mirror of
https://forge.chapril.org/tykayn/orgmode-to-gemini-blog
synced 2025-06-20 09:04:42 +02:00
up regen
This commit is contained in:
Tykayn
tykayn
parent
3bf5856bab
commit
dbbae888f4
15 changed files with 172 additions and 149 deletions
|
|
@ -12,6 +12,7 @@ import argparse
|
|||
import pypandoc
|
||||
from jinja2 import Environment, FileSystemLoader
|
||||
import time
|
||||
import sys
|
||||
|
||||
|
||||
# Démarrer le chronomètre
|
||||
|
|
@ -28,7 +29,7 @@ parser.add_argument('--enable_roam_id_rewrite', type=str, help='Activer ou non l
|
|||
parser.add_argument('--generate_html_pages', type=str, help='Activer ou non la génération des pages html', default=True)
|
||||
parser.add_argument('--generate_linkings_json', type=str, help='Activer ou non la génération du json des liens entre articles', default=True)
|
||||
parser.add_argument('--force_html_regen', action='store_true', help='Forcer la régénération des fichiers HTML même s\'ils existent déjà', default=False)
|
||||
parser.add_argument('--rebuild_articles_info_json', type=str, help='Chemin du fichier JSON des articles', default=True)
|
||||
parser.add_argument('--rebuild_articles_info_json', type=str, help='Reconstruire le fichier de données JSON des articles', default=False)
|
||||
|
||||
args = parser.parse_args()
|
||||
run_gemini = args.run_gemini
|
||||
|
|
@ -70,7 +71,7 @@ if rebuild_articles_info_json:
|
|||
|
||||
files_dict = {}
|
||||
if os.path.exists(json_file):
|
||||
#print(f"Chargement du fichier JSON existant: {json_file}")
|
||||
print(f" ============== Chargement du fichier JSON existant: {json_file}")
|
||||
try:
|
||||
with open(json_file, 'r', encoding='utf-8') as f:
|
||||
files_dict = json.load(f)
|
||||
|
|
@ -79,7 +80,7 @@ if os.path.exists(json_file):
|
|||
print(f"Erreur lors du chargement du fichier JSON: {e}")
|
||||
|
||||
else:
|
||||
print(f"Aucun fichier articles_info.json existant trouvé, reconstruction des informations du blog {args.blog}")
|
||||
print(f"\033[91m ============== Aucun fichier articles_info.json existant trouvé, reconstruction des informations du blog {args.blog}\033[0m")
|
||||
|
||||
|
||||
|
||||
|
|
@ -151,6 +152,7 @@ if generate_linkings_json :
|
|||
basename = get_basename(file_name)
|
||||
date_str, annee, slug = find_year_and_slug_on_filename(basename)
|
||||
slug = slugify_title(title)
|
||||
slug_with_year = f"{annee}/{slug}"
|
||||
tags = extract_tags_from_file(file_path, global_config['excluded_tags'])
|
||||
|
||||
# Convertir les tags en liste si c'est un set
|
||||
|
|
@ -206,10 +208,10 @@ if generate_linkings_json :
|
|||
|
||||
rebuild_this_article_html = file_modified_time > last_html_build_time
|
||||
if rebuild_this_article_html:
|
||||
print(f"--------- article modifié après le build de son rendu html: {file_path}, {rebuild_this_article_html}")
|
||||
print(f"\033[91m--------- article modifié après le build de son rendu html: {file_path}, {rebuild_this_article_html}\033[0m")
|
||||
else:
|
||||
print(f"--------- article non modifié après le build de son rendu html: {file_path}, {rebuild_this_article_html}, on ne recrée pas")
|
||||
rebuild_this_article_html = False
|
||||
print(f"\033[91m--------- article non modifié après le build de son rendu html: {file_path}, {rebuild_this_article_html}, on ne recrée pas\033[0m")
|
||||
|
||||
else:
|
||||
# si il n'y a pas de fichier html, on le construit pour la première fois
|
||||
print('on reconstruit le html de l\'article', file_name)
|
||||
|
|
@ -219,17 +221,25 @@ if generate_linkings_json :
|
|||
rebuild_counter += 1
|
||||
|
||||
|
||||
print(f"slug_with_year: {slug_with_year}")
