crawlo 1.4.5__py3-none-any.whl → 1.4.7__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 +90 -89
- 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 +186 -186
- crawlo/commands/help.py +140 -138
- crawlo/commands/list.py +155 -155
- crawlo/commands/run.py +379 -341
- crawlo/commands/startproject.py +460 -460
- crawlo/commands/stats.py +187 -187
- crawlo/commands/utils.py +196 -196
- crawlo/config.py +320 -312
- crawlo/config_validator.py +277 -277
- crawlo/core/__init__.py +52 -52
- crawlo/core/engine.py +451 -438
- crawlo/core/processor.py +47 -47
- crawlo/core/scheduler.py +290 -291
- crawlo/crawler.py +698 -657
- crawlo/data/__init__.py +5 -5
- crawlo/data/user_agents.py +194 -194
- crawlo/downloader/__init__.py +280 -276
- crawlo/downloader/aiohttp_downloader.py +233 -233
- crawlo/downloader/cffi_downloader.py +250 -245
- crawlo/downloader/httpx_downloader.py +265 -259
- crawlo/downloader/hybrid_downloader.py +212 -212
- crawlo/downloader/playwright_downloader.py +425 -402
- crawlo/downloader/selenium_downloader.py +486 -472
- crawlo/event.py +45 -11
- crawlo/exceptions.py +215 -82
- crawlo/extension/__init__.py +65 -64
- crawlo/extension/health_check.py +141 -141
- crawlo/extension/log_interval.py +94 -94
- crawlo/extension/log_stats.py +70 -70
- crawlo/extension/logging_extension.py +53 -61
- 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 +27 -27
- crawlo/factories/base.py +68 -68
- crawlo/factories/crawler.py +104 -103
- crawlo/factories/registry.py +84 -84
- crawlo/factories/utils.py +135 -0
- crawlo/filters/__init__.py +170 -153
- crawlo/filters/aioredis_filter.py +348 -264
- crawlo/filters/memory_filter.py +261 -276
- crawlo/framework.py +306 -292
- crawlo/initialization/__init__.py +44 -44
- crawlo/initialization/built_in.py +391 -434
- crawlo/initialization/context.py +141 -141
- crawlo/initialization/core.py +240 -194
- crawlo/initialization/phases.py +230 -149
- crawlo/initialization/registry.py +143 -145
- crawlo/initialization/utils.py +49 -0
- crawlo/interfaces.py +23 -23
- crawlo/items/__init__.py +23 -23
- crawlo/items/base.py +23 -23
- crawlo/items/fields.py +52 -52
- crawlo/items/items.py +104 -104
- crawlo/logging/__init__.py +42 -46
- crawlo/logging/config.py +277 -197
- crawlo/logging/factory.py +175 -171
- crawlo/logging/manager.py +104 -112
- crawlo/middleware/__init__.py +87 -24
- crawlo/middleware/default_header.py +132 -132
- crawlo/middleware/download_delay.py +104 -104
- crawlo/middleware/middleware_manager.py +142 -142
- crawlo/middleware/offsite.py +123 -123
- crawlo/middleware/proxy.py +209 -386
- crawlo/middleware/request_ignore.py +86 -86
- crawlo/middleware/response_code.py +150 -150
- crawlo/middleware/response_filter.py +136 -136
- crawlo/middleware/retry.py +124 -124
- crawlo/mode_manager.py +287 -253
- crawlo/network/__init__.py +21 -21
- crawlo/network/request.py +375 -379
- crawlo/network/response.py +569 -664
- crawlo/pipelines/__init__.py +53 -22
- crawlo/pipelines/base_pipeline.py +452 -0
- crawlo/pipelines/bloom_dedup_pipeline.py +146 -146
- crawlo/pipelines/console_pipeline.py +39 -39
- crawlo/pipelines/csv_pipeline.py +316 -316
- crawlo/pipelines/database_dedup_pipeline.py +197 -197
- crawlo/pipelines/json_pipeline.py +218 -218
- crawlo/pipelines/memory_dedup_pipeline.py +105 -105
- crawlo/pipelines/mongo_pipeline.py +140 -132
- crawlo/pipelines/mysql_pipeline.py +470 -326
- crawlo/pipelines/pipeline_manager.py +100 -100
- crawlo/pipelines/redis_dedup_pipeline.py +155 -156
- crawlo/project.py +347 -347
- crawlo/queue/__init__.py +10 -0
- crawlo/queue/pqueue.py +38 -38
- crawlo/queue/queue_manager.py +591 -525
- crawlo/queue/redis_priority_queue.py +519 -370
- crawlo/settings/__init__.py +7 -7
- crawlo/settings/default_settings.py +285 -270
- crawlo/settings/setting_manager.py +219 -219
- crawlo/spider/__init__.py +657 -657
- crawlo/stats_collector.py +82 -73
- crawlo/subscriber.py +129 -129
- crawlo/task_manager.py +138 -138
- crawlo/templates/crawlo.cfg.tmpl +10 -10
- crawlo/templates/project/__init__.py.tmpl +2 -4
- crawlo/templates/project/items.py.tmpl +13 -17
- crawlo/templates/project/middlewares.py.tmpl +38 -38
- crawlo/templates/project/pipelines.py.tmpl +35 -36
- crawlo/templates/project/settings.py.tmpl +110 -157
- crawlo/templates/project/settings_distributed.py.tmpl +156 -161
- crawlo/templates/project/settings_gentle.py.tmpl +170 -171
- crawlo/templates/project/settings_high_performance.py.tmpl +171 -172
- crawlo/templates/project/settings_minimal.py.tmpl +99 -77
- crawlo/templates/project/settings_simple.py.tmpl +168 -169
- crawlo/templates/project/spiders/__init__.py.tmpl +9 -9
- crawlo/templates/run.py.tmpl +23 -30
- crawlo/templates/spider/spider.py.tmpl +33 -144
- crawlo/templates/spiders_init.py.tmpl +5 -10
- crawlo/tools/__init__.py +86 -189
- crawlo/tools/date_tools.py +289 -289
- crawlo/tools/distributed_coordinator.py +384 -384
- crawlo/tools/scenario_adapter.py +262 -262
- crawlo/tools/text_cleaner.py +232 -232
- crawlo/utils/__init__.py +50 -50
- crawlo/utils/batch_processor.py +276 -259
- crawlo/utils/config_manager.py +442 -0
- crawlo/utils/controlled_spider_mixin.py +439 -439
- crawlo/utils/db_helper.py +250 -244
- crawlo/utils/error_handler.py +410 -410
- crawlo/utils/fingerprint.py +121 -121
- crawlo/utils/func_tools.py +82 -82
- crawlo/utils/large_scale_helper.py +344 -344
- crawlo/utils/leak_detector.py +335 -0
- crawlo/utils/log.py +79 -79
- crawlo/utils/misc.py +81 -81
- crawlo/utils/mongo_connection_pool.py +157 -0
- crawlo/utils/mysql_connection_pool.py +197 -0
- crawlo/utils/performance_monitor.py +285 -285
- crawlo/utils/queue_helper.py +175 -175
- crawlo/utils/redis_checker.py +91 -0
- crawlo/utils/redis_connection_pool.py +578 -388
- crawlo/utils/redis_key_validator.py +198 -198
- crawlo/utils/request.py +278 -256
- crawlo/utils/request_serializer.py +225 -225
- crawlo/utils/resource_manager.py +337 -0
- crawlo/utils/selector_helper.py +137 -137
- crawlo/utils/singleton.py +70 -0
- crawlo/utils/spider_loader.py +201 -201
- crawlo/utils/text_helper.py +94 -94
- crawlo/utils/{url.py → url_utils.py} +39 -39
- crawlo-1.4.7.dist-info/METADATA +689 -0
- crawlo-1.4.7.dist-info/RECORD +347 -0
- examples/__init__.py +7 -7
- tests/__init__.py +7 -7
- tests/advanced_tools_example.py +217 -275
- tests/authenticated_proxy_example.py +110 -106
- tests/baidu_performance_test.py +108 -108
- tests/baidu_test.py +59 -59
- tests/bug_check_test.py +250 -250
- tests/cleaners_example.py +160 -160
- tests/comprehensive_framework_test.py +212 -212
- tests/comprehensive_test.py +81 -81
- tests/comprehensive_testing_summary.md +186 -186
- 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 +69 -69
- tests/debug_framework_logger.py +84 -84
- tests/debug_log_config.py +126 -126
- tests/debug_log_levels.py +63 -63
- tests/debug_pipelines.py +66 -66
- tests/detailed_log_test.py +233 -233
- tests/direct_selector_helper_test.py +96 -96
- tests/distributed_dedup_test.py +467 -0
- tests/distributed_test.py +66 -66
- tests/distributed_test_debug.py +76 -76
- tests/dynamic_loading_example.py +523 -523
- tests/dynamic_loading_test.py +104 -104
- tests/error_handling_example.py +171 -171
- tests/explain_mysql_update_behavior.py +77 -0
- tests/final_comprehensive_test.py +151 -151
- tests/final_log_test.py +260 -260
- tests/final_validation_test.py +182 -182
- tests/fix_log_test.py +142 -142
- tests/framework_performance_test.py +202 -202
- tests/log_buffering_test.py +111 -111
- tests/log_generation_timing_test.py +153 -153
- tests/monitor_redis_dedup.sh +72 -0
- tests/ofweek_scrapy/ofweek_scrapy/items.py +12 -12
- tests/ofweek_scrapy/ofweek_scrapy/middlewares.py +100 -100
- tests/ofweek_scrapy/ofweek_scrapy/pipelines.py +13 -13
- tests/ofweek_scrapy/ofweek_scrapy/settings.py +84 -84
- tests/ofweek_scrapy/scrapy.cfg +11 -11
- tests/optimized_performance_test.py +211 -211
- tests/performance_comparison.py +244 -244
- tests/queue_blocking_test.py +113 -113
- tests/queue_test.py +89 -89
- 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 +138 -138
- tests/scrapy_comparison/scrapy_test.py +133 -133
- tests/simple_cli_test.py +55 -0
- tests/simple_command_test.py +119 -119
- tests/simple_crawlo_test.py +126 -126
- tests/simple_follow_test.py +38 -38
- tests/simple_log_test2.py +137 -137
- tests/simple_optimization_test.py +128 -128
- tests/simple_queue_type_test.py +41 -41
- tests/simple_response_selector_test.py +94 -94
- tests/simple_selector_helper_test.py +154 -154
- tests/simple_selector_test.py +207 -207
- tests/simple_spider_test.py +49 -49
- tests/simple_url_test.py +73 -73
- tests/simulate_mysql_update_test.py +140 -0
- tests/spider_log_timing_test.py +177 -177
- tests/test_advanced_tools.py +148 -148
- tests/test_all_commands.py +230 -230
- tests/test_all_pipeline_fingerprints.py +133 -133
- tests/test_all_redis_key_configs.py +145 -145
- tests/test_asyncmy_usage.py +57 -0
