convoviz 0.2.6__tar.gz → 0.2.8__tar.gz

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 (60) hide show
  1. {convoviz-0.2.6 → convoviz-0.2.8}/PKG-INFO +2 -8
  2. {convoviz-0.2.6 → convoviz-0.2.8}/README.md +1 -7
  3. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/cli.py +9 -1
  4. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/config.py +9 -0
  5. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/io/assets.py +16 -0
  6. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/io/writers.py +58 -2
  7. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/models/message.py +70 -9
  8. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/pipeline.py +1 -0
  9. {convoviz-0.2.6 → convoviz-0.2.8}/pyproject.toml +1 -1
  10. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/__init__.py +0 -0
  11. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/__main__.py +0 -0
  12. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/analysis/__init__.py +0 -0
  13. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/analysis/graphs.py +0 -0
  14. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/analysis/wordcloud.py +0 -0
  15. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/colormaps.txt +0 -0
  16. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/AmaticSC-Regular.ttf +0 -0
  17. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/ArchitectsDaughter-Regular.ttf +0 -0
  18. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/BebasNeue-Regular.ttf +0 -0
  19. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/Borel-Regular.ttf +0 -0
  20. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/Courgette-Regular.ttf +0 -0
  21. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/CroissantOne-Regular.ttf +0 -0
  22. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/Handjet-Regular.ttf +0 -0
  23. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/IndieFlower-Regular.ttf +0 -0
  24. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/Kalam-Regular.ttf +0 -0
  25. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/Lobster-Regular.ttf +0 -0
  26. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/MartianMono-Regular.ttf +0 -0
  27. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/MartianMono-Thin.ttf +0 -0
  28. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/Montserrat-Regular.ttf +0 -0
  29. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/Mooli-Regular.ttf +0 -0
  30. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/Pacifico-Regular.ttf +0 -0
  31. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/PlayfairDisplay-Regular.ttf +0 -0
  32. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/Raleway-Regular.ttf +0 -0
  33. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/RobotoMono-Regular.ttf +0 -0
  34. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/RobotoMono-Thin.ttf +0 -0
  35. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/RobotoSlab-Regular.ttf +0 -0
  36. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/RobotoSlab-Thin.ttf +0 -0
  37. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/Ruwudu-Regular.ttf +0 -0
  38. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/Sacramento-Regular.ttf +0 -0
  39. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/SedgwickAveDisplay-Regular.ttf +0 -0
  40. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/ShadowsIntoLight-Regular.ttf +0 -0
  41. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/TitilliumWeb-Regular.ttf +0 -0
  42. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/Yellowtail-Regular.ttf +0 -0
  43. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/YsabeauOffice-Regular.ttf +0 -0
  44. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/YsabeauSC-Regular.ttf +0 -0
  45. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/YsabeauSC-Thin.ttf +0 -0
  46. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/fonts/Zeyada-Regular.ttf +0 -0
  47. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/assets/stopwords.txt +0 -0
  48. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/exceptions.py +0 -0
  49. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/interactive.py +0 -0
  50. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/io/__init__.py +0 -0
  51. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/io/loaders.py +0 -0
  52. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/models/__init__.py +0 -0
  53. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/models/collection.py +0 -0
  54. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/models/conversation.py +0 -0
  55. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/models/node.py +0 -0
  56. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/py.typed +0 -0
  57. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/renderers/__init__.py +0 -0
  58. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/renderers/markdown.py +0 -0
  59. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/renderers/yaml.py +0 -0
  60. {convoviz-0.2.6 → convoviz-0.2.8}/convoviz/utils.py +0 -0
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: convoviz
3
- Version: 0.2.6
3
+ Version: 0.2.8
4
4
  Summary: Get analytics and visualizations on your ChatGPT data!
5
5
  Keywords: markdown,chatgpt,openai,visualization,analytics,json,export,data-analysis,obsidian
6
6
  Author: Mohamed Cheikh Sidiya
@@ -34,8 +34,7 @@ Convert your ChatGPT history into well-formatted Markdown files. Additionally, v
34
34
  ## Features
35
35
 