|
||||
# Afficher les clés de files_dict pour débogage
|
||||
# print("\nClés disponibles dans files_dict:")
|
||||
# for key in files_dict.keys():
|
||||
# print(f"- {key}")
|
||||
# print("\n")
|
||||
# Garder le contenu HTML existant si déjà présent
|
||||
if f"{annee}/{slug}" in files_dict and 'html_content' in files_dict[f"{annee}/{slug}"]:
|
||||
#print('on reprend le contenu html existant')
|
||||
if len(files_dict[f"{annee}/{slug}"]['html_content']) > 0 :
|
||||
html_content = files_dict[f"{annee}/{slug}"]['html_content']
|
||||
if len(files_dict[f"{annee}/{slug}"]['html_content_without_h1']) > 0 :
|
||||
html_content_without_h1 = files_dict[f"{annee}/{slug}"]['html_content_without_h1']
|
||||
else:
|
||||
if not rebuild_this_article_html and slug_with_year in files_dict and 'html_content' in files_dict[slug_with_year]:
|
||||
print('========= on reprend le contenu html existant')
|
||||
len_html = len(files_dict[slug_with_year]['html_content'])
|
||||
print(f"len_html: {len_html}")
|
||||
if len_html > 0 :
|
||||
html_content = files_dict[slug_with_year]['html_content']
|
||||
html_content_without_h1 = re.sub(r'<h1>.*?</h1>', '', html_content)
|
||||
if len(files_dict[slug_with_year]['html_content_without_h1']) > 0 :
|
||||
html_content_without_h1 = files_dict[slug_with_year]['html_content_without_h1']
|
||||
else:
|
||||
print(f"\033[91m {time.strftime('%H:%M:%S')} BRRRRRRRRRRRRR pandoc : {title} en html\033[0m")
|
||||
print('========= pas de contenu html existant')
|
||||
print(f"\033[91m {time.strftime('%H:%M:%S')} BRRRRRRRRRRRRR pandoc html_content : {title} en html\033[0m")
|
||||
pandoc_runs_counter += 1
|
||||
html_content = pypandoc.convert_text(content_without_h1, 'html', format='org')
|
||||
html_content_without_h1 = re.sub(r'<h1>.*?</h1>', '', html_content)
|
||||
|
|
@ -237,6 +247,9 @@ if generate_linkings_json :
|
|||
|
||||
|
||||
if run_pandoc and rebuild_this_article_html or force_html_regen:
|
||||
print(f"run_pandoc: {run_pandoc}")
|
||||
print(f"rebuild_this_article_html: {rebuild_this_article_html}")
|
||||
print(f"force_html_regen: {force_html_regen}")
|
||||
# convertir le contenu d'article org vers html
|
||||
print(f"\033[91m {time.strftime('%H:%M:%S')} BRRRRRRRRRRRRR pandoc : {title} en html\033[0m")
|
||||
# print(f"\033[91m.\033[0m", end='', flush=True)
|
||||
|
|
@ -244,26 +257,18 @@ if generate_linkings_json :
|
|||
html_content = pypandoc.convert_text(content_without_h1, 'html', format='org')
|
||||
html_content_without_h1 = re.sub(r'<h1>.*?</h1>', '', html_content)
|
||||
pandoc_runs_counter += 1
|
||||
# else:
|
||||
# html_content = content_without_h1
|
||||
|
||||
# if run_gemini and rebuild_this_article_gemini:
|
||||
# #print('-----------on régénère le gemini')
|
||||
# # convertir le contenu d'article org vers gmi pour la capsule gemini
|
||||
# gemini_content = org_to_gmi(content_without_h1)
|
||||
# #print('len(gemini_content)', len(gemini_content))
|
||||
# else:
|
||||
# print('-----------on ne régénère pas le gemini')
|
||||
if rebuild_this_article_gemini:
|
||||
print(f"\033[91m {time.strftime('%H:%M:%S')} BRRRRRRRRRRRRR gemini : {title} en gmi\033[0m")
|
||||
pandoc_runs_counter += 1
|
||||
gemini_content = org_to_gmi(content_without_h1)
|
||||
|
||||
files_dict[f"{annee}/{slug}"] = {
|
||||
files_dict[slug_with_year] = {
|
||||
'path': file_path,
|
||||
'basename': basename,
|
||||
'roam_id': find_org_roam_id(content),
|
||||
'slug': f"{slug}/",
|
||||
'slug_with_year': f"{annee}/{slug}",
|
||||
'slug_with_year': slug_with_year,
|
||||
'date': boom[0],
|
||||
'lang': lang_folder,
|
||||