- tests/test_batch_processor.py +178 -178
- tests/test_cleaners.py +54 -54
- tests/test_cli_arguments.py +119 -0
- tests/test_component_factory.py +174 -174
- 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 +79 -79
- tests/test_crawler_process_import.py +38 -38
- tests/test_crawler_process_spider_modules.py +47 -47
- tests/test_crawlo_proxy_integration.py +114 -108
- tests/test_date_tools.py +123 -123
- tests/test_dedup_fix.py +220 -220
- tests/test_dedup_pipeline_consistency.py +124 -124
- tests/test_default_header_middleware.py +313 -313
- tests/test_distributed.py +65 -65
- tests/test_double_crawlo_fix.py +204 -204
- tests/test_double_crawlo_fix_simple.py +124 -124
- tests/test_download_delay_middleware.py +221 -221
- tests/test_downloader_proxy_compatibility.py +272 -268
- tests/test_edge_cases.py +305 -305
- tests/test_encoding_core.py +56 -56
- tests/test_encoding_detection.py +126 -126
- tests/test_enhanced_error_handler.py +270 -270
- tests/test_enhanced_error_handler_comprehensive.py +245 -245
- tests/test_error_handler_compatibility.py +112 -112
- tests/test_factories.py +252 -252
- tests/test_factory_compatibility.py +196 -196
- tests/test_final_validation.py +153 -153
- tests/test_fingerprint_consistency.py +135 -135
- tests/test_fingerprint_simple.py +51 -51
- tests/test_get_component_logger.py +83 -83
- tests/test_hash_performance.py +99 -99
- tests/test_integration.py +169 -169
- tests/test_item_dedup_redis_key.py +122 -122
- tests/test_large_scale_helper.py +235 -235
- tests/test_logging_enhancements.py +374 -374
- tests/test_logging_final.py +184 -184
- tests/test_logging_integration.py +312 -312
- tests/test_logging_system.py +282 -282
- tests/test_middleware_debug.py +141 -141
- tests/test_mode_consistency.py +51 -51
- tests/test_multi_directory.py +67 -67
- tests/test_multiple_spider_modules.py +80 -80
- tests/test_mysql_pipeline_config.py +165 -0
- tests/test_mysql_pipeline_error.py +99 -0
- tests/test_mysql_pipeline_init_log.py +83 -0
- tests/test_mysql_pipeline_integration.py +133 -0
- tests/test_mysql_pipeline_refactor.py +144 -0
- tests/test_mysql_pipeline_refactor_simple.py +86 -0
- tests/test_mysql_pipeline_robustness.py +196 -0
- tests/test_mysql_pipeline_types.py +89 -0
- tests/test_mysql_update_columns.py +94 -0
- tests/test_offsite_middleware.py +244 -244
- tests/test_offsite_middleware_simple.py +203 -203
- tests/test_optimized_selector_naming.py +100 -100
- tests/test_parsel.py +29 -29
- tests/test_performance.py +327 -327
- tests/test_performance_monitor.py +115 -115
- tests/test_pipeline_fingerprint_consistency.py +86 -86
- tests/test_priority_behavior.py +211 -211
- tests/test_priority_consistency.py +151 -151
- tests/test_priority_consistency_fixed.py +249 -249
- tests/test_proxy_health_check.py +32 -32
- tests/test_proxy_middleware.py +217 -121
- tests/test_proxy_middleware_enhanced.py +212 -216
- tests/test_proxy_middleware_integration.py +142 -137
- tests/test_proxy_middleware_refactored.py +207 -184
- tests/test_proxy_only.py +84 -0
- tests/test_proxy_providers.py +56 -56
- tests/test_proxy_stats.py +19 -19
- tests/test_proxy_strategies.py +59 -59
- tests/test_proxy_with_downloader.py +153 -0
- tests/test_queue_empty_check.py +41 -41
- tests/test_queue_manager_double_crawlo.py +173 -173
- tests/test_queue_manager_redis_key.py +179 -179
- tests/test_queue_naming.py +154 -154
- tests/test_queue_type.py +106 -106
- tests/test_queue_type_redis_config_consistency.py +130 -130
- tests/test_random_headers_default.py +322 -322
- tests/test_random_headers_necessity.py +308 -308
- tests/test_random_user_agent.py +72 -72
- 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_redis_queue_name_fix.py +175 -175
- tests/test_redis_queue_type_fallback.py +129 -129
- 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_follow.py +104 -104
- tests/test_response_improvements.py +152 -152
- tests/test_response_selector_methods.py +92 -92
- tests/test_response_url_methods.py +70 -70
- tests/test_response_urljoin.py +86 -86
- tests/test_retry_middleware.py +333 -333
- tests/test_retry_middleware_realistic.py +273 -273
- tests/test_scheduler.py +252 -252
- tests/test_scheduler_config_update.py +133 -133
- tests/test_scrapy_style_encoding.py +112 -112
- tests/test_selector_helper.py +100 -100
- tests/test_selector_optimizations.py +146 -146
- tests/test_simple_response.py +61 -61
- tests/test_spider_loader.py +49 -49
- tests/test_spider_loader_comprehensive.py +69 -69
- tests/test_spider_modules.py +84 -84
- tests/test_spiders/test_spider.py +9 -9
- 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_agent_randomness.py +176 -176
- tests/test_user_agents.py +96 -96
- tests/untested_features_report.md +138 -138
- tests/verify_debug.py +51 -51
- tests/verify_distributed.py +117 -117
- tests/verify_log_fix.py +111 -111
- tests/verify_mysql_warnings.py +110 -0
- crawlo/logging/async_handler.py +0 -181
- crawlo/logging/monitor.py +0 -153
- crawlo/logging/sampler.py +0 -167
- crawlo/middleware/simple_proxy.py +0 -65
- crawlo/tools/authenticated_proxy.py +0 -241
- crawlo/tools/data_formatter.py +0 -226
- crawlo/tools/data_validator.py +0 -181
- crawlo/tools/encoding_converter.py +0 -127
- crawlo/tools/network_diagnostic.py +0 -365
- crawlo/tools/request_tools.py +0 -83
- crawlo/tools/retry_mechanism.py +0 -224
- crawlo/utils/env_config.py +0 -143
- crawlo/utils/large_scale_config.py +0 -287
- crawlo/utils/system.py +0 -11
- crawlo/utils/tools.py +0 -5
- crawlo-1.4.5.dist-info/METADATA +0 -329
- crawlo-1.4.5.dist-info/RECORD +0 -347
- tests/env_config_example.py +0 -134
- tests/ofweek_scrapy/ofweek_scrapy/spiders/ofweek_spider.py +0 -162
- tests/test_authenticated_proxy.py +0 -142
- tests/test_comprehensive.py +0 -147
- tests/test_dynamic_downloaders_proxy.py +0 -125
- tests/test_dynamic_proxy.py +0 -93
- tests/test_dynamic_proxy_config.py +0 -147
- tests/test_dynamic_proxy_real.py +0 -110
- tests/test_env_config.py +0 -122
- tests/test_framework_env_usage.py +0 -104
- tests/test_large_scale_config.py +0 -113
- tests/test_proxy_api.py +0 -265
- tests/test_real_scenario_proxy.py +0 -196
- tests/tools_example.py +0 -261
- {crawlo-1.4.5.dist-info → crawlo-1.4.7.dist-info}/WHEEL +0 -0
- {crawlo-1.4.5.dist-info → crawlo-1.4.7.dist-info}/entry_points.txt +0 -0
- {crawlo-1.4.5.dist-info → crawlo-1.4.7.dist-info}/top_level.txt +0 -0
crawlo/spider/__init__.py
CHANGED
|
@@ -1,657 +1,657 @@
|
|
|
1
|
-
#!/usr/bin/python
|
|
2
|
-
# -*- coding:UTF-8 -*-
|
|
3
|
-
"""
|
|
4
|
-
Crawlo Spider Module
|
|
5
|
-
==================
|
|
6
|
-
提供爬虫基类和相关功能。
|
|
7
|
-
|
|
8
|
-
核心功能:
|
|
9
|
-
- Spider基类:所有爬虫的基础类
|
|
10
|
-
- 自动注册机制:通过元类自动注册爬虫
|
|
11
|
-
- 配置管理:支持自定义设置和链式调用
|
|
12
|
-
- 生命周期管理:开启/关闭钩子函数
|
|
13
|
-
- 分布式支持:智能检测运行模式
|
|
14
|
-
|
|
15
|
-
使用示例:
|
|
16
|
-
class MySpider(Spider):
|
|
17
|
-
name = 'my_spider'
|
|
18
|
-
start_urls = ['http://example.com']
|
|
19
|
-
|
|
20
|
-
# 自定义配置
|
|
21
|
-
custom_settings = {
|
|
22
|
-
'DOWNLOADER_TYPE': 'httpx',
|
|
23
|
-
'CONCURRENCY': 10
|
|
24
|
-
}
|
|
25
|
-
|
|
26
|
-
def parse(self, response):
|
|
27
|
-
# 解析逻辑
|
|
28
|
-
yield Item(data=response.json())
|
|
29
|
-
"""
|
|
30
|
-
from __future__ import annotations
|
|
31
|
-
from typing import Type, Any, Optional, List, Dict, Union, Iterator, AsyncIterator
|
|
32
|
-
from ..network.request import Request
|
|
33
|
-
from ..utils.log import get_logger
|
|
34
|
-
|
|
35
|
-
|
|
36
|
-
# 全局爬虫注册表
|
|
37
|
-
_DEFAULT_SPIDER_REGISTRY: dict[str, Type[Spider]] = {}
|
|
38
|
-
|
|
39
|
-
|
|
40
|
-
class SpiderMeta(type):
|
|
41
|
-
"""
|
|
42
|
-
爬虫元类,提供自动注册功能
|
|
43
|
-
|
|
44
|
-
功能:
|
|
45
|
-
- 自动注册爬虫到全局注册表
|
|
46
|
-
- 验证爬虫名称的唯一性
|
|
47
|
-
- 提供完整的错误提示
|
|
48
|
-
"""
|
|
49
|
-
|
|
50
|
-
def __new__(mcs, name: str, bases: tuple[type], namespace: dict[str, Any], **kwargs):
|
|
51
|
-
cls = super().__new__(mcs, name, bases, namespace)
|
|
52
|
-
|
|
53
|
-
# 检查是否为Spider子类
|
|
54
|
-
is_spider_subclass = any(
|
|
55
|
-
base is Spider or (isinstance(base, type) and issubclass(base, Spider))
|
|
56
|
-
for base in bases
|
|
57
|
-
)
|
|
58
|
-
if not is_spider_subclass:
|
|
59
|
-
return cls
|
|
60
|
-
|
|
61
|
-
# 验证爬虫名称
|
|
62
|
-
spider_name = namespace.get('name')
|
|
63
|
-
if not isinstance(spider_name, str):
|
|
64
|
-
raise AttributeError(
|
|
65
|
-
f"爬虫类 '{cls.__name__}' 必须定义字符串类型的 'name' 属性。\n"
|
|
66
|
-
f"示例: name = 'my_spider'"
|
|
67
|
-
)
|
|
68
|
-
|
|
69
|
-
# 检查名称唯一性
|
|
70
|
-
if spider_name in _DEFAULT_SPIDER_REGISTRY:
|
|
71
|
-
existing_class = _DEFAULT_SPIDER_REGISTRY[spider_name]
|
|
72
|
-
raise ValueError(
|
|
73
|
-
f"爬虫名称 '{spider_name}' 已被 {existing_class.__name__} 占用。\n"
|
|
74
|
-
f"请确保每个爬虫的 name 属性全局唯一。\n"
|
|
75
|
-
f"建议使用格式: 'project_module_function'"
|
|
76
|
-
)
|
|
77
|
-
|
|
78
|
-
# 注册爬虫
|
|
79
|
-
_DEFAULT_SPIDER_REGISTRY[spider_name] = cls
|
|
80
|
-
# 延迟初始化logger避免模块级别阻塞
|
|
81
|
-
try:
|
|
82
|
-
from crawlo.
