crawlo 1.3.3__py3-none-any.whl → 1.3.4__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Potentially problematic release.
This version of crawlo might be problematic. Click here for more details.
- crawlo/__init__.py +87 -63
- crawlo/__version__.py +1 -1
- crawlo/cli.py +75 -75
- crawlo/commands/__init__.py +14 -14
- crawlo/commands/check.py +594 -594
- crawlo/commands/genspider.py +151 -151
- crawlo/commands/help.py +138 -138
- crawlo/commands/list.py +155 -155
- crawlo/commands/run.py +341 -323
- crawlo/commands/startproject.py +436 -436
- crawlo/commands/stats.py +187 -187
- crawlo/commands/utils.py +196 -196
- crawlo/config.py +312 -312
- crawlo/config_validator.py +277 -277
- crawlo/core/__init__.py +46 -2
- crawlo/core/engine.py +439 -365
- crawlo/core/processor.py +40 -40
- crawlo/core/scheduler.py +257 -256
- crawlo/crawler.py +639 -1167
- crawlo/data/__init__.py +5 -5
- crawlo/data/user_agents.py +194 -194
- crawlo/downloader/__init__.py +273 -273
- crawlo/downloader/aiohttp_downloader.py +228 -226
- crawlo/downloader/cffi_downloader.py +245 -245
- crawlo/downloader/httpx_downloader.py +259 -259
- crawlo/downloader/hybrid_downloader.py +212 -212
- crawlo/downloader/playwright_downloader.py +402 -402
- crawlo/downloader/selenium_downloader.py +472 -472
- crawlo/event.py +11 -11
- crawlo/exceptions.py +81 -81
- crawlo/extension/__init__.py +39 -39
- crawlo/extension/health_check.py +141 -141
- crawlo/extension/log_interval.py +57 -57
- crawlo/extension/log_stats.py +81 -81
- crawlo/extension/logging_extension.py +61 -52
- crawlo/extension/memory_monitor.py +104 -104
- crawlo/extension/performance_profiler.py +133 -133
- crawlo/extension/request_recorder.py +107 -107
- crawlo/factories/__init__.py +28 -0
- crawlo/factories/base.py +69 -0
- crawlo/factories/crawler.py +104 -0
- crawlo/factories/registry.py +85 -0
- crawlo/filters/__init__.py +154 -154
- crawlo/filters/aioredis_filter.py +257 -234
- crawlo/filters/memory_filter.py +269 -269
- crawlo/framework.py +292 -0
- crawlo/initialization/__init__.py +40 -0
- crawlo/initialization/built_in.py +426 -0
- crawlo/initialization/context.py +142 -0
- crawlo/initialization/core.py +194 -0
- crawlo/initialization/phases.py +149 -0
- crawlo/initialization/registry.py +146 -0
- crawlo/items/__init__.py +23 -23
- crawlo/items/base.py +23 -22
- crawlo/items/fields.py +52 -52
- crawlo/items/items.py +104 -104
- crawlo/logging/__init__.py +38 -0
- crawlo/logging/config.py +97 -0
- crawlo/logging/factory.py +129 -0
- crawlo/logging/manager.py +112 -0
- crawlo/middleware/__init__.py +21 -21
- crawlo/middleware/default_header.py +132 -132
- crawlo/middleware/download_delay.py +104 -104
- crawlo/middleware/middleware_manager.py +135 -135
- crawlo/middleware/offsite.py +123 -123
- crawlo/middleware/proxy.py +386 -386
- crawlo/middleware/request_ignore.py +86 -86
- crawlo/middleware/response_code.py +163 -163
- crawlo/middleware/response_filter.py +136 -136
- crawlo/middleware/retry.py +124 -124
- crawlo/middleware/simple_proxy.py +65 -65
- crawlo/mode_manager.py +212 -187
- crawlo/network/__init__.py +21 -21
- crawlo/network/request.py +379 -379
- crawlo/network/response.py +359 -359
- crawlo/pipelines/__init__.py +21 -21
- crawlo/pipelines/bloom_dedup_pipeline.py +156 -156
- crawlo/pipelines/console_pipeline.py +39 -39
- crawlo/pipelines/csv_pipeline.py +316 -316
- crawlo/pipelines/database_dedup_pipeline.py +222 -222
- crawlo/pipelines/json_pipeline.py +218 -218
- crawlo/pipelines/memory_dedup_pipeline.py +115 -115
- crawlo/pipelines/mongo_pipeline.py +131 -131
- crawlo/pipelines/mysql_pipeline.py +318 -318
- crawlo/pipelines/pipeline_manager.py +76 -75
- crawlo/pipelines/redis_dedup_pipeline.py +166 -166
- crawlo/project.py +327 -325
- crawlo/queue/pqueue.py +43 -37
- crawlo/queue/queue_manager.py +503 -379
- crawlo/queue/redis_priority_queue.py +326 -306
- crawlo/settings/__init__.py +7 -7
- crawlo/settings/default_settings.py +321 -225
- crawlo/settings/setting_manager.py +214 -198
- crawlo/spider/__init__.py +657 -639
- crawlo/stats_collector.py +73 -59
- crawlo/subscriber.py +129 -129
- crawlo/task_manager.py +139 -30
- crawlo/templates/crawlo.cfg.tmpl +10 -10
- crawlo/templates/project/__init__.py.tmpl +3 -3