36
36
  - **YAML Headers**: Optional and included by default.
37
- - **Track message versions**: prompt/response edits included.
38
- - **Code Interpreter**: Environment code blocks and execution results.
37
+ - **Inline Images**: Media attachments rendered directly in Markdown.
39
38
  - **Data Visualizations**: Word clouds, graphs, and more.
40
39
  - **Custom Instructions**: All your custom instructions in one JSON file.
41
40
 
@@ -84,11 +83,6 @@ You can provide arguments directly to skip the prompts:
84
83
  convoviz --input path/to/your/export.zip --output path/to/output/folder
85
84
  ```
86
85
 
87
- Inputs can be any of:
88
- - A ChatGPT export ZIP (downloaded from OpenAI)
89
- - An extracted export directory containing `conversations.json`
90
- - A `conversations.json` file directly
91
-
92
86
  Notes:
93
87
  - `--zip` / `-z` is kept as an alias for `--input` for convenience.
94
88
  - You can force non-interactive mode with `--no-interactive`.
@@ -8,8 +8,7 @@ Convert your ChatGPT history into well-formatted Markdown files. Additionally, v
8
8
  ## Features
9
9
 
10
10
  - **YAML Headers**: Optional and included by default.
11
- - **Track message versions**: prompt/response edits included.
12
- - **Code Interpreter**: Environment code blocks and execution results.
11
+ - **Inline Images**: Media attachments rendered directly in Markdown.
13
12
  - **Data Visualizations**: Word clouds, graphs, and more.
14
13
  - **Custom Instructions**: All your custom instructions in one JSON file.
15
14
 
@@ -58,11 +57,6 @@ You can provide arguments directly to skip the prompts:
58
57
  convoviz --input path/to/your/export.zip --output path/to/output/folder
59
58
  ```
60
59
 
61
- Inputs can be any of:
62
- - A ChatGPT export ZIP (downloaded from OpenAI)
63
- - An extracted export directory containing `conversations.json`
64
- - A `conversations.json` file directly
65
-
66
60
  Notes:
67
61
  - `--zip` / `-z` is kept as an alias for `--input` for convenience.
68
62
  - You can force non-interactive mode with `--no-interactive`.
@@ -5,7 +5,7 @@ from pathlib import Path
5
5
  import typer
6
6
  from rich.console import Console
7
7
 
8
- from convoviz.config import get_default_config
8
+ from convoviz.config import FolderOrganization, get_default_config
9
9
  from convoviz.exceptions import ConfigurationError, InvalidZipError
10
10
  from convoviz.interactive import run_interactive_config
11
11
  from convoviz.io.loaders import find_latest_zip
@@ -38,6 +38,12 @@ def run(
38
38
  "-o",
39
39
  help="Path to the output directory.",
40
40
  ),
41
+ flat: bool = typer.Option(
42
+ False,
43
+ "--flat",
44
+ "-f",
45
+ help="Put all markdown files in a single folder (disables date organization).",
46
+ ),
41
47
  interactive: bool | None = typer.Option(
42
48
  None,
43
49
  "--interactive/--no-interactive",
@@ -57,6 +63,8 @@ def run(
57
63
  config.input_path = input_path
58
64
  if output_dir:
59
65
  config.output_folder = output_dir
66
+ if flat:
67
+ config.folder_organization = FolderOrganization.FLAT
60
68
 
61
69
  # Determine mode: interactive if explicitly requested or no input provided
62
70
  use_interactive = interactive if interactive is not None else (input_path is None)
@@ -1,11 +1,19 @@
1
1
  """Configuration models using Pydantic v2."""
2
2
 
3
+ from enum import Enum
3
4
  from pathlib import Path
4
5
  from typing import Literal
5
6
 
