tumblrbot 1.0.0__py3-none-any.whl → 1.1.1__py3-none-any.whl

tumblrbot/__main__.py

@@ -1,49 +1,19 @@
-from collections.abc import Generator
-
-import rich
 from openai import OpenAI
-from pydantic import Secret
-from rich.prompt import Confirm, Prompt
+from rich.prompt import Confirm
 from rich.traceback import install
 
 from tumblrbot.flow.download import PostDownloader
 from tumblrbot.flow.examples import ExamplesWriter
 from tumblrbot.flow.fine_tune import FineTuner
 from tumblrbot.flow.generate import DraftGenerator
+from tumblrbot.utils.common import TumblrClient
 from tumblrbot.utils.settings import Tokens
-from tumblrbot.utils.tumblr import TumblrClient
-
-
-def online_token_prompt(url: str, *tokens: str) -> Generator[Secret[str]]:
-    formatted_tokens = [f"[cyan]{token}[/]" for token in tokens]
-    formatted_token_string = " and ".join(formatted_tokens)
-
-    rich.print(f"Retrieve your {formatted_token_string} from: {url}")
-    for token in formatted_tokens:
-        prompt = f"Enter your {token} [yellow](hidden)"
-        yield Secret(Prompt.ask(prompt, password=True).strip())
-
-    rich.print()
-
-
-def verify_tokens() -> Tokens:
-    tokens = Tokens()
-
-    if not tokens.openai.api_key.get_secret_value():
-        (tokens.openai.api_key,) = online_token_prompt("https://platform.openai.com/api-keys", "API key")
-        tokens.model_post_init()
-
-    if not (tokens.tumblr.client_id.get_secret_value() and tokens.tumblr.client_secret.get_secret_value()):
-        tokens.tumblr.client_id, tokens.tumblr.client_secret = online_token_prompt("https://tumblr.com/oauth/apps", "consumer key", "consumer secret")
-        tokens.model_post_init()
-
-    return tokens
 
 
 def main() -> None:
     install()
-    tokens = verify_tokens()
-    with OpenAI(api_key=tokens.openai.api_key.get_secret_value()) as openai, TumblrClient(tokens) as tumblr:
+    tokens = Tokens()
+    with OpenAI(api_key=tokens.openai_api_key.get_secret_value()) as openai, TumblrClient(tokens) as tumblr:
         post_downloader = PostDownloader(openai, tumblr)
         if Confirm.ask("Download latest posts?", default=False):
             post_downloader.download()
@@ -56,7 +26,7 @@ def main() -> None:
 
         fine_tuner = FineTuner(openai, tumblr, estimated_tokens)
         fine_tuner.print_estimates()
-        if Confirm.ask("Upload data to OpenAI for fine-tuning?", default=False):
+        if Confirm.ask("Upload data to OpenAI for fine-tuning? [bold]You must do this to set the model to generate drafts from. Alternatively, manually enter a model into the config.", default=False):
             fine_tuner.fine_tune()
 
         if Confirm.ask("Generate drafts?", default=False):

tumblrbot/flow/download.py

@@ -2,19 +2,17 @@ from io import TextIOBase
 from json import dump
 from pathlib import Path
 
-from more_itertools import last
-
 from tumblrbot.utils.common import PreviewLive, UtilClass
 from tumblrbot.utils.models import Post
 
 
 class PostDownloader(UtilClass):
-    def paginate_posts(self, blog_name: str, before: int, completed: int, fp: TextIOBase, live: PreviewLive) -> None:
-        task_id = live.progress.add_task(f"Downloading posts from '{blog_name}'...", total=None, completed=completed)
+    def paginate_posts(self, blog_identifier: str, offset: int, fp: TextIOBase, live: PreviewLive) -> None:
+        task_id = live.progress.add_task(f"Downloading posts from '{blog_identifier}'...", total=None, completed=offset)
 
         while True:
-            response = self.tumblr.retrieve_published_posts(blog_name, before).json()["response"]
-            live.progress.update(task_id, total=response["blog"]["posts"])
+            response = self.tumblr.retrieve_published_posts(blog_identifier, offset).json()["response"]
+            live.progress.update(task_id, total=response["blog"]["posts"], completed=offset)
 
             if posts := response["posts"]:
                 for post in posts:
@@ -22,32 +20,29 @@ class PostDownloader(UtilClass):
                     fp.write("\n")
 
                     model = Post.model_validate(post)
-                    before = model.timestamp
-
-                    live.progress.update(task_id, advance=1)
                     live.custom_update(model)
+
+                offset += len(posts)
             else:
                 break
 
-    def get_data_path(self, blog_name: str) -> Path:
-        return (self.config.training.data_directory / blog_name).with_suffix(".jsonl")
+    def get_data_path(self, blog_identifier: str) -> Path:
+        return (self.config.data_directory / blog_identifier).with_suffix(".jsonl")
 
     def get_data_paths(self) -> list[Path]:
-        return list(map(self.get_data_path, self.config.training.blog_names))
+        return list(map(self.get_data_path, self.config.download_blog_identifiers))
 
     def download(self) -> None:
-        self.config.training.data_directory.mkdir(parents=True, exist_ok=True)
+        self.config.data_directory.mkdir(parents=True, exist_ok=True)
 
         with PreviewLive() as live:
-            for blog_name in self.config.training.blog_names:
-                data_path = self.get_data_path(blog_name)
-                lines = data_path.read_text("utf_8").splitlines() if data_path.exists() else []
+            for blog_identifier in self.config.download_blog_identifiers:
+                data_path = self.get_data_path(blog_identifier)
 
                 with data_path.open("a", encoding="utf_8") as fp:
                     self.paginate_posts(
-                        blog_name,
-                        Post.model_validate_json(last(lines, "{}")).timestamp,
-                        len(lines),
+                        blog_identifier,
+                        len(data_path.read_text("utf_8").splitlines()) if data_path.exists() else 0,
                         fp,
                         live,
                     )

tumblrbot/flow/examples.py

@@ -28,7 +28,7 @@ class ExamplesWriter(UtilClass):
             encoding = get_encoding("o200k_base")
             Console(stderr=True, style="logging.level.warning").print(f"[Warning] Using encoding '{encoding.name}': {''.join(error.args)}\n")
 
