edsl 0.1.36.dev6__py3-none-any.whl → 0.1.37.dev2__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 (257) hide show
  1. edsl/Base.py +303 -303
  2. edsl/BaseDiff.py +260 -260
  3. edsl/TemplateLoader.py +24 -24
  4. edsl/__init__.py +48 -47
  5. edsl/__version__.py +1 -1
  6. edsl/agents/Agent.py +804 -804
  7. edsl/agents/AgentList.py +345 -337
  8. edsl/agents/Invigilator.py +222 -222
  9. edsl/agents/InvigilatorBase.py +305 -294
  10. edsl/agents/PromptConstructor.py +312 -312
  11. edsl/agents/__init__.py +3 -3
  12. edsl/agents/descriptors.py +86 -86
  13. edsl/agents/prompt_helpers.py +129 -129
  14. edsl/auto/AutoStudy.py +117 -117
  15. edsl/auto/StageBase.py +230 -230
  16. edsl/auto/StageGenerateSurvey.py +178 -178
  17. edsl/auto/StageLabelQuestions.py +125 -125
  18. edsl/auto/StagePersona.py +61 -61
  19. edsl/auto/StagePersonaDimensionValueRanges.py +88 -88
  20. edsl/auto/StagePersonaDimensionValues.py +74 -74
  21. edsl/auto/StagePersonaDimensions.py +69 -69
  22. edsl/auto/StageQuestions.py +73 -73
  23. edsl/auto/SurveyCreatorPipeline.py +21 -21
  24. edsl/auto/utilities.py +224 -224
  25. edsl/base/Base.py +289 -289
  26. edsl/config.py +149 -149
  27. edsl/conjure/AgentConstructionMixin.py +152 -152
  28. edsl/conjure/Conjure.py +62 -62
  29. edsl/conjure/InputData.py +659 -659
  30. edsl/conjure/InputDataCSV.py +48 -48
  31. edsl/conjure/InputDataMixinQuestionStats.py +182 -182
  32. edsl/conjure/InputDataPyRead.py +91 -91
  33. edsl/conjure/InputDataSPSS.py +8 -8
  34. edsl/conjure/InputDataStata.py +8 -8
  35. edsl/conjure/QuestionOptionMixin.py +76 -76
  36. edsl/conjure/QuestionTypeMixin.py +23 -23
  37. edsl/conjure/RawQuestion.py +65 -65
  38. edsl/conjure/SurveyResponses.py +7 -7
  39. edsl/conjure/__init__.py +9 -9
  40. edsl/conjure/naming_utilities.py +263 -263
  41. edsl/conjure/utilities.py +201 -201
  42. edsl/conversation/Conversation.py +238 -238
  43. edsl/conversation/car_buying.py +58 -58
  44. edsl/conversation/mug_negotiation.py +81 -81
  45. edsl/conversation/next_speaker_utilities.py +93 -93
  46. edsl/coop/PriceFetcher.py +54 -54
  47. edsl/coop/__init__.py +2 -2
  48. edsl/coop/coop.py +824 -849
  49. edsl/coop/utils.py +131 -131
  50. edsl/data/Cache.py +527 -527
  51. edsl/data/CacheEntry.py +228 -228
  52. edsl/data/CacheHandler.py +149 -149
  53. edsl/data/RemoteCacheSync.py +97 -84
  54. edsl/data/SQLiteDict.py +292 -292
  55. edsl/data/__init__.py +4 -4
  56. edsl/data/orm.py +10 -10
  57. edsl/data_transfer_models.py +73 -73
  58. edsl/enums.py +173 -173
  59. edsl/exceptions/__init__.py +50 -50
  60. edsl/exceptions/agents.py +40 -40
  61. edsl/exceptions/configuration.py +16 -16
  62. edsl/exceptions/coop.py +10 -10
  63. edsl/exceptions/data.py +14 -14
  64. edsl/exceptions/general.py +34 -34
  65. edsl/exceptions/jobs.py +33 -33
  66. edsl/exceptions/language_models.py +63 -63
  67. edsl/exceptions/prompts.py +15 -15
  68. edsl/exceptions/questions.py +91 -91
  69. edsl/exceptions/results.py +26 -26
  70. edsl/exceptions/surveys.py +34 -34
  71. edsl/inference_services/AnthropicService.py +87 -87
  72. edsl/inference_services/AwsBedrock.py +115 -115
  73. edsl/inference_services/AzureAI.py +217 -217
  74. edsl/inference_services/DeepInfraService.py +18 -18
  75. edsl/inference_services/GoogleService.py +156 -156
  76. edsl/inference_services/GroqService.py +20 -20
  77. edsl/inference_services/InferenceServiceABC.py +147 -147
  78. edsl/inference_services/InferenceServicesCollection.py +74 -72
  79. edsl/inference_services/MistralAIService.py +123 -123
  80. edsl/inference_services/OllamaService.py +18 -18
  81. edsl/inference_services/OpenAIService.py +224 -224
  82. edsl/inference_services/TestService.py +89 -89
  83. edsl/inference_services/TogetherAIService.py +170 -170
  84. edsl/inference_services/models_available_cache.py +118 -118
  85. edsl/inference_services/rate_limits_cache.py +25 -25
  86. edsl/inference_services/registry.py +39 -39
  87. edsl/inference_services/write_available.py +10 -10
  88. edsl/jobs/Answers.py +56 -56
  89. edsl/jobs/Jobs.py +1112 -1112
  90. edsl/jobs/__init__.py +1 -1
  91. edsl/jobs/buckets/BucketCollection.py +63 -63
  92. edsl/jobs/buckets/ModelBuckets.py +65 -65
  93. edsl/jobs/buckets/TokenBucket.py +248 -248
  94. edsl/jobs/interviews/Interview.py +661 -651
  95. edsl/jobs/interviews/InterviewExceptionCollection.py +99 -99
  96. edsl/jobs/interviews/InterviewExceptionEntry.py +182 -182
  97. edsl/jobs/interviews/InterviewStatistic.py +63 -63
  98. edsl/jobs/interviews/InterviewStatisticsCollection.py +25 -25
  99. edsl/jobs/interviews/InterviewStatusDictionary.py +78 -78
  100. edsl/jobs/interviews/InterviewStatusLog.py +92 -92
  101. edsl/jobs/interviews/ReportErrors.py +66 -66
  102. edsl/jobs/interviews/interview_status_enum.py +9 -9
  103. edsl/jobs/runners/JobsRunnerAsyncio.py +338 -337