'article_type': article_type,
|
||||
|
|
@ -336,7 +341,11 @@ os.makedirs(destination_json, exist_ok=True)
|
|||
|
||||
# sauver le json de tous les articles et pages
|
||||
if pandoc_runs_counter > 0 or not os.path.exists(json_file) or rebuild_articles_info_json:
|
||||
print(f"\033[91m Les articles ont changé, Génération du json {json_file} \033[0m")
|
||||
print(f"pandoc_runs_counter: {pandoc_runs_counter}")
|
||||
print(f"rebuild_articles_info_json: {rebuild_articles_info_json}")
|
||||
|
||||
print(f"\033[94m Génération du json {json_file} \033[0m")
|
||||
|
||||
with open( json_file, 'w', encoding='utf-8') as json_file:
|
||||
files_dict_serialized = json.dumps(files_dict, ensure_ascii=False, indent=4)
|
||||
json_file.write(files_dict_serialized)
|
||||
|
|
@ -445,11 +454,11 @@ Pages:
|
|||
|
||||
|
||||
# Générer la page d'index seulement si des articles ont été convertis
|
||||
if pandoc_runs_counter > 0 or run_gemini:
|
||||
if pandoc_runs_counter > 0 or run_gemini or force_html_regen:
|
||||
# Appel de la fonction pour générer la page d'index
|
||||
generate_blog_index(destination_json + '/articles_info.json', 'templates/html/index.html.jinja', destination_html + 'index.html')
|
||||
|
||||
print(f"\033[91m index régénéré {destination_html}index.html \033[0m")
|
||||
print(f"\033[94m index régénéré {destination_html}index.html \033[0m")
|
||||
# else:
|
||||
# #print("Aucun article n'a été converti, la page d'index n'est pas régénérée")
|
||||
|
||||
|
|
@ -464,57 +473,64 @@ def generate_article_pages(json_file, template_file, output_dir):
|
|||
"""
|
||||
counter_gemini = 0
|
||||
print(f"generate_article_pages: ouverture du json {json_file}")
|
||||
# Charger les données JSON
|
||||
with open(json_file, 'r', encoding='utf-8') as f:
|
||||
print('----------------------- yay json chargé')
|
||||
articles_info = json.load(f)
|
||||
# Configurer Jinja2
|
||||
env = Environment(loader=FileSystemLoader('.'))
|
||||
template = env.get_template(template_file)
|
||||
template_content = get_blog_template_conf(args.blog)
|
||||
|
||||
print(f"articles count: {len(articles_info.values())}")
|
||||
# Générer les pages pour chaque article
|
||||
for article in articles_info.values():
|
||||
print('----------------------- article', article['title'])
|
||||
|
||||
if article['first_picture_url']:
|
||||
template_content['OG_IMAGE'] = article['first_picture_url']
|
||||
else:
|
||||
template_content['OG_IMAGE'] = template_content['SITE_ICON']
|
||||
|
||||
output_html = template.render(
|
||||
template_content=template_content,
|
||||
article=article,
|
||||
all_articles=articles_info
|
||||
)
|
||||
slug_to_use = article['slug_with_year']
|
||||
|
||||
# Déterminer le slug à utiliser selon le type d'article
|
||||
if 'article_type' in article and article['article_type'] == 'article':
|
||||
slug_to_use = article['slug_with_year']
|
||||
else:
|
||||
slug_to_use = article['slug']
|
||||
# Construire le chemin de sortie html en fonction du slug avec l'année
|
||||
output_subdir = os.path.join(output_dir, slug_to_use)
|
||||
print(f"output_subdir: {output_subdir}")
|
||||
os.makedirs(output_subdir, exist_ok=True)
|
||||
output_file = os.path.join(output_subdir ,"index.html")
|
||||
|
||||
print(f"output_file: {output_file}")
|
||||
# Écrire le fichier de sortie en HTML pour un article
|
||||
with open(output_file, 'w', encoding='utf-8') as f:
|
||||
f.write(output_html)
|
||||
try:
|
||||
with open(json_file, 'r', encoding='utf-8') as f:
|
||||
articles_info = json.load(f)
|
||||
# Configurer Jinja2
|
||||
env = Environment(loader=FileSystemLoader('.'))
