     """Return language information from a specified content.
 
     Args:
         q: query string of the form <hash_algo:hash>
 
     Yields:
         language information (dict) list if the content is found.
 
     """
     sha1 = _lookup_content_sha1(q)
     if not sha1:
         return None
     lang = _first_element(list(idx_storage.content_language_get([sha1])))
     if not lang:
         return None
     return converters.from_swh(lang, hashess={'id'})
 
 
 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
     lic = _first_element(idx_storage.content_fossology_license_get([sha1]))
 
     if not lic:
         return None
     return converters.from_swh({'id': sha1, 'facts': lic[sha1]},
                                hashess={'id'})
 
 
 def lookup_origin(origin):
     """Return information about the origin matching dict origin.
 
     Args:
         origin: origin's dict with 'url' key
 
     Returns:
         origin information as dict.
 
     """
     origin_info = storage.origin_get(origin)
     if not origin_info:
         msg = 'Origin with url %s not found!' % origin['url']
         raise NotFoundExc(msg)
     return converters.from_origin(origin_info)
 
 
 def lookup_origins(origin_from=1, origin_count=100):
     """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
 
     Yields:
         origins information as dicts
     """
     origins = storage.origin_get_range(origin_from, origin_count)
     return map(converters.from_origin, origins)
 
 
 def search_origin(url_pattern, limit=50, with_visit=False, page_token=None):
     """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.
 
     """
-    offset = int(page_token) if page_token else 0
-    regexp = True
-    search_words = [re.escape(word) for word in url_pattern.split()]
-    if len(search_words) >= 7:
-        url_pattern = '.*'.join(search_words)
+    if search:
+        results = search.origin_search(url_pattern=url_pattern, count=limit,
+                                       page_token=page_token,
+                                       with_visit=with_visit)
+        origins = list(map(converters.from_origin, results['results']))
+        return (origins, results['next_page_token'])
     else:
-        pattern_parts = []
-        for permut in itertools.permutations(search_words):
-            pattern_parts.append('.*'.join(permut))
-        url_pattern = '|'.join(pattern_parts)
-
-    origins = storage.origin_search(url_pattern, offset, limit, regexp,
-                                    with_visit)
-    origins = list(map(converters.from_origin, origins))
-    if len(origins) >= limit:
-        page_token = str(offset + len(origins))
-    else:
-        page_token = None
-    return (origins, page_token)
+        # Fallback to swh-storage if swh-search is not configured
+        offset = int(page_token) if page_token else 0
+        regexp = True
+        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)
+
+        origins = storage.origin_search(url_pattern, offset, limit, regexp,
+                                        with_visit)
+        origins = list(map(converters.from_origin, origins))
+        if len(origins) >= limit:
+            page_token = str(offset + len(origins))
+        else:
+            page_token = None
+        return (origins, page_token)
 
 
 def search_origin_metadata(fulltext, limit=50):
     """Search for origins whose metadata match a provided string pattern.
 
     Args:
         fulltext: the string pattern to search for in origin metadata
         offset: number of found origins to skip before returning results
         limit: the maximum number of found origins to return
 
     Returns:
         list of origin metadata as dict.
 
     """
     matches = idx_storage.origin_intrinsic_metadata_search_fulltext(
         conjunction=[fulltext], limit=limit)
     results = []
 
     for match in matches:
         match['from_revision'] = hashutil.hash_to_hex(match['from_revision'])
 
         origin = storage.origin_get({'url': match['id']})
         del match['id']
 
         result = converters.from_origin(origin)
         if result:
             result['metadata'] = match
             results.append(result)
 
     return results
 
 
 def lookup_origin_intrinsic_metadata(origin_dict):
     """Return intrinsic metadata for origin whose origin matches given
     origin.
 
     Args:
         origin_dict: origin's dict with keys ('type' AND 'url')
 
     Returns:
         origin metadata.
 
     """
     origin_info = storage.origin_get(origin_dict)
     if not origin_info:
         msg = 'Origin with url %s not found!' % origin_dict['url']
         raise NotFoundExc(msg)
 
     origins = [origin_info['url']]
     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'],  # HACK: sha1_git really
         'Only sha1_git is supported.')
     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):
     """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('Release with sha1_git %s not found.'
                           % release_sha1_git)
     return converters.from_release(release)
 
 
 def lookup_release_multiple(sha1_git_list):
     """Return information about the revisions identified with
     their sha1_git identifiers.
 
     Args:
         sha1_git_list: A list of revision sha1_git identifiers
 
     Returns:
         Release 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) or []
     return (converters.from_release(r) for r in releases)
 
 
 def lookup_revision(rev_sha1_git):
     """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 = _first_element(storage.revision_get([sha1_git_bin]))
     if not revision:
         raise NotFoundExc('Revision with sha1_git %s not found.'
                           % rev_sha1_git)
     return converters.from_revision(revision)
 
 
 def lookup_revision_multiple(sha1_git_list):
     """Return information about the revisions identified with
     their sha1_git identifiers.
 
     Args:
         sha1_git_list: A list of revision sha1_git identifiers
 
     Returns:
         Generator of revisions 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)
     revisions = storage.revision_get(sha1_bin_list) or []
     return (converters.from_revision(r) for r in revisions)
 
 
 def lookup_revision_message(rev_sha1_git):
     """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 = _first_element(storage.revision_get([sha1_git_bin]))
     if not revision:
         raise NotFoundExc('Revision with sha1_git %s not found.'
                           % rev_sha1_git)
     if 'message' not in revision:
         raise NotFoundExc('No message for revision with sha1_git %s.'
                           % rev_sha1_git)
     res = {'message': revision['message']}
     return res
 
 
 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.')
 
     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.
         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:
         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 = _first_element(storage.revision_get([rev_root_id_bin]))
 
     return (converters.from_revision(rev_root),
             lookup_revision_with_context(rev_root, sha1_git, limit))
 
 
 def lookup_revision_with_context(sha1_git_root, sha1_git, limit=100):
     """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 = _first_element(storage.revision_get([sha1_git_bin]))
     if not revision:
         raise NotFoundExc('Revision %s not found' % sha1_git)
 
     if isinstance(sha1_git_root, str):
         sha1_git_root_bin = _to_sha1_bin(sha1_git_root)
 
         revision_root = _first_element(storage.revision_get([sha1_git_root_bin])) # noqa
         if not revision_root:
             raise NotFoundExc('Revision root %s not found' % sha1_git_root)
     else:
         sha1_git_root_bin = sha1_git_root['id']
 
     revision_log = storage.revision_log([sha1_git_root_bin], limit)
 
     parents = {}
     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('Revision %s is not an ancestor of %s' %
                           (sha1_git, sha1_git_root))
 
     revision['children'] = children[revision['id']]
 
     return converters.from_revision(revision)
 
 
 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 = _first_element(storage.revision_get([sha1_git_bin]))
     if not revision:
         raise NotFoundExc('Revision %s not found' % sha1_git)
     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('Content not found for revision %s'
                               % sha1_git)
         if with_data:
             c = _first_element(storage.content_get([content['sha1']]))
             content['data'] = c['data']
         return {'type': 'file',
                 'path': '.' if not dir_path else dir_path,
                 'revision': sha1_git,
                 'content': converters.from_content(content)}
     elif entity['type'] == 'rev':  # revision
         revision = next(storage.revision_get([entity['target']]))
         return {'type': 'rev',
                 'path': '.' if not dir_path else dir_path,
                 'revision': sha1_git,
                 'content': converters.from_revision(revision)}
     else:
         raise NotImplementedError('Entity of type %s not implemented.'
                                   % entity['type'])
 
 
 def lookup_content(q):
     """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:
         raise NotFoundExc('Content with %s checksum equals to %s not found!' %
                           (algo, hashutil.hash_to_hex(hash)))
     return converters.from_content(c)
 
 
 def lookup_content_raw(q):
     """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 = _first_element(storage.content_get([content_sha1_bytes]))
     if not content:
         algo, hash = query.parse_hash(q)
         raise NotFoundExc('Bytes of content with %s checksum equals to %s '
                           'are not available!' %
                           (algo, hashutil.hash_to_hex(hash)))
     return converters.from_content(content)
 
 
 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, last_visit=None, limit=10):
     """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:
        Dictionaries of origin_visit for that origin
 
     """
     limit = min(limit, MAX_LIMIT)
     for visit in storage.origin_visit_get(
             origin_url, last_visit=last_visit, limit=limit):
         visit['origin'] = origin_url
         yield visit
 
 
 def lookup_origin_visits(origin, last_visit=None, per_page=10):
     """Yields the origin origins' visits.
 