  104. edsl/jobs/runners/JobsRunnerStatus.py +332 -332
  105. edsl/jobs/tasks/QuestionTaskCreator.py +242 -242
  106. edsl/jobs/tasks/TaskCreators.py +64 -64
  107. edsl/jobs/tasks/TaskHistory.py +441 -441
  108. edsl/jobs/tasks/TaskStatusLog.py +23 -23
  109. edsl/jobs/tasks/task_status_enum.py +163 -163
  110. edsl/jobs/tokens/InterviewTokenUsage.py +27 -27
  111. edsl/jobs/tokens/TokenUsage.py +34 -34
  112. edsl/language_models/LanguageModel.py +718 -718
  113. edsl/language_models/ModelList.py +102 -102
  114. edsl/language_models/RegisterLanguageModelsMeta.py +184 -184
  115. edsl/language_models/__init__.py +2 -2
  116. edsl/language_models/fake_openai_call.py +15 -15
  117. edsl/language_models/fake_openai_service.py +61 -61
  118. edsl/language_models/registry.py +137 -137
  119. edsl/language_models/repair.py +156 -156
  120. edsl/language_models/unused/ReplicateBase.py +83 -83
  121. edsl/language_models/utilities.py +64 -64
  122. edsl/notebooks/Notebook.py +259 -259
  123. edsl/notebooks/__init__.py +1 -1
  124. edsl/prompts/Prompt.py +353 -358
  125. edsl/prompts/__init__.py +2 -2
  126. edsl/questions/AnswerValidatorMixin.py +289 -289
  127. edsl/questions/QuestionBase.py +616 -616
  128. edsl/questions/QuestionBaseGenMixin.py +161 -161
  129. edsl/questions/QuestionBasePromptsMixin.py +266 -266
  130. edsl/questions/QuestionBudget.py +227 -227
  131. edsl/questions/QuestionCheckBox.py +359 -359
  132. edsl/questions/QuestionExtract.py +183 -183
  133. edsl/questions/QuestionFreeText.py +114 -113
  134. edsl/questions/QuestionFunctional.py +159 -159
  135. edsl/questions/QuestionList.py +231 -231
  136. edsl/questions/QuestionMultipleChoice.py +286 -286
  137. edsl/questions/QuestionNumerical.py +153 -153
  138. edsl/questions/QuestionRank.py +324 -324
  139. edsl/questions/Quick.py +41 -41
  140. edsl/questions/RegisterQuestionsMeta.py +71 -71
  141. edsl/questions/ResponseValidatorABC.py +174 -174
  142. edsl/questions/SimpleAskMixin.py +73 -73
  143. edsl/questions/__init__.py +26 -26
  144. edsl/questions/compose_questions.py +98 -98
  145. edsl/questions/decorators.py +21 -21
  146. edsl/questions/derived/QuestionLikertFive.py +76 -76
  147. edsl/questions/derived/QuestionLinearScale.py +87 -87
  148. edsl/questions/derived/QuestionTopK.py +91 -91
  149. edsl/questions/derived/QuestionYesNo.py +82 -82
  150. edsl/questions/descriptors.py +418 -418
  151. edsl/questions/prompt_templates/question_budget.jinja +13 -13
  152. edsl/questions/prompt_templates/question_checkbox.jinja +32 -32
  153. edsl/questions/prompt_templates/question_extract.jinja +11 -11
  154. edsl/questions/prompt_templates/question_free_text.jinja +3 -3
  155. edsl/questions/prompt_templates/question_linear_scale.jinja +11 -11
  156. edsl/questions/prompt_templates/question_list.jinja +17 -17
  157. edsl/questions/prompt_templates/question_multiple_choice.jinja +33 -33
  158. edsl/questions/prompt_templates/question_numerical.jinja +36 -36
  159. edsl/questions/question_registry.py +147 -147
  160. edsl/questions/settings.py +12 -12
  161. edsl/questions/templates/budget/answering_instructions.jinja +7 -7
  162. edsl/questions/templates/budget/question_presentation.jinja +7 -7
  163. edsl/questions/templates/checkbox/answering_instructions.jinja +10 -10
  164. edsl/questions/templates/checkbox/question_presentation.jinja +22 -22
  165. edsl/questions/templates/extract/answering_instructions.jinja +7 -7
  166. edsl/questions/templates/likert_five/answering_instructions.jinja +10 -10
  167. edsl/questions/templates/likert_five/question_presentation.jinja +11 -11
  168. edsl/questions/templates/linear_scale/answering_instructions.jinja +5 -5
  169. edsl/questions/templates/linear_scale/question_presentation.jinja +5 -5
  170. edsl/questions/templates/list/answering_instructions.jinja +3 -3
  171. edsl/questions/templates/list/question_presentation.jinja +5 -5
  172. edsl/questions/templates/multiple_choice/answering_instructions.jinja +9 -9
  173. edsl/questions/templates/multiple_choice/question_presentation.jinja +11 -11
  174. edsl/questions/templates/numerical/answering_instructions.jinja +6 -6
  175. edsl/questions/templates/numerical/question_presentation.jinja +6 -6
  176. edsl/questions/templates/rank/answering_instructions.jinja +11 -11
  177. edsl/questions/templates/rank/question_presentation.jinja +15 -15
  178. edsl/questions/templates/top_k/answering_instructions.jinja +8 -8
  179. edsl/questions/templates/top_k/question_presentation.jinja +22 -22
  180. edsl/questions/templates/yes_no/answering_instructions.jinja +6 -6
  181. edsl/questions/templates/yes_no/question_presentation.jinja +11 -11
  182. edsl/results/Dataset.py +293 -293
  183. edsl/results/DatasetExportMixin.py +693 -693
  184. edsl/results/DatasetTree.py +145 -145
  185. edsl/results/Result.py +435 -433