|
||||
template = env.get_template(template_file)
|
||||
template_content = get_blog_template_conf(args.blog)
|
||||
|
||||
print(f"Génération de la page gemini pour {article['title']}")
|
||||
if 'gemini_content' in article and len(article['gemini_content']) > 0:
|
||||
# Construire le chemin de sortie gmi en fonction du slug avec l'année
|
||||
save_gemini_file(args.blog, article, articles_info, template_content)
|
||||
counter_gemini += 1
|
||||
else:
|
||||
print(f"----------- on ne génère pas le gemini pour {article['slug']}")
|
||||
print(f"articles count: {len(articles_info.values())}")
|
||||
# Générer les pages pour chaque article
|
||||
for article in articles_info.values():
|
||||
print('----------------------- article', article['title'])
|
||||
|
||||
if article['first_picture_url']:
|
||||
template_content['OG_IMAGE'] = article['first_picture_url']
|
||||
else:
|
||||
template_content['OG_IMAGE'] = template_content['SITE_ICON']
|
||||
|
||||
output_html = template.render(
|
||||
template_content=template_content,
|
||||
article=article,
|
||||
all_articles=articles_info
|
||||
)
|
||||
slug_to_use = article['slug_with_year']
|
||||
|
||||
# Déterminer le slug à utiliser selon le type d'article
|
||||
if 'article_type' in article and article['article_type'] == 'article':
|
||||
slug_to_use = article['slug_with_year']
|
||||
else:
|
||||
slug_to_use = article['slug']
|
||||
# Construire le chemin de sortie html en fonction du slug avec l'année
|
||||
output_subdir = os.path.join(output_dir, slug_to_use)
|
||||
print(f"output_subdir: {output_subdir}")
|
||||
os.makedirs(output_subdir, exist_ok=True)
|
||||
output_file = os.path.join(output_subdir ,"index.html")
|
||||
|
||||
print(f"output_file: {output_file}")
|
||||
# Écrire le fichier de sortie en HTML pour un article
|
||||
with open(output_file, 'w', encoding='utf-8') as f:
|
||||
f.write(output_html)
|
||||
|
||||
print(f"Génération de la page gemini pour {article['title']}")
|
||||
if 'gemini_content' in article and len(article['gemini_content']) > 0:
|
||||
# Construire le chemin de sortie gmi en fonction du slug avec l'année
|
||||
save_gemini_file(args.blog, article, articles_info, template_content)
|
||||
counter_gemini += 1
|
||||
else:
|
||||
print(f"----------- on ne génère pas le gemini pour {article['slug']}")
|
||||
|
||||
return
|
||||
print(f"\033[94m Nombre d'articles gemini générés : {counter_gemini}\033[0m")
|
||||
return
|
||||
except IOError as e:
|
||||
print(f"Erreur lors de la lecture du fichier {json_file}: {e}")
|
||||
sys.exit(1)
|
||||
except Exception as e:
|
||||
print(f"Erreur inattendue lors de la lecture du fichier {json_file}: {e}")
|
||||
sys.exit(1)
|
||||
|
||||
|
||||
|
||||
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue