crieur 1.7.0__tar.gz → 1.9.0__tar.gz

{crieur-1.7.0 → crieur-1.9.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: crieur
-Version: 1.7.0
+Version: 1.9.0
 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
@@ -693,6 +693,7 @@ Requires-Dist: pillow
 Requires-Dist: python-slugify
 Requires-Dist: pytz>=2025.2
 Requires-Dist: pyyaml>=6.0.2
+Requires-Dist: regex>=2024.11.6
 Requires-Dist: tzdata>=2025.2
 Requires-Dist: unidecode
 Provides-Extra: dev

{crieur-1.7.0 → crieur-1.9.0}/crieur/__init__.py

@@ -1,4 +1,4 @@
 from pathlib import Path
 
-VERSION = "1.7.0"
+VERSION = "1.9.0"
 ROOT_DIR = Path(__file__).parent

{crieur-1.7.0 → crieur-1.9.0}/crieur/generator.py

@@ -11,10 +11,12 @@ from jinja2 import FileSystemLoader
 from slugify import slugify
 
 from . import VERSION
+from .typography import typographie
 from .utils import neighborhood
 
 locale.setlocale(locale.LC_ALL, "")
-md = mistune.create_markdown(plugins=["footnotes", "superscript"])
+mistune_plugins = ["footnotes", "superscript", "table"]
+md = mistune.create_markdown(plugins=mistune_plugins, escape=False)
 
 
 def slugify_(value):
@@ -25,6 +27,11 @@ def markdown(value):
     return md(value) if value else ""
 
 
+def typography(value):
+    value = value.replace("\\ ", " ")
+    return typographie(value) if value else ""
+
+
 def generate_html(
     title, base_url, numeros, keywords, authors, extra_vars, target_path, templates_path
 ):
@@ -35,6 +42,7 @@ def generate_html(
     )
     environment.filters["slugify"] = slugify_
     environment.filters["markdown"] = markdown
+    environment.filters["typography"] = typography
 
     extra_vars = json.loads(extra_vars) if extra_vars else {}
 

{crieur-1.7.0 → crieur-1.9.0}/crieur/models.py

@@ -10,8 +10,23 @@ from PIL import Image, UnidentifiedImageError
 from slugify import slugify
 from yaml.composer import ComposerError
 
+from .generator import mistune_plugins
+from .typography import typographie
 
-class ImgsWithSizesRenderer(mistune.HTMLRenderer):
+
+class FrenchTypographyRenderer(mistune.HTMLRenderer):
+    """Apply French typographic rules to text."""
+
+    def text(self, text):
+        text = text.replace("\\ ", " ")
+        return typographie(super().text(text), html=True)
+
+    def block_html(self, html):
+        html = html.replace("\\ ", " ")
+        return typographie(super().block_html(html), html=True)
+
+
+class ImgsWithSizesRenderer(FrenchTypographyRenderer):
     """Renders images as <figure>s and add sizes."""
 
     def __init__(
@@ -21,7 +36,7 @@ class ImgsWithSizesRenderer(mistune.HTMLRenderer):
         base_url=None,
         article=None,
     ):
-        super().__init__()
+        super().__init__(escape, allow_harmful_protocols)
         self._base_url = base_url
         self._article = article
 
@@ -31,7 +46,7 @@ class ImgsWithSizesRenderer(mistune.HTMLRenderer):
             return text
         return super().paragraph(text)
 
-    def image(self, alt, url):
+    def image(self, text, url, title=None):
         if self._article.images_path is None:
             print(f"Image with URL `{url}` is discarded.")
             return ""
@@ -42,7 +57,7 @@ class ImgsWithSizesRenderer(mistune.HTMLRenderer):
             print(f"`{full_path}` is not a valid image.")
             return ""
         width, height = image.size
-        caption = f"<figcaption>{alt}</figcaption>" if alt else ""
+        caption = f"<figcaption>{text}</figcaption>" if text else ""
         full_url = f"{self._base_url}{self._article.url}{url}"
         return dedent(
             f"""\
@@ -54,7 +69,7 @@ class ImgsWithSizesRenderer(mistune.HTMLRenderer):
                         width="{width}" height="{height}"
                         loading="lazy"
                         decoding="async"
-                        alt="{alt}">
+                        alt="{text}">
                 </a>
                 {caption}
             </figure>
@@ -121,7 +136,8 @@ class Numero(YAMLWizard):
                     base_url=base_url,
                     article=loaded_article,
                 ),
-                plugins=["footnotes", "superscript"],
+                plugins=mistune_plugins,
+                escape=False,
             )
             loaded_article.content_html = md(loaded_article.content_md)
             loaded_articles.append(loaded_article)