6
7
  from pydantic import BaseModel, Field
7
8
 
8
9
 
10
+ class FolderOrganization(str, Enum):
11
+ """How to organize markdown output files in folders."""
12
+
13
+ FLAT = "flat" # All files in one directory (default)
14
+ DATE = "date" # Nested by year/month/week
15
+
16
+
9
17
  class AuthorHeaders(BaseModel):
10
18
  """Headers for different message authors in markdown output."""
11
19
 
@@ -83,6 +91,7 @@ class ConvovizConfig(BaseModel):
83
91
 
84
92
  input_path: Path | None = None
85
93
  output_folder: Path = Field(default_factory=lambda: Path.home() / "Documents" / "ChatGPT-Data")
94
+ folder_organization: FolderOrganization = FolderOrganization.DATE
86
95
  message: MessageConfig = Field(default_factory=MessageConfig)
87
96
  conversation: ConversationConfig = Field(default_factory=ConversationConfig)
88
97
  wordcloud: WordCloudConfig = Field(default_factory=WordCloudConfig)
@@ -57,6 +57,22 @@ def resolve_asset_path(source_dir: Path, asset_id: str) -> Path | None:
57
57
  except Exception:
58
58
  pass
59
59
 
60
+ # 4. Try prefix match in user-* directories (new 2025 format)
61
+ try:
62
+ for user_dir in source_dir.glob("user-*"):
63
+ if user_dir.is_dir():
64
+ user_dir = user_dir.resolve()
65
+ candidates = list(user_dir.glob(f"{asset_id}*"))
66
+ files = [
67
+ p.resolve()
68
+ for p in candidates
69
+ if p.is_file() and p.resolve().is_relative_to(user_dir)
70
+ ]
71
+ if files:
72
+ return files[0]
73
+ except Exception:
74
+ pass
75
+
60
76
  return None
61
77
 
62
78
 
@@ -6,12 +6,59 @@ from pathlib import Path
6
6
  from orjson import OPT_INDENT_2, dumps
7
7
  from tqdm import tqdm
8
8
 
9
- from convoviz.config import AuthorHeaders, ConversationConfig
9
+ from convoviz.config import AuthorHeaders, ConversationConfig, FolderOrganization
10
10
  from convoviz.io.assets import copy_asset, resolve_asset_path
11
11
  from convoviz.models import Conversation, ConversationCollection
12
12
  from convoviz.renderers import render_conversation
13
13
  from convoviz.utils import sanitize
14
14
 
15
+ # Month names for folder naming
16
+ _MONTH_NAMES = [
17
+ "January",
18
+ "February",
19
+ "March",
20
+ "April",
21
+ "May",
22
+ "June",
23
+ "July",
24
+ "August",
25
+ "September",
26
+ "October",
27
+ "November",
28
+ "December",
29
+ ]
30
+
31
+
32
+ def get_date_folder_path(conversation: Conversation) -> Path:
33
+ """Get the date-based folder path for a conversation.
34
+
35
+ Creates a nested structure: year/month/week
36
+ Example: 2024/03-March/Week-02/
37
+
38
+ Args:
39
+ conversation: The conversation to get the path for
40
+
41
+ Returns:
42
+ Relative path for the date-based folder structure
43
+ """
44
+ create_time = conversation.create_time
45
+
46
+ # Year folder: "2024"
47
+ year = str(create_time.year)
48
+
49
+ # Month folder: "03-March"
50
+ month_num = create_time.month
51
+ month_name = _MONTH_NAMES[month_num - 1]
52
+ month = f"{month_num:02d}-{month_name}"
53
+
54
+ # Week folder: "Week-01" through "Week-05" (week of the month)
55
+ # Calculate which week of the month this date falls into
56
+ day = create_time.day
57
+ week_of_month = (day - 1) // 7 + 1
58
+ week = f"Week-{week_of_month:02d}"
59
+
60
+ return Path(year) / month / week
61
+
15
62
 