|
|
83
|
-
get_logger(__name__).debug(f"自动注册爬虫: {spider_name} -> {cls.__name__}")
|
|
84
|
-
except:
|
|
85
|
-
# 如果日志系统未初始化,静默失败
|
|
86
|
-
pass
|
|
87
|
-
|
|
88
|
-
return cls
|
|
89
|
-
|
|
90
|
-
|
|
91
|
-
class Spider(metaclass=SpiderMeta):
|
|
92
|
-
"""
|
|
93
|
-
爬虫基类 - 所有爬虫实现的基础
|
|
94
|
-
|
|
95
|
-
必须定义的属性:
|
|
96
|
-
- name: 爬虫名称,必须全局唯一
|
|
97
|
-
|
|
98
|
-
可选配置:
|
|
99
|
-
- start_urls: 起始 URL 列表
|
|
100
|
-
- custom_settings: 自定义设置字典
|
|
101
|
-
- allowed_domains: 允许的域名列表
|
|
102
|
-
|
|
103
|
-
必须实现的方法:
|
|
104
|
-
- parse(response): 解析响应的主方法
|
|
105
|
-
|
|
106
|
-
可选实现的方法:
|
|
107
|
-
- spider_opened(): 爬虫开启时调用
|
|
108
|
-
- spider_closed(): 爬虫关闭时调用
|
|
109
|
-
- start_requests(): 生成初始请求(默认使用start_urls)
|
|
110
|
-
|
|
111
|
-
示例:
|
|
112
|
-
class MySpider(Spider):
|
|
113
|
-
name = 'example_spider'
|
|
114
|
-
start_urls = ['https://example.com']
|
|
115
|
-
|
|
116
|
-
custom_settings = {
|
|
117
|
-
'DOWNLOADER_TYPE': 'httpx',
|
|
118
|
-
'CONCURRENCY': 5,
|
|
119
|
-
'DOWNLOAD_DELAY': 1.0
|
|
120
|
-
}
|
|
121
|
-
|
|
122
|
-
def parse(self, response):
|
|
123
|
-
# 提取数据
|
|
124
|
-
data = response.css('title::text').get()
|
|
125
|
-
yield {'title': data}
|
|
126
|
-
|
|
127
|
-
# 生成新请求
|
|
128
|
-
for link in response.css('a::attr(href)').getall():
|
|
129
|
-
yield Request(url=link, callback=self.parse_detail)
|
|
130
|
-
"""
|
|
131
|
-
|
|
132
|
-
# 必须定义的属性
|
|
133
|
-
name: str = None
|
|
134
|
-
|
|
135
|
-
# 可选属性
|
|
136
|
-
start_urls: List[str] = None
|
|
137
|
-
custom_settings: Dict[str, Any] = None
|
|
138
|
-
allowed_domains: List[str] = None
|
|
139
|
-
|
|
140
|
-
def __init__(self, name: str = None, **kwargs):
|
|
141
|
-
"""
|
|
142
|
-
初始化爬虫实例
|
|
143
|
-
|
|
144
|
-
:param name: 爬虫名称(可选,默认使用类属性)
|
|
145
|
-
:param kwargs: 其他初始化参数
|
|
146
|
-
"""
|
|
147
|
-
# 初始化基本属性
|
|
148
|
-
if not hasattr(self, 'start_urls') or self.start_urls is None:
|
|
149
|
-
self.start_urls = []
|
|
150
|
-
if not hasattr(self, 'custom_settings') or self.custom_settings is None:
|
|
151
|
-
self.custom_settings = {}
|
|
152
|
-
if not hasattr(self, 'allowed_domains') or self.allowed_domains is None:
|
|
153
|
-
self.allowed_domains = []
|
|
154
|
-
|
|
155
|
-
# 设置爬虫名称
|
|
156
|
-
self.name = name or self.name
|
|
157
|
-
if not self.name:
|
|
158
|
-
raise ValueError(f"爬虫 {self.__class__.__name__} 必须指定 name 属性")
|
|
159
|
-
|
|
160
|
-
# 初始化其他属性
|
|
161
|
-
self.crawler = None
|
|
162
|
-
# 延迟初始化logger避免阻塞
|
|
163
|
-
self._logger = None
|
|
164
|
-
self.stats = None
|
|
165
|
-
|
|
166
|
-
# 应用额外参数
|
|
167
|
-
for key, value in kwargs.items():
|
|
168
|
-
setattr(self, key, value)
|
|
169
|
-
|
|
170
|
-
@property
|
|
171
|
-
def logger(self):
|
|
172
|
-
"""延迟初始化logger"""
|
|
173
|
-
if self._logger is None:
|
|
174
|
-
from crawlo.
|
|
175
|
-
self._logger = get_logger(self.name)
|
|
176
|
-
return self._logger
|
|
177
|
-
|
|
178
|
-
@classmethod
|
|
179
|
-
def create_instance(cls, crawler) -> 'Spider':
|
|
180
|
-
"""
|
|
181
|
-
创建爬虫实例并绑定 crawler
|
|
182
|
-
|
|
183
|
-
:param crawler: Crawler 实例
|
|
184
|
-
:return: 爬虫实例
|
|
185
|
-
"""
|
|
186
|
-
spider = cls()
|
|
187
|
-
spider.crawler = crawler
|
|
188
|
-
spider.stats = getattr(crawler, 'stats', None)
|
|
189
|
-
|
|
190
|
-
# 合并自定义设置 - 使用延迟应用避免初始化时的循环依赖
|
|
191
|
-
if hasattr(spider, 'custom_settings') and spider.custom_settings:
|
|
192
|
-
# 延迟到真正需要时才应用设置
|
|
193
|
-
spider._pending_settings = spider.custom_settings.copy()
|
|
194
|
-
spider.logger.debug(f"准备应用 {len(spider.custom_settings)} 项自定义设置")
|
|
195
|
-
|
|
196
|
-
return spider
|
|
197
|
-
|
|
198
|
-
def apply_pending_settings(self):
|
|
199
|
-
"""应用待处理的设置(在初始化完成后调用)"""
|
|
200
|
-
if hasattr(self, '_pending_settings') and self._pending_settings:
|
|
201
|
-
for key, value in self._pending_settings.items():
|
|
202
|
-
if self.crawler and hasattr(self.crawler, 'settings'):
|
|
203
|
-
self.crawler.settings.set(key, value)
|
|
204
|
-
self.logger.debug(f"应用自定义设置: {key} = {value}")
|
|
205
|
-
# 清除待处理的设置
|
|
206
|
-
delattr(self, '_pending_settings')
|
|
207
|
-
|
|
208
|
-
def start_requests(self) -> Iterator[Request]:
|
|
209
|
-
"""
|
|
210
|
-
生成初始请求
|
|
211
|
-
|
|
212
|
-
默认行为:
|
|
213
|
-
- 使用 start_urls 生成请求
|
|
214
|
-
- 智能检测分布式模式决定是否去重
|
|
215
|
-
- 支持单个 start_url 属性(兼容性)
|
|
216
|
-
- 支持批量生成优化(大规模URL场景)
|
|
217
|
-
|
|
218
|
-
:return: Request 迭代器
|
|
219
|
-
"""
|
|
220
|
-
# 检测是否为分布式模式
|
|
221
|
-
is_distributed = self._is_distributed_mode()
|
|
222
|
-
|
|
223
|
-
# 获取批量处理配置
|
|
224
|
-
batch_size = self._get_batch_size()
|
|
225
|
-
|
|
226
|
-
# 从 start_urls 生成请求
|
|
227
|
-
if self.start_urls:
|
|
228
|
-
generated_count = 0
|
|
229
|
-
for url in self.start_urls:
|
|
230
|
-
if self._is_allowed_domain(url):
|
|
231
|
-
yield Request(
|
|
232
|
-
url=url,
|
|
233
|
-
callback=self.parse,
|
|
234
|
-
dont_filter=not is_distributed,
|
|
235
|
-
meta={'spider_name': self.name}
|
|
236
|
-
)
|
|
237
|
-
generated_count += 1
|
|
238
|
-
|
|
239
|
-
# 大规模URL时进行批量控制
|
|
240
|
-
if batch_size > 0 and generated_count % batch_size == 0:
|
|
241
|
-
self.logger.debug(f"已生成 {generated_count} 个请求(批量大小: {batch_size})")
|
|
242
|
-
else:
|
|
243
|
-
self.logger.warning(f"跳过不允许的域名: {url}")
|
|
244
|
-
|
|
245
|
-
# 兼容单个 start_url 属性
|
|
246
|
-
elif hasattr(self, 'start_url') and isinstance(getattr(self, 'start_url'), str):
|
|
247
|
-
url = getattr(self, 'start_url')
|
|
248
|
-
if self._is_allowed_domain(url):
|
|
249
|
-
yield Request(
|
|
250
|
-
url=url,
|
|
251
|
-
callback=self.parse,
|
|
252
|
-
dont_filter=not is_distributed,
|
|
253
|
-
meta={'spider_name': self.name}
|
|
254
|
-
)
|
|
255
|
-
else:
|
|
256
|
-
self.logger.warning(f"跳过不允许的域名: {url}")
|
|
257
|
-
|
|
258
|
-
else:
|
|
259
|
-
self.logger.warning(
|
|
260
|
-
f"爬虫 {self.name} 没有定义 start_urls 或 start_url。\n"
|
|
261
|
-
f"请在爬虫类中定义或重写 start_requests() 方法。"
|
|
262
|
-
)
|
|
263
|
-
|
|
264
|
-
def _get_batch_size(self) -> int:
|
|
265
|
-
"""
|
|
266
|
-
获取批量处理大小配置
|
|
267
|
-
|
|
268
|
-
用于大规模URL场景的性能优化
|
|
269
|
-
|
|
270
|
-
:return: 批量大小(0表示无限制)
|
|
271
|
-
"""
|
|
272
|
-
if not self.crawler:
|
|
273
|
-
return 0
|
|
274
|
-
|
|
275
|
-
# 从设置中获取批量大小
|
|
276
|
-
batch_size = self.crawler.settings.get_int('SPIDER_BATCH_SIZE', 0)
|
|
277
|
-
|
|
278
|
-
# 如果start_urls超过一定数量,自动启用批量模式
|
|
279
|
-
if batch_size == 0 and self.start_urls and len(self.start_urls) > 1000:
|
|
280
|
-
batch_size = 500 # 默认批量大小
|
|
281
|
-
self.logger.info(f"检测到大量start_urls ({len(self.start_urls)}),启用批量模式 (批量大小: {batch_size})")
|
|
282
|
-
|
|
283
|
-
return batch_size
|
|
284
|
-
|
|
285
|
-
