outcrawl 0.1.0__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.
Files changed (15) hide show
  1. outcrawl/__init__.py +177 -0
  2. outcrawl/agent.py +697 -0
  3. outcrawl/availability.py +198 -0
  4. outcrawl/client.py +158 -0
  5. outcrawl/errors.py +1082 -0
  6. outcrawl/py.typed +0 -0
  7. outcrawl/registry.py +453 -0
  8. outcrawl/resources.py +447 -0
  9. outcrawl/result.py +182 -0
  10. outcrawl/scrape.py +212 -0
  11. outcrawl/transport.py +278 -0
  12. outcrawl-0.1.0.dist-info/METADATA +156 -0
  13. outcrawl-0.1.0.dist-info/RECORD +15 -0
  14. outcrawl-0.1.0.dist-info/WHEEL +4 -0
  15. outcrawl-0.1.0.dist-info/licenses/LICENSE +21 -0
outcrawl/__init__.py ADDED
@@ -0,0 +1,177 @@
1
+ """``outcrawl`` — the Python client for Outcrawl.
2
+
3
+ Scrape, crawl, search and run agent tasks against real stealth browsers, on one
4
+ credit balance. Pure Python: one ``py3-none-any`` wheel for every operating
5
+ system, and no compiler on any of them.
6
+
7
+ pip install outcrawl
8
+
9
+ The surface mirrors ``@outcrawl/sdk`` on npm, in Python casing, and the two are
10
+ held to the same contract by the same source of truth: every one of the 34 rows
11
+ in ``packages/core/src/registry.ts`` is reachable from this client, and
12
+ ``tests/test_registry.py`` fails when one is not.
13
+
14
+ Errors are the TypeScript SDK's, class for class, field for field, sentence for
15
+ sentence — branch on ``error.code``, never on ``str(error)``.
16
+
17
+ WHAT IS NOT HERE, and it is one thing rather than a missing capability: the live
18
+ browser (``oc.browser()`` in TypeScript). It is deliberately not a registry row
19
+ — it needs a CDP connection driven in the caller's own process — and it reaches
20
+ the same page capabilities through ``agent`` and ``scrape``, which run the loop
21
+ on our side. That is a difference in kind, not an omission.
22
+ """
23
+
24
+ from __future__ import annotations
25
+
26
+ from .agent import (
27
+ AGENT_EVENT_LEVELS,
28
+ AGENT_REQUEST_FIELDS,
29
+ AgentApi,
30
+ AgentEvents,
31
+ AgentHandle,
32
+ AgentJob,
33
+ DEFAULT_AGENT_EVENT_LEVEL,
34
+ InvalidAgentRequestError,
35
+ validate_agent_request,
36
+ )
37
+ from .availability import CAPABILITY_AVAILABILITY, Unavailable, assert_available
38
+ from .client import Outcrawl
39
+ from .errors import (
40
+ OUTCRAWL_ERROR_CODES,
41
+ AnswerLostError,
42
+ BadRequestError,
43
+ CapabilityUnavailableError,
44
+ DeadlineExceededError,
45
+ DecodeUnavailableError,
46
+ DocumentUnrepresentableError,
47
+ CapExceededError,
48
+ ConcurrencyLimitError,
49
+ ExitUnavailableError,
50
+ ExtractionFailedError,
51
+ InternalError,
52
+ MethodNotAllowedError,
53
+ NoCapacityError,
54
+ NotFoundError,
55
+ OutcrawlApiError,
56
+ OutcrawlError,
57
+ PageLostError,
58
+ PageUnreachableError,
59
+ PlanRequiredError,
60
+ ProfileInUseError,
61
+ ProfileNotFoundError,
62
+ QuotaExceededError,
63
+ TenantScopeError,
64
+ UnauthorizedError,
65
+ error_from_wire,
66
+ is_known_error_code,
67
+ is_outcrawl_error,
68
+ )
69
+ from .registry import (
70
+ CAPABILITIES,
71
+ CAPABILITY_NAMES,
72
+ CLI_CAPABILITIES,
73
+ MCP_CAPABILITIES,
74
+ Capability,
75
+ get_capability,
76
+ )
77
+ from .resources import (
78
+ IdentityOverrideError,
79
+ Integrations,
80
+ Monitors,
81
+ Profiles,
82
+ Rules,
83
+ Secrets,
84
+ Session,
85
+ Sessions,
86
+ UnsupportedSessionFilterError,
87
+ )
88
+ from .result import Deleted, Paged, Result
89
+ from .scrape import CrawlApi, CrawlHandle, CrawlJob
90
+ from .transport import (
91
+ API_KEY_ENV,
92
+ API_URL_ENV,
93
+ DEFAULT_BASE_URL,
94
+ HttpTransport,
95
+ RequestSpec,
96
+ Transport,
97
+ always_streams,
98
+ resolve_request,
99
+ stream_switch,
100
+ )
101
+
102
+ __version__ = "0.1.0"
103
+
104
+ __all__ = [
105
+ "AGENT_EVENT_LEVELS",
106
+ "AGENT_REQUEST_FIELDS",
107
+ "API_KEY_ENV",
108
+ "API_URL_ENV",
109
+ "AgentApi",
110
+ "AgentEvents",
111
+ "AgentHandle",
112
+ "AgentJob",
113
+ "AnswerLostError",
114
+ "BadRequestError",
115
+ "CAPABILITIES",
116
+ "CAPABILITY_AVAILABILITY",
117
+ "CAPABILITY_NAMES",
118
+ "CLI_CAPABILITIES",
119
+ "Capability",
120
+ "CapExceededError",
121
+ "CapabilityUnavailableError",
122
+ "DeadlineExceededError",
123
+ "DecodeUnavailableError",
124
+ "DocumentUnrepresentableError",
125
+ "ConcurrencyLimitError",
126
+ "CrawlApi",
127
+ "CrawlHandle",
128
+ "CrawlJob",
129
+ "DEFAULT_AGENT_EVENT_LEVEL",
130
+ "DEFAULT_BASE_URL",
131
+ "Deleted",
132
+ "ExitUnavailableError",
133
+ "ExtractionFailedError",
134
+ "HttpTransport",
135
+ "IdentityOverrideError",
136
+ "Integrations",
137
+ "InternalError",
138
+ "InvalidAgentRequestError",
139
+ "MCP_CAPABILITIES",
140
+ "MethodNotAllowedError",
141
+ "Monitors",
142
+ "NoCapacityError",
143
+ "NotFoundError",
144
+ "OUTCRAWL_ERROR_CODES",
145
+ "Outcrawl",
146
+ "OutcrawlApiError",
147
+ "OutcrawlError",
148
+ "Paged",
149
+ "PageLostError",
150
+ "PageUnreachableError",
151
+ "PlanRequiredError",
152
+ "Profiles",
153
+ "ProfileInUseError",
154
+ "ProfileNotFoundError",
155
+ "QuotaExceededError",
156
+ "RequestSpec",
157
+ "Result",
158
+ "Rules",
159
+ "Secrets",
160
+ "Session",
161
+ "Sessions",
162
+ "TenantScopeError",
163
+ "Transport",
164
+ "UnauthorizedError",
165
+ "Unavailable",
166
+ "UnsupportedSessionFilterError",
167
+ "__version__",
168
+ "always_streams",
169
+ "assert_available",
170
+ "error_from_wire",
171
+ "get_capability",
172
+ "is_known_error_code",
173
+ "is_outcrawl_error",
174
+ "resolve_request",
175
+ "stream_switch",
176
+ "validate_agent_request",
177
+ ]