16
63
  def save_conversation(
17
64
  conversation: Conversation,
@@ -74,6 +121,7 @@ def save_collection(
74
121
  config: ConversationConfig,
75
122
  headers: AuthorHeaders,
76
123
  *,
124
+ folder_organization: FolderOrganization = FolderOrganization.FLAT,
77
125
  progress_bar: bool = False,
78
126
  ) -> None:
79
127
  """Save all conversations in a collection to markdown files.
@@ -83,6 +131,7 @@ def save_collection(
83
131
  directory: Target directory
84
132
  config: Conversation rendering configuration
85
133
  headers: Author header configuration
134
+ folder_organization: How to organize files in folders (flat or by date)
86
135
  progress_bar: Whether to show a progress bar
87
136
  """
88
137
  directory.mkdir(parents=True, exist_ok=True)
@@ -92,7 +141,14 @@ def save_collection(
92
141
  desc="Writing Markdown 📄 files",
93
142
  disable=not progress_bar,
94
143
  ):
95
- filepath = directory / f"{sanitize(conv.title)}.md"
144
+ # Determine target directory based on organization setting
145
+ if folder_organization == FolderOrganization.DATE:
146
+ target_dir = directory / get_date_folder_path(conv)
147
+ target_dir.mkdir(parents=True, exist_ok=True)
148
+ else:
149
+ target_dir = directory
150
+
151
+ filepath = target_dir / f"{sanitize(conv.title)}.md"
96
152
  save_conversation(conv, filepath, config, headers, source_path=collection.source_path)
97
153
 
98
154
 
@@ -28,6 +28,14 @@ class MessageContent(BaseModel):
28
28
  parts: list[Any] | None = None
29
29
  text: str | None = None
30
30
  result: str | None = None
31
+ # reasoning_recap content type
32
+ content: str | None = None
33
+ # thoughts content type (list of thought objects with summary/content/finished)
34
+ thoughts: list[Any] | None = None
35
+ # tether_quote content type
36
+ url: str | None = None
37
+ domain: str | None = None
38
+ title: str | None = None
31
39
 
32
40
 
33
41
  class MessageMetadata(BaseModel):
@@ -36,6 +44,7 @@ class MessageMetadata(BaseModel):
36
44
  model_slug: str | None = None
37
45
  invoked_plugin: dict[str, Any] | None = None
38
46
  is_user_system_message: bool | None = None
47
+ is_visually_hidden_from_conversation: bool | None = None
39
48
  user_context_message_data: dict[str, Any] | None = None
40
49
 
41
50
  model_config = ConfigDict(protected_namespaces=())
@@ -105,12 +114,48 @@ class Message(BaseModel):
105
114
  if self.content.parts:
106
115
  return ""
107
116
 
117
+ # tether_quote: render as a blockquote with attribution (check before .text)
118
+ if self.content.content_type == "tether_quote":
119
+ return self._render_tether_quote()
108
120
  if self.content.text is not None:
109
121
  return self.content.text
110
122
  if self.content.result is not None:
111
123
  return self.content.result
124
+ # reasoning_recap content type uses 'content' field
125
+ if self.content.content is not None:
126
+ return self.content.content
127
+ # thoughts content type uses 'thoughts' field (list of thought objects)
128
+ if self.content.thoughts is not None:
129
+ return self._render_thoughts()
112
130
  raise MessageContentError(self.id)
113
131
 