def _is_distributed_mode(self) -> bool:
|
|
286
|
-
"""
|
|
287
|
-
智能检测是否为分布式模式
|
|
288
|
-
|
|
289
|
-
检测条件:
|
|
290
|
-
- QUEUE_TYPE = 'redis'
|
|
291
|
-
- FILTER_CLASS 包含 'aioredis_filter'
|
|
292
|
-
- RUN_MODE = 'distributed'
|
|
293
|
-
|
|
294
|
-
:return: 是否为分布式模式
|
|
295
|
-
"""
|
|
296
|
-
if not self.crawler:
|
|
297
|
-
return False
|
|
298
|
-
|
|
299
|
-
settings = self.crawler.settings
|
|
300
|
-
|
|
301
|
-
# 检查多个条件来判断是否为分布式模式
|
|
302
|
-
queue_type = settings.get('QUEUE_TYPE', 'memory')
|
|
303
|
-
filter_class = settings.get('FILTER_CLASS', '')
|
|
304
|
-
run_mode = settings.get('RUN_MODE', 'standalone')
|
|
305
|
-
|
|
306
|
-
# 分布式模式的标志
|
|
307
|
-
is_redis_queue = queue_type == 'redis'
|
|
308
|
-
is_redis_filter = 'aioredis_filter' in filter_class.lower()
|
|
309
|
-
is_distributed_run_mode = run_mode == 'distributed'
|
|
310
|
-
|
|
311
|
-
distributed = is_redis_queue or is_redis_filter or is_distributed_run_mode
|
|
312
|
-
|
|
313
|
-
if distributed:
|
|
314
|
-
self.logger.debug("检测到分布式模式,启用请求去重")
|
|
315
|
-
else:
|
|
316
|
-
self.logger.debug("检测到单机模式,禁用请求去重")
|
|
317
|
-
|
|
318
|
-
return distributed
|
|
319
|
-
|
|
320
|
-
def _is_allowed_domain(self, url: str) -> bool:
|
|
321
|
-
"""
|
|
322
|
-
检查URL是否在允许的域名列表中
|
|
323
|
-
|
|
324
|
-
:param url: 要检查的URL
|
|
325
|
-
:return: 是否允许
|
|
326
|
-
"""
|
|
327
|
-
if not self.allowed_domains:
|
|
328
|
-
return True
|
|
329
|
-
|
|
330
|
-
from urllib.parse import urlparse
|
|
331
|
-
try:
|
|
332
|
-
domain = urlparse(url).netloc.lower()
|
|
333
|
-
return any(
|
|
334
|
-
domain == allowed.lower() or domain.endswith('.' + allowed.lower())
|
|
335
|
-
for allowed in self.allowed_domains
|
|
336
|
-
)
|
|
337
|
-
except Exception as e:
|
|
338
|
-
self.logger.warning(f"URL解析失败: {url} - {e}")
|
|
339
|
-
return False
|
|
340
|
-
|
|
341
|
-
def parse(self, response):
|
|
342
|
-
"""
|
|
343
|
-
解析响应的主方法(必须实现)
|
|
344
|
-
|
|
345
|
-
:param response: 响应对象
|
|
346
|
-
:return: 生成的 Item 或 Request
|
|
347
|
-
"""
|
|
348
|
-
raise NotImplementedError(
|
|
349
|
-
f"爬虫 {self.__class__.__name__} 必须实现 parse() 方法\n"
|
|
350
|
-
f"示例:\n"
|
|
351
|
-
f"def parse(self, response):\n"
|
|
352
|
-
f" # 提取数据\n"
|
|
353
|
-
f" yield {{'title': response.css('title::text').get()}}\n"
|
|
354
|
-
f" # 生成新请求\n"
|
|
355
|
-
f" for link in response.css('a::attr(href)').getall():\n"
|
|
356
|
-
f" yield Request(url=link)"
|
|
357
|
-
)
|
|
358
|
-
|
|
359
|
-
async def spider_opened(self):
|
|
360
|
-
"""
|
|
361
|
-
爬虫开启时调用的钩子函数
|
|
362
|
-
|
|
363
|
-
可用于:
|
|
364
|
-
- 初始化资源
|
|
365
|
-
- 连接数据库
|
|
366
|
-
- 设置初始状态
|
|
367
|
-
"""
|
|
368
|
-
self.logger.info(f"Spider {self.name} opened")
|
|
369
|
-
|
|
370
|
-
async def spider_closed(self):
|
|
371
|
-
"""
|
|
372
|
-
爬虫关闭时调用的钩子函数
|
|
373
|
-
|
|
374
|
-
可用于:
|
|
375
|
-
- 清理资源
|
|
376
|
-
- 关闭数据库连接
|
|
377
|
-
"""
|
|
378
|
-
# 不再输出任何信息,避免与统计信息重复
|
|
379
|
-
# 统计信息由StatsCollector负责输出
|
|
380
|
-
pass
|
|
381
|
-
|
|
382
|
-
def __str__(self) -> str:
|
|
383
|
-
return f"{self.__class__.__name__}(name='{self.name}')"
|
|
384
|
-
|
|
385
|
-
def __repr__(self) -> str:
|
|
386
|
-
return self.__str__()
|
|
387
|
-
|
|
388
|
-
def set_custom_setting(self, key: str, value: Any) -> 'Spider':
|
|
389
|
-
"""
|
|
390
|
-
设置自定义配置(链式调用)
|
|
391
|
-
|
|
392
|
-
:param key: 配置键名
|
|
393
|
-
:param value: 配置值
|
|
394
|
-
:return: self(支持链式调用)
|
|
395
|
-
|
|
396
|
-
示例:
|
|
397
|
-
spider.set_custom_setting('CONCURRENCY', 10)\
|
|
398
|
-
.set_custom_setting('DOWNLOAD_DELAY', 1.0)
|
|
399
|
-
"""
|
|
400
|
-
if not hasattr(self, 'custom_settings') or self.custom_settings is None:
|
|
401
|
-
self.custom_settings = {}
|
|
402
|
-
|
|
403
|
-
self.custom_settings[key] = value
|
|
404
|
-
self.logger.debug(f"设置自定义配置: {key} = {value}")
|
|
405
|
-
|
|
406
|
-
# 如果已绑定crawler,立即应用设置
|
|
407
|
-
if self.crawler:
|
|
408
|
-
self.crawler.settings.set(key, value)
|
|
409
|
-
|
|
410
|
-
return self
|
|
411
|
-
|
|
412
|
-
def get_custom_setting(self, key: str, default: Any = None) -> Any:
|
|
413
|
-
"""
|
|
414
|
-
获取自定义配置值
|
|
415
|
-
|
|
416
|
-
:param key: 配置键名
|
|
417
|
-
:param default: 默认值
|
|
418
|
-
:return: 配置值
|
|
419
|
-
"""
|
|
420
|
-
if hasattr(self, 'custom_settings') and self.custom_settings:
|
|
421
|
-
return self.custom_settings.get(key, default)
|
|
422
|
-
return default
|
|
423
|
-
|
|
424
|
-
def get_spider_info(self) -> Dict[str, Any]:
|
|
425
|
-
"""
|
|
426
|
-
获取爬虫详细信息
|
|
427
|
-
|
|
428
|
-
:return: 爬虫信息字典
|
|
429
|
-
"""
|
|
430
|
-
info = {
|
|
431
|
-
'name': self.name,
|
|
432
|
-
'class_name': self.__class__.__name__,
|
|
433
|
-
'module': self.__module__,
|
|
434
|
-
'start_urls_count': len(self.start_urls) if self.start_urls else 0,
|
|
435
|
-
'allowed_domains_count': len(self.allowed_domains) if self.allowed_domains else 0,
|
|
436
|
-
'custom_settings_count': len(self.custom_settings) if self.custom_settings else 0,
|
|
437
|
-
'is_distributed': self._is_distributed_mode() if self.crawler else None,
|
|
438
|
-
'has_crawler': self.crawler is not None,
|
|
439
|
-
'logger_name': self.logger.name if hasattr(self, 'logger') else None
|
|
440
|
-
}
|
|
441
|
-
|
|
442
|
-
# 添加方法检查
|
|
443
|
-
info['methods'] = {
|
|
444
|
-
'has_parse': callable(getattr(self, 'parse', None)),
|
|
445
|
-
'has_spider_opened': callable(getattr(self, 'spider_opened', None)),
|
|
446
|
-
'has_spider_closed': callable(getattr(self, 'spider_closed', None)),
|
|
447
|
-
'has_start_requests': callable(getattr(self, 'start_requests', None))
|
|
448
|
-
}
|
|
449
|
-
|
|
450
|
-
return info
|
|
451
|
-
|
|
452
|
-
def make_request(self, url: str, callback=None, **kwargs) -> Request:
|
|
453
|
-
"""
|
|
454
|
-
便捷方法:创建 Request 对象
|
|
455
|
-
|
|
456
|
-
:param url: 请求URL
|
|
457
|
-
:param callback: 回调函数(默认为parse)
|
|
458
|
-
:param kwargs: 其他Request参数
|
|
459
|
-
:return: Request对象
|
|
460
|
-
"""
|
|
461
|
-
return Request(
|
|
462
|
-
url=url,
|
|
463
|
-
callback=callback or self.parse,
|
|
464
|
-
meta={'spider_name': self.name},
|
|
465
|
-
**kwargs
|
|
466
|
-
)
|
|
467
|
-
|
|
468
|
-
|
|
469
|
-
# === 高级爬虫功能扩展 ===
|
|
470
|
-
|
|
471
|
-
class SpiderStatsTracker:
|
|
472
|
-
"""
|
|
473
|
-
爬虫统计跟踪器
|
|
474
|
-
提供详细的性能监控功能
|
|
475
|
-
"""
|
|
476
|
-
|
|
477
|
-
def __init__(self, spider_name: str):
|
|
478
|
-
self.spider_name = spider_name
|
|
479
|
-
self.start_time = None
|
|
480
|
-
self.end_time = None
|
|
481
|
-
self.request_count = 0
|
|
482
|
-
self.response_count = 0
|
|
483
|
-
self.item_count = 0
|
|
484
|
-
self.error_count = 0
|
|
485
|
-
self.domain_stats = {}
|
|
486
|
-
|
|
487
|
-