- crawlo/templates/project/items.py.tmpl +17 -17
- crawlo/templates/project/middlewares.py.tmpl +118 -118
- crawlo/templates/project/pipelines.py.tmpl +96 -96
- crawlo/templates/project/settings.py.tmpl +168 -267
- crawlo/templates/project/settings_distributed.py.tmpl +167 -180
- crawlo/templates/project/settings_gentle.py.tmpl +167 -61
- crawlo/templates/project/settings_high_performance.py.tmpl +168 -131
- crawlo/templates/project/settings_minimal.py.tmpl +66 -35
- crawlo/templates/project/settings_simple.py.tmpl +165 -102
- crawlo/templates/project/spiders/__init__.py.tmpl +10 -6
- crawlo/templates/run.py.tmpl +34 -38
- crawlo/templates/spider/spider.py.tmpl +143 -143
- crawlo/templates/spiders_init.py.tmpl +10 -0
- crawlo/tools/__init__.py +200 -200
- crawlo/tools/anti_crawler.py +268 -268
- crawlo/tools/authenticated_proxy.py +240 -240
- crawlo/tools/data_formatter.py +225 -225
- crawlo/tools/data_validator.py +180 -180
- crawlo/tools/date_tools.py +289 -289
- crawlo/tools/distributed_coordinator.py +388 -388
- crawlo/tools/encoding_converter.py +127 -127
- crawlo/tools/network_diagnostic.py +365 -0
- crawlo/tools/request_tools.py +82 -82
- crawlo/tools/retry_mechanism.py +224 -224
- crawlo/tools/scenario_adapter.py +262 -262
- crawlo/tools/text_cleaner.py +232 -232
- crawlo/utils/__init__.py +34 -34
- crawlo/utils/batch_processor.py +259 -259
- crawlo/utils/class_loader.py +26 -0
- crawlo/utils/controlled_spider_mixin.py +439 -439
- crawlo/utils/db_helper.py +343 -343
- crawlo/utils/enhanced_error_handler.py +356 -356
- crawlo/utils/env_config.py +142 -142
- crawlo/utils/error_handler.py +165 -124
- crawlo/utils/func_tools.py +82 -82
- crawlo/utils/large_scale_config.py +286 -286
- crawlo/utils/large_scale_helper.py +344 -344
- crawlo/utils/log.py +44 -200
- crawlo/utils/performance_monitor.py +285 -285
- crawlo/utils/queue_helper.py +175 -175
- crawlo/utils/redis_connection_pool.py +388 -351
- crawlo/utils/redis_key_validator.py +198 -198
- crawlo/utils/request.py +267 -267
- crawlo/utils/request_serializer.py +225 -218
- crawlo/utils/spider_loader.py +61 -61
- crawlo/utils/system.py +11 -11
- crawlo/utils/tools.py +4 -4
- crawlo/utils/url.py +39 -39
- {crawlo-1.3.3.dist-info → crawlo-1.3.4.dist-info}/METADATA +1126 -1020
- crawlo-1.3.4.dist-info/RECORD +278 -0
- examples/__init__.py +7 -7
- tests/__init__.py +7 -7
- tests/advanced_tools_example.py +275 -275
- tests/authenticated_proxy_example.py +107 -107
- tests/baidu_performance_test.py +109 -0
- tests/baidu_test.py +60 -0
- tests/cleaners_example.py +160 -160
- tests/comprehensive_framework_test.py +213 -0
- tests/comprehensive_test.py +82 -0
- tests/comprehensive_testing_summary.md +187 -0
- tests/config_validation_demo.py +142 -142
- tests/controlled_spider_example.py +205 -205
- tests/date_tools_example.py +180 -180
- tests/debug_configure.py +70 -0
- tests/debug_framework_logger.py +85 -0
- tests/debug_log_levels.py +64 -0
- tests/debug_pipelines.py +66 -66
- tests/distributed_test.py +67 -0
- tests/distributed_test_debug.py +77 -0
- tests/dynamic_loading_example.py +523 -523
- tests/dynamic_loading_test.py +104 -104
- tests/env_config_example.py +133 -133
- tests/error_handling_example.py +171 -171
- tests/final_command_test_report.md +0 -0
- tests/final_comprehensive_test.py +152 -0
- tests/final_validation_test.py +183 -0
- tests/framework_performance_test.py +203 -0
- tests/optimized_performance_test.py +212 -0
- tests/performance_comparison.py +246 -0
- tests/queue_blocking_test.py +114 -0
- tests/queue_test.py +90 -0
- tests/redis_key_validation_demo.py +130 -130
- tests/request_params_example.py +150 -150
- tests/response_improvements_example.py +144 -144
- tests/scrapy_comparison/ofweek_scrapy.py +139 -0
- tests/scrapy_comparison/scrapy_test.py +134 -0
- tests/simple_command_test.py +120 -0
- tests/simple_crawlo_test.py +128 -0
- tests/simple_log_test.py +58 -0
- tests/simple_optimization_test.py +129 -0
- tests/simple_spider_test.py +50 -0
- tests/simple_test.py +48 -0
- tests/test_advanced_tools.py +148 -148
- tests/test_all_commands.py +231 -0
- tests/test_all_redis_key_configs.py +145 -145
- tests/test_authenticated_proxy.py +141 -141
