Closes internetarchive#5080. Adds basic ability to search/filter one'…

…s reading log.

openlibrary/core/bookshelves.py

@@ -177,6 +177,7 @@ def get_users_logged_books(
  limit: int = 100,
  page: int = 1, # Not zero-based counting!
  sort: Literal['created asc', 'created desc'] = 'created desc',
+ q: str = "",
  ) -> list[dict]:
  """Returns a list of Reading Log database records for books which
  the user has logged. Records are described in core/schema.py
@@ -188,6 +189,9 @@ def get_users_logged_books(
  """
  oldb = db.get_db()
  page = int(page or 1)
+ # Filtering by title/author needs a larger limit as as the results are
+ # used to query Solr to get the title/author to do the actual filtering.
+ limit = 1000 if q else limit
  data = {
  'username': username,
  'limit': limit,
@@ -213,7 +217,34 @@ def get_users_logged_books(
  # XXX Removing limit, offset, etc from data looks like a bug
  # unrelated / not fixing in this PR.
  data = {'username': username}
- return list(oldb.query(query, vars=data))
+ logged_books = list(oldb.query(query, vars=data))
+
+ # Filtering won't occur unless q >= 3, as limited in mybooks.my_books_view().
+ if q:
+ from openlibrary.plugins.worksearch.code import do_search
+
+ def filter_key(key: str) -> int:
+ """
+ Get the int() representation of an OL work key.
+ filter_key("/works/OL123W")
+ >>> 123
+ """
+ return int("".join(char for char in key if char.isdigit()))
+
+ work_ids = ['/works/OL%sW' % i['work_id'] for i in logged_books]
+ solr_query = {'q': f'{q} key:({" OR ".join(work_ids)})'}
+ solr_resp = do_search(param=solr_query, page=1, rows=100, sort=None)
+ solr_work_keys_in_oldb_form = {
+ filter_key(doc["key"]) for doc in solr_resp.docs
+ }
+ filtered_books = [
+ book
+ for book in logged_books
+ if book["work_id"] in solr_work_keys_in_oldb_form
+ ]
+ return filtered_books
+ else:
+ return logged_books
 
  @classmethod
  def iterate_users_logged_books(cls, username: str) -> Iterable[dict]:

openlibrary/core/models.py

@@ -445,7 +445,7 @@ def edition_count(self):
  def get_lists(self, limit=50, offset=0, sort=True):
  return self._get_lists(limit=limit, offset=offset, sort=sort)
 
- def get_users_rating(self, username):
+ def get_users_rating(self, username: str) -> int | None:
  if not username:
  return None
  work_id = extract_numeric_id_from_olid(self.key)

openlibrary/core/ratings.py

@@ -111,12 +111,13 @@ def get_all_works_ratings(cls, work_id) -> list:
  return list(oldb.query(query, vars={'work_id': int(work_id)}))
 
  @classmethod
- def get_users_rating_for_work(cls, username, work_id):
+ def get_users_rating_for_work(cls, username: str, work_id: str | int) -> int | None:
+ """work_id must be convetrible to int."""
  oldb = db.get_db()
  data = {'username': username, 'work_id': int(work_id)}
  query = 'SELECT * from ratings where username=$username AND work_id=$work_id'
  results = list(oldb.query(query, vars=data))
- rating = results[0].rating if results else None
+ rating: int | None = results[0].rating if results else None
  return rating
 
  @classmethod

openlibrary/plugins/upstream/mybooks.py

@@ -28,8 +28,13 @@ class my_books_view(delegate.page):
  path = r"/people/([^/]+)/books/([a-zA-Z_-]+)"
 
  def GET(self, username, key):
- i = web.input(page=1, sort='desc')
- return MyBooksTemplate(username, key).render(page=i.page, sort=i.sort)
+ i = web.input(page=1, sort='desc', q=None)
+ # Limit reading log filtering to queries of 3+ characters because filtering the
+ # reading log requires comparing potentially large queries, which can be expensive
+ # and are not paginated.
+ if i.q:
+ i.q = i.q if len(i.q) >= 3 else None
+ return MyBooksTemplate(username, key).render(page=i.page, sort=i.sort, q=i.q)
 
 
 class public_my_books_json(delegate.page):
@@ -189,7 +194,7 @@ def __init__(self, username, key):
  self.lists = self.readlog.lists
  self.counts = self.readlog.reading_log_counts
 
