feat: add docling step #16
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
sam-hey
reviewed
Mar 21, 2025
|
I think it would be good to have a test for this new step. @iamdeepank |
tweigel-dev
reviewed
Mar 21, 2025
| }, | ||
| ) | ||
|
|
||
| def run(self, inpt: None) -> list[MarkdownDataContract]: |
There was a problem hiding this comment.
unittests missing, trying docling on pdf pptx excel or website
tweigel-dev
reviewed
Mar 21, 2025
Comment on lines
+29
to
+45
| def get_paths(self) -> list[Path]: | ||
| """Retrieve all Markdown file paths. | ||
|
|
||
| Returns: | ||
| List[Path]: List of valid file paths. | ||
|
|
||
| """ | ||
| path = Path(self.settings.FILE_PATHS) | ||
|
|
||
| if not path.exists() or not path.is_dir(): | ||
| raise FileNotFoundError(f"Invalid path: {path}") | ||
|
|
||
| files = [file for file in path.iterdir() if file.is_file()] | ||
| if not files: | ||
| raise ValueError(f"No valid files found in {path}") | ||
|
|
||
| return files |
There was a problem hiding this comment.
we can not rely on having them local, isn't it? @merren-fx so we need to retrieve/download them first
There was a problem hiding this comment.
make it also configureable what to download through the settings
tweigel-dev
reviewed
Mar 21, 2025
Comment on lines
+55
to
+62
| InputFormat.PDF, | ||
| InputFormat.IMAGE, | ||
| InputFormat.DOCX, | ||
| InputFormat.HTML, | ||
| InputFormat.PPTX, | ||
| InputFormat.ASCIIDOC, | ||
| InputFormat.CSV, | ||
| InputFormat.MD, |
There was a problem hiding this comment.
should be configurable through settings
tweigel-dev
reviewed
Mar 21, 2025
| @@ -43,7 +43,8 @@ dependencies= [ | |||
| "mdformat==0.7.17", | |||
| "spacy==3.7.5", | |||
| "tiktoken==0.7.0", | |||
| "joblib>=1.4.0" | |||
| "joblib>=1.4.0", | |||
| "docling==2.26.0" | |||
There was a problem hiding this comment.
add docling like qdrant as optional. its to big to have it as mandatory dep.
tweigel-dev
reviewed
Mar 21, 2025
Comment on lines
+59
to
+77
| @classmethod | ||
| def from_docling_file( | ||
| cls, contract: DocumentConverter, paths: Path, url_prefix: str = "" | ||
| ) -> Self: | ||
| """ | ||
| Creates a `MarkdownDataContract` instance from a file. | ||
| """ | ||
|
|
||
| md = "\n\n".join(res.document.export_to_markdown() for res in contract) | ||
|
|
||
| def find_first(pattern: _re_pattern, text: str, fallback: str): | ||
| x = pattern.findall(text) | ||
| return x[0] if len(x) >= 1 else fallback | ||
|
|
||
| return MarkdownDataContract( | ||
| md=str(find_first(_RE_BODY, md, md)), | ||
| url=str(find_first(_RE_URL, md, url_prefix + paths.as_posix())), | ||
| keywords=str(find_first(_RE_TOPIC, md, paths.name.split(".")[0])), | ||
| ) |
There was a problem hiding this comment.
opt. depdency should not be part of the defintion of the main DataContract
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
Checklist