- tests/test_batch_processor.py +179 -0
- tests/test_cleaners.py +54 -54
- tests/test_component_factory.py +175 -0
- tests/test_comprehensive.py +146 -146
- tests/test_config_consistency.py +80 -80
- tests/test_config_merge.py +152 -152
- tests/test_config_validator.py +182 -182
- tests/test_controlled_spider_mixin.py +80 -0
- tests/test_crawlo_proxy_integration.py +108 -108
- tests/test_date_tools.py +123 -123
- tests/test_default_header_middleware.py +158 -158
- tests/test_distributed.py +65 -65
- tests/test_double_crawlo_fix.py +207 -207
- tests/test_double_crawlo_fix_simple.py +124 -124
- tests/test_download_delay_middleware.py +221 -221
- tests/test_downloader_proxy_compatibility.py +268 -268
- tests/test_dynamic_downloaders_proxy.py +124 -124
- tests/test_dynamic_proxy.py +92 -92
- tests/test_dynamic_proxy_config.py +146 -146
- tests/test_dynamic_proxy_real.py +109 -109
- tests/test_edge_cases.py +303 -303
- tests/test_enhanced_error_handler.py +270 -270
- tests/test_enhanced_error_handler_comprehensive.py +246 -0
- tests/test_env_config.py +121 -121
- tests/test_error_handler_compatibility.py +112 -112
- tests/test_factories.py +253 -0
- tests/test_final_validation.py +153 -153
- tests/test_framework_env_usage.py +103 -103
- tests/test_framework_logger.py +67 -0
- tests/test_framework_startup.py +65 -0
- tests/test_integration.py +169 -169
- tests/test_item_dedup_redis_key.py +122 -122
- tests/test_large_scale_config.py +113 -0
- tests/test_large_scale_helper.py +236 -0
- tests/test_mode_change.py +73 -0
- tests/test_mode_consistency.py +51 -51
- tests/test_offsite_middleware.py +221 -221
- tests/test_parsel.py +29 -29
- tests/test_performance.py +327 -327
- tests/test_performance_monitor.py +116 -0
- tests/test_proxy_api.py +264 -264
- tests/test_proxy_health_check.py +32 -32
- tests/test_proxy_middleware.py +121 -121
- tests/test_proxy_middleware_enhanced.py +216 -216
- tests/test_proxy_middleware_integration.py +136 -136
- tests/test_proxy_middleware_refactored.py +184 -184
- tests/test_proxy_providers.py +56 -56
- tests/test_proxy_stats.py +19 -19
- tests/test_proxy_strategies.py +59 -59
- tests/test_queue_empty_check.py +42 -0
- tests/test_queue_manager_double_crawlo.py +173 -173
- tests/test_queue_manager_redis_key.py +176 -176
- tests/test_random_user_agent.py +72 -72
- tests/test_real_scenario_proxy.py +195 -195
- tests/test_redis_config.py +28 -28
- tests/test_redis_connection_pool.py +294 -294
- tests/test_redis_key_naming.py +181 -181
- tests/test_redis_key_validator.py +123 -123
- tests/test_redis_queue.py +224 -224
- tests/test_request_ignore_middleware.py +182 -182
- tests/test_request_params.py +111 -111
- tests/test_request_serialization.py +70 -70
- tests/test_response_code_middleware.py +349 -349
- tests/test_response_filter_middleware.py +427 -427
- tests/test_response_improvements.py +152 -152
- tests/test_retry_middleware.py +241 -241
- tests/test_scheduler.py +252 -252
- tests/test_scheduler_config_update.py +133 -133
- tests/test_simple_response.py +61 -61
- tests/test_telecom_spider_redis_key.py +205 -205
- tests/test_template_content.py +87 -87
- tests/test_template_redis_key.py +134 -134
- tests/test_tools.py +159 -159
- tests/test_user_agents.py +96 -96
- tests/tools_example.py +260 -260
- tests/untested_features_report.md +139 -0
- tests/verify_debug.py +52 -0
- tests/verify_distributed.py +117 -117
- tests/verify_log_fix.py +112 -0
- crawlo-1.3.3.dist-info/RECORD +0 -219
- tests/DOUBLE_CRAWLO_PREFIX_FIX_REPORT.md +0 -82
- {crawlo-1.3.3.dist-info → crawlo-1.3.4.dist-info}/WHEEL +0 -0
- {crawlo-1.3.3.dist-info → crawlo-1.3.4.dist-info}/entry_points.txt +0 -0
- {crawlo-1.3.3.dist-info → crawlo-1.3.4.dist-info}/top_level.txt +0 -0
crawlo/stats_collector.py
CHANGED
|
@@ -1,59 +1,73 @@
|
|
|
1
|
-
#!/usr/bin/python
|
|
2
|
-
# -*- coding:UTF-8 -*-
|
|
3
|
-
"""
|
|
4
|
-
# @Time : 2025-05-17 09:57
|
|
5
|
-
# @Author : crawl-coder
|
|
6
|
-
# @Desc : 统计信息收集器
|
|
7
|
-
"""
|
|
8
|
-
from pprint import pformat
|
|
9
|
-
from crawlo.utils.log import get_logger
|
|
10
|
-
|
|
11
|
-
|
|
12
|
-
class StatsCollector(object):
|
|
13
|
-
|
|
14
|
-
def __init__(self, crawler):
|
|
15
|
-
self.crawler = crawler
|
|
16
|
-