-        with self.config.training.output_file.open(encoding="utf_8") as fp:
+        with self.config.examples_file.open(encoding="utf_8") as fp:
             for line in fp:
                 example = Example.model_validate_json(line)
                 yield len(encoding.encode("assistant"))  # every reply is primed with <|start|>assistant<|message|>
@@ -77,8 +77,8 @@ class ExamplesWriter(UtilClass):
             yield from posts
 
     def write_examples(self) -> None:
-        self.config.training.output_file.parent.mkdir(parents=True, exist_ok=True)
-        with self.config.training.output_file.open("w", encoding="utf_8") as fp:
+        self.config.examples_file.parent.mkdir(parents=True, exist_ok=True)
+        with self.config.examples_file.open("w", encoding="utf_8") as fp:
             for post in self.get_filtered_posts():
                 example = Example(
                     messages=[
@@ -89,4 +89,4 @@ class ExamplesWriter(UtilClass):
                 )
                 fp.write(f"{example.model_dump_json()}\n")
 
-        rich.print(f"[bold]The examples file can be found at: '{self.config.training.output_file}'\n")
+        rich.print(f"[bold]The examples file can be found at: '{self.config.examples_file}'\n")

tumblrbot/flow/fine_tune.py

@@ -24,22 +24,19 @@ class FineTuner(UtilClass):
                 Cost: {self.get_cost_string(job.trained_tokens)}
             """)
 
-        self.config.training.job_id = ""
-        self.config.model_post_init()
+        self.config.job_id = ""
 
         if job.status == "failed" and job.error is not None:
             raise RuntimeError(job.error.message)
 
         if job.fine_tuned_model:
-            self.config.generation.fine_tuned_model = job.fine_tuned_model or ""
-            self.config.model_post_init()
+            self.config.fine_tuned_model = job.fine_tuned_model or ""
 
     def poll_job_status(self) -> FineTuningJob:
-        job = self.openai.fine_tuning.jobs.retrieve(self.config.training.job_id)
+        job = self.openai.fine_tuning.jobs.retrieve(self.config.job_id)
 
-        if self.config.training.expected_epochs != job.hyperparameters.n_epochs and isinstance(job.hyperparameters.n_epochs, int):
-            self.config.training.expected_epochs = job.hyperparameters.n_epochs
-            self.config.model_post_init()
+        if self.config.expected_epochs != job.hyperparameters.n_epochs and isinstance(job.hyperparameters.n_epochs, int):
+            self.config.expected_epochs = job.hyperparameters.n_epochs
 
             self.dedent_print(f"""
                 The number of epochs has been updated to {job.hyperparameters.n_epochs}!
@@ -50,11 +47,11 @@ class FineTuner(UtilClass):
         return job
 
     def create_job(self) -> FineTuningJob:
-        if self.config.training.job_id:
+        if self.config.job_id:
             return self.poll_job_status()
 
         file = self.openai.files.create(
-            file=self.config.training.output_file,
+            file=self.config.examples_file,
             purpose="fine-tune",
         )
         job = self.openai.fine_tuning.jobs.create(
@@ -62,8 +59,7 @@ class FineTuner(UtilClass):
             training_file=file.id,
         )
 
-        self.config.training.job_id = job.id
-        self.config.model_post_init()
+        self.config.job_id = job.id
         return job
 
     def fine_tune(self) -> None:
@@ -86,7 +82,7 @@ class FineTuner(UtilClass):
 
                 live.progress.update(
                     task_id,
-                    description=f"Fine-tuning: {job.status}...",
+                    description=f"Fine-tuning: [italic]{job.status.replace('_', ' ').title()}[/]...",
                 )
 
                 sleep(1)
@@ -94,16 +90,16 @@ class FineTuner(UtilClass):
         self.process_completed_job(job)
 
     def get_cost_string(self, total_tokens: int) -> str:
-        return f"${self.config.training.token_price / 1000000 * total_tokens:.2f}"
+        return f"${self.config.token_price / 1000000 * total_tokens:.2f}"
 
     def print_estimates(self) -> None:
-        total_tokens = self.config.training.expected_epochs * self.estimated_tokens
+        total_tokens = self.config.expected_epochs * self.estimated_tokens
         cost_string = self.get_cost_string(total_tokens)
 
         self.dedent_print(f"""
             Tokens {self.estimated_tokens:,}:
-            Total tokens for [bold orange1]{self.config.training.expected_epochs}[/] epoch(s): {total_tokens:,}
+            Total tokens for [bold orange1]{self.config.expected_epochs}[/] epoch(s): {total_tokens:,}
             Expected cost when trained with [bold purple]{self.config.base_model}[/]: {cost_string}
             NOTE: Token values are approximate and may not be 100% accurate, please be aware of this when using the data.
-                    [italic red]Neither Amelia nor Mutsumi are responsible for any inaccuracies in the token count or estimated price.[/]
+                    [italic red]Amelia, Mutsumi, and Marin are not responsible for any inaccuracies in the token count or estimated price.[/]
         """)

tumblrbot/flow/generate.py

@@ -8,7 +8,7 @@ from tumblrbot.utils.models import Post
 
 class DraftGenerator(UtilClass):
     def generate_tags(self, content: Post.Block) -> Post | None:
-        if random() < self.config.generation.tags_chance:  # noqa: S311
+        if random() < self.config.tags_chance:  # noqa: S311
             return self.openai.responses.parse(
                 input=content.text,
                 model=self.config.base_model,
@@ -23,29 +23,32 @@ class DraftGenerator(UtilClass):
         content = self.openai.responses.create(
             input=self.config.user_input,
             instructions=self.config.developer_message,
-            model=self.config.generation.fine_tuned_model,
+            model=self.config.fine_tuned_model,
         ).output_text
 
