| 1 | """Tests for the Telegram source adapter (lib/telegram.py). |
| 2 | |
| 3 | Covers handle parsing, availability gating, pagination, date filtering, |
| 4 | and the channels-required constraint. |
| 5 | """ |
| 6 | |
| 7 | from __future__ import annotations |
| 8 | |
| 9 | import pytest |
| 10 | |
| 11 | from lib import pipeline, telegram |
| 12 | |
| 13 | |
| 14 | # ---- handle parsing ---- |
| 15 | |
| 16 | @pytest.mark.parametrize("raw,expected", [ |
| 17 | ("aipost", "aipost"), |
| 18 | ("@aipost", "aipost"), |
| 19 | ("https://t.me/aipost", "aipost"), |
| 20 | ("https://t.me/s/aipost", "aipost"), |
| 21 | ("http://t.me/durov", "durov"), |
| 22 | ("t.me/channel", "channel"), |
| 23 | (" @spaced ", "spaced"), |
| 24 | ]) |
| 25 | def test_parse_channel_handle_valid(raw, expected): |
| 26 | assert telegram.parse_channel_handle(raw) == expected |
| 27 | |
| 28 | |
| 29 | @pytest.mark.parametrize("raw", [ |
| 30 | "", |
| 31 | " ", |
| 32 | "-1001234567890", |
| 33 | "-100999888777", |
| 34 | "https://t.me/joinchat/abcdef", |
| 35 | "joinchat/xyz", |
| 36 | ]) |
| 37 | def test_parse_channel_handle_invalid(raw): |
| 38 | with pytest.raises(telegram.InvalidChannelHandle): |
| 39 | telegram.parse_channel_handle(raw) |
| 40 | |
| 41 | |
| 42 | def test_parse_channel_sources_filters_invalid(): |
| 43 | raw = "aipost, @durov, https://t.me/joinchat/secret, -1001234, valid" |
| 44 | handles = telegram.parse_channel_sources(raw) |
| 45 | assert handles == ["aipost", "durov", "valid"] |
| 46 | |
| 47 | |
| 48 | def test_parse_channel_sources_dedupes(): |
| 49 | raw = "aipost, @aipost, https://t.me/aipost, AIPOST" |
| 50 | handles = telegram.parse_channel_sources(raw) |
| 51 | assert handles == ["aipost"] |
| 52 | |
| 53 | |
| 54 | # ---- opt-in availability gating ---- |
| 55 | |
| 56 | def test_not_available_by_default(): |
| 57 | config = {"SCRAPECREATORS_API_KEY": "test-key"} |
| 58 | avail = pipeline.available_sources(config) |
| 59 | assert "telegram" not in avail |
| 60 | |
| 61 | |
| 62 | def test_not_available_without_channels(): |
| 63 | config = { |
| 64 | "SCRAPECREATORS_API_KEY": "test-key", |
| 65 | "INCLUDE_SOURCES": "telegram", |
| 66 | } |
| 67 | avail = pipeline.available_sources(config) |
| 68 | assert "telegram" not in avail |
| 69 | |
| 70 | |
| 71 | def test_available_when_fully_configured(): |
| 72 | config = { |
| 73 | "SCRAPECREATORS_API_KEY": "test-key", |
| 74 | "INCLUDE_SOURCES": "telegram", |
| 75 | "TELEGRAM_SOURCES": "aipost,durov", |
| 76 | } |
| 77 | avail = pipeline.available_sources(config) |
| 78 | assert "telegram" in avail |
| 79 | |
| 80 | |
| 81 | def test_available_when_requested(): |
| 82 | config = { |
| 83 | "SCRAPECREATORS_API_KEY": "test-key", |
| 84 | "TELEGRAM_SOURCES": "aipost", |
| 85 | } |
| 86 | avail = pipeline.available_sources(config, requested_sources=["telegram"]) |
| 87 | assert "telegram" in avail |
| 88 | |
| 89 | |
| 90 | def test_excluded_when_in_exclude_sources(): |
| 91 | config = { |
| 92 | "SCRAPECREATORS_API_KEY": "test-key", |
| 93 | "INCLUDE_SOURCES": "telegram", |
| 94 | "TELEGRAM_SOURCES": "aipost", |
| 95 | "EXCLUDE_SOURCES": "telegram", |