self._dump = self.crawler.settings.get_bool('STATS_DUMP')
|
|
17
|
-
self._stats = {}
|
|
18
|
-
self.logger = get_logger(self.__class__.__name__, "INFO")
|
|
19
|
-
|
|
20
|
-
def inc_value(self, key, count=1, start=0):
|
|
21
|
-
self._stats[key] = self._stats.setdefault(key, start) + count
|
|
22
|
-
|
|
23
|
-
def get_value(self, key, default=None):
|
|
24
|
-
return self._stats.get(key, default)
|
|
25
|
-
|
|
26
|
-
def get_stats(self):
|
|
27
|
-
return self._stats
|
|
28
|
-
|
|
29
|
-
def set_stats(self, stats):
|
|
30
|
-
self._stats = stats
|
|
31
|
-
|
|
32
|
-
def clear_stats(self):
|
|
33
|
-
self._stats.clear()
|
|
34
|
-
|
|
35
|
-
def close_spider(self, spider, reason):
|
|
36
|
-
self._stats['reason'] = reason
|
|
37
|
-
|
|
38
|
-
# 首选:使用 spider.name
|
|
39
|
-
# 次选:使用实例的类名
|
|
40
|
-
# 最后:使用一个完全未知的占位符
|
|
41
|
-
spider_name = (
|
|
42
|
-
getattr(spider, 'name', None) or
|
|
43
|
-
spider.__class__.__name__ or
|
|
44
|
-
'<Unknown>'
|
|
45
|
-
)
|
|
46
|
-
|
|
47
|
-
self._stats['spider_name'] = spider_name
|
|
48
|
-
|
|
49
|
-
|
|
50
|
-
|
|
51
|
-
|
|
52
|
-
def
|
|
53
|
-
|
|
54
|
-
|
|
55
|
-
def
|
|
56
|
-
self._stats[key]
|
|
57
|
-
|
|
58
|
-
def
|
|
59
|
-
|
|
1
|
+
#!/usr/bin/python
|
|
2
|
+
# -*- coding:UTF-8 -*-
|
|
3
|
+
"""
|
|
4
|
+
# @Time : 2025-05-17 09:57
|
|
5
|
+
# @Author : crawl-coder
|
|
6
|
+
# @Desc : 统计信息收集器
|
|
7
|
+
"""
|
|
8
|
+
from pprint import pformat
|
|
9
|
+
from crawlo.utils.log import get_logger
|
|
10
|
+
|
|
11
|
+
|
|
12
|
+
class StatsCollector(object):
|
|
13
|
+
|
|
14
|
+
def __init__(self, crawler):
|
|
15
|
+
self.crawler = crawler
|
|
16
|
+
self._dump = self.crawler.settings.get_bool('STATS_DUMP')
|
|
17
|
+
self._stats = {}
|
|
18
|
+
self.logger = get_logger(self.__class__.__name__, "INFO")
|
|
19
|
+
|
|
20
|
+
def inc_value(self, key, count=1, start=0):
|
|
21
|
+
self._stats[key] = self._stats.setdefault(key, start) + count
|
|
22
|
+
|
|
23
|
+
def get_value(self, key, default=None):
|
|
24
|
+
return self._stats.get(key, default)
|
|
25
|
+
|
|
26
|
+
def get_stats(self):
|
|
27
|
+
return self._stats
|
|
28
|
+
|
|
29
|
+
def set_stats(self, stats):
|
|
30
|
+
self._stats = stats
|
|
31
|
+
|
|
32
|
+
def clear_stats(self):
|
|
33
|
+
self._stats.clear()
|
|
34
|
+
|
|
35
|
+
def close_spider(self, spider, reason):
|
|
36
|
+
self._stats['reason'] = reason
|
|
37
|
+
|
|
38
|
+
# 首选:使用 spider.name
|
|
39
|
+
# 次选:使用实例的类名
|
|
40
|
+
# 最后:使用一个完全未知的占位符
|
|
41
|
+
spider_name = (
|
|
42
|
+
getattr(spider, 'name', None) or
|
|
43
|
+
spider.__class__.__name__ or
|
|
44
|
+
'<Unknown>'
|
|
45
|
+
)
|
|
46
|
+
|
|
47
|
+
self._stats['spider_name'] = spider_name
|
|
48
|
+
|
|
49
|
+
def __getitem__(self, item):
|
|
50
|
+
return self._stats[item]
|
|
51
|
+
|
|
52
|
+
def __setitem__(self, key, value):
|
|
53
|
+
self._stats[key] = value
|
|
54
|
+
|
|
55
|
+
def __delitem__(self, key):
|
|
56
|
+
del self._stats[key]
|
|
57
|
+
|
|
58
|
+
def close(self):
|
|
59
|
+
"""关闭统计收集器并输出统计信息"""
|
|
60
|
+
if self._dump:
|
|
61
|
+
# 获取爬虫名称
|
|
62
|
+
spider_name = self._stats.get('spider_name', 'unknown')
|
|
63
|
+
|
|
64
|
+
# 如果还没有设置爬虫名称,尝试从crawler中获取
|
|
65
|
+
if spider_name == 'unknown' and hasattr(self, 'crawler') and self.crawler:
|
|
66
|
+
spider = getattr(self.crawler, 'spider', None)
|
|
67
|
+
if spider and hasattr(spider, 'name'):
|
|
68
|
+
spider_name = spider.name
|
|
69
|
+
# 同时更新_stats中的spider_name
|
|
70
|
+
self._stats['spider_name'] = spider_name
|
|
71
|
+
|
|
72
|
+
# 输出统计信息(这是唯一输出统计信息的地方)
|
|
73
|
+
self.logger.info(f'{spider_name} stats: \n{pformat(self._stats)}')
|
crawlo/subscriber.py
CHANGED
|
@@ -1,130 +1,130 @@
|
|
|
1
|
-
#!/usr/bin/python
|
|
2
|
-
# -*- coding:UTF-8 -*-
|
|
3
|
-
import asyncio
|
|
4
|
-
from collections import defaultdict
|
|
5
|
-
from inspect import iscoroutinefunction
|
|
6
|
-
from typing import Dict, Callable, Coroutine, Any, TypeAlias, List, Tuple
|
|
7
|
-
|
|
8
|
-
|
|
9
|
-
class ReceiverTypeError(TypeError):
|
|
10
|
-
"""当订阅的接收者不是一个协程函数时抛出。"""
|
|
11
|
-
pass
|
|
12
|
-
|
|
13
|
-
|
|
14
|
-
ReceiverCoroutine: TypeAlias = Callable[..., Coroutine[Any, Any, Any]]
|
|
15
|
-
|
|
16
|
-
|
|
17
|
-
class Subscriber:
|
|
18
|
