crieur 1.3.1__tar.gz → 1.4.0__tar.gz

{crieur-1.3.1 → crieur-1.4.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: crieur
-Version: 1.3.1
+Version: 1.4.0
 Summary: A Static Revue Generator.
 Project-URL: Homepage, https://gitlab.huma-num.fr/ecrinum/crieur
 Project-URL: Issues, https://gitlab.huma-num.fr/ecrinum/crieur/-/issues
@@ -684,6 +684,7 @@ Classifier: Topic :: Software Development :: Build Tools
 Classifier: Topic :: Text Processing :: Markup :: Markdown
 Requires-Python: >=3.8
 Requires-Dist: dataclass-wizard
+Requires-Dist: feedgen>=1.0.0
 Requires-Dist: httpx
 Requires-Dist: jinja2
 Requires-Dist: minicli

{crieur-1.3.1 → crieur-1.4.0}/crieur/__init__.py

@@ -1,4 +1,4 @@
 from pathlib import Path
 
-VERSION = "1.3.1"
+VERSION = "1.4.0"
 ROOT_DIR = Path(__file__).parent

{crieur-1.3.1 → crieur-1.4.0}/crieur/cli.py

@@ -9,7 +9,7 @@ import httpx
 from minicli import cli, run
 
 from . import VERSION
-from .generator import generate_html
+from .generator import generate_feed, generate_html
 from .models import collect_authors, collect_keywords, configure_numero
 from .utils import each_file_from, each_folder_from
 
@@ -58,6 +58,9 @@ def generate(
         target_path,
         templates_path,
     )
+    generate_feed(
+        title, base_url, numeros, extra_vars, target_path, number=10, lang="fr"
+    )
 
     if not without_statics:
         static_path_local = Path(__file__).parent / "statics"
@@ -101,14 +104,27 @@ def stylo(
                 with httpx.stream("GET", url, timeout=None) as r:
                     for data in r.iter_bytes():
                         fd.write(data)
+        else:
+            print(
+                f"Source already exists: `{zip_path}` (no download). "
+                "Use the `--force` option to download it again"
+            )
 
         target_path = sources_path / f"{i + 1}-{stylo_id}"
         try:
             with zipfile.ZipFile(zip_path, "r") as zip_ref:
                 zip_ref.extractall(target_path)
-            print(f"Data downloaded and extracted to {target_path}")
+            print(f"Data extracted to {target_path}")
         except zipfile.BadZipFile:
+            zip_problematic_path = Path() / f"problematic-export-{i + 1}-{stylo_id}.zip"
+            zip_path.rename(zip_problematic_path)
             print(f"Unable to find corpus with id {stylo_id}!")
+            print(
+                f"Check out the content of {zip_problematic_path} to try to understand."
+            )
+            print(
+                "Either you use a wrong corpus id or there is an issue with the export."
+            )
             return
 
 

{crieur-1.3.1 → crieur-1.4.0}/crieur/generator.py

@@ -1,8 +1,10 @@
 import json
 import locale
 import shutil
+from datetime import datetime, timedelta, timezone
 
 import mistune
+from feedgen.feed import FeedGenerator
 from jinja2 import Environment as Env
 from jinja2 import FileSystemLoader
 from slugify import slugify
@@ -83,3 +85,40 @@ def generate_html(
         author_folder = target_path / "auteur" / author.slug
         author_folder.mkdir(parents=True, exist_ok=True)
         (author_folder / "index.html").write_text(content)
+
+
+def generate_feed(
+    title, base_url, numeros, extra_vars, target_path, number=10, lang="fr"
+):
+    feed = FeedGenerator()
+    feed.id(base_url)
+    feed.title(title)
+    feed.link(href=base_url, rel="alternate")
+    feed.link(href=f"{base_url}feed.xml", rel="self")
+    feed.language(lang)
+
+    articles = sorted(
+        [article for numero in numeros for article in numero.articles], reverse=True
+    )
+
+    for article in articles[:number]:
+        feed_entry = feed.add_entry(order="append")
+        feed_entry.id(f"{base_url}{article.url}")
+        feed_entry.title(article.title_f)
+        feed_entry.link(href=f"{base_url}{article.url}")
+        feed_entry.updated(
+            datetime.combine(
+                article.date,
+                datetime.min.time(),
+                tzinfo=timezone(timedelta(hours=-4), "ET"),
+            )
+        )
+        for author in article.authors:
+            feed_entry.author(name=str(author))
+        feed_entry.summary(summary=md(article.content_md), type="html")
+        if article.keywords:
+            for keyword in article.keywords:
+                feed_entry.category(term=keyword.name)
+
+    feed.atom_file(target_path / "feed.xml", pretty=True)
+    print(f"Generated meta-feed with {number} items.")