         return Post.Block(type="text", text=content)
 
     def generate_post(self) -> Post:
         content = self.generate_content()
-        post = Post(content=[content])
+        post = Post(
+            content=[content],
+            state="draft",
+        )
         if tags := self.generate_tags(content):
             post.tags = tags.tags
         return post
 
     def create_drafts(self) -> None:
-        message = f"View drafts here: https://tumblr.com/blog/{self.config.generation.blog_name}/drafts"
+        message = f"View drafts here: https://tumblr.com/blog/{self.config.upload_blog_identifier}/drafts"
 
         with PreviewLive() as live:
-            for i in live.progress.track(range(self.config.generation.draft_count), description="Generating drafts..."):
+            for i in live.progress.track(range(self.config.draft_count), description="Generating drafts..."):
                 try:
                     post = self.generate_post()
-                    self.tumblr.create_draft_post(self.config.generation.blog_name, post)
+                    self.tumblr.create_post(self.config.upload_blog_identifier, post)
                     live.custom_update(post)
                 except BaseException as exc:
                     exc.add_note(f"📉 An error occurred! Generated {i} draft(s) before failing. {message}")
                     raise
 
-        rich.print(f":chart_increasing: [bold green]Generated {self.config.generation.draft_count} draft(s).[/] {message}")
+        rich.print(f":chart_increasing: [bold green]Generated {self.config.draft_count} draft(s).[/] {message}")

tumblrbot/utils/models.py

@@ -18,16 +18,15 @@ class FullyValidatedModel(BaseModel):
 
 class Post(FullyValidatedModel):
     class Block(FullyValidatedModel):
-        type: str = "text"
+        type: str = ""
         text: str = ""
-        blocks: set[int] = set()  # noqa: RUF012
+        blocks: list[int] = []  # noqa: RUF012
 
-    tags: Annotated[set[str], PlainSerializer(",".join)] = set()  # noqa: RUF012
+    tags: Annotated[list[str], PlainSerializer(",".join)] = []  # noqa: RUF012
     content: SkipJsonSchema[list[Block]] = []  # noqa: RUF012
     layout: SkipJsonSchema[list[Block]] = []  # noqa: RUF012
     trail: SkipJsonSchema[list[Any]] = []  # noqa: RUF012
-    state: SkipJsonSchema[Literal["published", "queued", "draft", "private", "unapproved"]] = "draft"
-    timestamp: SkipJsonSchema[int] = 0
+    state: SkipJsonSchema[Literal["published", "queued", "draft", "private", "unapproved"]] = "published"
     is_submission: SkipJsonSchema[bool] = False
 
     def __rich__(self) -> Panel:
@@ -45,7 +44,7 @@ class Post(FullyValidatedModel):
         indices: set[int] = set()
         for block in self.layout:
             if block.type == "ask":
-                indices |= block.blocks
+                indices.update(block.blocks)
 
         self.content = [block for i, block in enumerate(self.content) if i not in indices and block.type == "text"]
 
@@ -55,7 +54,7 @@ class Post(FullyValidatedModel):
 
 class Example(FullyValidatedModel):
     class Message(FullyValidatedModel):
-        role: str
+        role: Literal["developer", "user", "assistant"]
         content: str
 
     messages: list[Message]

tumblrbot/utils/settings.py

@@ -1,18 +1,19 @@
-from collections.abc import Sequence
+from collections.abc import Generator, Sequence
 from pathlib import Path
-from typing import TYPE_CHECKING, Any, override
+from typing import TYPE_CHECKING, Any, Self, override
 
+import rich
 from openai.types import ChatModel
-from pydantic import Field, PositiveFloat, PositiveInt, Secret, field_serializer
+from pydantic import Field, PositiveFloat, PositiveInt, Secret, model_validator
 from pydantic_settings import BaseSettings, PydanticBaseSettingsSource, SettingsConfigDict, TomlConfigSettingsSource
-from tomlkit import comment, dumps, table
-from tomlkit.items import Table
+from rich.prompt import Prompt
+from tomlkit import comment, document, dumps  # pyright: ignore[reportUnknownVariableType]
 
 if TYPE_CHECKING:
     from _typeshed import StrPath
 
 
-class TomlSettings(BaseSettings):
+class TOMLSettings(BaseSettings):
     model_config = SettingsConfigDict(
         extra="ignore",
         validate_assignment=True,
@@ -20,38 +21,13 @@ class TomlSettings(BaseSettings):
         validate_by_name=True,
     )
 
-    @field_serializer("*", when_used="json-unless-none")
-    @staticmethod
-    def serialize_secret(value: object) -> object:
-        if isinstance(value, Secret):
-            return value.get_secret_value()
-        return value
-
-    def get_toml_table(self) -> Table:
-        toml_table = table()
-
-        dumped_model = self.model_dump(mode="json")
-        for name, field in self.__class__.model_fields.items():
-            if field.description:
-                for line in field.description.split(". "):
-                    toml_table.add(comment(f"{line.removesuffix('.')}."))
-
-            value = getattr(self, name)
-            toml_table[name] = value.get_toml_table() if isinstance(value, TomlSettings) else dumped_model[name]
-
-        return toml_table
-
-
-class AutoGenerateTomlSettings(TomlSettings):
     @override
     @classmethod
     def settings_customise_sources(cls, settings_cls: type[BaseSettings], *args: PydanticBaseSettingsSource, **kwargs: PydanticBaseSettingsSource) -> tuple[PydanticBaseSettingsSource, ...]:
         return (TomlConfigSettingsSource(settings_cls),)
 
-    @override
-    def model_post_init(self, context: object = None) -> None:
-        super().model_post_init(context)
-
+    @model_validator(mode="after")
+    def write_to_file(self) -> Self:
         # Make sure to call this if updating values in nested models.
         toml_files = self.model_config.get("toml_file")
         if isinstance(toml_files, (Path, str)):
@@ -60,14 +36,27 @@ class AutoGenerateTomlSettings(TomlSettings):
             for toml_file in toml_files:
                 self.dump_toml(toml_file)
 