| 96 | } |
| 97 | avail = pipeline.available_sources(config) |
| 98 | assert "telegram" not in avail |
| 99 | |
| 100 | |
| 101 | # ---- is_telegram_configured ---- |
| 102 | |
| 103 | def test_is_telegram_configured_true(): |
| 104 | config = { |
| 105 | "SCRAPECREATORS_API_KEY": "test-key", |
| 106 | "TELEGRAM_SOURCES": "aipost", |
| 107 | } |
| 108 | assert telegram.is_telegram_configured(config) is True |
| 109 | |
| 110 | |
| 111 | def test_is_telegram_configured_false_no_key(): |
| 112 | config = {"TELEGRAM_SOURCES": "aipost"} |
| 113 | assert telegram.is_telegram_configured(config) is False |
| 114 | |
| 115 | |
| 116 | def test_is_telegram_configured_false_no_channels(): |
| 117 | config = {"SCRAPECREATORS_API_KEY": "test-key"} |
| 118 | assert telegram.is_telegram_configured(config) is False |
| 119 | |
| 120 | |
| 121 | # ---- search without channels returns error ---- |
| 122 | |
| 123 | def test_search_returns_error_without_channels(): |
| 124 | result = telegram.search_telegram( |
| 125 | "test topic", |
| 126 | "2026-08-01", |
| 127 | "2026-08-21", |
| 128 | token="test-key", |
| 129 | config={}, |
| 130 | ) |
| 131 | assert result["items"] == [] |
| 132 | assert "channel list required" in result.get("error", "").lower() |
| 133 | |
| 134 | |
| 135 | def test_search_returns_error_without_token(): |
| 136 | result = telegram.search_telegram( |
| 137 | "test topic", |
| 138 | "2026-08-01", |
| 139 | "2026-08-21", |
| 140 | token=None, |
| 141 | config={"TELEGRAM_SOURCES": "aipost"}, |
| 142 | ) |
| 143 | assert result["items"] == [] |
| 144 | assert "SCRAPECREATORS_API_KEY" in result.get("error", "") |
| 145 | |
| 146 | |
| 147 | # ---- pagination and date filtering (mocked HTTP) ---- |
| 148 | |
| 149 | MOCK_CHANNEL = { |
| 150 | "handle": "testchannel", |
| 151 | "name": "Test Channel", |
| 152 | "subscriber_count": 10000, |
| 153 | } |
| 154 | |
| 155 | MOCK_POSTS_PAGE1 = [ |
| 156 | { |
| 157 | "id": "100", |
| 158 | "channel_handle": "testchannel", |
| 159 | "url": "https://t.me/testchannel/100", |
| 160 | "author_name": "Test Channel", |
| 161 | "text": "Recent post about AI agents", |
| 162 | "published_at": "2026-08-20T10:00:00+00:00", |
| 163 | "view_count": 5000, |
| 164 | "reaction_count": 100, |
| 165 | }, |
| 166 | { |
| 167 | "id": "99", |
| 168 | "channel_handle": "testchannel", |
| 169 | "url": "https://t.me/testchannel/99", |
| 170 | "author_name": "Test Channel", |
| 171 | "text": "Another recent post", |
| 172 | "published_at": "2026-08-19T10:00:00+00:00", |
| 173 | "view_count": 3000, |
| 174 | "reaction_count": 50, |
| 175 | }, |
| 176 | ] |
| 177 | |
| 178 | MOCK_POSTS_PAGE2 = [ |
| 179 | { |
| 180 | "id": "98", |
| 181 | "channel_handle": "testchannel", |
| 182 | "url": "https://t.me/testchannel/98", |
| 183 | "author_name": "Test Channel", |
| 184 | "text": "Old post", |
| 185 | "published_at": "2026-07-01T10:00:00+00:00", |
| 186 | "view_count": 1000, |
| 187 | "reaction_count": 10, |
| 188 | }, |
| 189 | ] |
| 190 | |
| 191 | |
| 192 | def test_pagination_stops_on_old_page(monkeypatch): |
| 193 | pages_fetched = [] |
| 194 | |
| 195 | def mock_get(url, **kwargs): |