     Args:
         origin: origin to list visits for
 
     Yields:
        Dictionaries of origin_visit for that origin
 
     """
     visits = _lookup_origin_visits(origin, last_visit=last_visit,
                                    limit=per_page)
     for visit in visits:
         yield converters.from_origin_visit(visit)
 
 
 def lookup_origin_visit_latest(origin_url, require_snapshot):
     """Return the origin's latest visit
 
     Args:
         origin_url (str): origin to list visits for
         require_snapshot (bool): filter out origins without a snapshot
 
     Returns:
        dict: The origin_visit concerned
 
     """
     visit = storage.origin_visit_get_latest(
         origin_url, require_snapshot=require_snapshot)
     return converters.from_origin_visit(visit)
 
 
 def lookup_origin_visit(origin_url, visit_id):
     """Return information about visit visit_id with origin origin.
 
     Args:
         origin (str): 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)
     if not visit:
         raise NotFoundExc('Origin %s or its visit '
                           'with id %s not found!' % (origin_url, visit_id))
     visit['origin'] = origin_url
     return converters.from_origin_visit(visit)
 
 
 def lookup_snapshot_sizes(snapshot_id):
     """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():
             if lookup_snapshot(snapshot_id,
                                branches_from=alias['target'],
                                branches_count=1,
                                target_types=['revision']):
                 snapshot_sizes['revision'] += 1
             else:
                 snapshot_sizes['release'] += 1
         del snapshot_sizes['alias']
     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)
     snapshot = storage.snapshot_get_branches(snapshot_id_bin,
                                              branches_from.encode(),
                                              branches_count, target_types)
     if not snapshot:
         raise NotFoundExc('Snapshot with id %s not found!' % snapshot_id)
     return converters.from_snapshot(snapshot)
 
 
 def lookup_latest_origin_snapshot(origin, allowed_statuses=None):
     """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.
     """
     snapshot = storage.snapshot_get_latest(origin, allowed_statuses)
     return converters.from_snapshot(snapshot)
 
 
 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_cook(obj_type, obj_id, email=None):
     """Cook a vault bundle.
     """
     return vault.cook(obj_type, obj_id, email=email)
 
 
 def vault_fetch(obj_type, obj_id):
     """Fetch a vault bundle.
     """
     return vault.fetch(obj_type, obj_id)
 
 
 def vault_progress(obj_type, obj_id):
     """Get the current progress of a vault bundle.
     """
     return 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:
         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}.'))
diff --git a/swh/web/config.py b/swh/web/config.py
index ab251ad0..2a5cb790 100644
--- a/swh/web/config.py
+++ b/swh/web/config.py
@@ -1,165 +1,184 @@
 # 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 os
 
 from typing import Any, Dict
 
 from swh.core import config
 from swh.indexer.storage import get_indexer_storage
 from swh.scheduler import get_scheduler
+from swh.search import get_search
 from swh.storage import get_storage
 from swh.vault import get_vault
 from swh.web import settings
 
 SETTINGS_DIR = os.path.dirname(settings.__file__)
 
 DEFAULT_CONFIG = {
     'allowed_hosts': ('list', []),
+    'search': ('dict', {
+        'cls': 'remote',
+        'args': {
+            'url': 'http://127.0.0.1:5010/',
+            'timeout': 10,
+        },
+    }),
     'storage': ('dict', {
         'cls': 'remote',
         'url': 'http://127.0.0.1:5002/',
         'timeout': 10,
     }),
     'indexer_storage': ('dict', {
         'cls': 'remote',
         'args': {
             'url': 'http://127.0.0.1:5007/',
             'timeout': 1,
         }
     }),
     'log_dir': ('string', '/tmp/swh/log'),
     'debug': ('bool', False),
     'serve_assets': ('bool', False),
     'host': ('string', '127.0.0.1'),
     'port': ('int', 5004),
     'secret_key': ('string', 'development key'),
     # do not display code highlighting for content > 1MB
     'content_display_max_size': ('int', 5 * 1024 * 1024),
     'snapshot_content_max_size': ('int', 1000),
     'throttling': ('dict', {
         'cache_uri': None,  # production: memcached as cache (127.0.0.1:11211)
                             # development: in-memory cache so None
         'scopes': {
             'swh_api': {
                 'limiter_rate': {
                     'default': '120/h'
                 },
                 'exempted_networks': ['127.0.0.0/8']
             },
             'swh_api_origin_search': {
                 'limiter_rate': {
                     'default': '10/m'
                 },
                 'exempted_networks': ['127.0.0.0/8']
             },
             'swh_vault_cooking': {
                 'limiter_rate': {
                     'default': '120/h',
                     'GET': '60/m'
                 },
                 'exempted_networks': ['127.0.0.0/8']
             },
             'swh_save_origin': {
                 'limiter_rate': {
                     'default': '120/h',
                     'POST': '10/h'
                 },
                 'exempted_networks': ['127.0.0.0/8']
             },
             'swh_api_origin_visit_latest': {
                 'limiter_rate': {
                     'default': '700/m'
                 },
                 'exempted_networks': ['127.0.0.0/8'],
             },
         }
     }),
     'vault': ('dict', {
         'cls': 'remote',
         'args': {
             'url': 'http://127.0.0.1:5005/',
         }
     }),
     'scheduler': ('dict', {
         'cls': 'remote',
         'args': {
             'url': 'http://127.0.0.1:5008/'
         }
     }),
     'development_db': ('string', os.path.join(SETTINGS_DIR, 'db.sqlite3')),
     'test_db': ('string', os.path.join(SETTINGS_DIR, 'testdb.sqlite3')),
     'production_db': ('string', '/var/lib/swh/web.sqlite3'),
     'deposit': ('dict', {
         'private_api_url': 'https://deposit.softwareheritage.org/1/private/',
         'private_api_user': 'swhworker',
         'private_api_password': ''
     }),
     'coverage_count_origins': ('bool', False),
     'e2e_tests_mode': ('bool', False),
     'es_workers_index_url': ('string', ''),
     'history_counters_url': ('string', 'https://stats.export.softwareheritage.org/history_counters.json'), # noqa
     'client_config': ('dict', {}),
 }
 
 swhweb_config = {}  # type: Dict[str, Any]
 
 
 def get_config(config_file='web/web'):
     """Read the configuration file `config_file`.
 
        If an environment variable SWH_CONFIG_FILENAME is defined, this
        takes precedence over the config_file parameter.
 
        In any case, update the app with parameters (secret_key, conf)
        and return the parsed configuration as a dict.
 
        If no configuration file is provided, return a default
        configuration.
 
     """
 
     if not swhweb_config:
         config_filename = os.environ.get('SWH_CONFIG_FILENAME')
         if config_filename:
             config_file = config_filename
         cfg = config.load_named_config(config_file, DEFAULT_CONFIG)
         swhweb_config.update(cfg)
         config.prepare_folders(swhweb_config, 'log_dir')
+        if swhweb_config.get('search'):
+            swhweb_config['search'] = get_search(**swhweb_config['search'])
+        else:
+            swhweb_config['search'] = None
         swhweb_config['storage'] = get_storage(**swhweb_config['storage'])
         swhweb_config['vault'] = get_vault(**swhweb_config['vault'])
         swhweb_config['indexer_storage'] = \
             get_indexer_storage(**swhweb_config['indexer_storage'])
         swhweb_config['scheduler'] = get_scheduler(
             **swhweb_config['scheduler'])
     return swhweb_config
 
 
+def search():
+    """Return the current application's search.
+
+    """
+    return get_config()['search']
+
+
 def storage():
     """Return the current application's storage.
 