  186. edsl/results/Results.py +1160 -1158
  187. edsl/results/ResultsDBMixin.py +238 -238
  188. edsl/results/ResultsExportMixin.py +43 -43
  189. edsl/results/ResultsFetchMixin.py +33 -33
  190. edsl/results/ResultsGGMixin.py +121 -121
  191. edsl/results/ResultsToolsMixin.py +98 -98
  192. edsl/results/Selector.py +118 -118
  193. edsl/results/__init__.py +2 -2
  194. edsl/results/tree_explore.py +115 -115
  195. edsl/scenarios/FileStore.py +458 -443
  196. edsl/scenarios/Scenario.py +510 -507
  197. edsl/scenarios/ScenarioHtmlMixin.py +59 -59
  198. edsl/scenarios/ScenarioList.py +1101 -1101
  199. edsl/scenarios/ScenarioListExportMixin.py +52 -52
  200. edsl/scenarios/ScenarioListPdfMixin.py +261 -261
  201. edsl/scenarios/__init__.py +4 -2
  202. edsl/shared.py +1 -1
  203. edsl/study/ObjectEntry.py +173 -173
  204. edsl/study/ProofOfWork.py +113 -113
  205. edsl/study/SnapShot.py +80 -80
  206. edsl/study/Study.py +528 -528
  207. edsl/study/__init__.py +4 -4
  208. edsl/surveys/DAG.py +148 -148
  209. edsl/surveys/Memory.py +31 -31
  210. edsl/surveys/MemoryPlan.py +244 -244
  211. edsl/surveys/Rule.py +324 -324
  212. edsl/surveys/RuleCollection.py +387 -387
  213. edsl/surveys/Survey.py +1772 -1772
  214. edsl/surveys/SurveyCSS.py +261 -261
  215. edsl/surveys/SurveyExportMixin.py +259 -259
  216. edsl/surveys/SurveyFlowVisualizationMixin.py +121 -121
  217. edsl/surveys/SurveyQualtricsImport.py +284 -284
  218. edsl/surveys/__init__.py +3 -3
  219. edsl/surveys/base.py +53 -53
  220. edsl/surveys/descriptors.py +56 -56
  221. edsl/surveys/instructions/ChangeInstruction.py +47 -47
  222. edsl/surveys/instructions/Instruction.py +51 -51
  223. edsl/surveys/instructions/InstructionCollection.py +77 -77
  224. edsl/templates/error_reporting/base.html +23 -23
  225. edsl/templates/error_reporting/exceptions_by_model.html +34 -34
  226. edsl/templates/error_reporting/exceptions_by_question_name.html +16 -16
  227. edsl/templates/error_reporting/exceptions_by_type.html +16 -16
  228. edsl/templates/error_reporting/interview_details.html +115 -115
  229. edsl/templates/error_reporting/interviews.html +9 -9
  230. edsl/templates/error_reporting/overview.html +4 -4
  231. edsl/templates/error_reporting/performance_plot.html +1 -1
  232. edsl/templates/error_reporting/report.css +73 -73
  233. edsl/templates/error_reporting/report.html +117 -117
  234. edsl/templates/error_reporting/report.js +25 -25
  235. edsl/tools/__init__.py +1 -1
  236. edsl/tools/clusters.py +192 -192
  237. edsl/tools/embeddings.py +27 -27
  238. edsl/tools/embeddings_plotting.py +118 -118
  239. edsl/tools/plotting.py +112 -112
  240. edsl/tools/summarize.py +18 -18
  241. edsl/utilities/SystemInfo.py +28 -28
  242. edsl/utilities/__init__.py +22 -22
  243. edsl/utilities/ast_utilities.py +25 -25
  244. edsl/utilities/data/Registry.py +6 -6
  245. edsl/utilities/data/__init__.py +1 -1
  246. edsl/utilities/data/scooter_results.json +1 -1
  247. edsl/utilities/decorators.py +77 -77
  248. edsl/utilities/gcp_bucket/cloud_storage.py +96 -96
  249. edsl/utilities/interface.py +627 -627
  250. edsl/utilities/repair_functions.py +28 -28
  251. edsl/utilities/restricted_python.py +70 -70
  252. edsl/utilities/utilities.py +391 -391
  253. {edsl-0.1.36.dev6.dist-info → edsl-0.1.37.dev2.dist-info}/LICENSE +21 -21
  254. {edsl-0.1.36.dev6.dist-info → edsl-0.1.37.dev2.dist-info}/METADATA +1 -1
  255. edsl-0.1.37.dev2.dist-info/RECORD +279 -0
  256. edsl-0.1.36.dev6.dist-info/RECORD +0 -279
  257. {edsl-0.1.36.dev6.dist-info → edsl-0.1.37.dev2.dist-info}/WHEEL +0 -0
@@ -1,443 +1,458 @@
1
- import base64
2
- import io
3
- import tempfile
4
- import mimetypes
5
- import os
6
- from typing import Dict, Any, IO, Optional
7
- import requests
8
- from urllib.parse import urlparse
9
-
10
- import google.generativeai as genai
11
-
12
- from edsl import Scenario
13
- from edsl.utilities.decorators import add_edsl_version, remove_edsl_version
14
- from edsl.utilities.utilities import is_notebook
15
-
16
-
17
- def view_pdf(pdf_path):
18
- import os
19
- import subprocess
20
-
21
- if is_notebook():
22
- from IPython.display import IFrame
23
- from IPython.display import display, HTML
24
-
25
- # Replace 'path/to/your/file.pdf' with the actual path to your PDF file
26
- IFrame(pdf_path, width=700, height=600)
27
- display(HTML(f'<a href="{pdf_path}" target="_blank">Open PDF</a>'))
28
- return
29
-
30
- if os.path.exists(pdf_path):
31
- try:
32
- if (os_name := os.name) == "posix":
33
- # for cool kids
34
- subprocess.run(["open", pdf_path], check=True) # macOS
35
- elif os_name == "nt":
36
- os.startfile(pdf_path) # Windows
37
- else:
38
- subprocess.run(["xdg-open", pdf_path], check=True) # Linux
39
- except Exception as e:
40
- print(f"Error opening PDF: {e}")
41
- else:
42
- print("PDF file was not created successfully.")