+        return self
+
     def dump_toml(self, toml_file: "StrPath") -> None:
+        toml_table = document()
+
+        dumped_model = self.model_dump(mode="json")
+        for name, field in self.__class__.model_fields.items():
+            if field.description:
+                for line in field.description.split(". "):
+                    toml_table.add(comment(f"{line.removesuffix('.')}."))
+
+            value = getattr(self, name)
+            toml_table[name] = value.get_secret_value() if isinstance(value, Secret) else dumped_model[name]
+
         Path(toml_file).write_text(
-            dumps(self.get_toml_table()),
+            dumps(toml_table),
             encoding="utf_8",
         )
 
 
-class Config(AutoGenerateTomlSettings):
+class Config(TOMLSettings):
     model_config = SettingsConfigDict(
         cli_parse_args=True,
         cli_avoid_json=True,
@@ -75,44 +64,68 @@ class Config(AutoGenerateTomlSettings):
         toml_file="config.toml",
     )
 
-    class Generation(TomlSettings):
-        fine_tuned_model: str = Field("", description="The name of the OpenAI model that was fine-tuned with your posts.")
-        blog_name: str = Field(
-            "",
-            description='The name of the blog which generated drafts will be uploaded to that appears in the URL. This must be a blog associated with the same account as the configured Tumblr secret values. Examples: "staff" for https://staff.tumblr.com and "changes" for https://tumblr.com/changes or https://tumblr.com/@changes',
-        )
-        draft_count: PositiveInt = Field(150, description="The number of drafts to process. This will affect the number of tokens used with OpenAI")
-        tags_chance: float = Field(0.1, description="The chance to generate tags for any given post. This will incur extra calls to OpenAI.")
+    fine_tuned_model: str = Field("", description="The name of the OpenAI model that was fine-tuned with your posts.")
+    upload_blog_identifier: str = Field(
+        "",
+        description="The identifier of the blog which generated drafts will be uploaded to. This must be a blog associated with the same account as the configured Tumblr secret tokens.",
+    )
+    draft_count: PositiveInt = Field(150, description="The number of drafts to process. This will affect the number of tokens used with OpenAI")
+    tags_chance: float = Field(0.1, description="The chance to generate tags for any given post. This will incur extra calls to OpenAI.")
 
-    class Training(TomlSettings):
-        blog_names: list[str] = Field(
-            [],
-            description='The names of the blogs which post data will be downloaded from that appears in the URL. This must be a blog associated with the same account as the configured Tumblr secret values. Examples: ["staff", "changes"] for https://staff.tumblr.com and https://www.tumblr.com/changes or https://www.tumblr.com/@changes',
-        )
-        data_directory: Path = Field(Path("data"), description="Where to store downloaded post data.")
-        output_file: Path = Field(Path("training.jsonl"), description="Where to output the training data that will be used to fine-tune the model.")
-        job_id: str = Field("", description="The fine-tuning job ID that will be polled on next run.")
-        expected_epochs: PositiveInt = Field(3, description="The expected number of epochs fine-tuning will be run for. This will be updated during fine-tuning.")
-        token_price: PositiveFloat = Field(1.50, description="The expected price in USD per million tokens during fine-tuning for the current model.")
+    download_blog_identifiers: list[str] = Field(
+        [],
+        description="The identifiers of the blogs which post data will be downloaded from. These must be blogs associated with the same account as the configured Tumblr secret tokens.",
+    )
+    data_directory: Path = Field(Path("data"), description="Where to store downloaded post data.")
+    examples_file: Path = Field(Path("examples.jsonl"), description="Where to output the examples that will be used to fine-tune the model.")
+    job_id: str = Field("", description="The fine-tuning job ID that will be polled on next run.")
+    expected_epochs: PositiveInt = Field(3, description="The expected number of epochs fine-tuning will be run for. This will be updated during fine-tuning.")
+    token_price: PositiveFloat = Field(1.50, description="The expected price in USD per million tokens during fine-tuning for the current model.")
 
     base_model: ChatModel = Field("gpt-4.1-nano-2025-04-14", description="The name of the model that will be fine-tuned by the generated training data.")
     developer_message: str = Field("You are a Tumblr post bot. Please generate a Tumblr post in accordance with the user's request.", description="The developer message used by the OpenAI API to generate drafts.")
     user_input: str = Field("Please write a comical Tumblr post.", description="The user input used by the OpenAI API to generate drafts.")
 
-    generation: Generation = Generation()  # pyright: ignore[reportCallIssue]
-    training: Training = Training()  # pyright: ignore[reportCallIssue]
+    @override
+    def model_post_init(self, context: object) -> None:
+        super().model_post_init(context)
 
+        if not self.download_blog_identifiers:
+            rich.print("Enter the [cyan]identifiers of your blogs[/] that data should be [bold purple]downloaded[/] from, separated by commas.")
+            self.download_blog_identifiers = list(map(str.strip, Prompt.ask("[bold]Example: staff.tumblr.com,changes").split(",")))
 
-class Tokens(AutoGenerateTomlSettings):
+        if not self.upload_blog_identifier:
+            rich.print("Enter the [cyan]identifier of your blog[/] that drafts should be [bold purple]uploaded[/] to.")
+            self.upload_blog_identifier = Prompt.ask("[bold]Examples: staff.tumblr.com or changes").strip()
+
+
+class Tokens(TOMLSettings):
     model_config = SettingsConfigDict(toml_file="env.toml")
 