     """
     return get_config()['storage']
 
 
 def vault():
     """Return the current application's vault.
 
     """
     return get_config()['vault']
 
 
 def indexer_storage():
     """Return the current application's indexer storage.
 
     """
     return get_config()['indexer_storage']
 
 
 def scheduler():
     """Return the current application's scheduler.
 
     """
     return get_config()['scheduler']
diff --git a/swh/web/settings/tests.py b/swh/web/settings/tests.py
index 87c8c7f1..10bb42c7 100644
--- a/swh/web/settings/tests.py
+++ b/swh/web/settings/tests.py
@@ -1,109 +1,110 @@
 # 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
 
 """
 Django tests settings for swh-web.
 """
 
 import sys
 
 from swh.web.config import get_config
 
 scope1_limiter_rate = 3
 scope1_limiter_rate_post = 1
 scope2_limiter_rate = 5
 scope2_limiter_rate_post = 2
 scope3_limiter_rate = 1
 scope3_limiter_rate_post = 1
 save_origin_rate_post = 10
 
 swh_web_config = get_config()
 
 swh_web_config.update({
     'debug': False,
     'secret_key': 'test',
     'history_counters_url': '',
     'throttling': {
         'cache_uri': None,
         'scopes': {
             'swh_api': {
                 'limiter_rate': {
                     'default': '60/min'
                 },
                 'exempted_networks': ['127.0.0.0/8']
             },
             'swh_api_origin_search': {
                 'limiter_rate': {
                     'default': '100/min'
                 },
                 'exempted_networks': ['127.0.0.0/8']
             },
             'swh_api_origin_visit_latest': {
                 'limiter_rate': {
                     'default': '6000/min'
                 },
                 'exempted_networks': ['127.0.0.0/8']
             },
             'swh_vault_cooking': {
                 'limiter_rate': {
                     'default': '120/h',
                     'GET': '60/m'
                 },
                 'exempted_networks': ['127.0.0.0/8']
             },
             'swh_save_origin': {
                 'limiter_rate': {
                     'default': '120/h',
                     'POST': '%s/h' % save_origin_rate_post,
                 }
             },
             'scope1': {
                 'limiter_rate': {
                     'default': '%s/min' % scope1_limiter_rate,
                     'POST': '%s/min' % scope1_limiter_rate_post,
                 }
             },
             'scope2': {
                 'limiter_rate': {
                     'default': '%s/min' % scope2_limiter_rate,
                     'POST': '%s/min' % scope2_limiter_rate_post
                 }
             },
             'scope3': {
                 'limiter_rate': {
                     'default': '%s/min' % scope3_limiter_rate,
                     'POST': '%s/min' % scope3_limiter_rate_post
                 },
                 'exempted_networks': ['127.0.0.0/8']
             }
         }
     }
 })
 
 
 from .common import * # noqa
 from .common import ALLOWED_HOSTS, LOGGING # noqa
 
 DATABASES = {
     'default': {
         'ENGINE': 'django.db.backends.sqlite3',
         'NAME': swh_web_config['test_db'],
     }
 }
 
 # when not running unit tests, make the webapp fetch data from memory storages
 if 'pytest' not in sys.argv[0]:
     swh_web_config.update({
         'debug': True,
         'e2e_tests_mode': True
     })
     from swh.web.tests.data import get_tests_data, override_storages # noqa
     test_data = get_tests_data()
-    override_storages(test_data['storage'], test_data['idx_storage'])
+    override_storages(test_data['storage'], test_data['idx_storage'],
+                      test_data['search'])
 else:
     ALLOWED_HOSTS += ['testserver']
 
     # Silent DEBUG output when running unit tests
     LOGGING['handlers']['console']['level'] = 'INFO'  # type: ignore
diff --git a/swh/web/tests/api/views/test_origin.py b/swh/web/tests/api/views/test_origin.py
index 57f823f1..c57885ca 100644
--- a/swh/web/tests/api/views/test_origin.py
+++ b/swh/web/tests/api/views/test_origin.py
@@ -1,641 +1,669 @@
 # Copyright (C) 2015-2019  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
 import pytest
 from requests.utils import parse_header_links
 
 from swh.storage.exc import StorageDBError, StorageAPIError
 
 from swh.web.api.utils import enrich_origin_visit, enrich_origin
 from swh.web.common.exc import BadInputExc
 from swh.web.common.utils import reverse
 from swh.web.common.origin_visits import get_origin_visits
 from swh.web.tests.strategies import (
     origin, new_origin, visit_dates, new_snapshots
 )
 
 
 def _scroll_results(api_client, url):
     """Iterates through pages of results, and returns them all."""
     results = []
 
     while True:
         rv = api_client.get(url)
         assert rv.status_code == 200, rv.data
         assert rv['Content-Type'] == 'application/json'
 
         results.extend(rv.data)
 
         if 'Link' in rv:
             for link in parse_header_links(rv['Link']):
                 if link['rel'] == 'next':
                     # Found link to next page of results
                     url = link['url']
                     break
             else:
                 # No link with 'rel=next'
                 break
         else:
             # No Link header
             break
 
     return results
 
 
 def test_api_lookup_origin_visits_raise_error(api_client, mocker):
     mock_get_origin_visits = mocker.patch(
         'swh.web.api.views.origin.get_origin_visits')
     err_msg = 'voluntary error to check the bad request middleware.'
 
     mock_get_origin_visits.side_effect = BadInputExc(err_msg)
 
     url = reverse('api-1-origin-visits', url_args={'origin_url': 'http://foo'})
     rv = api_client.get(url)
 
     assert rv.status_code == 400, rv.data
     assert rv['Content-Type'] == 'application/json'
     assert rv.data == {
         'exception': 'BadInputExc',
         'reason': err_msg
     }
 
 
 def test_api_lookup_origin_visits_raise_swh_storage_error_db(api_client,
                                                              mocker):
     mock_get_origin_visits = mocker.patch(
         'swh.web.api.views.origin.get_origin_visits')
     err_msg = 'Storage exploded! Will be back online shortly!'
 
     mock_get_origin_visits.side_effect = StorageDBError(err_msg)
 
     url = reverse('api-1-origin-visits', url_args={'origin_url': 'http://foo'})
     rv = api_client.get(url)
 
     assert rv.status_code == 503, rv.data
     assert rv['Content-Type'] == 'application/json'
     assert rv.data == {
         'exception': 'StorageDBError',
         'reason':
         'An unexpected error occurred in the backend: %s' % err_msg
     }
 
 
 def test_api_lookup_origin_visits_raise_swh_storage_error_api(api_client,
                                                               mocker):
     mock_get_origin_visits = mocker.patch(
         'swh.web.api.views.origin.get_origin_visits')
     err_msg = 'Storage API dropped dead! Will resurrect asap!'
 
     mock_get_origin_visits.side_effect = StorageAPIError(err_msg)
 
     url = reverse(
         'api-1-origin-visits', url_args={'origin_url': 'http://foo'})
     rv = api_client.get(url)
 
     assert rv.status_code == 503, rv.data
     assert rv['Content-Type'] == 'application/json'
     assert rv.data == {
         'exception': 'StorageAPIError',
         'reason':
         'An unexpected error occurred in the api backend: %s' % err_msg
     }
 