| 196 | params = kwargs.get("params", {}) |
| 197 | cursor = params.get("cursor") |
| 198 | pages_fetched.append(cursor) |
| 199 | |
| 200 | if cursor is None: |
| 201 | return { |
| 202 | "success": True, |
| 203 | "channel": MOCK_CHANNEL, |
| 204 | "posts": MOCK_POSTS_PAGE1, |
| 205 | "cursor": "98", |
| 206 | "has_more": True, |
| 207 | } |
| 208 | else: |
| 209 | return { |
| 210 | "success": True, |
| 211 | "channel": MOCK_CHANNEL, |
| 212 | "posts": MOCK_POSTS_PAGE2, |
| 213 | "cursor": "50", |
| 214 | "has_more": True, |
| 215 | } |
| 216 | |
| 217 | monkeypatch.setattr(telegram.http, "get", mock_get) |
| 218 | |
| 219 | result = telegram.search_telegram( |
| 220 | "AI agents", |
| 221 | "2026-08-01", |
| 222 | "2026-08-21", |
| 223 | depth="default", |
| 224 | token="test-key", |
| 225 | config={"TELEGRAM_SOURCES": "testchannel"}, |
| 226 | ) |
| 227 | |
| 228 | assert len(pages_fetched) == 2 |
| 229 | assert len(result["items"]) == 2 |
| 230 | |
| 231 | |
| 232 | def test_date_filtering(monkeypatch): |
| 233 | def mock_get(url, **kwargs): |
| 234 | return { |
| 235 | "success": True, |
| 236 | "channel": MOCK_CHANNEL, |
| 237 | "posts": MOCK_POSTS_PAGE1 + MOCK_POSTS_PAGE2, |
| 238 | "has_more": False, |
| 239 | } |
| 240 | |
| 241 | monkeypatch.setattr(telegram.http, "get", mock_get) |
| 242 | |
| 243 | result = telegram.search_telegram( |
| 244 | "AI agents", |
| 245 | "2026-08-15", |
| 246 | "2026-08-21", |
| 247 | token="test-key", |
| 248 | config={"TELEGRAM_SOURCES": "testchannel"}, |
| 249 | ) |
| 250 | |
| 251 | dates = [item["date"] for item in result["items"]] |
| 252 | assert all(d >= "2026-08-15" for d in dates) |
| 253 | assert "2026-07-01" not in dates |
| 254 | |
| 255 | |
| 256 | def test_topic_relevance_scoring(monkeypatch): |
| 257 | def mock_get(url, **kwargs): |
| 258 | return { |
| 259 | "success": True, |
| 260 | "channel": MOCK_CHANNEL, |
| 261 | "posts": [ |
| 262 | { |
| 263 | "id": "1", |
| 264 | "text": "AI agents are transforming software development", |
| 265 | "published_at": "2026-08-20T10:00:00+00:00", |
| 266 | "view_count": 100, |
| 267 | "reaction_count": 10, |
| 268 | }, |
| 269 | { |
| 270 | "id": "2", |
| 271 | "text": "Unrelated post about cooking recipes", |
| 272 | "published_at": "2026-08-20T10:00:00+00:00", |
| 273 | "view_count": 100, |
| 274 | "reaction_count": 10, |
| 275 | }, |
| 276 | ], |
| 277 | "has_more": False, |
| 278 | } |
| 279 | |
| 280 | monkeypatch.setattr(telegram.http, "get", mock_get) |
| 281 | |
| 282 | result = telegram.search_telegram( |
| 283 | "AI agents", |
| 284 | "2026-08-01", |
| 285 | "2026-08-21", |
| 286 | token="test-key", |
| 287 | config={"TELEGRAM_SOURCES": "testchannel"}, |
| 288 | ) |
| 289 | |
| 290 | items = result["items"] |
| 291 | assert len(items) == 2 |
| 292 | ai_item = next(i for i in items if "AI agents" in i["text"]) |
| 293 | cook_item = next(i for i in items if "cooking" in i["text"]) |
| 294 | assert ai_item["relevance"] > cook_item["relevance"] |
| 295 | |
| 296 | |
| 297 | def test_depth_page_caps(): |
| 298 | assert telegram.DEPTH_PAGE_CAPS["quick"] == 1 |
| 299 | assert telegram.DEPTH_PAGE_CAPS["default"] == 3 |