-
"""
|
|
19
|
-
一个支持异步协程的发布/订阅(Pub/Sub)模式实现。
|
|
20
|
-
|
|
21
|
-
这个类允许你注册(订阅)协程函数来监听特定事件,并在事件发生时
|
|
22
|
-
以并发的方式异步地通知所有订阅者。
|
|
23
|
-
"""
|
|
24
|
-
|
|
25
|
-
def __init__(self):
|
|
26
|
-
"""初始化一个空的订阅者字典。"""
|
|
27
|
-
# 使用弱引用字典避免内存泄漏
|
|
28
|
-
self._subscribers: Dict[str, Dict[ReceiverCoroutine, int]] = defaultdict(dict)
|
|
29
|
-
# 用于缓存排序后的订阅者列表,提高频繁事件的处理性能
|
|
30
|
-
self._sorted_subscribers_cache: Dict[str, List[Tuple[ReceiverCoroutine, int]]] = {}
|
|
31
|
-
|
|
32
|
-
def subscribe(self, receiver: ReceiverCoroutine, *, event: str, priority: int = 0) -> None:
|
|
33
|
-
"""
|
|
34
|
-
订阅一个事件。
|
|
35
|
-
|
|
36
|
-
Args:
|
|
37
|
-
receiver: 一个协程函数 (例如 async def my_func(...))。
|
|
38
|
-
event: 要订阅的事件名称。
|
|
39
|
-
priority: 订阅者优先级,数值越小优先级越高,默认为0。
|
|
40
|
-
|
|
41
|
-
Raises:
|
|
42
|
-
ReceiverTypeError: 如果提供的 `receiver` 不是一个协程函数。
|
|
43
|
-
"""
|
|
44
|
-
if not iscoroutinefunction(receiver):
|
|
45
|
-
raise ReceiverTypeError(f"接收者 '{receiver.__qualname__}' 必须是一个协程函数。")
|
|
46
|
-
|
|
47
|
-
# 使用弱引用避免内存泄漏
|
|
48
|
-
self._subscribers[event][receiver] = priority
|
|
49
|
-
# 清除缓存
|
|
50
|
-
self._sorted_subscribers_cache.pop(event, None)
|
|
51
|
-
|
|
52
|
-
def unsubscribe(self, receiver: ReceiverCoroutine, *, event: str) -> None:
|
|
53
|
-
"""
|
|
54
|
-
取消订阅一个事件。
|
|
55
|
-
|
|
56
|
-
如果事件或接收者不存在,将静默处理。
|
|
57
|
-
|
|
58
|
-
Args:
|
|
59
|
-
receiver: 要取消订阅的协程函数。
|
|
60
|
-
event: 事件名称。
|
|
61
|
-
"""
|
|
62
|
-
if event in self._subscribers:
|
|
63
|
-
self._subscribers[event].pop(receiver, None)
|
|
64
|
-
# 清除缓存
|
|
65
|
-
self._sorted_subscribers_cache.pop(event, None)
|
|
66
|
-
|
|
67
|
-
def _get_sorted_subscribers(self, event: str) -> List[Tuple[ReceiverCoroutine, int]]:
|
|
68
|
-
"""
|
|
69
|
-
获取按优先级排序的订阅者列表。
|
|
70
|
-
|
|
71
|
-
Args:
|
|
72
|
-
event: 事件名称。
|
|
73
|
-
|
|
74
|
-
Returns:
|
|
75
|
-
按优先级排序的订阅者列表。
|
|
76
|
-
"""
|
|
77
|
-
# 检查缓存
|
|
78
|
-
if event in self._sorted_subscribers_cache:
|
|
79
|
-
return self._sorted_subscribers_cache[event]
|
|
80
|
-
|
|
81
|
-
# 获取有效的订阅者(使用弱引用检查)
|
|
82
|
-
valid_subscribers = {}
|
|
83
|
-
for receiver, priority in list(self._subscribers[event].items()):
|
|
84
|
-
# 检查弱引用是否仍然有效
|
|
85
|
-
if isinstance(receiver, Callable):
|
|
86
|
-
valid_subscribers[receiver] = priority
|
|
87
|
-
|
|
88
|
-
# 更新订阅者字典
|
|
89
|
-
self._subscribers[event] = valid_subscribers
|
|
90
|
-
|
|
91
|
-
# 按优先级排序(数值小的优先级高)
|
|
92
|
-
sorted_subscribers = sorted(valid_subscribers.items(), key=lambda x: x[1])
|
|
93
|
-
# 缓存结果
|
|
94
|
-
self._sorted_subscribers_cache[event] = sorted_subscribers
|
|
95
|
-
|
|
96
|
-
return sorted_subscribers
|
|
97
|
-
|
|
98
|
-
async def notify(self, event: str, *args, **kwargs) -> List[Any]:
|
|
99
|
-
"""
|
|
100
|
-
异步地、并发地通知所有订阅了该事件的接收者。
|
|
101
|
-
|
|
102
|
-
此方法会等待所有订阅者任务完成后再返回,并收集所有结果或异常。
|
|
103
|
-
订阅者按优先级顺序执行,优先级高的先执行。
|
|
104
|
-
|
|
105
|
-
Args:
|
|
106
|
-
event: 要触发的事件名称。
|
|
107
|
-
*args: 传递给接收者的位置参数。
|
|
108
|
-
**kwargs: 传递给接收者的关键字参数。
|
|
109
|
-
|
|
110
|
-
Returns:
|
|
111
|
-
一个列表,包含每个订阅者任务的返回结果或在执行期间捕获的异常。
|
|
112
|
-
"""
|
|
113
|
-
sorted_subscribers = self._get_sorted_subscribers(event)
|
|
114
|
-
if not sorted_subscribers:
|
|
115
|
-
return []
|
|
116
|
-
|
|
117
|
-
# 为频繁触发的事件重用任务对象以提高性能
|
|
118
|
-
tasks = []
|
|
119
|
-
for receiver, _ in sorted_subscribers:
|
|
120
|
-
try:
|
|
121
|
-
# 创建任务并添加到列表
|
|
122
|
-
task = asyncio.create_task(receiver(*args, **kwargs))
|
|
123
|
-
tasks.append(task)
|
|
124
|
-
except Exception as e:
|
|
125
|
-
# 如果创建任务失败,记录异常并继续处理其他订阅者
|
|
126
|
-
tasks.append(asyncio.Future()) # 添加一个已完成的Future表示错误
|
|
127
|
-
tasks[-1].set_exception(e)
|
|
128
|
-
|
|
129
|
-
# 并发执行所有任务并返回结果列表(包括异常)
|
|
1
|
+
#!/usr/bin/python
|
|
2
|
+
# -*- coding:UTF-8 -*-
|
|
3
|
+
import asyncio
|
|
4
|
+
from collections import defaultdict
|
|
5
|
+
from inspect import iscoroutinefunction
|
|
6
|
+
from typing import Dict, Callable, Coroutine, Any, TypeAlias, List, Tuple
|
|
7
|
+
|
|
8
|
+
|
|
9
|
+
class ReceiverTypeError(TypeError):
|
|
10
|
+
"""当订阅的接收者不是一个协程函数时抛出。"""
|
|
11
|
+
pass
|
|
12
|
+
|
|
13
|
+
|
|
14
|
+
ReceiverCoroutine: TypeAlias = Callable[..., Coroutine[Any, Any, Any]]
|
|
15
|
+
|
|
16