43
-
44
-
45
- class FileStore(Scenario):
46
- def __init__(
47
- self,
48
- path: Optional[str] = None,
49
- mime_type: Optional[str] = None,
50
- binary: Optional[bool] = None,
51
- suffix: Optional[str] = None,
52
- base64_string: Optional[str] = None,
53
- external_locations: Optional[Dict[str, str]] = None,
54
- **kwargs,
55
- ):
56
- if path is None and "filename" in kwargs:
57
- path = kwargs["filename"]
58
- self.path = path
59
- self.suffix = suffix or path.split(".")[-1]
60
- self.binary = binary or False
61
- self.mime_type = (
62
- mime_type or mimetypes.guess_type(path)[0] or "application/octet-stream"
63
- )
64
- self.base64_string = base64_string or self.encode_file_to_base64_string(path)
65
- self.external_locations = external_locations or {}
66
- super().__init__(
67
- {
68
- "path": self.path,
69
- "base64_string": self.base64_string,
70
- "binary": self.binary,
71
- "suffix": self.suffix,
72
- "mime_type": self.mime_type,
73
- "external_locations": self.external_locations,
74
- }
75
- )
76
-
77
- def __str__(self):
78
- return "FileStore: self.path"
79
-
80
- @property
81
- def size(self) -> int:
82
- return os.path.getsize(self.path)
83
-
84
- def upload_google(self, refresh: bool = False) -> None:
85
- genai.configure(api_key=os.getenv("GOOGLE_API_KEY"))
86
- google_info = genai.upload_file(self.path, mime_type=self.mime_type)
87
- self.external_locations["google"] = google_info.to_dict()
88
-
89
- @classmethod
90
- @remove_edsl_version
91
- def from_dict(cls, d):
92
- # return cls(d["filename"], d["binary"], d["suffix"], d["base64_string"])
93
- return cls(**d)
94
-
95
- def __repr__(self):
96
- return f"FileStore({self.path})"
97
-
98
- def encode_file_to_base64_string(self, file_path: str):
99
- try:
100
- # Attempt to open the file in text mode
101
- with open(file_path, "r") as text_file:
102
- # Read the text data
103
- text_data = text_file.read()
104
- # Encode the text data to a base64 string
105
- base64_encoded_data = base64.b64encode(text_data.encode("utf-8"))
106
- except UnicodeDecodeError:
107
- # If reading as text fails, open the file in binary mode
108
- with open(file_path, "rb") as binary_file:
109
- # Read the binary data
110
- binary_data = binary_file.read()
111
- # Encode the binary data to a base64 string
112
- base64_encoded_data = base64.b64encode(binary_data)
113
- self.binary = True
114
- # Convert the base64 bytes to a string
115
- base64_string = base64_encoded_data.decode("utf-8")
116
-
117
- return base64_string
118
-
119
- def open(self) -> "IO":
120
- if self.binary:
121
- return self.base64_to_file(self["base64_string"], is_binary=True)
122
- else:
123
- return self.base64_to_text_file(self["base64_string"])
124
-
125
- @staticmethod
126
- def base64_to_text_file(base64_string) -> "IO":
127
- # Decode the base64 string to bytes
128
- text_data_bytes = base64.b64decode(base64_string)
129
-
130
- # Convert bytes to string
131
- text_data = text_data_bytes.decode("utf-8")
132
-
133
- # Create a StringIO object from the text data
134
- text_file = io.StringIO(text_data)
135
-
136
- return text_file
137
-
138
- @staticmethod
139
- def base64_to_file(base64_string, is_binary=True):
140
- # Decode the base64 string to bytes
141
- file_data = base64.b64decode(base64_string)
142
-
143
- if is_binary:
144
- # Create a BytesIO object for binary data
145
- return io.BytesIO(file_data)
146
- else:
147
- # Convert bytes to string for text data
148
- text_data = file_data.decode("utf-8")
149
- # Create a StringIO object for text data
150
- return io.StringIO(text_data)
151
-
152
- def to_tempfile(self, suffix=None):
153
- if suffix is None:
154
- suffix = self.suffix
155
- if self.binary:
156
- file_like_object = self.base64_to_file(
157
- self["base64_string"], is_binary=True
158
- )
159
- else:
160
- file_like_object = self.base64_to_text_file(self["base64_string"])
161
-
162
- # Create a named temporary file
163
- mode = "wb" if self.binary else "w"
164
- temp_file = tempfile.NamedTemporaryFile(
165
- delete=False, suffix="." + suffix, mode=mode
166
- )
167
-
168
- if self.binary:
169
- temp_file.write(file_like_object.read())
170
- else:
171
- temp_file.write(file_like_object.read())
172
-
173
- temp_file.close()
174
-
175
- return temp_file.name
176
-
177
- def view(self, max_size: int = 300) -> None:
178
- if self.suffix == "pdf":
179
- view_pdf(self.path)
180
-
181
- if self.suffix == "png" or self.suffix == "jpg" or self.suffix == "jpeg":
182
- if is_notebook():
183
- from IPython.display import Image
184
- from PIL import Image as PILImage
185
-
186
- if max_size:
187
- # Open the image using Pillow
188
- with PILImage.open(self.path) as img:
189
- # Get original width and height
190
- original_width, original_height = img.size
191
-
192
- # Calculate the scaling factor
193
- scale = min(
194
- max_size / original_width, max_size / original_height
195
- )
196
-
197
- # Calculate new dimensions
198
- new_width = int(original_width * scale)
199
- new_height = int(original_height * scale)
200
-
201
- return Image(self.path, width=new_width, height=new_height)
202
- else:
203
- return Image(self.path)
204
-
205
- def push(
206
- self, description: Optional[str] = None, visibility: str = "unlisted"
207
- ) -> dict:
208
- """
209
- Push the object to Coop.
210
- :param description: The description of the object to push.
211
- :param visibility: The visibility of the object to push.