def start_tracking(self):
|
|
488
|
-
"""开始统计"""
|
|
489
|
-
import time
|
|
490
|
-
self.start_time = time.time()
|
|
491
|
-
|
|
492
|
-
def stop_tracking(self):
|
|
493
|
-
"""停止统计"""
|
|
494
|
-
import time
|
|
495
|
-
self.end_time = time.time()
|
|
496
|
-
|
|
497
|
-
def record_request(self, url: str):
|
|
498
|
-
"""记录请求"""
|
|
499
|
-
self.request_count += 1
|
|
500
|
-
from urllib.parse import urlparse
|
|
501
|
-
domain = urlparse(url).netloc
|
|
502
|
-
self.domain_stats[domain] = self.domain_stats.get(domain, 0) + 1
|
|
503
|
-
|
|
504
|
-
def record_response(self):
|
|
505
|
-
"""记录响应"""
|
|
506
|
-
self.response_count += 1
|
|
507
|
-
|
|
508
|
-
def record_item(self):
|
|
509
|
-
"""记录Item"""
|
|
510
|
-
self.item_count += 1
|
|
511
|
-
|
|
512
|
-
def record_error(self):
|
|
513
|
-
"""记录错误"""
|
|
514
|
-
self.error_count += 1
|
|
515
|
-
|
|
516
|
-
def get_summary(self) -> Dict[str, Any]:
|
|
517
|
-
"""获取统计摘要"""
|
|
518
|
-
duration = (self.end_time - self.start_time) if (self.start_time and self.end_time) else 0
|
|
519
|
-
|
|
520
|
-
return {
|
|
521
|
-
'spider_name': self.spider_name,
|
|
522
|
-
'duration_seconds': round(duration, 2),
|
|
523
|
-
'requests': self.request_count,
|
|
524
|
-
'responses': self.response_count,
|
|
525
|
-
'items': self.item_count,
|
|
526
|
-
'errors': self.error_count,
|
|
527
|
-
'success_rate': round((self.response_count / max(1, self.request_count)) * 100, 2),
|
|
528
|
-
'requests_per_second': round(self.request_count / max(1, duration), 2),
|
|
529
|
-
'top_domains': sorted(
|
|
530
|
-
self.domain_stats.items(),
|
|
531
|
-
key=lambda x: x[1],
|
|
532
|
-
reverse=True
|
|
533
|
-
)[:5]
|
|
534
|
-
}
|
|
535
|
-
|
|
536
|
-
|
|
537
|
-
def create_spider_from_template(name: str, start_urls: List[str], **options) -> Type[Spider]:
|
|
538
|
-
"""
|
|
539
|
-
从模板快速创建爬虫类
|
|
540
|
-
|
|
541
|
-
:param name: 爬虫名称
|
|
542
|
-
:param start_urls: 起始URL列表
|
|
543
|
-
:param options: 其他选项
|
|
544
|
-
:return: 新创建的爬虫类
|
|
545
|
-
|
|
546
|
-
示例:
|
|
547
|
-
MySpider = create_spider_from_template(
|
|
548
|
-
name='quick_spider',
|
|
549
|
-
start_urls=['http://example.com'],
|
|
550
|
-
allowed_domains=['example.com'],
|
|
551
|
-
custom_settings={'CONCURRENCY': 5}
|
|
552
|
-
)
|
|
553
|
-
"""
|
|
554
|
-
|
|
555
|
-
# 动态创建爬虫类
|
|
556
|
-
class_attrs = {
|
|
557
|
-
'name': name,
|
|
558
|
-
'start_urls': start_urls,
|
|
559
|
-
'allowed_domains': options.get('allowed_domains', []),
|
|
560
|
-
'custom_settings': options.get('custom_settings', {})
|
|
561
|
-
}
|
|
562
|
-
|
|
563
|
-
# 添加自定义parse方法
|
|
564
|
-
if 'parse_function' in options:
|
|
565
|
-
class_attrs['parse'] = options['parse_function']
|
|
566
|
-
else:
|
|
567
|
-
def default_parse(self, response):
|
|
568
|
-
"""默认解析方法"""
|
|
569
|
-
yield {'url': response.url, 'title': getattr(response, 'title', 'N/A')}
|
|
570
|
-
class_attrs['parse'] = default_parse
|
|
571
|
-
|
|
572
|
-
# 创建类名
|
|
573
|
-
class_name = options.get('class_name', f"Generated{name.replace('_', '').title()}Spider")
|
|
574
|
-
|
|
575
|
-
# 动态创建类
|
|
576
|
-
spider_class = type(class_name, (Spider,), class_attrs)
|
|
577
|
-
|
|
578
|
-
get_logger(__name__).info(f"动态创建爬虫类: {class_name} (name='{name}')")
|
|
579
|
-
|
|
580
|
-
return spider_class
|
|
581
|
-
|
|
582
|
-
|
|
583
|
-
# === 公共只读接口 ===
|
|
584
|
-
def get_global_spider_registry() -> dict[str, Type[Spider]]:
|
|
585
|
-
"""
|
|
586
|
-
获取全局爬虫注册表的副本
|
|
587
|
-
|
|
588
|
-
:return: 爬虫注册表的副本
|
|
589
|
-
"""
|
|
590
|
-
return _DEFAULT_SPIDER_REGISTRY.copy()
|
|
591
|
-
|
|
592
|
-
|
|
593
|
-
def get_spider_by_name(name: str) -> Optional[Type[Spider]]:
|
|
594
|
-
"""
|
|
595
|
-
根据名称获取爬虫类
|
|
596
|
-
|
|
597
|
-
:param name: 爬虫名称
|
|
598
|
-
:return: 爬虫类或None
|
|
599
|
-
"""
|
|
600
|
-
return _DEFAULT_SPIDER_REGISTRY.get(name)
|
|
601
|
-
|
|
602
|
-
|
|
603
|
-
def get_all_spider_classes() -> List[Type[Spider]]:
|
|
604
|
-
"""
|
|
605
|
-
获取所有注册的爬虫类
|
|
606
|
-
|
|
607
|
-
:return: 爬虫类列表
|
|
608
|
-
"""
|
|
609
|
-
return list(set(_DEFAULT_SPIDER_REGISTRY.values()))
|
|
610
|
-
|
|
611
|
-
|
|
612
|
-
def get_spider_names() -> List[str]:
|
|
613
|
-
"""
|
|
614
|
-
获取所有爬虫名称
|
|
615
|
-
|
|
616
|
-
:return: 爬虫名称列表
|
|
617
|
-
"""
|
|
618
|
-
return list(_DEFAULT_SPIDER_REGISTRY.keys())
|
|
619
|
-
|
|
620
|
-
|
|
621
|
-
def is_spider_registered(name: str) -> bool:
|
|
622
|
-
"""
|
|
623
|
-
检查爬虫是否已注册
|
|
624
|
-
|
|
625
|
-
:param name: 爬虫名称
|
|
626
|
-
:return: 是否已注册
|
|
627
|
-
"""
|
|
628
|
-
return name in _DEFAULT_SPIDER_REGISTRY
|
|
629
|
-
|
|
630
|
-
|
|
631
|
-
def unregister_spider(name: str) -> bool:
|
|
632
|
-
"""
|
|
633
|
-
取消注册爬虫(仅用于测试)
|
|
634
|
-
|
|
635
|
-
:param name: 爬虫名称
|
|
636
|
-
:return: 是否成功取消注册
|
|
637
|
-
"""
|
|
638
|
-
if name in _DEFAULT_SPIDER_REGISTRY:
|
|
639
|
-
del _DEFAULT_SPIDER_REGISTRY[name]
|
|
640
|
-
return True
|
|
641
|
-
return False
|
|
642
|
-
|
|
643
|
-
|
|
644
|
-
# 导出的公共接口
|
|
645
|
-
__all__ = [
|
|
646
|
-
'Spider',
|
|
647
|
-
'SpiderMeta',
|
|
648
|
-
'SpiderStatsTracker',
|
|
649
|
-
'create_spider_from_template',
|
|
650
|
-
'get_global_spider_registry',
|
|
651
|
-
'get_spider_by_name',
|
|
652
|
-
'get_all_spider_classes',
|
|
653
|
-
'get_spider_names',
|
|
654
|
-
'is_spider_registered',
|
|
655
|
-
'unregister_spider'
|
|
656
|
-
]
|
|
657
|
-
|
|
1
|
+
#!/usr/bin/python
|
|
2
|
+
# -*- coding:UTF-8 -*-
|
|
3
|
+
"""
|
|
4
|
+
Crawlo Spider Module
|
|
5
|
+
==================
|
|
6
|
+
提供爬虫基类和相关功能。
|
|
7
|
+
|
|
8
|
+
核心功能:
|
|
9
|
+
- Spider基类:所有爬虫的基础类
|
|
10
|
+
- 自动注册机制:通过元类自动注册爬虫
|
|
11
|
+
- 配置管理:支持自定义设置和链式调用
|
|
12
|
+
- 生命周期管理:开启/关闭钩子函数
|
|
13
|
+
- 分布式支持:智能检测运行模式
|
|
14
|
+
|
|
15
|
+
使用示例:
|
|
16
|
+
class MySpider(Spider):
|
|
17
|
+
name = 'my_spider'
|
|
18
|
+
start_urls = ['http://example.com']
|
|
19
|
+
|
|
20
|
+
# 自定义配置
|
|
21
|
+
custom_settings = {
|
|
22
|
+
'DOWNLOADER_TYPE': 'httpx',
|
|
23
|
+
'CONCURRENCY': 10
|
|
24
|
+
}
|
|
25
|
+
|
|
26
|
+
def parse(self, response):
|
|
27
|
+
# 解析逻辑
|
|
28
|
+
yield Item(data=response.json())
|
|
29
|
+
"""
|
|
30
|
+
from __future__ import annotations
|
|
31
|
+
from typing import Type, Any, Optional, List, Dict, Union, Iterator, AsyncIterator
|
|
32
|
+
from ..network.request import Request
|
|
33
|
+
from ..utils.log import get_logger
|
|
34
|
+
|
|
35
|
+
|
|
36
|
+
# 全局爬虫注册表
|
|
37
|
+
_DEFAULT_SPIDER_REGISTRY: dict[str, Type[Spider]] = {}
|
|
38
|
+
|
|
39
|
+
|
|
40
|
+
class SpiderMeta(type):
|
|
41
|
+
"""
|
|
42
|
+
爬虫元类,提供自动注册功能
|
|
43
|
+
|
|
44
|
+
功能:
|
|
45
|
+
- 自动注册爬虫到全局注册表
|
|
46
|
+
- 验证爬虫名称的唯一性
|
|
47
|
+
- 提供完整的错误提示