|
+
|
|
17
|
+
class Subscriber:
|
|
18
|
+
"""
|
|
19
|
+
一个支持异步协程的发布/订阅(Pub/Sub)模式实现。
|
|
20
|
+
|
|
21
|
+
这个类允许你注册(订阅)协程函数来监听特定事件,并在事件发生时
|
|
22
|
+
以并发的方式异步地通知所有订阅者。
|
|
23
|
+
"""
|
|
24
|
+
|
|
25
|
+
def __init__(self):
|
|
26
|
+
"""初始化一个空的订阅者字典。"""
|
|
27
|
+
# 使用弱引用字典避免内存泄漏
|
|
28
|
+
self._subscribers: Dict[str, Dict[ReceiverCoroutine, int]] = defaultdict(dict)
|
|
29
|
+
# 用于缓存排序后的订阅者列表,提高频繁事件的处理性能
|
|
30
|
+
self._sorted_subscribers_cache: Dict[str, List[Tuple[ReceiverCoroutine, int]]] = {}
|
|
31
|
+
|
|
32
|
+
def subscribe(self, receiver: ReceiverCoroutine, *, event: str, priority: int = 0) -> None:
|
|
33
|
+
"""
|
|
34
|
+
订阅一个事件。
|
|
35
|
+
|
|
36
|
+
Args:
|
|
37
|
+
receiver: 一个协程函数 (例如 async def my_func(...))。
|
|
38
|
+
event: 要订阅的事件名称。
|
|
39
|
+
priority: 订阅者优先级,数值越小优先级越高,默认为0。
|
|
40
|
+
|
|
41
|
+
Raises:
|
|
42
|
+
ReceiverTypeError: 如果提供的 `receiver` 不是一个协程函数。
|
|
43
|
+
"""
|
|
44
|
+
if not iscoroutinefunction(receiver):
|
|
45
|
+
raise ReceiverTypeError(f"接收者 '{receiver.__qualname__}' 必须是一个协程函数。")
|
|
46
|
+
|
|
47
|
+
# 使用弱引用避免内存泄漏
|
|
48
|
+
self._subscribers[event][receiver] = priority
|
|
49
|
+
# 清除缓存
|
|
50
|
+
self._sorted_subscribers_cache.pop(event, None)
|
|
51
|
+
|
|
52
|
+
def unsubscribe(self, receiver: ReceiverCoroutine, *, event: str) -> None:
|
|
53
|
+
"""
|
|
54
|
+
取消订阅一个事件。
|
|
55
|
+
|
|
56
|
+
如果事件或接收者不存在,将静默处理。
|
|
57
|
+
|
|
58
|
+
Args:
|
|
59
|
+
receiver: 要取消订阅的协程函数。
|
|
60
|
+
event: 事件名称。
|
|
61
|
+
"""
|
|
62
|
+
if event in self._subscribers:
|
|
63
|
+
self._subscribers[event].pop(receiver, None)
|
|
64
|
+
# 清除缓存
|
|
65
|
+
self._sorted_subscribers_cache.pop(event, None)
|
|
66
|
+
|
|
67
|
+
def _get_sorted_subscribers(self, event: str) -> List[Tuple[ReceiverCoroutine, int]]:
|
|
68
|
+
"""
|
|
69
|
+
获取按优先级排序的订阅者列表。
|
|
70
|
+
|
|
71
|
+
Args:
|
|
72
|
+
event: 事件名称。
|
|
73
|
+
|
|
74
|
+
Returns:
|
|
75
|
+
按优先级排序的订阅者列表。
|
|
76
|
+
"""
|
|
77
|
+
# 检查缓存
|
|
78
|
+
if event in self._sorted_subscribers_cache:
|
|
79
|
+
return self._sorted_subscribers_cache[event]
|
|
80
|
+
|
|
81
|
+
# 获取有效的订阅者(使用弱引用检查)
|
|
82
|
+
valid_subscribers = {}
|
|
83
|
+
for receiver, priority in list(self._subscribers[event].items()):
|
|
84
|
+
# 检查弱引用是否仍然有效
|
|
85
|
+
if isinstance(receiver, Callable):
|
|
86
|
+
valid_subscribers[receiver] = priority
|
|
87
|
+
|
|
88
|
+
# 更新订阅者字典
|
|
89
|
+
self._subscribers[event] = valid_subscribers
|
|
90
|
+
|
|
91
|
+
# 按优先级排序(数值小的优先级高)
|
|
92
|
+
sorted_subscribers = sorted(valid_subscribers.items(), key=lambda x: x[1])
|
|
93
|
+
# 缓存结果
|
|
94
|
+
self._sorted_subscribers_cache[event] = sorted_subscribers
|
|
95
|
+
|
|
96
|
+
return sorted_subscribers
|
|
97
|
+
|
|
98
|
+
async def notify(self, event: str, *args, **kwargs) -> List[Any]:
|
|
99
|
+
"""
|
|
100
|
+
异步地、并发地通知所有订阅了该事件的接收者。
|
|
101
|
+
|
|
102
|
+
此方法会等待所有订阅者任务完成后再返回,并收集所有结果或异常。
|
|
103
|
+
订阅者按优先级顺序执行,优先级高的先执行。
|
|
104
|
+
|
|
105
|
+
Args:
|
|
106
|
+
event: 要触发的事件名称。
|
|
107
|
+
*args: 传递给接收者的位置参数。
|
|
108
|
+
**kwargs: 传递给接收者的关键字参数。
|
|
109
|
+
|
|
110
|
+
Returns:
|
|
111
|
+
一个列表,包含每个订阅者任务的返回结果或在执行期间捕获的异常。
|
|
112
|
+
"""
|
|
113
|
+
sorted_subscribers = self._get_sorted_subscribers(event)
|
|
114
|
+
if not sorted_subscribers:
|
|
115
|
+
return []
|
|
116
|
+
|
|
117
|
+
# 为频繁触发的事件重用任务对象以提高性能
|
|
118
|
+
tasks = []
|
|
119
|
+
for receiver, _ in sorted_subscribers:
|
|
120
|
+
try:
|
|
121
|
+
# 创建任务并添加到列表
|
|
122
|
+
task = asyncio.create_task(receiver(*args, **kwargs))
|
|
123
|
+
tasks.append(task)
|
|
124
|
+
except Exception as e:
|
|
125
|
+
# 如果创建任务失败,记录异常并继续处理其他订阅者
|
|
126
|
+
tasks.append(asyncio.Future()) # 添加一个已完成的Future表示错误
|
|
127
|
+
tasks[-1].set_exception(e)
|
|
128
|
+
|
|
129
|
+
# 并发执行所有任务并返回结果列表(包括异常)
|
|
130
130
|
return await asyncio.gather(*tasks, return_exceptions=True)
|
crawlo/task_manager.py
CHANGED
|
@@ -1,30 +1,139 @@
|
|
|
1
|
-
#!/usr/bin/python
|
|
2
|
-
# -*- coding:UTF-8 -*-
|
|
3
|
-
import asyncio
|
|
4
|
-
from asyncio import Task, Future, Semaphore
|
|
5
|
-
from typing import Set, Final
|
|
6
|
-
|
|
7
|
-
|
|
8
|
-
|
|
9
|
-
|
|
10
|
-
|
|
11
|
-
|
|
12
|
-
|
|
13
|
-
|
|
14
|
-
|
|
15
|
-
|
|
16
|
-
|
|
17
|
-
|
|
18
|
-
|
|
19
|
-
self.