 
 @given(new_origin(), visit_dates(3), new_snapshots(3))
 def test_api_lookup_origin_visits(api_client, archive_data, new_origin,
                                   visit_dates, new_snapshots):
     archive_data.origin_add_one(new_origin)
     for i, visit_date in enumerate(visit_dates):
         origin_visit = archive_data.origin_visit_add(
             new_origin['url'], visit_date, type='git')
         archive_data.snapshot_add([new_snapshots[i]])
         archive_data.origin_visit_update(
             new_origin['url'], origin_visit['visit'],
             snapshot=new_snapshots[i]['id'])
 
     all_visits = list(reversed(get_origin_visits(new_origin)))
 
     for last_visit, expected_visits in (
             (None, all_visits[:2]),
             (all_visits[1]['visit'], all_visits[2:])):
 
         url = reverse('api-1-origin-visits',
                       url_args={'origin_url': new_origin['url']},
                       query_params={'per_page': 2,
                                     'last_visit': last_visit})
 
         rv = api_client.get(url)
 
         assert rv.status_code == 200, rv.data
         assert rv['Content-Type'] == 'application/json'
 
         for i in range(len(expected_visits)):
             expected_visits[i] = enrich_origin_visit(
                 expected_visits[i], with_origin_link=False,
                 with_origin_visit_link=True, request=rv.wsgi_request)
 
         assert rv.data == expected_visits
 
 
 @given(new_origin(), visit_dates(3), new_snapshots(3))
 def test_api_lookup_origin_visits_by_id(api_client, archive_data, new_origin,
                                         visit_dates, new_snapshots):
     archive_data.origin_add_one(new_origin)
     for i, visit_date in enumerate(visit_dates):
         origin_visit = archive_data.origin_visit_add(
             new_origin['url'], visit_date, type='git')
         archive_data.snapshot_add([new_snapshots[i]])
         archive_data.origin_visit_update(
             new_origin['url'], origin_visit['visit'],
             snapshot=new_snapshots[i]['id'])
 
     all_visits = list(reversed(get_origin_visits(new_origin)))
 
     for last_visit, expected_visits in (
             (None, all_visits[:2]),
             (all_visits[1]['visit'], all_visits[2:4])):
 
         url = reverse('api-1-origin-visits',
                       url_args={'origin_url': new_origin['url']},
                       query_params={'per_page': 2,
                                     'last_visit': last_visit})
 
         rv = api_client.get(url)
 
         assert rv.status_code == 200, rv.data
         assert rv['Content-Type'] == 'application/json'
 
         for i in range(len(expected_visits)):
             expected_visits[i] = enrich_origin_visit(
                 expected_visits[i], with_origin_link=False,
                 with_origin_visit_link=True, request=rv.wsgi_request)
 
         assert rv.data == expected_visits
 
 
 @given(new_origin(), visit_dates(3), new_snapshots(3))
 def test_api_lookup_origin_visit(api_client, archive_data, new_origin,
                                  visit_dates, new_snapshots):
     archive_data.origin_add_one(new_origin)
     for i, visit_date in enumerate(visit_dates):
         origin_visit = archive_data.origin_visit_add(
             new_origin['url'], visit_date, type='git')
         visit_id = origin_visit['visit']
         archive_data.snapshot_add([new_snapshots[i]])
         archive_data.origin_visit_update(
             new_origin['url'], origin_visit['visit'],
             snapshot=new_snapshots[i]['id'])
         url = reverse('api-1-origin-visit',
                       url_args={'origin_url': new_origin['url'],
                                 'visit_id': visit_id})
 
         rv = api_client.get(url)
         assert rv.status_code == 200, rv.data
         assert rv['Content-Type'] == 'application/json'
 
         expected_visit = archive_data.origin_visit_get_by(
             new_origin['url'], visit_id)
 
         expected_visit = enrich_origin_visit(
             expected_visit, with_origin_link=True,
             with_origin_visit_link=False, request=rv.wsgi_request)
 
         assert rv.data == expected_visit
 
 
 @given(new_origin())
 def test_api_lookup_origin_visit_latest_no_visit(api_client, archive_data,
                                                  new_origin):
     archive_data.origin_add_one(new_origin)
 
     url = reverse('api-1-origin-visit-latest',
                   url_args={'origin_url': new_origin['url']})
 
     rv = api_client.get(url)
     assert rv.status_code == 404, rv.data
     assert rv.data == {
         'exception': 'NotFoundExc',
         'reason': 'No visit for origin %s found' % new_origin['url']
     }
 
 
 @given(new_origin(), visit_dates(2), new_snapshots(1))
 def test_api_lookup_origin_visit_latest(api_client, archive_data, new_origin,
                                         visit_dates, new_snapshots):
     archive_data.origin_add_one(new_origin)
     visit_dates.sort()
     visit_ids = []
     for i, visit_date in enumerate(visit_dates):
         origin_visit = archive_data.origin_visit_add(
             new_origin['url'], visit_date, type='git')
         visit_ids.append(origin_visit['visit'])
 
     archive_data.snapshot_add([new_snapshots[0]])
     archive_data.origin_visit_update(
         new_origin['url'], visit_ids[0],
         snapshot=new_snapshots[0]['id'])
 
     url = reverse('api-1-origin-visit-latest',
                   url_args={'origin_url': new_origin['url']})
 
     rv = api_client.get(url)
 
     assert rv.status_code == 200, rv.data
     assert rv['Content-Type'] == 'application/json'
 
     expected_visit = archive_data.origin_visit_get_by(
         new_origin['url'], visit_ids[1])
 
     expected_visit = enrich_origin_visit(
             expected_visit, with_origin_link=True,
             with_origin_visit_link=False, request=rv.wsgi_request)
 
     assert rv.data == expected_visit
 
 
 @given(new_origin(), visit_dates(2), new_snapshots(1))
 def test_api_lookup_origin_visit_latest_with_snapshot(api_client, archive_data,
                                                       new_origin, visit_dates,
                                                       new_snapshots):
     archive_data.origin_add_one(new_origin)
     visit_dates.sort()
     visit_ids = []
     for i, visit_date in enumerate(visit_dates):
         origin_visit = archive_data.origin_visit_add(
             new_origin['url'], visit_date, type='git')
         visit_ids.append(origin_visit['visit'])
 
     archive_data.snapshot_add([new_snapshots[0]])
     archive_data.origin_visit_update(
         new_origin['url'], visit_ids[0],
         snapshot=new_snapshots[0]['id'])
 
     url = reverse('api-1-origin-visit-latest',
                   url_args={'origin_url': new_origin['url']},
                   query_params={'require_snapshot': True})
 
     rv = api_client.get(url)
 
     assert rv.status_code == 200, rv.data
     assert rv['Content-Type'] == 'application/json'
 
     expected_visit = archive_data.origin_visit_get_by(
         new_origin['url'], visit_ids[0])
 
     expected_visit = enrich_origin_visit(
             expected_visit, with_origin_link=True,
             with_origin_visit_link=False, request=rv.wsgi_request)
 
     assert rv.data == expected_visit
 
 
 @given(origin())
 def test_api_lookup_origin_visit_not_found(api_client, origin):
 
     all_visits = list(reversed(get_origin_visits(origin)))
 
     max_visit_id = max([v['visit'] for v in all_visits])
 
     url = reverse('api-1-origin-visit',
                   url_args={'origin_url': origin['url'],
                             'visit_id': max_visit_id + 1})
 
     rv = api_client.get(url)
 
     assert rv.status_code == 404, rv.data
     assert rv['Content-Type'] == 'application/json'
     assert rv.data == {
         'exception': 'NotFoundExc',
         'reason': 'Origin %s or its visit with id %s not found!' %
         (origin['url'], max_visit_id+1)
     }
 
 
 def test_api_origins(api_client, archive_data):
     origins = list(archive_data.origin_get_range(0, 10000))
     origin_urls = {origin['url'] for origin in origins}
 
     # Get only one
     url = reverse('api-1-origins',
                   query_params={'origin_count': 1})
     rv = api_client.get(url)
     assert rv.status_code == 200, rv.data
     assert rv['Content-Type'] == 'application/json'
     assert len(rv.data) == 1
     assert {origin['url'] for origin in rv.data} <= origin_urls
 
     # Get all
     url = reverse('api-1-origins',
                   query_params={'origin_count': len(origins)})
     rv = api_client.get(url)
     assert rv.status_code == 200, rv.data
     assert rv['Content-Type'] == 'application/json'
     assert len(rv.data) == len(origins)
     assert {origin['url'] for origin in rv.data} == origin_urls
 
