1 files changed, 141 insertions, 0 deletions
diff --git a/src/article_handler.py b/src/article_handler.py
new file mode 100644
index 0000000..2655bcc
--- /dev/null
+++ b/src/article_handler.py
@@ -0,0 +1,141 @@
+import dataclasses
+import datetime
+import os
+import logging
+import pathlib
+import glob
+import typing
+import yaml
+
+
+# Known exceptions, these are raised when generating on program exiting error, and are displayed to the user nicely
+class ArticleHandlerException(Exception):
+    pass
+
+
+class ArticleMetaDataMalformed(ArticleHandlerException):
+    pass
+
+
+class ArticleNoMetaData(ArticleHandlerException):
+    pass
+
+
+@dataclasses.dataclass
+class MetaData:
+    description: str
+    created: datetime.date
+
+    @property
+    def pretty_print(self) -> str:
+        return f'''
+    Description: {self.description}
+    Created: {self.created}'''
+
+
+@dataclasses.dataclass
+class Article:
+    title: str
+    meta_data: MetaData
+    web_dir: tuple
+    source_path: str | os.PathLike
+    source: str
+    folder_path: typing.Union[None, str | os.PathLike] = None
+
+    @property
+    def web_path(self):
+        return f'{"/" if self.web_dir else ""}{"/".join(self.web_dir)}/{self.title}'
+
+    @property
+    def pretty_print(self) -> str:
+        return f'''
+Title:       {self.title}
+Meta data:   {self.meta_data.pretty_print}
+Web dir:     {self.web_dir}
+Web path:    {self.web_path}
+Source path: {self.source_path}
+Folder path: {self.folder_path}
+Source:      {f'{self.source[0:49]}...' if len(self.source) >= 50 else self.source}'''  # Truncate long source text
+
+
+@dataclasses.dataclass
+class Folder(list):
+    articles: typing.List[Article] = dataclasses.field(default_factory=list)
+    sub_folders: typing.Dict[str, typing.Self] = dataclasses.field(default_factory=dict)
+
+
+def get_web_dir(path, title) -> typing.Tuple[str]:
+    dir_structure = path.split(title)[0]
+    # Split into tuple, remove first part, assemble to path again, all to remove the first source dir agnostically 
+    dir_structure_prefix_striped = pathlib.Path(dir_structure).parts[1:]
+    return dir_structure_prefix_striped
+
+
+def parse_article_meta_data(source: str) -> typing.Tuple[str, MetaData]:
+    if source.startswith('---'):
+        meta_data_yml_end_char_index = source.find('---', 3)
+        meta_data_yml = source[3:meta_data_yml_end_char_index]
+        # Strip meta data text from source, before feeding it to the markdown reader
+        source = source[meta_data_yml_end_char_index + 3:]
+        meta_data = yaml.safe_load(meta_data_yml)
+
+        try:
+            meta_data = MetaData(**meta_data)
+        except TypeError as type_error_exception:
+            raise ArticleMetaDataMalformed(str(type_error_exception))
+        return source, meta_data
+    else:
+        raise ArticleNoMetaData(f'No metadata found')
+
+
+def get_article(path: str | os.PathLike) -> Article:
+    logging.debug(path)
+    article_args = {}
+    filename, file_extension = os.path.splitext(path)
+
+    basename = os.path.basename(filename)
+    if basename == 'index':
+        # Article is the folder
+        article_folder_name = os.path.dirname(path)
+        dir_basename = os.path.basename(article_folder_name)
+        article_args['title'] = dir_basename
+    else:
+        # Article one file
+        article_args['title'] = os.path.basename(filename)
+    article_args['web_dir'] = get_web_dir(path, article_args['title'])
+    article_args['source_path'] = path
+
+    with open(path, 'r') as file:
+        source = file.read()
+        article_args['source'], article_args['meta_data'] = parse_article_meta_data(source)
+
+    article = Article(**article_args)
+    return article
+
+
+def discover_folder_structure(article: Article, articles: typing.List[Article]):
+    previous_folder = articles
+    for folder_name in article.web_dir:
+        logging.debug(folder_name)
+
+        if folder_name not in previous_folder.sub_folders:
+            logging.debug('new')
+            current_folder = Folder()
+            previous_folder.sub_folders[folder_name] = current_folder
+        else:
+            logging.debug('reuse')
+            current_folder = previous_folder.sub_folders[folder_name]
+        previous_folder = current_folder
+        logging.debug('')
+
+    previous_folder.articles.append(article)
+
+
+def discover_articles(path):
+    articles_paths = glob.glob(f'{path}/**/*.md', recursive=True)  # Equivalent to ls ./**.md
+    articles = Folder()
+    for article_path in articles_paths:
+        article = get_article(article_path)
+        logging.debug(article.pretty_print)
+        discover_folder_structure(articles=articles, article=article)
+    return articles