-    class OpenAI(TomlSettings):
-        api_key: Secret[str] = Secret("")
+    openai_api_key: Secret[str] = Secret("")
+
+    tumblr_client_id: Secret[str] = Secret("")
+    tumblr_client_secret: Secret[str] = Secret("")
+    tumblr_token: Secret[Any] = Secret({})
+
+    @staticmethod
+    def online_token_prompt(url: str, *tokens: str) -> Generator[Secret[str]]:
+        formatted_tokens = [f"[cyan]{token}[/]" for token in tokens]
+        formatted_token_string = " and ".join(formatted_tokens)
+
+        rich.print(f"Retrieve your {formatted_token_string} from: {url}")
+        for token in formatted_tokens:
+            prompt = f"Enter your {token} [yellow](hidden)"
+            yield Secret(Prompt.ask(prompt, password=True).strip())
+
+        rich.print()
+
+    @override
+    def model_post_init(self, context: object) -> None:
+        super().model_post_init(context)
 
-    class Tumblr(TomlSettings):
-        client_id: Secret[str] = Secret("")
-        client_secret: Secret[str] = Secret("")
-        token: Secret[Any] = Secret({})
+        if not self.openai_api_key.get_secret_value():
+            (self.openai_api_key,) = self.online_token_prompt("https://platform.openai.com/api-keys", "API key")
 
-    openai: OpenAI = OpenAI()
-    tumblr: Tumblr = Tumblr()
+        if not (self.tumblr_client_id.get_secret_value() and self.tumblr_client_secret.get_secret_value()):
+            self.tumblr_client_id, self.tumblr_client_secret = self.online_token_prompt("https://tumblr.com/oauth/apps", "consumer key", "consumer secret")

tumblrbot/utils/tumblr.py

@@ -16,12 +16,16 @@ class TumblrClient(OAuth2Session):
     tokens: Tokens
 
     def __post_init__(self) -> None:
-        super().__init__(
-            self.tokens.tumblr.client_id.get_secret_value(),
+        super().__init__(  # pyright: ignore[reportUnknownMemberType]
+            self.tokens.tumblr_client_id.get_secret_value(),
             auto_refresh_url="https://api.tumblr.com/v2/oauth2/token",
-            auto_refresh_kwargs=self.tokens.tumblr.model_dump(mode="json"),
+            auto_refresh_kwargs={
+                "client_id": self.tokens.tumblr_client_id.get_secret_value(),
+                "client_secret": self.tokens.tumblr_client_secret.get_secret_value(),
+                "token": self.tokens.tumblr_token.get_secret_value(),
+            },
             scope=["basic", "write", "offline_access"],
-            token=self.tokens.tumblr.token.get_secret_value(),
+            token=self.tokens.tumblr_token.get_secret_value(),
             token_updater=self.token_saver,
         )
 
@@ -30,26 +34,25 @@ class TumblrClient(OAuth2Session):
     def __enter__(self) -> Self:
         super().__enter__()
 
-        if not self.tokens.tumblr.token.get_secret_value():
-            authorization_url, _ = self.authorization_url("https://tumblr.com/oauth2/authorize")
+        if not self.tokens.tumblr_token.get_secret_value():
+            authorization_url, _ = self.authorization_url("https://tumblr.com/oauth2/authorize")  # pyright: ignore[reportUnknownMemberType]
 
             rich.print(f"Please go to {authorization_url} and authorize access.")
             authorization_response = Prompt.ask("Enter the full callback URL")
             rich.print("\n")
 
             self.token_saver(
-                self.fetch_token(
+                self.fetch_token(  # pyright: ignore[reportUnknownMemberType]
                     "https://api.tumblr.com/v2/oauth2/token",
                     authorization_response=authorization_response,
-                    client_secret=self.tokens.tumblr.client_secret.get_secret_value(),
+                    client_secret=self.tokens.tumblr_client_secret.get_secret_value(),
                 ),
             )
 
         return self
 
     def token_saver(self, token: object) -> None:
-        self.tokens.tumblr.token = Secret(token)
-        self.tokens.model_post_init()
+        self.tokens.tumblr_token = Secret(token)
 
     def response_hook(self, response: Response, **_: object) -> None:
         try:
@@ -65,17 +68,18 @@ class TumblrClient(OAuth2Session):
                 error.add_note(str(json))
             raise
 
-    def create_draft_post(self, blog_name: str, post: Post) -> Response:
-        return self.post(
-            f"https://api.tumblr.com/v2/blog/{blog_name}/posts",
-            json=post.model_dump(mode="json"),
-        )
-
-    def retrieve_published_posts(self, blog_name: str, before: int) -> Response:
+    def retrieve_published_posts(self, blog_identifier: str, offset: int) -> Response:
         return self.get(
-            f"https://api.tumblr.com/v2/blog/{blog_name}/posts",
+            f"https://api.tumblr.com/v2/blog/{blog_identifier}/posts",
             params={
-                "before": before,
+                "offset": offset,
+                "sort": "asc",
                 "npf": True,
             },
         )
+
+    def create_post(self, blog_identifier: str, post: Post) -> Response:
+        return self.post(
+            f"https://api.tumblr.com/v2/blog/{blog_identifier}/posts",
+            json=post.model_dump(mode="json"),
+        )

tumblrbot-1.1.1.dist-info/METADATA

@@ -0,0 +1,15 @@
+Metadata-Version: 2.4
+Name: tumblrbot
+Version: 1.1.1
+Summary: An updated bot that posts to Tumblr, based on your very own blog!
+Requires-Python: >= 3.13
+Requires-Dist: more-itertools
+Requires-Dist: openai
+Requires-Dist: pydantic
+Requires-Dist: pydantic-settings
+Requires-Dist: requests
+Requires-Dist: requests-oauthlib
+Requires-Dist: rich
+Requires-Dist: tiktoken
+Requires-Dist: tomlkit
+Project-URL: Source, https://github.com/MaidThatPrograms/tumblrbot