     # Get "all + 10"
     url = reverse('api-1-origins',
                   query_params={'origin_count': len(origins)+10})
     rv = api_client.get(url)
     assert rv.status_code == 200, rv.data
     assert rv['Content-Type'] == 'application/json'
     assert len(rv.data) == len(origins)
     assert {origin['url'] for origin in rv.data} == origin_urls
 
 
 @pytest.mark.parametrize('origin_count', [1, 2, 10, 100])
 def test_api_origins_scroll(api_client, archive_data, origin_count):
     origins = list(archive_data.origin_get_range(0, 10000))
     origin_urls = {origin['url'] for origin in origins}
 
     url = reverse('api-1-origins',
                   query_params={'origin_count': origin_count})
 
     results = _scroll_results(api_client, url)
 
     assert len(results) == len(origins)
     assert {origin['url'] for origin in results} == origin_urls
 
 
 @given(origin())
 def test_api_origin_by_url(api_client, archive_data, origin):
     url = reverse('api-1-origin',
                   url_args={'origin_url': origin['url']})
     rv = api_client.get(url)
 
     expected_origin = archive_data.origin_get(origin)
 
     expected_origin = enrich_origin(expected_origin, rv.wsgi_request)
 
     assert rv.status_code == 200, rv.data
     assert rv['Content-Type'] == 'application/json'
     assert rv.data == expected_origin
 
 
 @given(new_origin())
 def test_api_origin_not_found(api_client, new_origin):
 
     url = reverse('api-1-origin',
                   url_args={'origin_url': new_origin['url']})
     rv = api_client.get(url)
 
     assert rv.status_code == 404, rv.data
     assert rv['Content-Type'] == 'application/json'
     assert rv.data == {
         'exception': 'NotFoundExc',
         'reason': 'Origin with url %s not found!' % new_origin['url']
     }
 
 
-def test_api_origin_search(api_client):
+@pytest.mark.parametrize('backend', ['swh-search', 'swh-storage'])
+def test_api_origin_search(api_client, mocker, backend):
+    if backend != 'swh-search':
+        # equivalent to not configuring search in the config
+        mocker.patch('swh.web.common.service.search', None)
+
     expected_origins = {
         'https://github.com/wcoder/highlightjs-line-numbers.js',
         'https://github.com/memononen/libtess2',
     }
 
     # Search for 'github.com', get only one
     url = reverse('api-1-origin-search',
                   url_args={'url_pattern': 'github.com'},
                   query_params={'limit': 1})
     rv = api_client.get(url)
     assert rv.status_code == 200, rv.data
     assert rv['Content-Type'] == 'application/json'
     assert len(rv.data) == 1
     assert {origin['url'] for origin in rv.data} <= expected_origins
 
     # Search for 'github.com', get all
     url = reverse('api-1-origin-search',
                   url_args={'url_pattern': 'github.com'},
                   query_params={'limit': 2})
     rv = api_client.get(url)
     assert rv.status_code == 200, rv.data
     assert rv['Content-Type'] == 'application/json'
     assert {origin['url'] for origin in rv.data} == expected_origins
 
     # Search for 'github.com', get more than available
     url = reverse('api-1-origin-search',
                   url_args={'url_pattern': 'github.com'},
                   query_params={'limit': 10})
     rv = api_client.get(url)
     assert rv.status_code == 200, rv.data
     assert rv['Content-Type'] == 'application/json'
     assert {origin['url'] for origin in rv.data} == expected_origins
 
 
-def test_api_origin_search_words(api_client):
+@pytest.mark.parametrize('backend', ['swh-search', 'swh-storage'])
+def test_api_origin_search_words(api_client, mocker, backend):
+    if backend != 'swh-search':
+        # equivalent to not configuring search in the config
+        mocker.patch('swh.web.common.service.search', None)
+
     expected_origins = {
         'https://github.com/wcoder/highlightjs-line-numbers.js',
         'https://github.com/memononen/libtess2',
     }
 
     url = reverse('api-1-origin-search',
                   url_args={'url_pattern': 'github com'},
                   query_params={'limit': 2})
     rv = api_client.get(url)
     assert rv.status_code == 200, rv.data
     assert rv['Content-Type'] == 'application/json'
     assert {origin['url'] for origin in rv.data} == expected_origins
 
     url = reverse('api-1-origin-search',
                   url_args={'url_pattern': 'com github'},
                   query_params={'limit': 2})
     rv = api_client.get(url)
     assert rv.status_code == 200, rv.data
     assert rv['Content-Type'] == 'application/json'
     assert {origin['url'] for origin in rv.data} == expected_origins
 
     url = reverse('api-1-origin-search',
                   url_args={'url_pattern': 'memononen libtess2'},
                   query_params={'limit': 2})
     rv = api_client.get(url)
     assert rv.status_code == 200, rv.data
     assert rv['Content-Type'] == 'application/json'
     assert len(rv.data) == 1
     assert {origin['url'] for origin in rv.data} \
         == {'https://github.com/memononen/libtess2'}
 
     url = reverse('api-1-origin-search',
                   url_args={'url_pattern': 'libtess2 memononen'},
                   query_params={'limit': 2})
     rv = api_client.get(url)
     assert rv.status_code == 200, rv.data
     assert rv['Content-Type'] == 'application/json'
     assert len(rv.data) == 1
     assert {origin['url'] for origin in rv.data} \
         == {'https://github.com/memononen/libtess2'}
 
 
+@pytest.mark.parametrize('backend', ['swh-search', 'swh-storage'])
 @pytest.mark.parametrize('limit', [1, 2, 3, 10])
-def test_api_origin_search_scroll(api_client, archive_data, limit):
+def test_api_origin_search_scroll(
+        api_client, archive_data, mocker, limit, backend):
+
+    if backend != 'swh-search':
+        # equivalent to not configuring search in the config
+        mocker.patch('swh.web.common.service.search', None)
+
     expected_origins = {
         'https://github.com/wcoder/highlightjs-line-numbers.js',
         'https://github.com/memononen/libtess2',
     }
 
     url = reverse('api-1-origin-search',
                   url_args={'url_pattern': 'github.com'},
                   query_params={'limit': limit})
 
     results = _scroll_results(api_client, url)
 
     assert {origin['url'] for origin in results} == expected_origins
 
 
-def test_api_origin_search_limit(api_client, archive_data):
-    archive_data.origin_add([
-        {'url': 'http://foobar/{}'.format(i)}
-        for i in range(2000)
-    ])
+@pytest.mark.parametrize('backend', ['swh-search', 'swh-storage'])
+def test_api_origin_search_limit(
+        api_client, archive_data, tests_data, mocker, backend):
+    if backend == 'swh-search':
+        tests_data['search'].origin_update([
+            {'url': 'http://foobar/{}'.format(i)}
+            for i in range(2000)
+        ])
+    else:
+        # equivalent to not configuring search in the config
+        mocker.patch('swh.web.common.service.search', None)
+
+        archive_data.origin_add([
+            {'url': 'http://foobar/{}'.format(i)}
+            for i in range(2000)
+        ])
 
     url = reverse('api-1-origin-search',
                   url_args={'url_pattern': 'foobar'},
                   query_params={'limit': 1050})
     rv = api_client.get(url)
     assert rv.status_code == 200, rv.data
     assert rv['Content-Type'] == 'application/json'
     assert len(rv.data) == 1000
 