212
- """
213
- scenario_version = Scenario.from_dict(self.to_dict())
214
- if description is None:
215
- description = "File: " + self.path
216
- info = scenario_version.push(description=description, visibility=visibility)
217
- return info
218
-
219
- @classmethod
220
- def pull(cls, uuid: str, expected_parrot_url: Optional[str] = None) -> "FileStore":
221
- """
222
- :param uuid: The UUID of the object to pull.
223
- :param expected_parrot_url: The URL of the Parrot server to use.
224
- :return: The object pulled from the Parrot server.
225
- """
226
- scenario_version = Scenario.pull(uuid, expected_parrot_url=expected_parrot_url)
227
- return cls.from_dict(scenario_version.to_dict())
228
-
229
- @classmethod
230
- def from_url(
231
- cls,
232
- url: str,
233
- download_path: Optional[str] = None,
234
- mime_type: Optional[str] = None,
235
- ) -> "FileStore":
236
- """
237
- :param url: The URL of the file to download.
238
- :param download_path: The path to save the downloaded file.
239
- :param mime_type: The MIME type of the file. If None, it will be guessed from the file extension.
240
- """
241
-
242
- response = requests.get(url, stream=True)
243
- response.raise_for_status() # Raises an HTTPError for bad responses
244
-
245
- # Get the filename from the URL if download_path is not provided
246
- if download_path is None:
247
- filename = os.path.basename(urlparse(url).path)
248
- if not filename:
249
- filename = "downloaded_file"
250
- # download_path = filename
251
- download_path = os.path.join(os.getcwd(), filename)
252
-
253
- # Ensure the directory exists
254
- os.makedirs(os.path.dirname(download_path), exist_ok=True)
255
-
256
- # Write the file
257
- with open(download_path, "wb") as file:
258
- for chunk in response.iter_content(chunk_size=8192):
259
- file.write(chunk)
260
-
261
- # Create and return a new File instance
262
- return cls(download_path, mime_type=mime_type)
263
-
264
-
265
- class CSVFileStore(FileStore):
266
- @classmethod
267
- def example(cls):
268
- from edsl.results.Results import Results
269
-
270
- r = Results.example()
271
- import tempfile
272
-
273
- with tempfile.NamedTemporaryFile(suffix=".csv", delete=False) as f:
274
- r.to_csv(filename=f.name)
275
- return cls(f.name)
276
-
277
- def view(self):
278
- import pandas as pd
279
-
280
- return pd.read_csv(self.to_tempfile())
281
-
282
-
283
- class PDFFileStore(FileStore):
284
- def view(self):
285
- pdf_path = self.to_tempfile()
286
- print(f"PDF path: {pdf_path}") # Print the path to ensure it exists
287
- import os
288
- import subprocess
289
-
290
- if os.path.exists(pdf_path):
291
- try:
292
- if os.name == "posix":
293
- # for cool kids
294
- subprocess.run(["open", pdf_path], check=True) # macOS
295
- elif os.name == "nt":
296
- os.startfile(pdf_path) # Windows
297
- else:
298
- subprocess.run(["xdg-open", pdf_path], check=True) # Linux
299
- except Exception as e:
300
- print(f"Error opening PDF: {e}")
301
- else:
302
- print("PDF file was not created successfully.")
303
-
304
- @classmethod
305
- def example(cls):
306
- import textwrap
307
-
308
- pdf_string = textwrap.dedent(
309
- """\
310
- %PDF-1.4
311
- 1 0 obj
312
- << /Type /Catalog /Pages 2 0 R >>
313
- endobj
314
- 2 0 obj
315
- << /Type /Pages /Kids [3 0 R] /Count 1 >>
316
- endobj
317
- 3 0 obj
318
- << /Type /Page /Parent 2 0 R /MediaBox [0 0 612 792] /Contents 4 0 R >>
319
- endobj
320
- 4 0 obj
321
- << /Length 44 >>
322
- stream
323
- BT
324
- /F1 24 Tf
325
- 100 700 Td
326
- (Hello, World!) Tj
327
- ET
328
- endstream
329
- endobj
330
- 5 0 obj
331
- << /Type /Font /Subtype /Type1 /BaseFont /Helvetica >>
332
- endobj
333
- 6 0 obj
334
- << /ProcSet [/PDF /Text] /Font << /F1 5 0 R >> >>
335
- endobj
336
- xref
337
- 0 7
338
- 0000000000 65535 f
339
- 0000000010 00000 n
340
- 0000000053 00000 n
341
- 0000000100 00000 n
342
- 0000000173 00000 n
343
- 0000000232 00000 n
344
- 0000000272 00000 n
345
- trailer
346
- << /Size 7 /Root 1 0 R >>
347
- startxref
348
- 318
349
- %%EOF"""
350
- )
351
- import tempfile
352
-
353
- with tempfile.NamedTemporaryFile(suffix=".pdf", delete=False) as f:
354
- f.write(pdf_string.encode())
355
- return cls(f.name)
356
-
357
-
358
- class PNGFileStore(FileStore):
359
- @classmethod
360
- def example(cls):
361
- import textwrap
362
-
363
- png_string = textwrap.dedent(
364
- """\x89PNG\r\n\x1a\n\x00\x00\x00\rIHDR\x00\x00\x01\x00\x00\x00\x01\x00\x08\x06\x00\x00\x00\x1f\x15\xc4\x89\x00\x00\x00\x0cIDAT\x08\xd7c\x00\x01"""
365
- )