tumblrbot-1.1.1.dist-info/RECORD

@@ -0,0 +1,16 @@
+tumblrbot/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+tumblrbot/__main__.py,sha256=RSvzROxs8hi_0sOyKsnZtV2-S3T-lPeJuDwULGN1-2U,1509
+tumblrbot/flow/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+tumblrbot/flow/download.py,sha256=FsiYJSyinFZi5wHvRL6WQCURmTDzrJ4XgThJkaRFPxw,1911
+tumblrbot/flow/examples.py,sha256=vCJ6KH-kqlmi7zW-jk6fQqSGDkYnXHplTmTmIUC-Xj0,4168
+tumblrbot/flow/fine_tune.py,sha256=AWBxlHfQDDIGku1oZaaMNQtNw3yOaScq6QMgI8sJhd0,3836
+tumblrbot/flow/generate.py,sha256=6b6-Hzqek0AO6i7ceX-mapJfOKLYx6FOqEvHItjg5kU,2088
+tumblrbot/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+tumblrbot/utils/common.py,sha256=3PPyutv7yyjvij5anyKQRDRn7lrr8Eu05m8XVhh74Hc,1342
+tumblrbot/utils/models.py,sha256=to4k0b1O0bYNXe_4zx5Vupxtb-739U3nMSGgJLJ3gto,1975
+tumblrbot/utils/settings.py,sha256=yC2srfEw8Kl0z-jDOkc9qWfRkutTzZFfg3uRzfjSGAQ,6507
+tumblrbot/utils/tumblr.py,sha256=tFrGwHY3FsX2kWNiLteettXpnWgqNpzPCQImtEpND2M,3283
+tumblrbot-1.1.1.dist-info/entry_points.txt,sha256=lTiN7PxAbyGY1fpCWApEw6NUIUgobfcOKhvn6cu3IQA,53
+tumblrbot-1.1.1.dist-info/WHEEL,sha256=G2gURzTEtmeR8nrdXUJfNiB3VYVxigPQ-bEQujpNiNs,82
+tumblrbot-1.1.1.dist-info/METADATA,sha256=CitrYF_L0yiPI_q4tmUPq_ascm2tDG1Uc_YXT8RnC3Q,453
+tumblrbot-1.1.1.dist-info/RECORD,,