- def render(self, page=1, sort='desc', list=None):
+ def render(self, page=1, sort='desc', list=None, q=None):
  if not self.user:
  return render.notfound("User %s" % self.username, create=False)
  logged_in_user = accounts.get_current_user()
@@ -221,7 +226,7 @@ def render(self, page=1, sort='desc', list=None):
  elif self.key in self.READING_LOG_KEYS:
  data = add_availability(
  self.readlog.get_works(
- self.key, page=page, sort='created', sort_order=sort
+ self.key, page=page, sort='created', sort_order=sort, q=q
  ),
  mode="openlibrary_work",
  )
@@ -239,6 +244,7 @@ def render(self, page=1, sort='desc', list=None):
  )
 
  if data is not None:
+
  return render['account/books'](
  data,
  self.key,
@@ -249,6 +255,7 @@ def render(self, page=1, sort='desc', list=None):
  public=is_public,
  owners_page=is_logged_in_user,
  sort_order=sort,
+ q=q,
  )
 
  raise web.seeother(self.user.key)
@@ -359,9 +366,15 @@ def get_waitlisted_editions(self):
  editions[i].waitlist_record = keyed_waitlists[editions[i].ocaid]
  return editions
 
- def process_logged_books(self, logged_books):
+ def process_logged_books(self, logged_books: list[web.storage]):
+ """
+ Turn logged_books into <Work> items, while adding the logge_date
+ and logged_edition, as that information is stored in a separate
+ database from that which holds returns works.
+ """
  work_ids = ['/works/OL%sW' % i['work_id'] for i in logged_books]
  works = web.ctx.site.get_many(work_ids)
+
  for i in range(len(works)):
  # insert the logged edition (if present) and logged date
  works[i].logged_date = logged_books[i]['created']
@@ -373,7 +386,7 @@ def process_logged_books(self, logged_books):
  return works
 
  def get_want_to_read(
- self, page=1, limit=RESULTS_PER_PAGE, sort='created', sort_order='desc'
+ self, page=1, limit=RESULTS_PER_PAGE, sort='created', sort_order='desc', q=None
  ):
  return self.process_logged_books(
  Bookshelves.get_users_logged_books(
@@ -382,11 +395,12 @@ def get_want_to_read(
  page=page,
  limit=limit,
  sort=sort + ' ' + sort_order,
- )
+ q=q,
+ ),
  )
 
  def get_currently_reading(
- self, page=1, limit=RESULTS_PER_PAGE, sort='created', sort_order='desc'
+ self, page=1, limit=RESULTS_PER_PAGE, sort='created', sort_order='desc', q=None
  ):
  return self.process_logged_books(
  Bookshelves.get_users_logged_books(
@@ -395,11 +409,12 @@ def get_currently_reading(
  page=page,
  limit=limit,
  sort=sort + ' ' + sort_order,
- )
+ q=q,
+ ),
  )
 
  def get_already_read(
- self, page=1, limit=RESULTS_PER_PAGE, sort='created', sort_order='desc'
+ self, page=1, limit=RESULTS_PER_PAGE, sort='created', sort_order='desc', q=None
  ):
  return self.process_logged_books(
  Bookshelves.get_users_logged_books(
@@ -408,19 +423,26 @@ def get_already_read(
  page=page,
  limit=limit,
  sort=sort + ' ' + sort_order,
- )
+ q=q,
+ ),
  )
 