366
- import tempfile
367
-
368
- with tempfile.NamedTemporaryFile(suffix=".png", delete=False) as f:
369
- f.write(png_string.encode())
370
- return cls(f.name)
371
-
372
- def view(self):
373
- import matplotlib.pyplot as plt
374
- import matplotlib.image as mpimg
375
-
376
- img = mpimg.imread(self.to_tempfile())
377
- plt.imshow(img)
378
- plt.show()
379
-
380
-
381
- class SQLiteFileStore(FileStore):
382
- @classmethod
383
- def example(cls):
384
- import sqlite3
385
- import tempfile
386
-
387
- with tempfile.NamedTemporaryFile(suffix=".sqlite", delete=False) as f:
388
- conn = sqlite3.connect(f.name)
389
- c = conn.cursor()
390
- c.execute("""CREATE TABLE stocks (date text)""")
391
- conn.commit()
392
-
393
- return cls(f.name)
394
-
395
- def view(self):
396
- import subprocess
397
- import os
398
-
399
- sqlite_path = self.to_tempfile()
400
- os.system(f"sqlite3 {sqlite_path}")
401
-
402
-
403
- class HTMLFileStore(FileStore):
404
- @classmethod
405
- def example(cls):
406
- import tempfile
407
-
408
- with tempfile.NamedTemporaryFile(suffix=".html", delete=False) as f:
409
- f.write("<html><body><h1>Test</h1></body></html>".encode())
410
- return cls(f.name)
411
-
412
- def view(self):
413
- import webbrowser
414
-
415
- html_path = self.to_tempfile()
416
- webbrowser.open("file://" + html_path)
417
-
418
-
419
- if __name__ == "__main__":
420
- # file_path = "../conjure/examples/Ex11-2.sav"
421
- # fs = FileStore(file_path)
422
- # info = fs.push()
423
- # print(info)
424
-
425
- # fs = CSVFileStore.example()
426
- # fs.to_tempfile()
427
- # print(fs.view())
428
-
429
- # fs = PDFFileStore.example()
430
- # fs.view()
431
-
432
- # fs = PDFFileStore("paper.pdf")
433
- # fs.view()
434
- # from edsl import Conjure
435
- pass
436
- # fs = PNGFileStore("logo.png")
437
- # fs.view()
438
- # fs.upload_google()
439
-
440
- # c = Conjure(datafile_name=fs.to_tempfile())
441
- # f = PDFFileStore("paper.pdf")
442
- # print(f.to_tempfile())
443
- # f.push()
1
+ import base64
2
+ import io
3
+ import tempfile
4
+ import mimetypes
5
+ import os
6
+ from typing import Dict, Any, IO, Optional
7
+ import requests
8
+ from urllib.parse import urlparse
9
+
10
+ import google.generativeai as genai
11
+
12
+ from edsl import Scenario
13
+ from edsl.utilities.decorators import add_edsl_version, remove_edsl_version
14
+ from edsl.utilities.utilities import is_notebook
15
+
16
+
17
+ def view_pdf(pdf_path):
18
+ import os
19
+ import subprocess
20
+
21
+ if is_notebook():
22
+ from IPython.display import IFrame
23
+ from IPython.display import display, HTML
24
+
25
+ # Replace 'path/to/your/file.pdf' with the actual path to your PDF file
26
+ IFrame(pdf_path, width=700, height=600)
27
+ display(HTML(f'<a href="{pdf_path}" target="_blank">Open PDF</a>'))
28
+ return
29
+
30
+ if os.path.exists(pdf_path):
31
+ try:
32
+ if (os_name := os.name) == "posix":
33
+ # for cool kids
34
+ subprocess.run(["open", pdf_path], check=True) # macOS
35
+ elif os_name == "nt":
36
+ os.startfile(pdf_path) # Windows
37
+ else:
38
+ subprocess.run(["xdg-open", pdf_path], check=True) # Linux
39
+ except Exception as e:
40
+ print(f"Error opening PDF: {e}")
41
+ else:
42
+ print("PDF file was not created successfully.")
43
+
44
+
45
+ class FileStore(Scenario):
46
+ def __init__(
47
+ self,
48
+ path: Optional[str] = None,
49
+ mime_type: Optional[str] = None,
50
+ binary: Optional[bool] = None,
51
+ suffix: Optional[str] = None,
52
+ base64_string: Optional[str] = None,
53
+ external_locations: Optional[Dict[str, str]] = None,
54
+ **kwargs,
55
+ ):
56
+ if path is None and "filename" in kwargs:
57
+ path = kwargs["filename"]
58
+ self.path = path
59
+ self.suffix = suffix or path.split(".")[-1]
60
+ self.binary = binary or False
61
+ self.mime_type = (
62
+ mime_type or mimetypes.guess_type(path)[0] or "application/octet-stream"
63
+ )
64
+ self.base64_string = base64_string or self.encode_file_to_base64_string(path)
65
+ self.external_locations = external_locations or {}
66
+ super().__init__(
67
+ {
68
+ "path": self.path,
69
+ "base64_string": self.base64_string,
70
+ "binary": self.binary,
71
+ "suffix": self.suffix,
72
+ "mime_type": self.mime_type,
73
+ "external_locations": self.external_locations,
74
+ }
75
+ )
76
+
77
+ def __str__(self):
78
+ return "FileStore: self.path"
79
+
80
+ @classmethod
81
+ def example(self):
82
+ import tempfile
83
+
84
+ with tempfile.NamedTemporaryFile(suffix=".txt", delete=False) as f:
85
+ f.write(b"Hello, World!")