tumblrbot-1.0.0.dist-info/METADATA

@@ -1,369 +0,0 @@
-Metadata-Version: 2.4
-Name: tumblrbot
-Version: 1.0.0
-Summary: An updated bot that posts to Tumblr, based on your very own blog!
-Requires-Python: >= 3.13
-Description-Content-Type: text/markdown
-License-Expression: Unlicense
-License-File: UNLICENSE
-Requires-Dist: more-itertools
-Requires-Dist: openai
-Requires-Dist: pydantic
-Requires-Dist: pydantic-settings
-Requires-Dist: requests
-Requires-Dist: requests-oauthlib
-Requires-Dist: rich
-Requires-Dist: tiktoken
-Requires-Dist: tomlkit
-Project-URL: Issues, https://github.com/MaidThatPrograms/tumblrbot/issues
-Project-URL: Repository, https://github.com/MaidThatPrograms/tumblrbot
-
-[OpenAI]: https://pypi.org/project/openai
-[Rich]: https://pypi.org/project/rich
-
-[gpt-4.1-nano-2025-04-14]: https://platform.openai.com/docs/models/gpt-4.1-nano
-[Moderation API]: https://platform.openai.com/docs/api-reference/moderations
-[New Post Format]: https://tumblr.com/docs/npf
-[OAuth 2.0]: https://www.tumblr.com/docs/en/api/v2#oauth2-authorization
-[pip]: https://pypi.org
-
-[Download]: tumblrbot/flow/download.py
-[Examples]: tumblrbot/flow/examples.py
-[Fine-Tune]: tumblrbot/flow/fine_tune.py
-[Generate]: tumblrbot/flow/generate.py
-[Utils]: tumblrbot/utils/common.py
-[Models]: tumblrbot/utils/models.py
-[Settings]: tumblrbot/utils/settings.py
-[Tumblr]: tumblrbot/utils/tumblr.py
-[Main]: __main__.py
-[README.md]: README.md
-
-# tumblrbot
-Description of original project:
-> 4tv-tumblrbot was a collaborative project I embarked on with my close friend Dima, who goes by @smoqueen on Tumblr. The aim of this endeavor was straightforward yet silly: to develop a Tumblr bot powered by a machine-learning model. This bot would be specifically trained on the content from a particular Tumblr blog or a selected set of blogs, allowing it to mimic the style, tone, and thematic essence of the original posts.
-
-This fork is largely a rewrite of the source code with similarities in its structure and process:
-- Updates:
-   - Updated to [OAuth 2.0].
-   - Updated to the [New Post Format].
-   - Updated to the latest version of [OpenAI].
-   - Updated the [base model version][Settings] to [gpt-4.1-nano-2025-04-14].
-- Removed features:
-   - [Generation][Generate]:
-      - Removed clearing drafts behavior.
-   - [Training][Examples]:
-      - Removed exports that had HTML or reblogs.
-      - Removed special word-replacement behavior.
-      - Removed filtering by year.
-   - Removed setup and related files.
-- Changed/Added features:
-   - [Generation][Generate]:
-      - Added a link to the blog's draft page.
-      - Added error checking for uploading drafts.
-   - [Training][Examples]:
-      - Added the option to [Download] the latest posts from the [specified blogs][Settings].
-      - Added the option to remove posts flagged by the [Moderation API].
-      - Added the option to automatically [Fine-Tune] the examples on the [specified base model][Settings].
-      - Changed to now escape examples automatically.
-      - Set encoding for reading post data to `UTF-8` to fix decoding errors.
-      - Added newlines between paragraphs.
-      - Removed "ALT", submission, ask, and poll text from posts.
-      - Improved the estimated token counts and costs.
-   - Changed to [Rich] for output.
-      - Added progress bars.
-      - Added post previews.
-      - Added color, formatting, and more information to output.
-   - Created a [guided utility][Main] for every step of building your bot blog.
-   - Maid scripts wait for user input before the console closes.
-   - Added comand-line options to override [Settings] options.
-   - Added behavior to regenerate the default [config.toml][Settings] and [env.toml][Settings] if missing.
-   - Renamed several files.
-   - Renamed several [Settings] options.
-   - Changed the value of several [Settings] options.
-   - Added full type-checking coverage (fully importable from third-party scripts).
-
-To-Do:
-- Add documentation.
-- Finish updating [README.md].
-- Look into places more-itertools can help.
-- Make this an installable [pip] package.
-- Add in-program configuration for non-defaulted [Settings].
-
-
-**Please submit an issue or contact us for features you want to added/reimplemented.**
-
-## Preparation with TumblThree
-
-**Download and Install TumblThree:**
-
-- Visit the official [TumblThree GitHub page](https://github.com/TumblThreeApp/TumblThree) and download the latest version of the application.
-- Extract the downloaded ZIP file and run the `TumblThree.exe` file to launch the application.
-
-**Add Tumblr Blogs:**
-
-- Copy the URLs of the Tumblr blogs you want to download.
-- In TumblThree, enter the blogs into the field marked **Enter URL**, then hit enter.
-- The blogs will be added to the list of blogs in the main interface.
-
-**Configure Download Settings:**
-
-- Click on a blog in the list to view its settings on the right panel.
-- Choose which content types you want to download, including text posts, answers, and more. 
-
-**Start Downloading:**
-
-- Click the **Download** button (represented by a download icon) to begin downloading the selected blogs.
-- The application will download all available posts from the blogs based on your configuration.
-
-
-
-## Setup
-
-#### 1. Move Text Post Files to the Data Folder
-
-- Organize your text post files by moving them to the `./data` directory within the project. Ensure that all your post files are stored here for easy access by the software.
-
-#### 2. Rename Text Post Files
-
-- Rename each text post file following the format:
-
-  ```
-  blogname_typeofpost.txt
-  ```
-
-  For example:
-
-  - If the blog name is "myblog" and the post is a "texts", rename it to `myblog_texts.txt`.
-
-#### 3. Install Python 3.11
-
-- **For Windows:**
-
-  1. Download Python 3.11 from the [official Python website](https://www.python.org/downloads/windows/).
-  2. Run the installer and select the checkbox to "Add Python to PATH" during installation.
-  3. Complete the installation process by following the prompts.
-
-- **For Linux:**
-
-  1. Open the terminal.
-
-  2. Run the following commands:
-
-     ```bash
-     sudo apt update
-     sudo apt install python3.11
-     ```
-
-#### 4. Install Pip
-
-- **For Windows:**
-
-  1. Open Command Prompt.
-
-  2. Run the following command to install pip:
-
-     ```bash
-     python -m ensurepip --upgrade
-     ```
-
-- **For Linux:**
-
-  1. Open the terminal.
-
-  2. Run the following command:
-
-     ```bash
-     sudo apt install python3-pip
-     ```
-
-#### 5. Install Virtualenv
-
-- **For Windows:**
-
-  1. Open Command Prompt.
-
-  2. Run the following command to install `virtualenv`:
-
-     ```bash
-     pip install virtualenv
-     ```
-
-- **For Linux:**
-
-  1. Open the terminal.
-
-  2. Run the following command:
-
-     ```bash
-     sudo pip install virtualenv
-     ```
-
-#### 6. Run the Setup Script
-
-- **For Windows:**
-
-  1. Navigate to the project directory in Command Prompt.
-
-  2. Run the following command:
-
-     ```bash
-     setup.bat
-     ```
-
-- **For Linux:**
-
-  1. Open the terminal and navigate to the project directory.
-
-  2. Run the following command:
-
-     ```bash
-     ./setup.sh
-     ```
-
-#### 7. Run Python Programs within the Virtual Environment
-
-- **For Windows:**
-
-  1. Activate the virtual environment by running:
-
-     ```bash
-     .\venv\Scripts\activate
-     ```
-
-  2. Once activated, run any of the Python programs with:
-
-     ```bash
-     python {program}.py
-     ```
-
-- **For Linux:**
-
-  1. Activate the virtual environment by running:
-
-     ```bash
-     source venv/bin/activate
-     ```
-
-  2. Once activated, run any of the Python programs with:
-
-     ```bash
-     python3 {program}.py
-     ```
-
-
-
-## Testing
-
-1. **Configure the Settings**: 
-   - Open the config file and fill in the required options. 
-   - Leave the `model` field blank for now, as it will be completed later after model training is complete.
-
-2. **Prepare Training Data**: 
-   - Place your correctly formatted post files into the `./data/` directory.
-   - Run the script `create_training_data.py` to generate the necessary training data.
-
-3. **Upload Training Data**: 
-   - Once the training files are generated, locate them in the `./output/` folder.
-   - Upload these files to OpenAI's fine-tuning web portal.
-   - **Ensure you select the most recent version of the `gpt-4o-mini` model** for the fine-tuning process.
-
-4. **Update the Config File**: 
-   - After the fine-tuning process is complete, copy the model identifier from OpenAI’s web portal.
-   - Paste the model identifier into the `model` field of your config file.
-
-5. **Test the Model**: 
-   - Run the script `4tv_tumblrbot.py` to generate posts. These posts will be saved in your Tumblr drafts.
-   - Review the output in your drafts. If needed, repeat the process with different blogs or make adjustments to the training data to improve the model.
-
-
-
-## Deployment
-
-### Linux Instructions
-
-1. **Place Your Python Script and Virtual Environment**
-
-   - Ensure `4tv_tumblrbot.py` and the virtual environment (`.venv/`) are located in `/home/user/4tv-tumblrbot-1.0/`.
-
-2. **Create a Shell Script to Run the Python Script**
-
-   - Create a shell script named `run_4tv_tumblrbot.sh` in the same directory:
-
-   ```bash
-   #!/bin/bash
-   cd /home/user/4tv-tumblrbot-1.0/
-   source .venv/bin/activate
-   python 4tv_tumblrbot.py
-   deactivate
-   ```
-
-3. **Make the Shell Script Executable**
-
-   - Run the following command to make the shell script executable:
-
-   ```bash
-   chmod +x /home/user/4tv-tumblrbot-1.0/run_4tv_tumblrbot.sh
-   ```
-
-4. **Schedule the Script Using Cron**
-
-   - Open the cron table:
-
-   ```bash
-   crontab -e
-   ```
-
-   - Add the following line to schedule the script to run daily at 2 AM:
-
-   ```bash
-   0 2 * * * /home/user/4tv-tumblrbot-1.0/run_4tv_tumblrbot.sh >> /home/user/4tv-tumblrbot-1.0/output.log 2>&1
-   ```
-
-   - This will log the output to `output.log` in the script's directory.
-
-5. **Verify the Cron Job**
-
-   - List your cron jobs to ensure it was added correctly:
-
-   ```bash
-   crontab -l
-   ```
-
-### Windows Instructions
-
-1. **Place Your Python Script and Virtual Environment**
-
-   - Ensure `4tv_tumblrbot.py` and the virtual environment (`.venv/`) are located in `C:\Users\user\4tv-tumblrbot-1.0\`.
-
-2. **Create a Batch File to Run the Python Script**
-
-   - Create a batch file named `run_4tv_tumblrbot.bat` in the same directory:
-
-   ```batch
-   @echo off
-   cd C:\Users\user\4tv-tumblrbot-1.0\
-   call .venv\Scripts\activate
-   python 4tv_tumblrbot.py
-   call .venv\Scripts\deactivate
-   ```
-
-3. **Schedule the Script Using Task Scheduler**
-
-   - Open **Task Scheduler** by searching for it in the Start Menu.
-   - Click **Create Basic Task** and name your task (e.g., “Run 4tv Tumblrbot Daily”).
-   - Set the trigger to "Daily" and select the time (e.g., 2:00 AM).
-   - Choose "Start a program" as the action.
-   - In the **Program/script** field, enter the path to the batch file (`C:\Users\user\4tv-tumblrbot-1.0\run_4tv_tumblrbot.bat`).
-   - Click **Finish** to create the task.
-
-4. **Verify the Task**
-
-   - Check the **Task Scheduler Library** to confirm your task is listed.
-   - Optionally, run the task manually to ensure it works by right-clicking the task and selecting **Run**.
-
-### Additional Notes:
-
-- Make sure Python and the virtual environment are properly set up on both systems.
-- Adjust paths and timings according to your preferences and system configurations.
-- Ensure the script and virtual environment have appropriate permissions and configurations to run correctly.
-