  def get_works(
- self, key, page=1, limit=RESULTS_PER_PAGE, sort='created', sort_order='desc'
+ self,
+ key,
+ page=1,
+ limit=RESULTS_PER_PAGE,
+ sort='created',
+ sort_order='desc',
+ q=None,
  ):
  """
  :rtype: list of openlibrary.plugins.upstream.models.Work
  """
  key = key.lower()
  if key in self.KEYS:
  return self.KEYS[key](
- page=page, limit=limit, sort=sort, sort_order=sort_order
+ page=page, limit=limit, sort=sort, sort_order=sort_order, q=q
  )
  else: # must be a list or invalid page!
  # works = web.ctx.site.get_many([ ... ])

openlibrary/templates/account/books.html

@@ -1,4 +1,4 @@
-$def with (items, key, counts, lists=None, user=None, logged_in_user=None, public=False, sort_order='desc', owners_page=False)
+$def with (items, key, counts, lists=None, user=None, logged_in_user=None, public=False, sort_order='desc', owners_page=False, q=None)
 
 $# Displays a user's reading log
 $# :param list items:
@@ -7,6 +7,7 @@
 $# :param list? lists:
 $# :param user:
 $# :param bool public:
+$# :param str q: search term.
 
 $ component_times = {}
 $ component_times['TotalTime'] = time()
@@ -106,9 +107,9 @@ <h1 class="details-title">
  $elif key == 'lists':
  $:render_template('lists/lists', items, lists, show_header=False)
  $elif key == 'already-read':
- $:render_template('account/reading_log', items, key, counts[key], owners_page, current_page, sort_order=sort_order, user=user, include_ratings=owners_page)
+ $:render_template('account/reading_log', items, key, counts[key], owners_page, current_page, sort_order=sort_order, user=user, include_ratings=owners_page, q=q)
  $elif key in {'currently-reading', 'want-to-read', 'sponsorships'}:
- $:render_template('account/reading_log', items, key, counts[key], owners_page, current_page, sort_order=sort_order, user=user)
+ $:render_template('account/reading_log', items, key, counts[key], owners_page, current_page, sort_order=sort_order, user=user, q=q)
  $else:
  $:render_template('type/list/view', items, check_owner=False, owns_page=True)
  </div>

openlibrary/templates/account/reading_log.html

@@ -1,4 +1,4 @@
-$def with (items, key, total_items, owners_page, current_page, sort_order='desc', user=None, include_ratings=False)
+$def with (items, key, total_items, owners_page, current_page, sort_order='desc', user=None, include_ratings=False, q=None)
 
 $ username = user.key.split('/')[-1]
 $ meta_photo_url = "https://archive.org/services/img/%s" % get_internet_archive_id(user.key)
@@ -24,14 +24,24 @@
 $add_metatag(property="og:description", content=og_description)
 $add_metatag(property="og:image", content=meta_photo_url)
 
+<form method="GET" class="olform pagesearchbox">
+ <input type="text" minlength="3" placeholder="Search by Title or Author" name="q" value="$(query_param('q', ''))"/>
+ <input type="submit"/>
+</form>
+
 <div class="mybooks-list">
+ $# If filtering results, override total_items to match the number of filtered results from bookshelves.get_users_logged_books().
+ $# This overrides pagination and will return all total_items on one page.
+ $if q:
+ $ total_items = len(items)
+ <span class="search-results-stats">$total_items hits</span>
  <span class="mybooks-tools"><img src="/images/icons/icon_sort.png" alt="$_('Sorting by')" style="margin-right:10px;" width="9" height="11">
  $if sort_order == 'desc':
  <strong class="lightgreen">$_("Date Added (newest)")</strong>
  |
- <a href="$ctx.path?sort=asc">$_("Date Added (oldest)")</a>
+ <a href="$ctx.path?sort=asc&q=$q">$_("Date Added (oldest)")</a>
  $else:
-  <a href="$ctx.path?sort=desc">$_("Date Added (newest)")</a>
+ <a href="$ctx.path?sort=desc&q=$q">$_("Date Added (newest)")</a>
  |
  <strong class="lightgreen">$_("Date Added (oldest)")</strong>
  </span>