| 300 | assert telegram.DEPTH_PAGE_CAPS["deep"] == 6 |
| 301 | |
| 302 | |
| 303 | def test_telegram_max_pages_override(monkeypatch): |
| 304 | pages_fetched = [] |
| 305 | |
| 306 | def mock_get(url, **kwargs): |
| 307 | params = kwargs.get("params", {}) |
| 308 | pages_fetched.append(params.get("cursor")) |
| 309 | return { |
| 310 | "success": True, |
| 311 | "channel": MOCK_CHANNEL, |
| 312 | "posts": MOCK_POSTS_PAGE1, |
| 313 | "cursor": str(len(pages_fetched) * 10), |
| 314 | "has_more": True, |
| 315 | } |
| 316 | |
| 317 | monkeypatch.setattr(telegram.http, "get", mock_get) |
| 318 | |
| 319 | result = telegram.search_telegram( |
| 320 | "AI agents", |
| 321 | "2026-08-01", |
| 322 | "2026-08-21", |
| 323 | depth="quick", |
| 324 | token="test-key", |
| 325 | config={ |
| 326 | "TELEGRAM_SOURCES": "testchannel", |
| 327 | "TELEGRAM_MAX_PAGES": "5", |
| 328 | }, |
| 329 | ) |
| 330 | |
| 331 | assert len(pages_fetched) == 5 |
| 332 | |
| 333 | |
| 334 | def test_http_error_gracefully_handled(monkeypatch): |
| 335 | def mock_get(url, **kwargs): |
| 336 | raise telegram.http.HTTPError("Connection failed", status_code=500) |
| 337 | |
| 338 | monkeypatch.setattr(telegram.http, "get", mock_get) |
| 339 | |
| 340 | result = telegram.search_telegram( |
| 341 | "AI agents", |
| 342 | "2026-08-01", |
| 343 | "2026-08-21", |
| 344 | token="test-key", |
| 345 | config={"TELEGRAM_SOURCES": "testchannel"}, |
| 346 | ) |
| 347 | |
| 348 | assert result["items"] == [] |
| 349 | |
| 350 | |
| 351 | def test_api_error_gracefully_handled(monkeypatch): |
| 352 | def mock_get(url, **kwargs): |
| 353 | return {"success": False, "error": "Channel not found"} |
| 354 | |
| 355 | monkeypatch.setattr(telegram.http, "get", mock_get) |
| 356 | |
| 357 | result = telegram.search_telegram( |
| 358 | "AI agents", |
| 359 | "2026-08-01", |
| 360 | "2026-08-21", |
| 361 | token="test-key", |
| 362 | config={"TELEGRAM_SOURCES": "testchannel"}, |
| 363 | ) |
| 364 | |
| 365 | assert result["items"] == [] |
| 366 | |
| 367 | |
| 368 | # ---- MAX_SOURCE_FETCHES cap ---- |
| 369 | |
| 370 | def test_telegram_capped_to_single_fetch(): |
| 371 | assert pipeline.MAX_SOURCE_FETCHES.get("telegram") == 1 |
| 372 | |
| 373 | |
| 374 | # ---- normalization ---- |
| 375 | |
| 376 | def test_normalize_telegram_posts(): |
| 377 | from lib import normalize |
| 378 | |
| 379 | items = [ |
| 380 | { |
| 381 | "id": "123", |
| 382 | "handle": "testchannel", |
| 383 | "display_name": "Test Channel", |
| 384 | "text": "Test post about AI", |
| 385 | "url": "https://t.me/testchannel/123", |
| 386 | "date": "2026-08-20", |
| 387 | "engagement": {"views": 5000, "reactions": 100}, |
| 388 | "relevance": 0.85, |
| 389 | "why_relevant": "Telegram @testchannel: Test post about AI", |
| 390 | }, |
| 391 | ] |
| 392 | |
| 393 | normalized = normalize.normalize_source_items("telegram", items, "2026-08-01", "2026-08-21") |
| 394 | assert len(normalized) == 1 |
| 395 | item = normalized[0] |
| 396 | assert item.source == "telegram" |
| 397 | assert item.item_id == "123" |
| 398 | assert "Test post about AI" in item.title |
| 399 |