|
|
20
|
-
|
|
21
|
-
|
|
22
|
-
|
|
23
|
-
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
|
|
27
|
-
|
|
28
|
-
|
|
29
|
-
|
|
30
|
-
|
|
1
|
+
#!/usr/bin/python
|
|
2
|
+
# -*- coding:UTF-8 -*-
|
|
3
|
+
import asyncio
|
|
4
|
+
from asyncio import Task, Future, Semaphore
|
|
5
|
+
from typing import Set, Final
|
|
6
|
+
from collections import deque
|
|
7
|
+
import time
|
|
8
|
+
from crawlo.utils.log import get_logger
|
|
9
|
+
|
|
10
|
+
|
|
11
|
+
class DynamicSemaphore(Semaphore):
|
|
12
|
+
"""支持动态调整的信号量"""
|
|
13
|
+
|
|
14
|
+
def __init__(self, initial_value: int = 8):
|
|
15
|
+
super().__init__(initial_value)
|
|
16
|
+
self._initial_value = initial_value
|
|
17
|
+
self._current_value = initial_value
|
|
18
|
+
self._response_times = deque(maxlen=10) # 存储最近10次响应时间
|
|
19
|
+
self._last_adjust_time = time.time()
|
|
20
|
+
|
|
21
|
+
def record_response_time(self, response_time: float):
|
|
22
|
+
"""记录响应时间"""
|
|
23
|
+
self._response_times.append(response_time)
|
|
24
|
+
|
|
25
|
+
def adjust_concurrency(self):
|
|
26
|
+
"""根据响应时间动态调整并发数"""
|
|
27
|
+
current_time = time.time()
|
|
28
|
+
# 限制调整频率,至少间隔1秒(从2秒减少到1秒)
|
|
29
|
+
if current_time - self._last_adjust_time < 1:
|
|
30
|
+
return
|
|
31
|
+
|
|
32
|
+
self._last_adjust_time = current_time
|
|
33
|
+
|
|
34
|
+
if len(self._response_times) < 2: # 从3减少到2
|
|
35
|
+
return
|
|
36
|
+
|
|
37
|
+
# 计算平均响应时间
|
|
38
|
+
avg_response_time = sum(self._response_times) / len(self._response_times)
|
|
39
|
+
|
|
40
|
+
# 根据响应时间调整并发数
|
|
41
|
+
if avg_response_time < 0.2: # 响应很快,增加并发(从0.3降到0.2)
|
|
42
|
+
new_concurrency = min(self._current_value + 5, self._initial_value * 3) # 增加幅度从3提高到5,最大值从2倍提高到3倍
|
|
43
|
+
elif avg_response_time > 1.0: # 响应较慢,减少并发(从1.5降到1.0)
|
|
44
|
+
new_concurrency = max(self._current_value - 5, max(1, self._initial_value // 3)) # 减少幅度从3提高到5,最小值从一半降低到三分之一
|
|
45
|
+
else:
|
|
46
|
+
return # 保持当前并发数
|
|
47
|
+
|
|
48
|
+
# 只有当变化较大时才调整
|
|
49
|
+
if abs(new_concurrency - self._current_value) > 1:
|
|
50
|
+
self._adjust_semaphore_value(new_concurrency)
|
|
51
|
+
|
|
52
|
+
def _adjust_semaphore_value(self, new_value: int):
|
|
53
|
+
"""调整信号量的值"""
|
|
54
|
+
if new_value > self._current_value:
|
|
55
|
+
# 增加信号量
|
|
56
|
+
for _ in range(new_value - self._current_value):
|
|
57
|
+
self.release()
|
|
58
|
+
elif new_value < self._current_value:
|
|
59
|
+
# 减少信号量,这里只是记录新的目标值
|
|
60
|
+
# 实际减少会在acquire时处理
|
|
61
|
+
pass
|
|
62
|
+
|
|
63
|
+
self._current_value = new_value
|
|
64
|
+
# 注意:Python的Semaphore没有直接修改内部计数器的方法
|
|
65
|
+
# 所以我们只能通过release()来增加,减少则需要在acquire时控制
|
|
66
|
+
|
|
67
|
+
|
|
68
|
+
class TaskManager:
|
|
69
|
+
|
|
70
|
+
def __init__(self, total_concurrency: int = 8):
|
|
71
|
+
self.current_task: Final[Set] = set()
|
|
72
|
+
# 使用动态信号量替代普通信号量
|
|
73
|
+
self.semaphore: DynamicSemaphore = DynamicSemaphore(max(1, total_concurrency))
|
|
74
|
+
self.logger = get_logger(self.__class__.__name__)
|
|
75
|
+
|
|
76
|
+
# 异常统计
|
|
77
|
+
self._exception_count = 0
|
|
78
|
+
self._total_tasks = 0
|
|
79
|
+
|
|
80
|
+
async def create_task(self, coroutine) -> Task:
|
|
81
|
+
# 等待信号量,控制并发数
|
|
82
|
+
await self.semaphore.acquire()
|
|
83
|
+
|
|
84
|
+
task = asyncio.create_task(coroutine)
|
|
85
|
+
self.current_task.add(task)
|
|
86
|
+
self._total_tasks += 1
|
|
87
|
+
|
|
88
|
+
def done_callback(_future: Future) -> None:
|
|
89
|
+
try:
|
|
90
|
+
self.current_task.discard(task) # 使用discard而不是remove,避免KeyError