86
+
87
+ return self(path=f.name)
88
+
89
+ @property
90
+ def size(self) -> int:
91
+ if self.base64_string != None:
92
+ return (len(self.base64_string) / 4.0) * 3 # from base64 to char size
93
+ return os.path.getsize(self.path)
94
+
95
+ def upload_google(self, refresh: bool = False) -> None:
96
+ genai.configure(api_key=os.getenv("GOOGLE_API_KEY"))
97
+ google_info = genai.upload_file(self.path, mime_type=self.mime_type)
98
+ self.external_locations["google"] = google_info.to_dict()
99
+
100
+ @classmethod
101
+ @remove_edsl_version
102
+ def from_dict(cls, d):
103
+ # return cls(d["filename"], d["binary"], d["suffix"], d["base64_string"])
104
+ return cls(**d)
105
+
106
+ def __repr__(self):
107
+ return f"FileStore(path='{self.path}')"
108
+
109
+ def encode_file_to_base64_string(self, file_path: str):
110
+ try:
111
+ # Attempt to open the file in text mode
112
+ with open(file_path, "r") as text_file:
113
+ # Read the text data
114
+ text_data = text_file.read()
115
+ # Encode the text data to a base64 string
116
+ base64_encoded_data = base64.b64encode(text_data.encode("utf-8"))
117
+ except UnicodeDecodeError:
118
+ # If reading as text fails, open the file in binary mode
119
+ with open(file_path, "rb") as binary_file:
120
+ # Read the binary data
121
+ binary_data = binary_file.read()
122
+ # Encode the binary data to a base64 string
123
+ base64_encoded_data = base64.b64encode(binary_data)
124
+ self.binary = True
125
+ # Convert the base64 bytes to a string
126
+ base64_string = base64_encoded_data.decode("utf-8")
127
+
128
+ return base64_string
129
+
130
+ def open(self) -> "IO":
131
+ if self.binary:
132
+ return self.base64_to_file(self["base64_string"], is_binary=True)
133
+ else:
134
+ return self.base64_to_text_file(self["base64_string"])
135
+
136
+ @staticmethod
137
+ def base64_to_text_file(base64_string) -> "IO":
138
+ # Decode the base64 string to bytes
139
+ text_data_bytes = base64.b64decode(base64_string)
140
+
141
+ # Convert bytes to string
142
+ text_data = text_data_bytes.decode("utf-8")
143
+
144
+ # Create a StringIO object from the text data
145
+ text_file = io.StringIO(text_data)
146
+
147
+ return text_file
148
+
149
+ @staticmethod
150
+ def base64_to_file(base64_string, is_binary=True):
151
+ # Decode the base64 string to bytes
152
+ file_data = base64.b64decode(base64_string)
153
+
154
+ if is_binary:
155
+ # Create a BytesIO object for binary data
156
+ return io.BytesIO(file_data)
157
+ else:
158
+ # Convert bytes to string for text data
159
+ text_data = file_data.decode("utf-8")
160
+ # Create a StringIO object for text data
161
+ return io.StringIO(text_data)
162
+
163
+ def to_tempfile(self, suffix=None):
164
+ if suffix is None:
165
+ suffix = self.suffix
166
+ if self.binary:
167
+ file_like_object = self.base64_to_file(
168
+ self["base64_string"], is_binary=True
169
+ )
170
+ else:
171
+ file_like_object = self.base64_to_text_file(self["base64_string"])
172
+
173
+ # Create a named temporary file
174
+ mode = "wb" if self.binary else "w"
175
+ temp_file = tempfile.NamedTemporaryFile(
176
+ delete=False, suffix="." + suffix, mode=mode
177
+ )
178
+
179
+ if self.binary:
180
+ temp_file.write(file_like_object.read())
181
+ else:
182
+ temp_file.write(file_like_object.read())
183
+
184
+ temp_file.close()
185
+
186
+ return temp_file.name
187
+
188
+ def view(self, max_size: int = 300) -> None:
189
+ if self.suffix == "pdf":
190
+ view_pdf(self.path)
191
+
192
+ if self.suffix == "png" or self.suffix == "jpg" or self.suffix == "jpeg":
193
+ if is_notebook():
194
+ from IPython.display import Image
195
+ from PIL import Image as PILImage
196
+
197
+ if max_size:
198
+ # Open the image using Pillow
199
+ with PILImage.open(self.path) as img:
200
+ # Get original width and height
201
+ original_width, original_height = img.size
202
+
203
+ # Calculate the scaling factor
204
+ scale = min(
205
+ max_size / original_width, max_size / original_height
206
+ )
207
+
208
+ # Calculate new dimensions
209
+ new_width = int(original_width * scale)
210
+ new_height = int(original_height * scale)
211
+
212
+ return Image(self.path, width=new_width, height=new_height)
213
+ else:
214
+ return Image(self.path)
215
+
216
+ def push(
217
+ self, description: Optional[str] = None, visibility: str = "unlisted"
218
+ ) -> dict:
219
+ """
220
+ Push the object to Coop.
221
+ :param description: The description of the object to push.
222
+ :param visibility: The visibility of the object to push.
223
+ """
224
+ scenario_version = Scenario.from_dict(self.to_dict())
225
+ if description is None:
226
+ description = "File: " + self.path
227
+ info = scenario_version.push(description=description, visibility=visibility)
228
+ return info
229
+
230
+ @classmethod
231
+ def pull(cls, uuid: str, expected_parrot_url: Optional[str] = None) -> "FileStore":
232
+ """
233
+ :param uuid: The UUID of the object to pull.
234
+ :param expected_parrot_url: The URL of the Parrot server to use.
235
+ :return: The object pulled from the Parrot server.
236
+ """
237
+ scenario_version = Scenario.pull(uuid, expected_parrot_url=expected_parrot_url)
238
+ return cls.from_dict(scenario_version.to_dict())
239
+
240
+ @classmethod
241
+ def from_url(
242
+ cls,
243
+ url: str,
244
+ download_path: Optional[str] = None,
245
+ mime_type: Optional[str] = None,
246
+ ) -> "FileStore":
247
+ """
248
+ :param url: The URL of the file to download.
249
+ :param download_path: The path to save the downloaded file.
250
+ :param mime_type: The MIME type of the file. If None, it will be guessed from the file extension.
251
+ """
252
+
253
+ response = requests.get(url, stream=True)
254
+ response.raise_for_status() # Raises an HTTPError for bad responses
255
+
256
+ # Get the filename from the URL if download_path is not provided
257
+ if download_path is None:
258
+ filename = os.path.basename(urlparse(url).path)
259
+ if not filename:
260
+ filename = "downloaded_file"
261
+ # download_path = filename
262
+ download_path = os.path.join(os.getcwd(), filename)
263
+
264
+ # Ensure the directory exists
265
+ os.makedirs(os.path.dirname(download_path), exist_ok=True)
266
+
267
+ # Write the file
268
+ with open(download_path, "wb") as file:
269
+ for chunk in response.iter_content(chunk_size=8192):
270
+ file.write(chunk)
271
+
272
+ # Create and return a new File instance
273
+ return cls(download_path, mime_type=mime_type)
274
+
275
+
276
+ class CSVFileStore(FileStore):
277
+ @classmethod
278
+ def example(cls):
279
+ from edsl.results.Results import Results
280
+
281
+ r = Results.example()
282
+ import tempfile
283
+
284
+ with tempfile.NamedTemporaryFile(suffix=".csv", delete=False) as f:
285
+ r.to_csv(filename=f.name)
286
+
287
+ return cls(f.name)
288
+
289
+ def view(self):
290
+ import pandas as pd
291
+
292
+ return pd.read_csv(self.to_tempfile())
293
+
294
+
295
+ class PDFFileStore(FileStore):
296
+ def view(self):
297
+ pdf_path = self.to_tempfile()
298
+ print(f"PDF path: {pdf_path}") # Print the path to ensure it exists
299
+ import os
300
+ import subprocess
301
+
302
+ if os.path.exists(pdf_path):
303
+ try:
304
+ if os.name == "posix":
305
+ # for cool kids
306
+ subprocess.run(["open", pdf_path], check=True) # macOS
307
+ elif os.name == "nt":
308
+ os.startfile(pdf_path) # Windows
309
+ else:
310
+ subprocess.run(["xdg-open", pdf_path], check=True) # Linux
311
+ except Exception as e:
312
+ print(f"Error opening PDF: {e}")
313
+ else:
314
+ print("PDF file was not created successfully.")