 
 @given(origin())
 def test_api_origin_metadata_search(api_client, mocker, origin):
     mock_idx_storage = mocker.patch('swh.web.common.service.idx_storage')
     oimsft = mock_idx_storage.origin_intrinsic_metadata_search_fulltext
     oimsft.side_effect = lambda conjunction, limit: [{
         'from_revision': (
             b'p&\xb7\xc1\xa2\xafVR\x1e\x95\x1c\x01\xed '
             b'\xf2U\xfa\x05B8'),
         'metadata': {'author': 'Jane Doe'},
         'id': origin['url'],
         'tool': {
             'configuration': {
                 'context': ['NpmMapping', 'CodemetaMapping'],
                 'type': 'local'
             },
             'id': 3,
             'name': 'swh-metadata-detector',
             'version': '0.0.1'
         }
     }]
 
     url = reverse('api-1-origin-metadata-search',
                   query_params={'fulltext': 'Jane Doe'})
     rv = api_client.get(url)
 
     assert rv.status_code == 200, rv.content
     assert rv['Content-Type'] == 'application/json'
     expected_data = [{
         'url': origin['url'],
         'metadata': {
             'metadata': {'author': 'Jane Doe'},
             'from_revision': (
                 '7026b7c1a2af56521e951c01ed20f255fa054238'),
             'tool': {
                 'configuration': {
                     'context': ['NpmMapping', 'CodemetaMapping'],
                     'type': 'local'
                 },
                 'id': 3,
                 'name': 'swh-metadata-detector',
                 'version': '0.0.1',
             }
         }
     }]
 
     assert rv.data == expected_data
     oimsft.assert_called_with(conjunction=['Jane Doe'], limit=70)
 
 
 @given(origin())
 def test_api_origin_metadata_search_limit(api_client, mocker, origin):
     mock_idx_storage = mocker.patch('swh.web.common.service.idx_storage')
     oimsft = mock_idx_storage.origin_intrinsic_metadata_search_fulltext
 
     oimsft.side_effect = lambda conjunction, limit: [{
         'from_revision': (
             b'p&\xb7\xc1\xa2\xafVR\x1e\x95\x1c\x01\xed '
             b'\xf2U\xfa\x05B8'),
         'metadata': {'author': 'Jane Doe'},
         'id': origin['url'],
         'tool': {
             'configuration': {
                 'context': ['NpmMapping', 'CodemetaMapping'],
                 'type': 'local'
             },
             'id': 3,
             'name': 'swh-metadata-detector',
             'version': '0.0.1'
         }
     }]
 
     url = reverse('api-1-origin-metadata-search',
                   query_params={'fulltext': 'Jane Doe'})
     rv = api_client.get(url)
 
     assert rv.status_code == 200, rv.content
     assert rv['Content-Type'] == 'application/json'
     assert len(rv.data) == 1
     oimsft.assert_called_with(conjunction=['Jane Doe'], limit=70)
 
     url = reverse('api-1-origin-metadata-search',
                   query_params={'fulltext': 'Jane Doe',
                                 'limit': 10})
     rv = api_client.get(url)
 
     assert rv.status_code == 200, rv.content
     assert rv['Content-Type'] == 'application/json'
     assert len(rv.data) == 1
     oimsft.assert_called_with(conjunction=['Jane Doe'], limit=10)
 
     url = reverse('api-1-origin-metadata-search',
                   query_params={'fulltext': 'Jane Doe',
                                 'limit': 987})
     rv = api_client.get(url)
 
     assert rv.status_code == 200, rv.content
     assert rv['Content-Type'] == 'application/json'
     assert len(rv.data) == 1
     oimsft.assert_called_with(conjunction=['Jane Doe'], limit=100)
 
 
 @given(origin())
 def test_api_origin_intrinsic_metadata(api_client, mocker, origin):
     mock_idx_storage = mocker.patch('swh.web.common.service.idx_storage')
     oimg = mock_idx_storage.origin_intrinsic_metadata_get
     oimg.side_effect = lambda origin_urls: [{
         'from_revision': (
             b'p&\xb7\xc1\xa2\xafVR\x1e\x95\x1c\x01\xed '
             b'\xf2U\xfa\x05B8'),
         'metadata': {'author': 'Jane Doe'},
         'id': origin['url'],
         'tool': {
             'configuration': {
                 'context': ['NpmMapping', 'CodemetaMapping'],
                 'type': 'local'
             },
             'id': 3,
             'name': 'swh-metadata-detector',
             'version': '0.0.1'
         }
     }]
 
     url = reverse('api-origin-intrinsic-metadata',
                   url_args={'origin_url': origin['url']})
     rv = api_client.get(url)
 
     oimg.assert_called_once_with([origin['url']])
     assert rv.status_code == 200, rv.content
     assert rv['Content-Type'] == 'application/json'
     expected_data = {'author': 'Jane Doe'}
     assert rv.data == expected_data
 
 
 def test_api_origin_metadata_search_invalid(api_client, mocker):
     mock_idx_storage = mocker.patch('swh.web.common.service.idx_storage')
     url = reverse('api-1-origin-metadata-search')
     rv = api_client.get(url)
 
     assert rv.status_code == 400, rv.content
     mock_idx_storage.assert_not_called()
diff --git a/swh/web/tests/conftest.py b/swh/web/tests/conftest.py
index 65b40e51..1c01295d 100644
--- a/swh/web/tests/conftest.py
+++ b/swh/web/tests/conftest.py
@@ -1,301 +1,301 @@
 # Copyright (C) 2018-2019  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 inspect
 import json
 import os
 import shutil
 
 from subprocess import run, PIPE
 
 import pytest
 
 from django.core.cache import cache
 from hypothesis import settings, HealthCheck
 from rest_framework.test import APIClient, APIRequestFactory
 
 from swh.model.hashutil import ALGORITHMS, hash_to_bytes
 from swh.web.common import converters
 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):
     # 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__)
     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(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'])
+    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
 def archive_data(tests_data):
     return _ArchiveData(tests_data)
 
 
 # Fixture to manipulate indexer data from a sample archive used in the tests
 @pytest.fixture
 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 _call_storage_method(method):
             def call_storage_method(*args, **kwargs):
                 return method(*args, **kwargs)
 
             return call_storage_method
 
         # Forward calls to non overridden Storage methods to wrapped
         # storage instance
         for method_name, method in inspect.getmembers(
                 self.storage, predicate=inspect.ismethod):
             if (not hasattr(self, method_name) and
                     not method_name.startswith('_')):
                 setattr(self, method_name, _call_storage_method(method))
 
     def content_find(self, content):
         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]) if cnt else cnt
 
     def content_get_metadata(self, cnt_id):
         cnt_id_bytes = hash_to_bytes(cnt_id)
         metadata = next(self.storage.content_get_metadata([cnt_id_bytes]))
         return converters.from_swh(metadata,
                                    hashess={'sha1', 'sha1_git', 'sha256',
                                             'blake2s256'})
 
     def content_get(self, cnt_id):
         cnt_id_bytes = hash_to_bytes(cnt_id)
         cnt = next(self.storage.content_get([cnt_id_bytes]))
         return converters.from_content(cnt)
 
     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):
         rel_id_bytes = hash_to_bytes(rel_id)
         rel_data = next(self.storage.release_get([rel_id_bytes]))
         return converters.from_release(rel_data)
 
     def revision_get(self, rev_id):
         rev_id_bytes = hash_to_bytes(rev_id)
         rev_data = next(self.storage.revision_get([rev_id_bytes]))
         return converters.from_revision(rev_data)
 
     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 = self.storage.snapshot_get_latest(origin_url)
         return converters.from_snapshot(snp)
 
     def origin_get(self, origin_info):
         origin = self.storage.origin_get(origin_info)
         return converters.from_origin(origin)
 
     def origin_visit_get(self, origin_url):
         visits = self.storage.origin_visit_get(origin_url)
         return list(map(converters.from_origin_visit, visits))
 
     def origin_visit_get_by(self, origin_url, visit_id):
         visit = self.storage.origin_visit_get_by(origin_url, visit_id)
         return converters.from_origin_visit(visit)
 
     def snapshot_get(self, snapshot_id):
         snp = self.storage.snapshot_get(hash_to_bytes(snapshot_id))
         return converters.from_snapshot(snp)
 
     def snapshot_get_branches(self, snapshot_id, branches_from='',
                               branches_count=1000, target_types=None):
         snp = self.storage.snapshot_get_branches(
             hash_to_bytes(snapshot_id), branches_from.encode(),
             branches_count, target_types)
         return converters.from_snapshot(snp)
 
