arcshiftwrap 0.4.0

Python client for the Arctic Shift API.

Arctic Shift API Client Reference

This page gives practical examples for every public function in arcshiftwrap/arctic_shift.py.

• API Base: https://arctic-shift.photon-reddit.com
• Install: pip install arcshiftwrap

Setup

from datetime import datetime, timezone

from arcshiftwrap import (
	ArcticShiftClient,
	collect_posts_by_subreddits_parallel,
	collect_comments_by_subreddits_parallel,
)
from arcshiftwrap.arctic_shift import (
	collect_comments_by_windows,
	collect_posts_by_windows,
	deduplicate_items,
	format_date,
	normalize_response,
	split_time_range,
	utc_now,
)

client = ArcticShiftClient(timeout=90, max_retries=4, backoff_factor=2.0)

request

Low-level method to send a GET request to the Arctic Shift API.

response = client.request(
        endpoint="posts/search",
        params={"subreddit": "technology", "limit": 10}
)

ID Lookup Methods

get_posts_by_ids

posts = client.get_posts_by_ids(
	ids=["1abcde", "1fghij"],
	fields=["id", "subreddit", "title", "created_utc"],
)

get_comments_by_ids

comments = client.get_comments_by_ids(
	ids=["k12345", "k67890"],
	fields=["id", "subreddit", "body", "created_utc"],
)

get_subreddits_by_ids

subreddits = client.get_subreddits_by_ids(
	ids=["technology", "MachineLearning"],
	fields=["display_name", "subscribers", "over18"],
)

get_users_by_ids

users = client.get_users_by_ids(
	ids=["spez", "AutoModerator"],
	fields=["author", "total_karma", "created_utc"],
)

Search Methods

search_posts

posts = client.search_posts(
	subreddit="technology",
	after="2026-04-01",
	before="2026-04-08",
	limit=100,
	sort="asc",
	query="llm",
	fields=["id", "title", "selftext", "score", "num_comments"],
)

search_comments

comments = client.search_comments(
	subreddit="technology",
	after="2026-04-01",
	before="2026-04-08",
	limit=100,
	sort="asc",
	body="chatgpt",
	fields=["id", "body", "score", "link_id", "parent_id"],
)

get_comment_tree

tree = client.get_comment_tree(
	link_id="t3_1abcde",
	limit=500,
	start_breadth=4,
	start_depth=4,
)

Aggregation Methods

aggregate_posts

post_agg = client.aggregate_posts(
	aggregate="subreddit",
	frequency="day",
	after="2026-04-01",
	before="2026-04-08",
	min_count=5,
)

aggregate_comments

comment_agg = client.aggregate_comments(
	aggregate="author",
	frequency="day",
	subreddit="technology",
	after="2026-04-01",
	before="2026-04-08",
)

Subreddit Methods

search_subreddits

subs = client.search_subreddits(
	subreddit_prefix="tech",
	min_subscribers=50000,
	limit=25,
	sort="desc",
	sort_type="subscribers",
	fields=["display_name", "subscribers", "over18"],
)

get_subreddit_rules

rules = client.get_subreddit_rules(subreddits=["technology", "MachineLearning"])

get_subreddit_wikis

wikis = client.get_subreddit_wikis(
	subreddit="technology",
	paths=["index", "faq"],
	limit=10,
)

list_subreddit_wikis

wiki_paths = client.list_subreddit_wikis(subreddit="technology")

User Methods

search_users

users = client.search_users(
	author_prefix="sam",
	min_karma=1000,
	limit=25,
	sort="desc",
	sort_type="total_karma",
)

user_user_interactions

interactions = client.user_user_interactions(
	author="spez",
	subreddit="technology",
	after="2026-04-01",
	before="2026-04-08",
	min_count=2,
	limit=100,
)

user_subreddit_interactions

subreddit_interactions = client.user_subreddit_interactions(
	author="spez",
	weight_posts=1.0,
	weight_comments=1.0,
	after="2026-04-01",
	before="2026-04-08",
	min_count=1,
	limit=100,
)

aggregate_flairs

flairs = client.aggregate_flairs(author="spez")

Utility Endpoints

resolve_short_links

resolved = client.resolve_short_links(paths=["3g1jfiw", "3h2kxyz"])

time_series

series = client.time_series(
	key="posts",
	precision="day",
	after="2026-04-01",
	before="2026-04-08",
)

Helper Functions

utc_now

now_utc = utc_now()

format_date

dt_str = format_date(datetime(2026, 4, 1, 12, 0, 0, tzinfo=timezone.utc))
# "2026-04-01T12:00:00Z"

split_time_range

start = datetime(2026, 4, 1, tzinfo=timezone.utc)
end = datetime(2026, 4, 3, tzinfo=timezone.utc)
windows = split_time_range(start=start, end=end, step_hours=12)

normalize_response

items = normalize_response({"data": [{"id": "1"}, {"id": "2"}]})

deduplicate_items

deduped = deduplicate_items(
	items=[{"id": "a"}, {"id": "a"}, {"id": "b"}],
	id_field="id",
)

collect_posts_by_windows

start = datetime(2026, 4, 1, tzinfo=timezone.utc)
end = datetime(2026, 4, 8, tzinfo=timezone.utc)

posts = collect_posts_by_windows(
	client=client,
	subreddit="technology",
	start=start,
	end=end,
	step_hours=24,
	limit=100,
	fields=["id", "title", "created_utc"],
)

collect_comments_by_windows

start = datetime(2026, 4, 1, tzinfo=timezone.utc)
end = datetime(2026, 4, 8, tzinfo=timezone.utc)

comments = collect_comments_by_windows(
	client=client,
	subreddit="technology",
	start=start,
	end=end,
	step_hours=24,
	limit=100,
	fields=["id", "body", "created_utc"],
)

collect_posts_by_subreddits_parallel

Collect posts from multiple subreddits concurrently using ProcessPoolExecutor. Significantly faster than sequential collection.

from arcshiftwrap import collect_posts_by_subreddits_parallel

start = datetime(2026, 4, 1, tzinfo=timezone.utc)
end = datetime(2026, 4, 8, tzinfo=timezone.utc)

subreddits = ["technology", "MachineLearning", "chatgpt", "OpenAI"]

posts = collect_posts_by_subreddits_parallel(
	client=client,
	subreddits=subreddits,
	start=start,
	end=end,
	step_hours=24,
	limit=100,
	max_workers=4,  # number of parallel processes
	fields=["id", "title", "subreddit", "created_utc"],
)

collect_comments_by_subreddits_parallel

Collect comments from multiple subreddits concurrently using ProcessPoolExecutor. Significantly faster than sequential collection.

from arcshiftwrap import collect_comments_by_subreddits_parallel

start = datetime(2026, 4, 1, tzinfo=timezone.utc)
end = datetime(2026, 4, 8, tzinfo=timezone.utc)

subreddits = ["technology", "MachineLearning", "chatgpt", "OpenAI"]

comments = collect_comments_by_subreddits_parallel(
	client=client,
	subreddits=subreddits,
	start=start,
	end=end,
	step_hours=24,
	limit=100,
	max_workers=4,  # number of parallel processes
	fields=["id", "body", "subreddit", "created_utc"],
)

Notes

• These examples reflect the current public API in arcshiftwrap/arctic_shift.py.
• Internal helper methods like _join, _bool, _clean, _parse_response, and _rate_limit_wait are intentionally not used directly.