def merge(src, dst, mode="no_move"):
assert_choice("mode", mode, MODES)
if not src.is_dir() or not dst.is_dir():
raise ValueError("src and dst must be directories")
for path in src.rglob("*"):
if path.is_dir():
relpath = path.relative_to(src)
(dst / relpath).mkdir(parents=True, exist_ok=True)
elif path.is_file():
relpath = path.relative_to(src)
(dst / relpath.parent).mkdir(parents=True, exist_ok=True)
target = dst / relpath
if target.exists():
if mode == "fail":
raise FileExistsError(fspath(path))
elif mode == "no_move":
pass
elif mode == "overwrite":
replace(fspath(path), fspath(target))
elif mode == "rename":
move_rename(path, target)
else:
path.rename(
target
) # race condition on linux, us renameat2 with RENAME_NOREPLACE?
else:
raise RuntimeError("Unhandled file: {}".format(path))
remove_empty_dirs(fspath(src), onerror=remove_empty_error_log)
def get_paths(self,
field: str,
token: str,
scoring: str = "unscored") -> Iterable[OptionalSearchResult]:
assert_choice("scoring", scoring, SCORING_METHODS)
docs = self.get_docs(field, token)
paths: Iterable[OptionalSearchResult]
if scoring == "unscored":
paths = ((self.ids2docs[doc_id], i)
for i, doc_id in enumerate(docs.keys()))
elif scoring == "term_freq":
paths = ((self.ids2docs[doc_id], term_freq)
for doc_id, term_freq in docs.items())
elif scoring == "tfidf":
if docs:
num_docs = len(self.table[field])
inv_doc_freq = log10(num_docs / len(docs))
paths = ((self.ids2docs[doc_id],
self._tfidf(term_freq, inv_doc_freq))
for doc_id, term_freq in docs.items())
return paths
def _sorted(self,
groupname: str,
paths: Iterable[OptionalSearchResult],
sortby="path") -> list[SearchResult]:
assert_choice("sortby", sortby, SORTBY_METHODS)
grouppaths = list(map(str, self.groups[groupname]))
try:
filtered_paths = cast(
Iterable[SearchResult],
(
(path, score) for path, score in paths
if any(fspath(path).startswith(gp)
for gp in grouppaths) # type: ignore[arg-type]
),
)
except TypeError as e:
raise RuntimeError(e)
if sortby == "path":
return sorted(filtered_paths, key=itemgetter(0), reverse=False)
elif sortby == "score":
return sorted(filtered_paths, key=itemgetter(1), reverse=True)
else:
assert False
def get_episodes_by_bohne_preview(self, bohne_id, order="ASC"):
# type: (int, str) -> Dict[str, Any]
""" Returns reduced information about all episodes for the given Bohne.
"""
assert_choice("order", order, {"ASC", "DESC"})
return self._request_single(
"/v1/media/episode/bybohne/preview/{}".format(bohne_id))
def get_newest_episodes_preview(self, order="ASC"):
# type: (str, ) -> Iterator[Dict[str, Any]]
assert_choice("order", order, {"ASC", "DESC"})
return self._request_paged("/v1/media/episode/preview/newest",
50,
False,
order=order)
def get_shows(self, sortby="LastEpisode", only=None):
# type: (str, Optional[str]) -> Iterator[Dict[str, Any]]
assert_choice("sortby", sortby, {"LastEpisode"})
assert_choice("only", only, {None, "podcast"})
return self._request_paged("/v1/media/show/all",
50,
sortby=sortby,
only=only)
def get_shows_mini(self, sortby="LastEpisode", only=None):
# type: (str, Optional[str]) -> List[Dict[str, Any]]
""" Returns minimal information about all shows.
"""
assert_choice("sortby", sortby, {"LastEpisode"})
assert_choice("only", only, {None, "podcast"})
return self._request_single("/v1/media/show/preview/mini/all",
sortby=sortby,
only=only)
def get_shows_preview(self, sortby="LastEpisode", only=None):
# type: (str, Optional[str]) -> Iterator[Dict[str, Any]]
""" Returns paginated, reduced information about all shows.
"""
assert_choice("sortby", sortby, {"LastEpisode"})
assert_choice("only", only, {None, "podcast"})
return self._request_paged("/v1/media/show/preview/all",
50,
sortby=sortby,
only=only)
def get_episodes_by_show_preview(self, show_id, order="ASC"):
# type: (int, str) -> Iterator[Dict[str, Any]]
""" Returns reduced information about all episodes for the given show.
"""
assert_choice("order", order, {"ASC", "DESC"})
return self._request_paged(
"/v1/media/episode/byshow/preview/{}".format(show_id),
50,
False,
order=order)
def get_episodes_by_season(self, season_id, order="ASC"):
# type: (int, str) -> Iterator[Dict[str, Any]]
""" Returns information about all episodes of a given season.
"""
assert_choice("order", order, {"ASC", "DESC"})
return self._request_paged(
"/v1/media/episode/byseason/{}".format(season_id),
50,
False,
order=order)
def get_episodes_by_bohne(self, bohne_id, order="ASC"):
# type: (int, str) -> Iterator[Dict[str, Any]]
""" Returns information about all episodes for the given Bohne.
"""
assert_choice("order", order, {"ASC", "DESC"})
return self._request_paged(
"/v1/media/episode/bybohne/{}".format(bohne_id),
50,
False,
order=order)
def get_paths_op(
self,
field: str,
tokens: Sequence[str],
setop: Callable[..., set[int]],
scoring: str = "unscored") -> Iterable[OptionalSearchResult]:
assert_choice("scoring", scoring, SCORING_METHODS)
sets = tuple(
set(self.get_docs(field, token).keys()) for token in tokens)
docs_op = setop(*sets)
paths: Iterable[OptionalSearchResult]
if scoring == "unscored":
paths = ((self.ids2docs[doc_id], i)
for i, doc_id in enumerate(docs_op))
elif scoring == "term_freq":
term_freqs = defaultdict(int) # type: DefaultDict[int, int]
for token in tokens:
docs = self.get_docs(field, token)
for doc_id, term_freq in docs.items():
term_freqs[doc_id] += term_freq
paths = ((self.ids2docs[doc_id], term_freqs[doc_id])
for doc_id in docs_op)
elif scoring == "tfidf":
tfidf = defaultdict(float) # type: DefaultDict[int, float]
for token in tokens:
docs = self.get_docs(field, token)
if docs:
num_docs = len(self.table[field])
inv_doc_freq = log10(num_docs / len(docs))
for doc_id, term_freq in docs.items():
tfidf[doc_id] += self._tfidf(term_freq, inv_doc_freq)
paths = ((self.ids2docs[doc_id], tfidf[doc_id])
for doc_id in docs_op)
return paths
