🚀 Add feature: WordPress conversion

backend/app/config.py

@@ -1,5 +1,6 @@
 import logging
 import os
+
 import dotenv
 from minio import Minio
 
@@ -72,7 +73,7 @@ available_sources = AvailableSourcesResponse(
             display_name="Markdown", name="markdown", format="markdown"
         ),
         AvailableSource(
-            display_name="Wordpress", name="wordpress_xml", format="xml"
+            display_name="Wordpress", name="wordpress", format="xml"
         ),
         AvailableSource(display_name="Ebook", name="ebook_pdf", format="pdf"),
         AvailableSource(

backend/app/convert/convert_wordpress_xml.py

@@ -0,0 +1,66 @@
+import datetime
+import re
+from typing import Dict, Union, List
+
+import markdownify
+import pandas as pd
+import xmltodict
+
+from app.config import logger
+from app.convert.base_converter import BaseConverter
+
+
+class WordPressXmlConverter(BaseConverter):
+    def __init__(self, content: Union[str, bytes]):
+        super().__init__(content)
+        self.wordpress_dict = None
+        self.df = None
+
+    def read_file(self) -> None:
+        self.wordpress_dict = xmltodict.parse(self.content)
+        self.df = pd.DataFrame(self.wordpress_dict['rss']['channel']['item'])
+
+    def add_metadata(self) -> None:
+        self.df['source'] = 'WordPress'
+        self.df['type'] = self.df['wp:post_type'].map(
+            {'post': 'Article WordPress', 'page': 'Page WordPress'})
+
+    def convert_columns(self) -> None:
+        self.df = self.df[
+            (self.df['wp:post_type'].isin(['page', 'post'])) &
+            (self.df['wp:status'] == 'publish')
+            ].copy()
+
+        self.df['date'] = self.df['wp:post_date'].apply(
+            lambda x: str(datetime.datetime.fromisoformat(x).isoformat())
+        )
+
+        self.df['content'] = self.df['content:encoded'].apply(
+            self.wp_to_markdown)
+
+    def rename_columns(self) -> None:
+        self.df.rename(columns={"link": "url"}, inplace=True)
+
+    def clean_data(self) -> None:
+        self.df = self.df[['source', 'type', 'date', 'url', 'content']].copy()
+        self.df['path'] = ""  # Add empty 'path' column
+        self.df.fillna(value="", inplace=True)
+
+    @staticmethod
+    def wp_to_markdown(x):
+        try:
+            md_text = re.sub(r'\n+', ' ', markdownify.markdownify(x,
+                                                                  heading_style='ATX')).strip()
+        except Exception as e:
+            logger.error(
+                f"Error converting WordPress content to Markdown: {str(e)}")
+            md_text = ""
+        return md_text
+
+
+def convert_wordpress_xml(content: Union[str, bytes]) -> List[Dict]:
+    logger.info(f"Starting conversion of WordPress XML content")
+    converter = WordPressXmlConverter(content)
+    result = converter.convert()
+
+    return converter.convert()

backend/app/routers/convert_router.py

@@ -1,7 +1,7 @@
 import datetime
 import json
 import os
-from typing import Any, Dict, Union, List
+from typing import Dict, Union, List
 
 from fastapi import APIRouter, HTTPException
 
@@ -37,6 +37,7 @@ from app.convert.convert_linkedin_comments_csv import (
 )
 from app.convert.convert_linkedin_shares_csv import convert_linkedin_shares_csv
 from app.convert.convert_markdown_txt import convert_markdown_txt
+from app.convert.convert_wordpress_xml import convert_wordpress_xml
 from app.convert.convert_youtube_shorts_video import (
     convert_youtube_shorts_video,
 )
@@ -90,6 +91,9 @@ CONVERTERS = {
     },
     "export": {
         "txt": convert_export_txt
+    },
+    "wordpress": {
+        "xml": convert_wordpress_xml
     }
 }
 

backend/requirements.txt

@@ -4,4 +4,6 @@ pydantic
 pytest
 pandas
 minio
-python-dotenv
+python-dotenv
+xmltodict
+markdownify