{crieur-1.3.1 → crieur-1.4.0}/crieur/models.py

@@ -2,6 +2,7 @@ from dataclasses import dataclass
 from typing import Optional
 
 from dataclass_wizard import DatePattern, DumpMeta, YAMLWizard
+from dataclass_wizard import errors as dw_errors
 from slugify import slugify
 from yaml.composer import ComposerError
 
@@ -29,11 +30,16 @@ class Numero(YAMLWizard):
             )
             article_yaml_path = article_folder / f"{article_slug}.yaml"
             try:
-                loaded_article = Article.from_yaml_file(article_yaml_path)
-            except ComposerError:
-                loaded_article = Article.from_yaml(
-                    article_yaml_path.read_text().split("---")[1]
-                )
+                try:
+                    loaded_article = Article.from_yaml_file(article_yaml_path)
+                except ComposerError:
+                    loaded_article = Article.from_yaml(
+                        article_yaml_path.read_text().split("---")[1]
+                    )
+            except dw_errors.ParseError as e:
+                print(f"Metadata error in `{article['article']['title']}`:")
+                print(e)
+                exit(1)
             if not loaded_article.id:
                 loaded_article.id = article_slug
             loaded_article.content_md = (
@@ -46,7 +52,7 @@ class Numero(YAMLWizard):
             )
             loaded_article.numero = self
             loaded_articles.append(loaded_article)
-        self.articles = loaded_articles
+        self.articles = sorted(loaded_articles, reverse=True)
 
 
 @dataclass
@@ -65,6 +71,18 @@ class Article(YAMLWizard):
     def __post_init__(self):
         self.slug = slugify(self.title)
 
+    def __eq__(self, other):
+        return self.id == other.id
+
+    def __lt__(self, other: "Article"):
+        if not isinstance(other, Article):
+            return NotImplemented
+        return self.date < other.date
+
+    @property
+    def url(self):
+        return f"numero/{self.numero.slug}/article/{self.id}/"
+
 
 def configure_numero(yaml_path):
     # Preserves abstract_fr key (vs. abstract-fr) when converting to_yaml()

{crieur-1.3.1 → crieur-1.4.0}/crieur/templates/base.html

@@ -17,6 +17,8 @@
   <link rel="icon" href="data:;base64,iVBORw0KGgo=" />
   <link rel="stylesheet"
         href="{{ base_url }}statics/pico.css" />
+  <link rel="alternate" type="application/atom+xml"
+        title="Feed" href="{{ base_url }}feed.xml">
   <style>
     h1 {
         text-align: center;

{crieur-1.3.1 → crieur-1.4.0}/crieur/templates/homepage.html

@@ -97,7 +97,7 @@
       </header>
       <ul>
         {% for article in numero.articles %}
-          <li><a href="{{ base_url }}numero/{{ numero.slug }}/article/{{ article.id }}/">{{ article.title_f }}</a></li>
+          <li><a href="{{ base_url }}{{ article.url }}">{{ article.title_f }}</a></li>
         {% endfor %}
       </ul>
     </article>

{crieur-1.3.1 → crieur-1.4.0}/crieur/templates/numero.html

@@ -59,7 +59,7 @@
   {% for article in numero.articles %}
     <article>
       <header>
-        <h3><a href="{{ base_url }}numero/{{ numero.slug }}/article/{{ article.id }}/">{{ article.title_f }}</a></h3>
+        <h3><a href="{{ base_url }}{{ article.url }}">{{ article.title_f }}</a></h3>
       </header>
       {{ article.subtitle_f|markdown }}
       {% if article.authors %}

{crieur-1.3.1 → crieur-1.4.0}/pyproject.toml

@@ -21,6 +21,7 @@ dependencies = [
   "tzdata>=2025.2",
   "pyyaml>=6.0.2",
   "pytz>=2025.2",
+  "feedgen>=1.0.0",
 ]
 classifiers = [
   "Development Status :: 5 - Production/Stable",