     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
 
 
 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 = next(self.idx_storage.content_mimetype_get(
                         [hash_to_bytes(cnt_id)]))
         return converters.from_filetype(mimetype)
 
     def content_add_language(self, cnt_id):
         raise NotImplementedError('Language indexer is disabled.')
         self.language_indexer.run([hash_to_bytes(cnt_id)],
                                   'update-dups')
 
     def content_get_language(self, cnt_id):
         lang = next(self.idx_storage.content_language_get(
                     [hash_to_bytes(cnt_id)]))
         return converters.from_swh(lang, hashess={'id'})
 
     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)
         lic = next(self.idx_storage.content_fossology_license_get(
                    [cnt_id_bytes]))
         return converters.from_swh({'id': cnt_id_bytes,
                                     'facts': lic[cnt_id_bytes]},
                                    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'})
diff --git a/swh/web/tests/data.py b/swh/web/tests/data.py
index dc01d364..3dcc4769 100644
--- a/swh/web/tests/data.py
+++ b/swh/web/tests/data.py
@@ -1,466 +1,479 @@
 # Copyright (C) 2018-2019  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 os
 import random
 
 from copy import deepcopy
 from typing import Dict
 
 from rest_framework.decorators import api_view
 from rest_framework.response import Response
 
 from swh.indexer.fossology_license import FossologyLicenseIndexer
 from swh.indexer.mimetype import MimetypeIndexer
 from swh.indexer.ctags import CtagsIndexer
 from swh.indexer.storage import get_indexer_storage
 from swh.model.from_disk import Directory
 from swh.model.hashutil import hash_to_hex, hash_to_bytes, DEFAULT_ALGORITHMS
 from swh.model.identifiers import directory_identifier
 from swh.loader.git.from_disk import GitLoaderFromArchive
+from swh.search import get_search
 from swh.storage.algos.dir_iterators import dir_iterator
 from swh.web import config
 from swh.web.browse.utils import (
     get_mimetype_and_encoding_for_content, prepare_content_for_display
 )
 from swh.web.common import service
 from swh.web.common.highlightjs import get_hljs_language_from_filename
 
 # Module used to initialize data that will be provided as tests input
 
 # Configuration for git loader
 _TEST_LOADER_CONFIG = {
     'storage': {
         'cls': 'memory'
     },
     'save_data': False,
     'max_content_size': 100 * 1024 * 1024,
 }
 
 # Base content indexer configuration
 _TEST_INDEXER_BASE_CONFIG = {
     'storage': {
         'cls': 'memory'
     },
     'objstorage': {
         'cls': 'memory',
         'args': {},
     },
     'indexer_storage': {
         'cls': 'memory',
         'args': {},
     }
 }
 
 
 def random_sha1():
     return hash_to_hex(bytes(random.randint(0, 255) for _ in range(20)))
 
 
 def random_sha256():
     return hash_to_hex(bytes(random.randint(0, 255) for _ in range(32)))
 
 
 def random_blake2s256():
     return hash_to_hex(bytes(random.randint(0, 255) for _ in range(32)))
 
 
 def random_content():
     return {
         'sha1': random_sha1(),
         'sha1_git': random_sha1(),
         'sha256': random_sha256(),
         'blake2s256': random_blake2s256(),
     }
 
 
 # MimetypeIndexer with custom configuration for tests
 class _MimetypeIndexer(MimetypeIndexer):
     def parse_config_file(self, *args, **kwargs):
         return {
             **_TEST_INDEXER_BASE_CONFIG,
             'tools': {
                 'name': 'file',
                 'version': '1:5.30-1+deb9u1',
                 'configuration': {
                     "type": "library",
                     "debian-package": "python3-magic"
                 }
             }
         }
 
 
 # FossologyLicenseIndexer with custom configuration for tests
 class _FossologyLicenseIndexer(FossologyLicenseIndexer):
     def parse_config_file(self, *args, **kwargs):
         return {
             **_TEST_INDEXER_BASE_CONFIG,
             'workdir': '/tmp/swh/indexer.fossology.license',
             'tools': {
                 'name': 'nomos',
                 'version': '3.1.0rc2-31-ga2cbb8c',
                 'configuration': {
                     'command_line': 'nomossa <filepath>',
                 },
             }
         }
 
 
 # CtagsIndexer with custom configuration for tests
 class _CtagsIndexer(CtagsIndexer):
     def parse_config_file(self, *args, **kwargs):
         return {
             **_TEST_INDEXER_BASE_CONFIG,
             'workdir': '/tmp/swh/indexer.ctags',
             'languages': {'c': 'c'},
             'tools': {
                 'name': 'universal-ctags',
                 'version': '~git7859817b',
                 'configuration': {
                     'command_line': '''ctags --fields=+lnz --sort=no --links=no ''' # noqa
                                     '''--output-format=json <filepath>'''
                 },
             }
         }
 
 
 # Lightweight git repositories that will be loaded to generate
 # input data for tests
 _TEST_ORIGINS = [
     {
         'type': 'git',
         'url': 'https://github.com/wcoder/highlightjs-line-numbers.js',
         'archives': ['highlightjs-line-numbers.js.zip',
                      'highlightjs-line-numbers.js_visit2.zip'],
         'visit_date': ['Dec 1 2018, 01:00 UTC',
                        'Jan 20 2019, 15:00 UTC']
     },
     {
         'type': 'git',
         'url': 'https://github.com/memononen/libtess2',
         'archives': ['libtess2.zip'],
         'visit_date': ['May 25 2018, 01:00 UTC']
     },
     {
         'type': 'git',
         'url': 'repo_with_submodules',
         'archives': ['repo_with_submodules.tgz'],
         'visit_date': ['Jan 1 2019, 01:00 UTC']
     }
 ]
 
 _contents = {}
 
 
 # Tests data initialization
 def _init_tests_data():
     # To hold reference to the memory storage
     storage = None
 
+    # Create search instance
+    search = get_search('memory', {})
+    search.initialize()
+    search.origin_update({'url': origin['url']} for origin in _TEST_ORIGINS)
+
     # Load git repositories from archives
     for origin in _TEST_ORIGINS:
         for i, archive in enumerate(origin['archives']):
             origin_repo_archive = \
                 os.path.join(os.path.dirname(__file__),
                              'resources/repos/%s' % archive)
             loader = GitLoaderFromArchive(origin['url'],
                                           archive_path=origin_repo_archive,
                                           config=_TEST_LOADER_CONFIG,
                                           visit_date=origin['visit_date'][i])
             if storage is None:
                 storage = loader.storage
             else:
                 loader.storage = storage
             loader.load()
 
         origin.update(storage.origin_get(origin))  # add an 'id' key if enabled
+        search.origin_update([{'url': origin['url'], 'has_visits': True}])
 
     for i in range(250):
         url = 'https://many.origins/%d' % (i+1)
         storage.origin_add([{'url': url}])
+        search.origin_update([{'url': url, 'has_visits': True}])
         visit = storage.origin_visit_add(url, '2019-12-03 13:55:05', 'tar')
         storage.origin_visit_update(
             url, visit['visit'],
             snapshot='1a8893e6a86f444e8be8e7bda6cb34fb1735a00e')
 
     contents = set()
     directories = set()
     revisions = set()
     releases = set()
     snapshots = set()
 
     content_path = {}
 
     # Get all objects loaded into the test archive
     for origin in _TEST_ORIGINS:
         snp = storage.snapshot_get_latest(origin['url'])
         snapshots.add(hash_to_hex(snp['id']))
         for branch_name, branch_data in snp['branches'].items():
             if branch_data['target_type'] == 'revision':
                 revisions.add(branch_data['target'])
             elif branch_data['target_type'] == 'release':
                 release = next(storage.release_get([branch_data['target']]))
                 revisions.add(release['target'])
                 releases.add(hash_to_hex(branch_data['target']))
 
         for rev_log in storage.revision_shortlog(set(revisions)):
             rev_id = rev_log[0]
             revisions.add(rev_id)
 
         for rev in storage.revision_get(revisions):
             dir_id = rev['directory']
             directories.add(hash_to_hex(dir_id))
             for entry in dir_iterator(storage, dir_id):
                 content_path[entry['sha1']] = '/'.join(
                     [hash_to_hex(dir_id), entry['path'].decode('utf-8')])
                 if entry['type'] == 'file':
                     contents.add(entry['sha1'])
                 elif entry['type'] == 'dir':
                     directories.add(hash_to_hex(entry['target']))
 