|
|
48
|
+
"""
|
|
49
|
+
|
|
50
|
+
def __new__(mcs, name: str, bases: tuple[type], namespace: dict[str, Any], **kwargs):
|
|
51
|
+
cls = super().__new__(mcs, name, bases, namespace)
|
|
52
|
+
|
|
53
|
+
# 检查是否为Spider子类
|
|
54
|
+
is_spider_subclass = any(
|
|
55
|
+
base is Spider or (isinstance(base, type) and issubclass(base, Spider))
|
|
56
|
+
for base in bases
|
|
57
|
+
)
|
|
58
|
+
if not is_spider_subclass:
|
|
59
|
+
return cls
|
|
60
|
+
|
|
61
|
+
# 验证爬虫名称
|
|
62
|
+
spider_name = namespace.get('name')
|
|
63
|
+
if not isinstance(spider_name, str):
|
|
64
|
+
raise AttributeError(
|
|
65
|
+
f"爬虫类 '{cls.__name__}' 必须定义字符串类型的 'name' 属性。\n"
|
|
66
|
+
f"示例: name = 'my_spider'"
|
|
67
|
+
)
|
|
68
|
+
|
|
69
|
+
# 检查名称唯一性
|
|
70
|
+
if spider_name in _DEFAULT_SPIDER_REGISTRY:
|
|
71
|
+
existing_class = _DEFAULT_SPIDER_REGISTRY[spider_name]
|
|
72
|
+
raise ValueError(
|
|
73
|
+
f"爬虫名称 '{spider_name}' 已被 {existing_class.__name__} 占用。\n"
|
|
74
|
+
f"请确保每个爬虫的 name 属性全局唯一。\n"
|
|
75
|
+
f"建议使用格式: 'project_module_function'"
|
|
76
|
+
)
|
|
77
|
+
|
|
78
|
+
# 注册爬虫
|
|
79
|
+
_DEFAULT_SPIDER_REGISTRY[spider_name] = cls
|
|
80
|
+
# 延迟初始化logger避免模块级别阻塞
|
|
81
|
+
try:
|
|
82
|
+
from crawlo.logging import get_logger
|
|
83
|
+
get_logger(__name__).debug(f"自动注册爬虫: {spider_name} -> {cls.__name__}")
|
|
84
|
+
except:
|
|
85
|
+
# 如果日志系统未初始化,静默失败
|
|
86
|
+
pass
|
|
87
|
+
|
|
88
|
+
return cls
|
|
89
|
+
|
|
90
|
+
|
|
91
|
+
class Spider(metaclass=SpiderMeta):
|
|
92
|
+
"""
|
|
93
|
+
爬虫基类 - 所有爬虫实现的基础
|
|
94
|
+
|
|
95
|
+
必须定义的属性:
|
|
96
|
+
- name: 爬虫名称,必须全局唯一
|
|
97
|
+
|
|
98
|
+
可选配置:
|
|
99
|
+
- start_urls: 起始 URL 列表
|
|
100
|
+
- custom_settings: 自定义设置字典
|
|
101
|
+
- allowed_domains: 允许的域名列表
|
|
102
|
+
|
|
103
|
+
必须实现的方法:
|
|
104
|
+
- parse(response): 解析响应的主方法
|
|
105
|
+
|
|
106
|
+
可选实现的方法:
|
|
107
|
+
- spider_opened(): 爬虫开启时调用
|
|
108
|
+
- spider_closed(): 爬虫关闭时调用
|
|
109
|
+
- start_requests(): 生成初始请求(默认使用start_urls)
|
|
110
|
+
|
|
111
|
+
示例:
|
|
112
|
+
class MySpider(Spider):
|
|
113
|
+
name = 'example_spider'
|
|
114
|
+
start_urls = ['https://example.com']
|
|
115
|
+
|
|
116
|
+
custom_settings = {
|
|
117
|
+
'DOWNLOADER_TYPE': 'httpx',
|
|
118
|
+
'CONCURRENCY': 5,
|
|
119
|
+
'DOWNLOAD_DELAY': 1.0
|
|
120
|
+
}
|
|
121
|
+
|
|
122
|
+
def parse(self, response):
|
|
123
|
+
# 提取数据
|
|
124
|
+
data = response.css('title::text').get()
|
|
125
|
+
yield {'title': data}
|
|
126
|
+
|
|
127
|
+
# 生成新请求
|
|
128
|
+
for link in response.css('a::attr(href)').getall():
|
|
129
|
+
yield Request(url=link, callback=self.parse_detail)
|
|
130
|
+
"""
|
|
131
|
+
|
|
132
|
+
# 必须定义的属性
|
|
133
|
+
name: str = None
|
|
134
|
+
|
|
135
|
+
# 可选属性
|
|
136
|
+
start_urls: List[str] = None
|
|
137
|
+
custom_settings: Dict[str, Any] = None
|
|
138
|
+
allowed_domains: List[str] = None
|
|
139
|
+
|
|
140
|
+
def __init__(self, name: str = None, **kwargs):
|
|
141
|
+
"""
|
|
142
|
+
初始化爬虫实例
|
|
143
|
+
|
|
144
|
+
:param name: 爬虫名称(可选,默认使用类属性)
|
|
145
|
+
:param kwargs: 其他初始化参数
|
|
146
|
+
"""
|
|
147
|
+
# 初始化基本属性
|
|
148
|
+
if not hasattr(self, 'start_urls') or self.start_urls is None:
|
|
149
|
+
self.start_urls = []
|
|
150
|
+
if not hasattr(self, 'custom_settings') or self.custom_settings is None:
|
|
151
|
+
self.custom_settings = {}
|
|
152
|
+
if not hasattr(self, 'allowed_domains') or self.allowed_domains is None:
|
|
153
|
+
self.allowed_domains = []
|
|
154
|
+
|
|
155
|
+
# 设置爬虫名称
|
|
156
|
+
self.name = name or self.name
|
|
157
|
+
if not self.name:
|
|
158
|
+
raise ValueError(f"爬虫 {self.__class__.__name__} 必须指定 name 属性")
|
|
159
|
+
|
|
160
|
+
# 初始化其他属性
|
|
161
|
+
self.crawler = None
|
|
162
|
+
# 延迟初始化logger避免阻塞
|
|
163
|
+
self._logger = None
|
|
164
|
+
self.stats = None
|
|
165
|
+
|
|
166
|
+
# 应用额外参数
|
|
167
|
+
for key, value in kwargs.items():
|
|
168
|
+
setattr(self, key, value)
|
|
169
|
+
|
|
170
|
+
@property
|
|
171
|
+
def logger(self):
|
|
172
|
+
"""延迟初始化logger"""
|
|
173
|
+
if self._logger is None:
|
|
174
|
+
from crawlo.logging import get_logger
|
|
175
|
+
self._logger = get_logger(self.name)
|
|
176
|
+
return self._logger
|
|
177
|
+
|
|
178
|
+
@classmethod
|
|
179
|
+
def create_instance(cls, crawler) -> 'Spider':
|
|
180
|
+
"""
|
|
181
|
+
创建爬虫实例并绑定 crawler
|
|
182
|
+
|
|
183
|
+
:param crawler: Crawler 实例
|
|
184
|
+
:return: 爬虫实例
|
|
185
|
+
"""
|
|
186
|
+
spider = cls()
|
|
187
|
+
spider.crawler = crawler
|
|
188
|
+
spider.stats = getattr(crawler, 'stats', None)
|
|
189
|
+
|
|
190
|
+
# 合并自定义设置 - 使用延迟应用避免初始化时的循环依赖
|
|
191
|
+
if hasattr(spider, 'custom_settings') and spider.custom_settings:
|
|
192
|
+
# 延迟到真正需要时才应用设置
|
|
193
|
+
spider._pending_settings = spider.custom_settings.copy()
|
|
194
|
+
spider.logger.debug(f"准备应用 {len(spider.custom_settings)} 项自定义设置")
|
|
195
|
+
|
|
196
|
+
return spider
|
|
197
|
+
|
|
198
|
+
def apply_pending_settings(self):
|
|
199
|
+
"""应用待处理的设置(在初始化完成后调用)"""
|
|
200
|
+
if hasattr(self, '_pending_settings') and self._pending_settings:
|
|
201
|
+
for key, value in self._pending_settings.items():
|
|
202
|
+
if self.crawler and hasattr(self.crawler, 'settings'):
|
|
203
|
+
self.crawler.settings.set(key, value)
|
|
204
|
+
self.logger.debug(f"应用自定义设置: {key} = {value}")
|
|
205
|
+
# 清除待处理的设置
|
|
206
|
+
delattr(self, '_pending_settings')
|
|
207
|
+
|
|
208
|
+
def start_requests(self) -> Iterator[Request]:
|
|
209
|
+
"""
|
|
210
|
+
生成初始请求
|
|
211
|
+
|
|
212
|
+
默认行为:
|
|
213
|
+
- 使用 start_urls 生成请求
|
|
214
|
+
- 智能检测分布式模式决定是否去重
|
|
215
|
+
- 支持单个 start_url 属性(兼容性)
|
|
216
|
+
- 支持批量生成优化(大规模URL场景)
|
|
217
|
+
|
|
218
|
+
:return: Request 迭代器
|
|
219
|
+
"""
|
|
220
|
+
# 检测是否为分布式模式
|
|
221
|
+
is_distributed = self._is_distributed_mode()
|
|
222
|
+
|
|
223
|
+
# 获取批量处理配置
|
|
224
|
+
batch_size = self._get_batch_size()
|
|
225
|
+
|
|
226
|
+
# 从 start_urls 生成请求
|
|
227
|
+
if self.start_urls:
|
|
228
|
+
generated_count = 0
|
|
229
|
+
for url in self.start_urls:
|
|
230
|
+
if self._is_allowed_domain(url):
|
|
231
|
+
yield Request(
|
|
232
|
+
url=url,
|
|
233
|
+
callback=self.parse,
|
|
234
|
+
dont_filter=not is_distributed,
|
|
235
|
+
meta={'spider_name': self.name}
|
|
236
|
+
)
|
|
237
|
+
generated_count += 1
|
|
238
|
+
|
|
239
|
+
# 大规模URL时进行批量控制
|
|
240
|
+
if batch_size > 0 and generated_count % batch_size == 0:
|
|
241
|
+
self.logger.debug(f"已生成 {generated_count} 个请求(批量大小: {batch_size})")
|
|
242
|
+
else:
|
|
243
|
+
self.logger.warning(f"跳过不允许的域名: {url}")
|
|
244
|
+
|
|
245
|
+
# 兼容单个 start_url 属性
|
|
246
|
+
elif hasattr(self, 'start_url') and isinstance(getattr(self, 'start_url'), str):
