fetch_blogs.py

import yaml
import json
from datetime import datetime
from operator import itemgetter
import re
import os 
from github import Github

FETCH_BLOGS = os.environ['FETCH_BLOGS']
g = Github(FETCH_BLOGS)


#print ('current working dir:', os.getcwd())
#github_api_url = 'https://api.github.com/repos/{}/{}/contents/{}'

repos = [
    ('xrdocs', 'design', '_blogs'), #this is the error repo
    ('xrdocs', 'design', '_tutorials'), 
    ('xrdocs', 'virtual-routing', '_blogs'),
    ('xrdocs', 'virtual-routing', '_tutorials'),
    ('xrdocs', 'ztp', '_blogs'),
    ('xrdocs', 'ztp', '_tutorials'),
    ('xrdocs', 'cisco-service-layer', '_blogs'),
    ('xrdocs', 'cisco-service-layer', '_tutorials'),
    ('xrdocs', 'security', '_blogs'),
    ('xrdocs', 'security', '_tutorials'),
    ('xrdocs', 'asr9k', '_blogs'),
    ('xrdocs', 'asr9k', '_tutorials'),
    ('xrdocs', 'cnbng', '_blogs'),
    ('xrdocs', 'cnbng', '_tutorials'),
    ('xrdocs', '8000', '_blogs'),
    ('xrdocs', '8000', '_tutorials'),
    ('xrdocs', 'packet-fronthaul', '_blogs'),
    ('xrdocs', 'packet-fronthaul', '_tutorials'),
    ('xrdocs', 'multicast', '_blogs'),
    ('xrdocs', 'multicast', '_tutorials'),
    ('xrdocs', 'programmability', '_blogs'),
    ('xrdocs', 'programmability', '_tutorials'),
    ('xrdocs', 'automation', '_blogs'),
    ('xrdocs', 'automation', '_tutorials'),
    ('xrdocs', 'application-hosting', '_blogs'),
    ('xrdocs', 'application-hosting', '_tutorials'),
    ('xrdocs', 'cloud-scale-networking', '_blogs'),
    ('xrdocs', 'cloud-scale-networking', '_tutorials'),
    ('xrdocs', 'device-lifecycle', '_blogs'),
    ('xrdocs', 'device-lifecycle', '_tutorials'),
    ('xrdocs', 'telemetry', '_blogs'),
    ('xrdocs', 'telemetry', '_tutorials'),
    ('xrdocs', 'tdm2ip', '_blogs'),
    ('xrdocs', 'tdm2ip', '_tutorials'),
    ('xrdocs', 'segment-routing', '_blogs'),
    ('xrdocs', 'segment-routing', '_tutorials'),
    ('xrdocs', 'routed-pon', '_blogs'),
    ('xrdocs', 'routed-pon', '_tutorials'),
    ('xrdocs', 'ncs5500', '_blogs'),
    ('xrdocs', 'ncs5500', '_tutorials')
]

def remove_date_from_title(title) :
    date = r'^\d{4}-\d{2}-\d{2}-(.*)$'
    match = re.match(date, title)
    if match:
        return match.group(1)
    else:
       return title

#have to add key here, key is only needed when fetching more than 60 requests an hour, only using for testing 
#authenticates the request from the bearer key for github
#session = requests.Session()
#session.headers.update(headers)

#parses through front yaml regardless of position of attributes
def get_published_info(content, path, user, repo, directory) :
    if content.startswith('---\n') :
        last_line = content.find('\n---\n', 4)
        if last_line != -1:
            first_lines_info = content[4:last_line]
            try:
                first_lines = yaml.safe_load(first_lines_info)
                if isinstance(first_lines, dict) :
                    first_lines['path'] = path
                    description = first_lines.get('excerpt', '')
                    #blank description edge case
                    if description is not None and isinstance(description, str):
                        first_lines['description'] = description.strip()
                    else :
                        first_lines['description'] = ''
                    return first_lines
            except yaml.YAMLError as error_message :
                print(f"Error retrieving first couple of lines: {error_message}")
    return {}

def main() :
    recent_posts = []

    for user, repo_name, directory in repos :
        repo_obj = g.get_repo(f"{user}/{repo_name}")
        contents = repo_obj.get_contents(directory)

        for content_file in contents :
            if content_file.name == '2019-02-02-modernizing-ixp-design.md' : #file that contains YAML error in design/_blogs repo
                continue 

            if content_file.type == 'file' and content_file.name.endswith('.md') :
                commits = list(repo_obj.get_commits(path = content_file.path))

                if commits:
                    last_commit_date = commits[0].commit.committer.date
                    date = last_commit_date.replace(tzinfo=None)
                    file_content = content_file.decoded_content.decode()
                    first_lines = get_published_info(file_content, content_file.name, user, repo_name, directory)
            
                    if first_lines and first_lines.get('published', False) and first_lines.get('position', '') == 'top' and first_lines.get('title') :
                        title = remove_date_from_title(content_file.name[:-3])
                        # Check if permalink exists in front matter, otherwise construct from filename
                        if 'permalink' in first_lines and first_lines['permalink']:
                            # Permalink is relative, prepend repo name
                            repo_url = f"https://xrdocs.io/{repo_name}{first_lines['permalink']}"
                        else:
                            repo_url = f'https://xrdocs.io/{repo_name}/{directory[1:]}/{content_file.name[:-3]}/'
                        description = first_lines.get('excerpt', '')
                        recent_posts.append({
                            'title' : title,
                            'date' : date,
                            'url' : repo_url,
                            'blog-description' : description
                        })

    recent_posts.sort(key = itemgetter('date'), reverse = True)

    recent_posts_data = [{
        'title' : file['title'],
        'date' : file['date'].isoformat(),
        'url' : file['url'],
        'blog-description' : file['blog-description']
    } for file in recent_posts]

    with open('latest-articles.json', 'w') as json_file :
        json.dump(recent_posts_data, json_file, indent = 4)

main()