|
|
91
|
+
|
|
92
|
+
# 获取任务结果或异常 - 这是关键,必须调用result()或exception()来"获取"异常
|
|
93
|
+
try:
|
|
94
|
+
# 尝试获取结果,如果有异常会被抛出
|
|
95
|
+
result = _future.result()
|
|
96
|
+
# 如果成功完成,可以在这里记录成功统计
|
|
97
|
+
except Exception as exception:
|
|
98
|
+
# 异常被正确"获取"了,不会再出现"never retrieved"警告
|
|
99
|
+
self._exception_count += 1
|
|
100
|
+
|
|
101
|
+
# 记录异常详情
|
|
102
|
+
self.logger.error(
|
|
103
|
+
f"Task completed with exception: {type(exception).__name__}: {exception}"
|
|
104
|
+
)
|
|
105
|
+
self.logger.debug("Task exception details:", exc_info=exception)
|
|
106
|
+
|
|
107
|
+
# 可以在这里添加更多的异常处理逻辑,如发送到监控系统
|
|
108
|
+
|
|
109
|
+
except Exception as e:
|
|
110
|
+
# 防止回调函数本身出现异常
|
|
111
|
+
self.logger.error(f"Error in task done callback: {e}")
|
|
112
|
+
finally:
|
|
113
|
+
# 确保信号量始终被释放
|
|
114
|
+
self.semaphore.release()
|
|
115
|
+
|
|
116
|
+
# 定期调整并发数(从每3个任务调整一次改为每2个任务调整一次)
|
|
117
|
+
if self._total_tasks % 2 == 0:
|
|
118
|
+
self.semaphore.adjust_concurrency()
|
|
119
|
+
|
|
120
|
+
task.add_done_callback(done_callback)
|
|
121
|
+
|
|
122
|
+
return task
|
|
123
|
+
|
|
124
|
+
def all_done(self) -> bool:
|
|
125
|
+
return len(self.current_task) == 0
|
|
126
|
+
|
|
127
|
+
def record_response_time(self, response_time: float):
|
|
128
|
+
"""记录任务的响应时间,用于动态调整并发数"""
|
|
129
|
+
self.semaphore.record_response_time(response_time)
|
|
130
|
+
|
|
131
|
+
def get_stats(self) -> dict:
|
|
132
|
+
"""获取任务管理器统计信息"""
|
|
133
|
+
return {
|
|
134
|
+
'active_tasks': len(self.current_task),
|
|
135
|
+
'total_tasks': self._total_tasks,
|
|
136
|
+
'exception_count': self._exception_count,
|
|
137
|
+
'success_rate': (self._total_tasks - self._exception_count) / max(1, self._total_tasks) * 100,
|
|
138
|
+
'current_concurrency': self.semaphore._current_value
|
|
139
|
+
}
|
crawlo/templates/crawlo.cfg.tmpl
CHANGED
|
@@ -1,11 +1,11 @@
|
|
|
1
|
-
# crawlo.cfg
|
|
2
|
-
# 项目的配置文件。
|
|
3
|
-
|
|
4
|
-
[settings]
|
|
5
|
-
# 指定 settings 模块的导入路径
|
|
6
|
-
default = {{project_name}}.settings
|
|
7
|
-
|
|
8
|
-
[deploy]
|
|
9
|
-
# (可选)用于部署配置
|
|
10
|
-
# url = http://localhost:6800/
|
|
1
|
+
# crawlo.cfg
|
|
2
|
+
# 项目的配置文件。
|
|
3
|
+
|
|
4
|
+
[settings]
|
|
5
|
+
# 指定 settings 模块的导入路径
|
|
6
|
+
default = {{project_name}}.settings
|
|
7
|
+
|
|
8
|
+
[deploy]
|
|
9
|
+
# (可选)用于部署配置
|
|
10
|
+
# url = http://localhost:6800/
|
|
11
11
|
# project = {{project_name}}
|
|
@@ -1,4 +1,4 @@
|
|
|
1
|
-
# -*- coding: UTF-8 -*-
|
|
2
|
-
"""
|
|
3
|
-
{{project_name}} 项目包
|
|
1
|
+
# -*- coding: UTF-8 -*-
|
|
2
|
+
"""
|
|
3
|
+
{{project_name}} 项目包
|
|
4
4
|
"""
|
|
@@ -1,18 +1,18 @@
|
|
|
1
|
-
# -*- coding: UTF-8 -*-
|
|
2
|
-
"""
|
|
3
|
-
{{project_name}}.items
|
|
4
|
-
======================
|
|
5
|
-
定义你抓取的数据结构。
|
|
6
|
-
"""
|
|
7
|
-
|
|
8
|
-
from crawlo.items import Item, Field
|
|
9
|
-
|
|
10
|
-
|
|
11
|
-
class ExampleItem(Item):
|
|
12
|
-
"""
|
|
13
|
-
一个示例数据项。
|
|
14
|
-
"""
|
|
15
|
-
id = Field()
|
|
16
|
-
# price = Field()
|
|
17
|
-
# description = Field()
|
|
1
|
+
# -*- coding: UTF-8 -*-
|
|
2
|
+
"""
|
|
3
|
+
{{project_name}}.items
|
|
4
|
+
======================
|
|
5
|
+
定义你抓取的数据结构。
|
|
6
|
+
"""
|
|
7
|
+
|
|
8
|
+
from crawlo.items import Item, Field
|
|
9
|
+
|
|
10
|
+
|
|
11
|
+
class ExampleItem(Item):
|
|
12
|
+
"""
|
|
13
|
+
一个示例数据项。
|
|
14
|
+
"""
|
|
15
|
+
id = Field()
|
|
16
|
+
# price = Field()
|
|
17
|
+
# description = Field()
|
|
18
18
|
pass
|