|
|
247
|
+
url = getattr(self, 'start_url')
|
|
248
|
+
if self._is_allowed_domain(url):
|
|
249
|
+
yield Request(
|
|
250
|
+
url=url,
|
|
251
|
+
callback=self.parse,
|
|
252
|
+
dont_filter=not is_distributed,
|
|
253
|
+
meta={'spider_name': self.name}
|
|
254
|
+
)
|
|
255
|
+
else:
|
|
256
|
+
self.logger.warning(f"跳过不允许的域名: {url}")
|
|
257
|
+
|
|
258
|
+
else:
|
|
259
|
+
self.logger.warning(
|
|
260
|
+
f"爬虫 {self.name} 没有定义 start_urls 或 start_url。\n"
|
|
261
|
+
f"请在爬虫类中定义或重写 start_requests() 方法。"
|
|
262
|
+
)
|
|
263
|
+
|
|
264
|
+
def _get_batch_size(self) -> int:
|
|
265
|
+
"""
|
|
266
|
+
获取批量处理大小配置
|
|
267
|
+
|
|
268
|
+
用于大规模URL场景的性能优化
|
|
269
|
+
|
|
270
|
+
:return: 批量大小(0表示无限制)
|
|
271
|
+
"""
|
|
272
|
+
if not self.crawler:
|
|
273
|
+
return 0
|
|
274
|
+
|
|
275
|
+
# 从设置中获取批量大小
|
|
276
|
+
batch_size = self.crawler.settings.get_int('SPIDER_BATCH_SIZE', 0)
|
|
277
|
+
|
|
278
|
+
# 如果start_urls超过一定数量,自动启用批量模式
|
|
279
|
+
if batch_size == 0 and self.start_urls and len(self.start_urls) > 1000:
|
|
280
|
+
batch_size = 500 # 默认批量大小
|
|
281
|
+
self.logger.info(f"检测到大量start_urls ({len(self.start_urls)}),启用批量模式 (批量大小: {batch_size})")
|
|
282
|
+
|
|
283
|
+
return batch_size
|
|
284
|
+
|
|
285
|
+
def _is_distributed_mode(self) -> bool:
|
|
286
|
+
"""
|
|
287
|
+
智能检测是否为分布式模式
|
|
288
|
+
|
|
289
|
+
检测条件:
|
|
290
|
+
- QUEUE_TYPE = 'redis'
|
|
291
|
+
- FILTER_CLASS 包含 'aioredis_filter'
|
|
292
|
+
- RUN_MODE = 'distributed'
|
|
293
|
+
|
|
294
|
+
:return: 是否为分布式模式
|
|
295
|
+
"""
|
|
296
|
+
if not self.crawler:
|
|
297
|
+
return False
|
|
298
|
+
|
|
299
|
+
settings = self.crawler.settings
|
|
300
|
+
|
|
301
|
+
# 检查多个条件来判断是否为分布式模式
|
|
302
|
+
queue_type = settings.get('QUEUE_TYPE', 'memory')
|
|
303
|
+
filter_class = settings.get('FILTER_CLASS', '')
|
|
304
|
+
run_mode = settings.get('RUN_MODE', 'standalone')
|
|
305
|
+
|
|
306
|
+
# 分布式模式的标志
|
|
307
|
+
is_redis_queue = queue_type == 'redis'
|
|
308
|
+
is_redis_filter = 'aioredis_filter' in filter_class.lower()
|
|
309
|
+
is_distributed_run_mode = run_mode == 'distributed'
|
|
310
|
+
|
|
311
|
+
distributed = is_redis_queue or is_redis_filter or is_distributed_run_mode
|
|
312
|
+
|
|
313
|
+
if distributed:
|
|
314
|
+
self.logger.debug("检测到分布式模式,启用请求去重")
|
|
315
|
+
else:
|
|
316
|
+
self.logger.debug("检测到单机模式,禁用请求去重")
|
|
317
|
+
|
|
318
|
+
return distributed
|
|
319
|
+
|
|
320
|
+
def _is_allowed_domain(self, url: str) -> bool:
|
|
321
|
+
"""
|
|
322
|
+
检查URL是否在允许的域名列表中
|
|
323
|
+
|
|
324
|
+
:param url: 要检查的URL
|
|
325
|
+
:return: 是否允许
|
|
326
|
+
"""
|
|
327
|
+
if not self.allowed_domains:
|
|
328
|
+
return True
|
|
329
|
+
|
|
330
|
+
from urllib.parse import urlparse
|
|
331
|
+
try:
|
|
332
|
+
domain = urlparse(url).netloc.lower()
|
|
333
|
+
return any(
|
|
334
|
+
domain == allowed.lower() or domain.endswith('.' + allowed.lower())
|
|
335
|
+
for allowed in self.allowed_domains
|
|
336
|
+
)
|
|
337
|
+
except Exception as e:
|
|
338
|
+
self.logger.warning(f"URL解析失败: {url} - {e}")
|
|
339
|
+
return False
|
|
340
|
+
|
|
341
|
+
def parse(self, response):
|
|
342
|
+
"""
|
|
343
|
+
解析响应的主方法(必须实现)
|
|
344
|
+
|
|
345
|
+
:param response: 响应对象
|
|
346
|
+
:return: 生成的 Item 或 Request
|
|
347
|
+
"""
|
|
348
|
+
raise NotImplementedError(
|
|
349
|
+
f"爬虫 {self.__class__.__name__} 必须实现 parse() 方法\n"
|
|
350
|
+
f"示例:\n"
|
|
351
|
+
f"def parse(self, response):\n"
|
|
352
|
+
f" # 提取数据\n"
|
|
353
|
+
f" yield {{'title': response.css('title::text').get()}}\n"
|
|
354
|
+
f" # 生成新请求\n"
|
|
355
|
+
f" for link in response.css('a::attr(href)').getall():\n"
|
|
356
|
+
f" yield Request(url=link)"
|
|
357
|
+
)
|
|
358
|
+
|
|
359
|
+
async def spider_opened(self):
|
|
360
|
+
"""
|
|
361
|
+
爬虫开启时调用的钩子函数
|
|
362
|
+
|
|
363
|
+
可用于:
|
|
364
|
+
- 初始化资源
|
|
365
|
+
- 连接数据库
|
|
366
|
+
- 设置初始状态
|
|
367
|
+
"""
|
|
368
|
+
self.logger.info(f"Spider {self.name} opened")
|
|
369
|
+
|
|
370
|
+
async def spider_closed(self):
|
|
371
|
+
"""
|
|
372
|
+
爬虫关闭时调用的钩子函数
|
|
373
|
+
|
|
374
|
+
可用于:
|
|
375
|
+
- 清理资源
|
|
376
|
+
- 关闭数据库连接
|
|
377
|
+
"""
|
|
378
|
+
# 不再输出任何信息,避免与统计信息重复
|
|
379
|
+
# 统计信息由StatsCollector负责输出
|
|
380
|
+
pass
|
|
381
|
+
|
|
382
|
+
def __str__(self) -> str:
|
|
383
|
+
return f"{self.__class__.__name__}(name='{self.name}')"
|
|
384
|
+
|
|
385
|
+
def __repr__(self) -> str:
|
|
386
|
+
return self.__str__()
|
|
387
|
+
|
|
388
|
+
def set_custom_setting(self, key: str, value: Any) -> 'Spider':
|
|
389
|
+
"""
|
|
390
|
+
设置自定义配置(链式调用)
|
|
391
|
+
|
|
392
|
+
:param key: 配置键名
|
|
393
|
+
:param value: 配置值
|
|
394
|
+
:return: self(支持链式调用)
|
|
395
|
+
|
|
396
|
+
示例:
|
|
397
|
+
spider.set_custom_setting('CONCURRENCY', 10)\
|
|
398
|
+
.set_custom_setting('DOWNLOAD_DELAY', 1.0)
|
|
399
|
+
"""
|
|
400
|
+
if not hasattr(self, 'custom_settings') or self.custom_settings is None:
|
|
401
|
+
self.custom_settings = {}
|
|
402
|
+
|
|
403
|
+
self.custom_settings[key] = value
|
|
404
|
+
self.logger.debug(f"设置自定义配置: {key} = {value}")
|
|
405
|
+
|
|
406
|
+
# 如果已绑定crawler,立即应用设置
|
|
407
|
+
if self.crawler:
|
|
408
|
+
self.crawler.settings.set(key, value)
|
|
409
|
+
|
|
410
|
+
return self
|
|
411
|
+
|
|
412
|
+
def get_custom_setting(self, key: str, default: Any = None) -> Any:
|
|
413
|
+
"""
|
|
414
|
+
获取自定义配置值
|
|
415
|
+
|
|
416
|
+
:param key: 配置键名
|
|
417
|
+
:param default: 默认值
|
|
418
|
+
:return: 配置值
|
|
419
|
+
"""
|
|
420
|
+
if hasattr(self, 'custom_settings') and self.custom_settings:
|
|
421
|
+
return self.custom_settings.get(key, default)
|
|
422
|
+
return default
|
|
423
|
+
|
|
424
|
+
def get_spider_info(self) -> Dict[str, Any]:
|
|
425
|
+
"""
|
|
426
|
+
获取爬虫详细信息
|
|
427
|
+
|
|
428
|
+
:return: 爬虫信息字典
|
|
429
|
+
"""
|
|
430
|
+
info = {
|
|
431
|
+
'name': self.name,
|
|
432
|
+
'class_name': self.__class__.__name__,
|
|
433
|
+
'module': self.__module__,
|
|
434
|
+
'start_urls_count': len(self.start_urls) if self.start_urls else 0,
|
|
435
|
+
'allowed_domains_count': len(self.allowed_domains) if self.allowed_domains else 0,
|
|
436
|
+
'custom_settings_count': len(self.custom_settings) if self.custom_settings else 0,
|
|
437
|
+
'is_distributed': self._is_distributed_mode() if self.crawler else None,
|
|
438
|
+
'has_crawler': self.crawler is not None,
|
|
439
|
+
'logger_name': self.logger.name if hasattr(self, 'logger') else None
|
|
440
|
+
}
|
|
441
|
+
|
|
442
|
+
# 添加方法检查
|
|
443
|
+
info['methods'] = {
|
|
444
|
+
'has_parse': callable(getattr(self, 'parse', None)),
|
|
445
|
+
'has_spider_opened': callable(getattr(self, 'spider_opened', None)),