315
+
316
+ @classmethod
317
+ def example(cls):
318
+ import textwrap
319
+
320
+ pdf_string = textwrap.dedent(
321
+ """\
322
+ %PDF-1.4
323
+ 1 0 obj
324
+ << /Type /Catalog /Pages 2 0 R >>
325
+ endobj
326
+ 2 0 obj
327
+ << /Type /Pages /Kids [3 0 R] /Count 1 >>
328
+ endobj
329
+ 3 0 obj
330
+ << /Type /Page /Parent 2 0 R /MediaBox [0 0 612 792] /Contents 4 0 R >>
331
+ endobj
332
+ 4 0 obj
333
+ << /Length 44 >>
334
+ stream
335
+ BT
336
+ /F1 24 Tf
337
+ 100 700 Td
338
+ (Hello, World!) Tj
339
+ ET
340
+ endstream
341
+ endobj
342
+ 5 0 obj
343
+ << /Type /Font /Subtype /Type1 /BaseFont /Helvetica >>
344
+ endobj
345
+ 6 0 obj
346
+ << /ProcSet [/PDF /Text] /Font << /F1 5 0 R >> >>
347
+ endobj
348
+ xref
349
+ 0 7
350
+ 0000000000 65535 f
351
+ 0000000010 00000 n
352
+ 0000000053 00000 n
353
+ 0000000100 00000 n
354
+ 0000000173 00000 n
355
+ 0000000232 00000 n
356
+ 0000000272 00000 n
357
+ trailer
358
+ << /Size 7 /Root 1 0 R >>
359
+ startxref
360
+ 318
361
+ %%EOF"""
362
+ )
363
+ import tempfile
364
+
365
+ with tempfile.NamedTemporaryFile(suffix=".pdf", delete=False) as f:
366
+ f.write(pdf_string.encode())
367
+
368
+ return cls(f.name)
369
+
370
+
371
+ class PNGFileStore(FileStore):
372
+ @classmethod
373
+ def example(cls):
374
+ import textwrap
375
+
376
+ png_string = textwrap.dedent(
377
+ """\x89PNG\r\n\x1a\n\x00\x00\x00\rIHDR\x00\x00\x01\x00\x00\x00\x01\x00\x08\x06\x00\x00\x00\x1f\x15\xc4\x89\x00\x00\x00\x0cIDAT\x08\xd7c\x00\x01"""
378
+ )
379
+ import tempfile
380
+
381
+ with tempfile.NamedTemporaryFile(suffix=".png", delete=False) as f:
382
+ f.write(png_string.encode())
383
+
384
+ return cls(f.name)
385
+
386
+ def view(self):
387
+ import matplotlib.pyplot as plt
388
+ import matplotlib.image as mpimg
389
+
390
+ img = mpimg.imread(self.to_tempfile())
391
+ plt.imshow(img)
392
+ plt.show()
393
+
394
+
395
+ class SQLiteFileStore(FileStore):
396
+ @classmethod
397
+ def example(cls):
398
+ import sqlite3
399
+ import tempfile
400
+
401
+ with tempfile.NamedTemporaryFile(suffix=".sqlite", delete=False) as f:
402
+ conn = sqlite3.connect(f.name)
403
+ c = conn.cursor()
404
+ c.execute("""CREATE TABLE stocks (date text)""")
405
+ conn.commit()
406
+
407
+ return cls(f.name)
408
+
409
+ def view(self):
410
+ import subprocess
411
+ import os
412
+
413
+ sqlite_path = self.to_tempfile()
414
+ os.system(f"sqlite3 {sqlite_path}")
415
+
416
+
417
+ class HTMLFileStore(FileStore):
418
+ @classmethod
419
+ def example(cls):
420
+ import tempfile
421
+
422
+ with tempfile.NamedTemporaryFile(suffix=".html", delete=False) as f:
423
+ f.write("<html><body><h1>Test</h1></body></html>".encode())
424
+
425
+ return cls(f.name)
426
+
427
+ def view(self):
428
+ import webbrowser
429
+
430
+ html_path = self.to_tempfile()
431
+ webbrowser.open("file://" + html_path)
432
+
433
+
434
+ if __name__ == "__main__":
435
+ # file_path = "../conjure/examples/Ex11-2.sav"
436
+ # fs = FileStore(file_path)
437
+ # info = fs.push()
438
+ # print(info)
439
+
440
+ # fs = CSVFileStore.example()
441
+ # fs.to_tempfile()
442
+ # print(fs.view())
443
+
444
+ # fs = PDFFileStore.example()
445
+ # fs.view()
446
+
447
+ # fs = PDFFileStore("paper.pdf")
448
+ # fs.view()
449
+ # from edsl import Conjure
450
+ pass
451
+ # fs = PNGFileStore("logo.png")
452
+ # fs.view()
453
+ # fs.upload_google()
454
+
455
+ # c = Conjure(datafile_name=fs.to_tempfile())
456
+ # f = PDFFileStore("paper.pdf")
457
+ # print(f.to_tempfile())
458
+ # f.push()