diff --git a/.github/workflows/ci.yaml b/.github/workflows/ci.yaml index f03dfd9a..b9e054cf 100644 --- a/.github/workflows/ci.yaml +++ b/.github/workflows/ci.yaml @@ -3,7 +3,7 @@ name: CI on: pull_request: branches: - - master + - main jobs: lint: diff --git a/.github/workflows/docs.yaml b/.github/workflows/docs.yaml index f1b1d285..aa0548f5 100644 --- a/.github/workflows/docs.yaml +++ b/.github/workflows/docs.yaml @@ -1,10 +1,10 @@ name: website -# Build the documentation whenever there are new commits on master +# Build the documentation whenever there are new commits on main on: push: branches: - - master + - main # Security: restrict permissions for CI jobs. permissions: diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md index 35dab6ce..59a7abf2 100644 --- a/CONTRIBUTING.md +++ b/CONTRIBUTING.md @@ -55,7 +55,7 @@ find an issue to work on, you are welcome to open a PR with a fix. ### Commit your update Commit the changes once you are happy with them. See [Atom's contributing -guide](https://github.com/atom/atom/blob/master/CONTRIBUTING.md#git-commit-messages) to +guide](https://github.com/atom/atom/blob/main/CONTRIBUTING.md#git-commit-messages) to know how to use emoji for commit messages. Once your changes are ready, don't forget to diff --git a/README.md b/README.md index da4bc7ba..5bc50610 100644 --- a/README.md +++ b/README.md @@ -1,11 +1,11 @@ # Domsdatabasen -Scraping og processering af [domsdatabasen](https://domsdatabasen.dk/#). +Scraping og processering af sager fra [Domsdatabasen](https://domsdatabasen.dk/#). Hver enkelt dom kan tilgås via https://domsdatabasen.dk/#sag/\, hvor \ er mellem 1 og 3821 (pr. 11-10-2023). -Datasættet ligger i en processeret udgave på [Huggingface](https://huggingface.co/datasets/oliverkinch/domsdatabasen). +Datasættet ligger i en processeret udgave på [Huggingface](https://huggingface.co/datasets/alexandrainst/domsdatabasen). ## Scraping af sager Se `src/scripts/scrape.py`. @@ -18,10 +18,10 @@ Se `src/scripts/finalize.py`. ______________________________________________________________________ [![Documentation](https://img.shields.io/badge/docs-passing-green)](https://alexandrainst.github.io/domsdatabasen/domsdatabasen.html) -[![License](https://img.shields.io/github/license/oliverkinch/domsdatabasen)](https://github.com/alexandrainst/domsdatabasen/blob/master/LICENSE) -[![LastCommit](https://img.shields.io/github/last-commit/oliverkinch/domsdatabasen)](https://github.com/alexandrainst/domsdatabasen/commits/master) -[![Code Coverage](https://img.shields.io/badge/Coverage-100%25-brightgreen.svg)](https://github.com/alexandrainst/domsdatabasen/tree/master/tests) -[![Contributor Covenant](https://img.shields.io/badge/Contributor%20Covenant-2.0-4baaaa.svg)](https://github.com/alexandrainst/domsdatabasen/blob/master/CODE_OF_CONDUCT.md) +[![License](https://img.shields.io/github/license/alexandrainst/domsdatabasen)](https://github.com/alexandrainst/domsdatabasen/blob/main/LICENSE) +[![LastCommit](https://img.shields.io/github/last-commit/alexandrainst/domsdatabasen)](https://github.com/alexandrainst/domsdatabasen/commits/main) +[![Code Coverage](https://img.shields.io/badge/Coverage-100%25-brightgreen.svg)](https://github.com/alexandrainst/domsdatabasen/tree/main/tests) +[![Contributor Covenant](https://img.shields.io/badge/Contributor%20Covenant-2.0-4baaaa.svg)](https://github.com/alexandrainst/domsdatabasen/blob/main/CODE_OF_CONDUCT.md) Developers: diff --git a/pyproject.toml b/pyproject.toml index cf0a4ba2..08f27d6b 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -1,7 +1,7 @@ [tool.poetry] name = "domsdatabasen" description = "Scraper and PDF text processor for domsdatabasen.dk" -version = "0.2.0" +version = "0.1.4" authors = [ "Oliver Kinch ", ] diff --git a/src/domsdatabasen/__init__.py b/src/domsdatabasen/__init__.py index 6a7970e6..8bb4f824 100644 --- a/src/domsdatabasen/__init__.py +++ b/src/domsdatabasen/__init__.py @@ -1 +1,4 @@ """__init__.py file for the domsdatabasen package.""" +from .dataset_builder import DatasetBuilder +from .processor import Processor +from .scraper import Scraper diff --git a/src/scripts/finalize.py b/src/scripts/finalize.py index 7a55397c..62dd6a42 100644 --- a/src/scripts/finalize.py +++ b/src/scripts/finalize.py @@ -11,7 +11,7 @@ import hydra -from domsdatabasen.dataset_builder import DatasetBuilder +from domsdatabasen import DatasetBuilder from omegaconf import DictConfig diff --git a/src/scripts/process.py b/src/scripts/process.py index f9818e68..44726f0b 100644 --- a/src/scripts/process.py +++ b/src/scripts/process.py @@ -17,7 +17,7 @@ import logging import hydra -from domsdatabasen.processor import Processor +from domsdatabasen import Processor from omegaconf import DictConfig logger = logging.getLogger(__name__) diff --git a/src/scripts/scrape.py b/src/scripts/scrape.py index 2134dd3a..8939616c 100644 --- a/src/scripts/scrape.py +++ b/src/scripts/scrape.py @@ -17,7 +17,7 @@ import logging import hydra -from domsdatabasen.scraper import Scraper +from domsdatabasen import Scraper from omegaconf import DictConfig logger = logging.getLogger(__name__)