|
|
446
|
+
'has_spider_closed': callable(getattr(self, 'spider_closed', None)),
|
|
447
|
+
'has_start_requests': callable(getattr(self, 'start_requests', None))
|
|
448
|
+
}
|
|
449
|
+
|
|
450
|
+
return info
|
|
451
|
+
|
|
452
|
+
def make_request(self, url: str, callback=None, **kwargs) -> Request:
|
|
453
|
+
"""
|
|
454
|
+
便捷方法:创建 Request 对象
|
|
455
|
+
|
|
456
|
+
:param url: 请求URL
|
|
457
|
+
:param callback: 回调函数(默认为parse)
|
|
458
|
+
:param kwargs: 其他Request参数
|
|
459
|
+
:return: Request对象
|
|
460
|
+
"""
|
|
461
|
+
return Request(
|
|
462
|
+
url=url,
|
|
463
|
+
callback=callback or self.parse,
|
|
464
|
+
meta={'spider_name': self.name},
|
|
465
|
+
**kwargs
|
|
466
|
+
)
|
|
467
|
+
|
|
468
|
+
|
|
469
|
+
# === 高级爬虫功能扩展 ===
|
|
470
|
+
|
|
471
|
+
class SpiderStatsTracker:
|
|
472
|
+
"""
|
|
473
|
+
爬虫统计跟踪器
|
|
474
|
+
提供详细的性能监控功能
|
|
475
|
+
"""
|
|
476
|
+
|
|
477
|
+
def __init__(self, spider_name: str):
|
|
478
|
+
self.spider_name = spider_name
|
|
479
|
+
self.start_time = None
|
|
480
|
+
self.end_time = None
|
|
481
|
+
self.request_count = 0
|
|
482
|
+
self.response_count = 0
|
|
483
|
+
self.item_count = 0
|
|
484
|
+
self.error_count = 0
|
|
485
|
+
self.domain_stats = {}
|
|
486
|
+
|
|
487
|
+
def start_tracking(self):
|
|
488
|
+
"""开始统计"""
|
|
489
|
+
import time
|
|
490
|
+
self.start_time = time.time()
|
|
491
|
+
|
|
492
|
+
def stop_tracking(self):
|
|
493
|
+
"""停止统计"""
|
|
494
|
+
import time
|
|
495
|
+
self.end_time = time.time()
|
|
496
|
+
|
|
497
|
+
def record_request(self, url: str):
|
|
498
|
+
"""记录请求"""
|
|
499
|
+
self.request_count += 1
|
|
500
|
+
from urllib.parse import urlparse
|
|
501
|
+
domain = urlparse(url).netloc
|
|
502
|
+
self.domain_stats[domain] = self.domain_stats.get(domain, 0) + 1
|
|
503
|
+
|
|
504
|
+
def record_response(self):
|
|
505
|
+
"""记录响应"""
|
|
506
|
+
self.response_count += 1
|
|
507
|
+
|
|
508
|
+
def record_item(self):
|
|
509
|
+
"""记录Item"""
|
|
510
|
+
self.item_count += 1
|
|
511
|
+
|
|
512
|
+
def record_error(self):
|
|
513
|
+
"""记录错误"""
|
|
514
|
+
self.error_count += 1
|
|
515
|
+
|
|
516
|
+
def get_summary(self) -> Dict[str, Any]:
|
|
517
|
+
"""获取统计摘要"""
|
|
518
|
+
duration = (self.end_time - self.start_time) if (self.start_time and self.end_time) else 0
|
|
519
|
+
|
|
520
|
+
return {
|
|
521
|
+
'spider_name': self.spider_name,
|
|
522
|
+
'duration_seconds': round(duration, 2),
|
|
523
|
+
'requests': self.request_count,
|
|
524
|
+
'responses': self.response_count,
|
|
525
|
+
'items': self.item_count,
|
|
526
|
+
'errors': self.error_count,
|
|
527
|
+
'success_rate': round((self.response_count / max(1, self.request_count)) * 100, 2),
|
|
528
|
+
'requests_per_second': round(self.request_count / max(1, duration), 2),
|
|
529
|
+
'top_domains': sorted(
|
|
530
|
+
self.domain_stats.items(),
|
|
531
|
+
key=lambda x: x[1],
|
|
532
|
+
reverse=True
|
|
533
|
+
)[:5]
|
|
534
|
+
}
|
|
535
|
+
|
|
536
|
+
|
|
537
|
+
def create_spider_from_template(name: str, start_urls: List[str], **options) -> Type[Spider]:
|
|
538
|
+
"""
|
|
539
|
+
从模板快速创建爬虫类
|
|
540
|
+
|
|
541
|
+
:param name: 爬虫名称
|
|
542
|
+
:param start_urls: 起始URL列表
|
|
543
|
+
:param options: 其他选项
|
|
544
|
+
:return: 新创建的爬虫类
|
|
545
|
+
|
|
546
|
+
示例:
|
|
547
|
+
MySpider = create_spider_from_template(
|
|
548
|
+
name='quick_spider',
|
|
549
|
+
start_urls=['http://example.com'],
|
|
550
|
+
allowed_domains=['example.com'],
|
|
551
|
+
custom_settings={'CONCURRENCY': 5}
|
|
552
|
+
)
|
|
553
|
+
"""
|
|
554
|
+
|
|
555
|
+
# 动态创建爬虫类
|
|
556
|
+
class_attrs = {
|
|
557
|
+
'name': name,
|
|
558
|
+
'start_urls': start_urls,
|
|
559
|
+
'allowed_domains': options.get('allowed_domains', []),
|
|
560
|
+
'custom_settings': options.get('custom_settings', {})
|
|
561
|
+
}
|
|
562
|
+
|
|
563
|
+
# 添加自定义parse方法
|
|
564
|
+
if 'parse_function' in options:
|
|
565
|
+
class_attrs['parse'] = options['parse_function']
|
|
566
|
+
else:
|
|
567
|
+
def default_parse(self, response):
|
|
568
|
+
"""默认解析方法"""
|
|
569
|
+
yield {'url': response.url, 'title': getattr(response, 'title', 'N/A')}
|
|
570
|
+
class_attrs['parse'] = default_parse
|
|
571
|
+
|
|
572
|
+
# 创建类名
|
|
573
|
+
class_name = options.get('class_name', f"Generated{name.replace('_', '').title()}Spider")
|
|
574
|
+
|
|
575
|
+
# 动态创建类
|
|
576
|
+
spider_class = type(class_name, (Spider,), class_attrs)
|
|
577
|
+
|
|
578
|
+
get_logger(__name__).info(f"动态创建爬虫类: {class_name} (name='{name}')")
|
|
579
|
+
|
|
580
|
+
return spider_class
|
|
581
|
+
|
|
582
|
+
|
|
583
|
+
# === 公共只读接口 ===
|
|
584
|
+
def get_global_spider_registry() -> dict[str, Type[Spider]]:
|
|
585
|
+
"""
|
|
586
|
+
获取全局爬虫注册表的副本
|
|
587
|
+
|
|
588
|
+
:return: 爬虫注册表的副本
|
|
589
|
+
"""
|
|
590
|
+
return _DEFAULT_SPIDER_REGISTRY.copy()
|
|
591
|
+
|
|
592
|
+
|
|
593
|
+
def get_spider_by_name(name: str) -> Optional[Type[Spider]]:
|
|
594
|
+
"""
|
|
595
|
+
根据名称获取爬虫类
|
|
596
|
+
|
|
597
|
+
:param name: 爬虫名称
|
|
598
|
+
:return: 爬虫类或None
|
|
599
|
+
"""
|
|
600
|
+
return _DEFAULT_SPIDER_REGISTRY.get(name)
|
|
601
|
+
|
|
602
|
+
|
|
603
|
+
def get_all_spider_classes() -> List[Type[Spider]]:
|
|
604
|
+
"""
|
|
605
|
+
获取所有注册的爬虫类
|
|
606
|
+
|
|
607
|
+
:return: 爬虫类列表
|
|
608
|
+
"""
|
|
609
|
+
return list(set(_DEFAULT_SPIDER_REGISTRY.values()))
|
|
610
|
+
|
|
611
|
+
|
|
612
|
+
def get_spider_names() -> List[str]:
|
|
613
|
+
"""
|
|
614
|
+
获取所有爬虫名称
|
|
615
|
+
|
|
616
|
+
:return: 爬虫名称列表
|
|
617
|
+
"""
|
|
618
|
+
return list(_DEFAULT_SPIDER_REGISTRY.keys())
|
|
619
|
+
|
|
620
|
+
|
|
621
|
+
def is_spider_registered(name: str) -> bool:
|
|
622
|
+
"""
|
|
623
|
+
检查爬虫是否已注册
|
|
624
|
+
|
|
625
|
+
:param name: 爬虫名称
|
|
626
|
+
:return: 是否已注册
|
|
627
|
+
"""
|
|
628
|
+
return name in _DEFAULT_SPIDER_REGISTRY
|
|
629
|
+
|
|
630
|
+
|
|
631
|
+
def unregister_spider(name: str) -> bool:
|
|
632
|
+
"""
|
|
633
|
+
取消注册爬虫(仅用于测试)
|
|
634
|
+
|
|
635
|
+
:param name: 爬虫名称
|
|
636
|
+
:return: 是否成功取消注册
|
|
637
|
+
"""
|
|
638
|
+
if name in _DEFAULT_SPIDER_REGISTRY:
|
|
639
|
+
del _DEFAULT_SPIDER_REGISTRY[name]
|
|
640
|
+
return True
|
|
641
|
+
return False
|
|
642
|
+
|
|
643
|
+
|
|
644
|
+
# 导出的公共接口
|
|
645
|
+
__all__ = [
|
|
646
|
+
'Spider',
|
|
647
|
+
'SpiderMeta',
|
|
648
|
+
'SpiderStatsTracker',
|
|
649
|
+
'create_spider_from_template',
|
|
650
|
+
'get_global_spider_registry',
|
|
651
|
+
'get_spider_by_name',
|
|
652
|
+
'get_all_spider_classes',
|
|
653
|
+
'get_spider_names',
|
|
654
|
+
'is_spider_registered',
|
|
655
|
+
'unregister_spider'
|
|
656
|
+
]
|
|
657
|
+
|