crieur-1.9.0/crieur/typography.py

@@ -0,0 +1,93 @@
+import html.entities
+import unicodedata
+from dataclasses import dataclass
+
+import regex  # pour le support de "\p{}"
+
+
+@dataclass
+class Caractere:
+    unicode: str
+    html: str
+
+    def __init__(self, name: str):
+        self.unicode = unicodedata.lookup(name)
+        codepoint = ord(self.unicode)
+        html_name = html.entities.codepoint2name.get(codepoint, f"#{codepoint}")
+        self.html = f"&{html_name};"
+
+
+ESPACE_INSECABLE = Caractere(name="NO-BREAK SPACE")
+ESPACE_FINE_INSECABLE = Caractere(name="NARROW NO-BREAK SPACE")
+
+
+def assemble_regexes(*regexes):
+    return "|".join(regexes)
+
+
+def build_regex(avant, apres):
+    # \p{} permet de reconnaître un caractère par sa catégorie Unicode
+    # "Zs" est la catégorie "Separator, space".
+    return (
+        rf"((?P<avant>{avant})"
+        + rf"(\p{{Zs}}|{ESPACE_INSECABLE.html})"
+        + rf"(?P<apres>{apres}))"
+        + r"(?!(.(?!<svg))*<\/svg>)"
+    )
+
+
+RE_ESPACE_FINE_INSECABLE = regex.compile(
+    assemble_regexes(
+        build_regex(r"\w?", r"[;\?!]"),  # Ponctuations doubles.
+        build_regex(
+            r"\d", r"([ghj]|min|sec|images|mm|hab|kg|mg|µg|L|km|°C|GHz)(\b|$)"
+        ),  # Unités.
+        build_regex(r"\d", r"(Mo|Ko|Go|Mb|Kb|Gb)(\b|$)"),  # Tailles de fichiers.
+        build_regex(r"\d", r"%"),  # Pourcentages.
+        build_regex(r"\d", r"€"),  # Symboles monétaires.
+        build_regex(r"\d", r"\d"),  # Séparateurs de milliers.
+    )
+)
+
+
+def insere_espaces_fines_insecables(texte):
+    return RE_ESPACE_FINE_INSECABLE.sub(
+        r"\g<avant>" + ESPACE_FINE_INSECABLE.unicode + r"\g<apres>", texte
+    )
+
+
+RE_ESPACE_INSECABLE = regex.compile(
+    assemble_regexes(
+        build_regex(r"\w?", r":"),  # Deux points.
+        build_regex(r"«", r""),  # Guillemets en chevrons.
+        build_regex(r"", r"»"),  # Guillemets en chevrons.
+        build_regex(
+            rf"\b(\d|{ESPACE_FINE_INSECABLE.html})+", r"(?!\d)\w"
+        ),  # Nombre suivi de lettres.
+        build_regex(r"(M\.|Mme)", r"\w"),  # Titres (Monsieur, Madame).
+    )
+)
+
+
+def insere_espaces_insecables(texte):
+    return RE_ESPACE_INSECABLE.sub(
+        r"\g<avant>" + ESPACE_INSECABLE.unicode + r"\g<apres>", texte
+    )
+
+
+def encode_espaces_insecables_en_html(texte):
+    for caractere in (ESPACE_INSECABLE, ESPACE_FINE_INSECABLE):
+        texte = texte.replace(caractere.unicode, caractere.html)
+    return texte
+
+
+def typographie(texte, html=False):
+    """
+    Utilise les espaces insécables fines ou normales lorsque c’est approprié
+
+    https://fr.wikipedia.org/wiki/Espace_ins%C3%A9cable#En_France
+    """
+    res = insere_espaces_fines_insecables(insere_espaces_insecables(texte))
+    if html:
+        res = encode_espaces_insecables_en_html(res)
+    return res

{crieur-1.7.0 → crieur-1.9.0}/pyproject.toml

@@ -23,6 +23,7 @@ dependencies = [
   "pyyaml>=6.0.2",
   "pytz>=2025.2",
   "feedgen>=1.0.0",
+  "regex>=2024.11.6",
 ]
 classifiers = [
   "Development Status :: 5 - Production/Stable",