     # Get all checksums for each content
     contents_metadata = storage.content_get_metadata(contents)
     contents = []
     for content_metadata in contents_metadata:
         contents.append({
             algo: hash_to_hex(content_metadata[algo])
             for algo in DEFAULT_ALGORITHMS
         })
         path = content_path[content_metadata['sha1']]
         cnt = next(storage.content_get([content_metadata['sha1']]))
         mimetype, encoding = get_mimetype_and_encoding_for_content(cnt['data'])
         content_display_data = prepare_content_for_display(
             cnt['data'], mimetype, path)
         contents[-1]['path'] = path
         contents[-1]['mimetype'] = mimetype
         contents[-1]['encoding'] = encoding
         contents[-1]['hljs_language'] = content_display_data['language']
         contents[-1]['data'] = content_display_data['content_data']
         _contents[contents[-1]['sha1']] = contents[-1]
 
     # Create indexer storage instance that will be shared by indexers
     idx_storage = get_indexer_storage('memory', {})
 
     # Add the empty directory to the test archive
     empty_dir_id = directory_identifier({'entries': []})
     empty_dir_id_bin = hash_to_bytes(empty_dir_id)
     storage.directory_add([{'id': empty_dir_id_bin, 'entries': []}])
 
     # Return tests data
     return {
+        'search': search,
         'storage': storage,
         'idx_storage': idx_storage,
         'origins': _TEST_ORIGINS,
         'contents': contents,
         'directories': list(directories),
         'releases': list(releases),
         'revisions': list(map(hash_to_hex, revisions)),
         'snapshots': list(snapshots),
         'generated_checksums': set(),
     }
 
 
 def _init_indexers(tests_data):
     # Instantiate content indexers that will be used in tests
     # and force them to use the memory storages
     indexers = {}
     for idx_name, idx_class in (('mimetype_indexer', _MimetypeIndexer),
                                 ('license_indexer', _FossologyLicenseIndexer),
                                 ('ctags_indexer', _CtagsIndexer)):
         idx = idx_class()
         idx.storage = tests_data['storage']
         idx.objstorage = tests_data['storage'].objstorage
         idx.idx_storage = tests_data['idx_storage']
         idx.register_tools(idx.config['tools'])
         indexers[idx_name] = idx
 
     return indexers
 
 
 def get_content(content_sha1):
     return _contents.get(content_sha1)
 
 
 _tests_data = None
 _current_tests_data = None
 _indexer_loggers = {}
 
 
 def get_tests_data(reset=False):
     """
     Initialize tests data and return them in a dict.
     """
     global _tests_data, _current_tests_data
     if _tests_data is None:
         _tests_data = _init_tests_data()
         indexers = _init_indexers(_tests_data)
         for (name, idx) in indexers.items():
             # pytest makes the loggers use a temporary file; and deepcopy
             # requires serializability. So we remove them, and add them
             # back after the copy.
             _indexer_loggers[name] = idx.log
             del idx.log
         _tests_data.update(indexers)
     if reset or _current_tests_data is None:
         _current_tests_data = deepcopy(_tests_data)
         for (name, logger) in _indexer_loggers.items():
             _current_tests_data[name].log = logger
     return _current_tests_data
 
 
-def override_storages(storage, idx_storage):
+def override_storages(storage, idx_storage, search):
     """
     Helper function to replace the storages from which archive data
     are fetched.
     """
     swh_config = config.get_config()
-    swh_config.update({'storage': storage})
-    service.storage = storage
+    swh_config.update({
+        'storage': storage,
+        'indexer_storage': idx_storage,
+        'search': search,
+    })
 
-    swh_config.update({'indexer_storage': idx_storage})
+    service.storage = storage
     service.idx_storage = idx_storage
+    service.search = search
 
 
 # Implement some special endpoints used to provide input tests data
 # when executing end to end tests with cypress
 
 _content_code_data_exts = {}  # type: Dict[str, Dict[str, str]]
 _content_code_data_filenames = {}  # type: Dict[str, Dict[str, str]]
 _content_other_data_exts = {}  # type: Dict[str, Dict[str, str]]
 
 
 def _init_content_tests_data(data_path, data_dict, ext_key):
     """
     Helper function to read the content of a directory, store it
     into a test archive and add some files metadata (sha1 and/or
     expected programming language) in a dict.
 
     Args:
         data_path (str): path to a directory relative to the tests
             folder of swh-web
         data_dict (dict): the dict that will store files metadata
         ext_key (bool): whether to use file extensions or filenames
             as dict keys
     """
     test_contents_dir = os.path.join(
         os.path.dirname(__file__), data_path).encode('utf-8')
     directory = Directory.from_disk(path=test_contents_dir, data=True,
                                     save_path=True)
     objects = directory.collect()
     for c in objects['content'].values():
         c['status'] = 'visible'
         sha1 = hash_to_hex(c['sha1'])
         if ext_key:
             key = c['path'].decode('utf-8').split('.')[-1]
             filename = 'test.' + key
         else:
             filename = c['path'].decode('utf-8').split('/')[-1]
             key = filename
         language = get_hljs_language_from_filename(filename)
         data_dict[key] = {'sha1': sha1,
                           'language': language}
         del c['path']
         del c['perms']
     storage = get_tests_data()['storage']
     storage.content_add(objects['content'].values())
 
 
 def _init_content_code_data_exts():
     """
     Fill a global dictionary which maps source file extension to
     a code content example.
     """
     global _content_code_data_exts
     _init_content_tests_data('resources/contents/code/extensions',
                              _content_code_data_exts, True)
 
 
 def _init_content_other_data_exts():
     """
     Fill a global dictionary which maps a file extension to
     a content example.
     """
     global _content_other_data_exts
     _init_content_tests_data('resources/contents/other/extensions',
                              _content_other_data_exts, True)
 
 
 def _init_content_code_data_filenames():
     """
     Fill a global dictionary which maps a filename to
     a content example.
     """
     global _content_code_data_filenames
     _init_content_tests_data('resources/contents/code/filenames',
                              _content_code_data_filenames, False)
 
 
 if config.get_config()['e2e_tests_mode']:
     _init_content_code_data_exts()
     _init_content_other_data_exts()
     _init_content_code_data_filenames()
 
 
 @api_view(['GET'])
 def get_content_code_data_all_exts(request):
     """
     Endpoint implementation returning a list of all source file
     extensions to test for highlighting using cypress.
     """
     return Response(sorted(_content_code_data_exts.keys()),
                     status=200, content_type='application/json')
 
 
 @api_view(['GET'])
 def get_content_code_data_by_ext(request, ext):
     """
     Endpoint implementation returning metadata of a code content example
     based on the source file extension.
     """
     data = None
     status = 404
     if ext in _content_code_data_exts:
         data = _content_code_data_exts[ext]
         status = 200
     return Response(data, status=status, content_type='application/json')
 
 
 @api_view(['GET'])
 def get_content_other_data_by_ext(request, ext):
     """
     Endpoint implementation returning metadata of a content example
     based on the file extension.
     """
     _init_content_other_data_exts()
     data = None
     status = 404
     if ext in _content_other_data_exts:
         data = _content_other_data_exts[ext]
         status = 200
     return Response(data, status=status, content_type='application/json')
 
 
 @api_view(['GET'])
 def get_content_code_data_all_filenames(request):
     """
     Endpoint implementation returning a list of all source filenames
     to test for highlighting using cypress.
     """
     return Response(sorted(_content_code_data_filenames.keys()),
                     status=200, content_type='application/json')
 
 
 @api_view(['GET'])
 def get_content_code_data_by_filename(request, filename):
     """
     Endpoint implementation returning metadata of a code content example
     based on the source filename.
     """
     data = None
     status = 404
     if filename in _content_code_data_filenames:
         data = _content_code_data_filenames[filename]
         status = 200
     return Response(data, status=status, content_type='application/json')