132
+ def _render_thoughts(self) -> str:
133
+ """Render thoughts content (list of thought objects with summary/content)."""
134
+ if not self.content.thoughts:
135
+ return ""
136
+ summaries = []
137
+ for thought in self.content.thoughts:
138
+ if isinstance(thought, dict) and (summary := thought.get("summary")):
139
+ summaries.append(summary)
140
+ return "\n".join(summaries) if summaries else ""
141
+
142
+ def _render_tether_quote(self) -> str:
143
+ """Render tether_quote content as a blockquote."""
144
+ quote_text = self.content.text or ""
145
+ if not quote_text.strip():
146
+ return ""
147
+ # Format as blockquote with source
148
+ lines = [f"> {line}" for line in quote_text.strip().split("\n")]
149
+ blockquote = "\n".join(lines)
150
+ # Add attribution if we have title/domain/url
151
+ if self.content.title and self.content.url:
152
+ blockquote += f"\n> — [{self.content.title}]({self.content.url})"
153
+ elif self.content.domain and self.content.url:
154
+ blockquote += f"\n> — [{self.content.domain}]({self.content.url})"
155
+ elif self.content.url:
156
+ blockquote += f"\n> — <{self.content.url}>"
157
+ return blockquote
158
+
114
159
  @property
115
160
  def has_content(self) -> bool:
116
161
  """Check if the message has extractable content."""
@@ -132,26 +177,42 @@ class Message(BaseModel):
132
177
 
133
178
  Hidden if:
134
179
  1. It is empty (no text, no images).
135
- 2. It is an internal system message (not custom instructions).
136
- 3. It is a browser tool output (intermediate search steps).
180
+ 2. Explicitly marked as visually hidden.
181
+ 3. It is an internal system message (not custom instructions).
182
+ 4. It is a browser tool output (intermediate search steps).
183
+ 5. It is an assistant message targeting a tool (internal call).
184
+ 6. It is code interpreter input (content_type="code").
185
+ 7. It is browsing status (tether_browsing_display).
186
+ 8. It is internal reasoning (thoughts, reasoning_recap from o1/o3).
137
187
  """
138
188
  if self.is_empty:
139
189
  return True
140
190
 
191
+ # Explicitly marked as hidden by OpenAI
192
+ if self.metadata.is_visually_hidden_from_conversation:
193
+ return True
194
+
141
195
  # Hide internal system messages
142
196
  if self.author.role == "system":
143
197
  # Only show if explicitly marked as user system message (Custom Instructions)
144
198
  return not self.metadata.is_user_system_message
145
199
 
146
- # Hide browser tool outputs (usually intermediate search steps)
200
+ # Hide browser tool outputs (intermediate search steps)
147
201
  if self.author.role == "tool" and self.author.name == "browser":
148
202
  return True
149
203
 
150
- # Hide assistant calls to browser tool (e.g. "search(...)") or code interpreter
151
- if self.author.role == "assistant" and (
152
- self.recipient == "browser" or self.content.content_type == "code"
153
- ):
204
+ # Hide assistant messages targeting tools (e.g., search(...), code input)
205
+ # recipient="all" or None means it's for the user; anything else is internal
206
+ if self.author.role == "assistant" and self.recipient not in ("all", None):
154
207
  return True
155
208
 
156
- # Hide browsing status messages
157
- return self.content.content_type == "tether_browsing_display"
209
+ # Hide code interpreter input (content_type="code")
210
+ if self.author.role == "assistant" and self.content.content_type == "code":
211
+ return True
212
+
213
+ # Hide browsing status and internal reasoning steps (o1/o3 models)
214
+ return self.content.content_type in (
215
+ "tether_browsing_display",
216
+ "thoughts",
217
+ "reasoning_recap",
218
+ )
@@ -113,6 +113,7 @@ def run_pipeline(config: ConvovizConfig) -> None:
113
113
  markdown_folder,
114
114
  config.conversation,
115
115
  config.message.author_headers,
116
+ folder_organization=config.folder_organization,
116
117
  progress_bar=True,
117
118
  )
118
119
  console.print(
@@ -1,6 +1,6 @@
1
1
  [project]
2
2
  name = "convoviz"
3
- version = "0.2.6"
3
+ version = "0.2.8"
4
4
  description = "Get analytics and visualizations on your ChatGPT data!"
5
5
  license = "MIT"
6
6
  keywords = [
File without changes
File without changes
File without changes
File without changes