tumblrbot-1.0.0.dist-info/RECORD

@@ -1,17 +0,0 @@
-tumblrbot/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-tumblrbot/__main__.py,sha256=elqi-tHjPr6B5u1PB1n1OqsbZQr_oKwF1D9XVi7XxTs,2519
-tumblrbot/flow/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-tumblrbot/flow/download.py,sha256=L16D2Jk9Yvm5Xc_AMMSscvZCUJdYQHdcr1w-bRz1cQA,2087
-tumblrbot/flow/examples.py,sha256=LxDyByKWJOczjDBmk6lj5-0sCwqGZrd3lPHthMZQgGQ,4196
-tumblrbot/flow/fine_tune.py,sha256=vk2sbqpp-F_j2757R6jnQSUIk5iKAO6FACMY2D8tCy8,4058
-tumblrbot/flow/generate.py,sha256=2CXUJN9_fdfAhdq1ckZDrAJLb13XFI-MgL0YtD8xlvk,2081
-tumblrbot/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-tumblrbot/utils/common.py,sha256=3PPyutv7yyjvij5anyKQRDRn7lrr8Eu05m8XVhh74Hc,1342
-tumblrbot/utils/models.py,sha256=rUi6bCjow0lhUr73vG5gMimz_hNOTrpvvbXjebFIHzg,1976
-tumblrbot/utils/settings.py,sha256=oTTR72WW73pZ2OEJl8JyRlKXtV4VKxVevcMeLx6YHJU,5734
-tumblrbot/utils/tumblr.py,sha256=AbRzQrOV2owahIWn2ekP2E0xjZpl2tFhI_dfP8n5L24,2929
-tumblrbot-1.0.0.dist-info/entry_points.txt,sha256=lTiN7PxAbyGY1fpCWApEw6NUIUgobfcOKhvn6cu3IQA,53
-tumblrbot-1.0.0.dist-info/licenses/UNLICENSE,sha256=8Bl77UGlO95Tuu1FjTzqAPr-UU_A11XBQdPct1-E3qE,1236
-tumblrbot-1.0.0.dist-info/WHEEL,sha256=G2gURzTEtmeR8nrdXUJfNiB3VYVxigPQ-bEQujpNiNs,82
-tumblrbot-1.0.0.dist-info/METADATA,sha256=H0wykWYLA-P2Fv374I8LRISxRppgqdWpUojHOiweKWA,11413
-tumblrbot-1.0.0.dist-info/RECORD,,

tumblrbot-1.0.0.dist-info/licenses/UNLICENSE

@@ -1,24 +0,0 @@
-This is free and unencumbered software released into the public domain.
-
-Anyone is free to copy, modify, publish, use, compile, sell, or
-distribute this software, either in source code form or as a compiled
-binary, for any purpose, commercial or non-commercial, and by any
-means.
-
-In jurisdictions that recognize copyright laws, the author or authors
-of this software dedicate any and all copyright interest in the
-software to the public domain. We make this dedication for the benefit
-of the public at large and to the detriment of our heirs and
-successors. We intend this dedication to be an overt act of
-relinquishment in perpetuity of all present and future rights to this
-software under copyright law.
-
-THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
-EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
-MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
-IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR
-OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
-ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
-OTHER DEALINGS IN THE SOFTWARE.
-
-For more information, please refer to <https://unlicense.org/>