use init instead of cli & temporarily disable <multi> blocks supressing

pyproject.toml

@@ -29,7 +29,7 @@ pymysql = "^1.0.3"
 peewee = "^3.16.2"
 
 [tool.poetry.scripts]
-spip2md = "spip2md.cli:main"
+spip2md = "spip2md:main"
 
 [build-system]
 requires = ["poetry-core"]

spip2md/__main__.py

@@ -1,4 +1,4 @@
 # SPIP website to plain Markdown files converter, Copyright (C) 2023 Guilhem Fauré
-from spip2md.cli import main
+from spip2md import main
 
 main()

spip2md/converters.py

@@ -106,13 +106,6 @@ SPIP_TO_MARKDOWN = (
         ),
         "```\n\\1\n\n```",
     ),
-    (  # WARNING Keep only the first language in multi-language blocks
-        compile(
-            r"<multi>\s*(?:\[.{2,4}\])?\s*(.*?)\s*(?:\s*\[.{2,4}\].*)*<\/multi>",
-            S | I,
-        ),
-        r"\1",
-    ),
     (  # WARNING remove every html tag
         compile(r"<\/?.*?>\s*", S | I),
         r"",
@@ -252,6 +245,12 @@ UNKNOWN_ISO = (
     r"∆",
 )
 
+# Multi language block, capture the first
+MULTILINGUAL = compile(
+    r"<multi>\s*(?:\[.{2,4}\])?\s*(.*?)\s*(?:\s*\[.{2,4}\].*)*<\/multi>",
+    S | I,
+)
+
 
 # Apply SPIP to Markdown & ISO to UTF conversions to a text, & eventually clean meta
 def convert(text: Optional[str], clean_meta: bool = False) -> str:

spip2md/spipobjects.py

@@ -84,7 +84,10 @@ class Document(SpipWritable, SpipDocuments):
     def filename(self, date: bool = False) -> str:
         name_type: tuple[str, str] = splitext(basename(str(self.fichier)))
         return (
-            slugify((self.date_publication + "-" if date else "") + name_type[0])
+            slugify(
+                (self.date_publication + "-" if date else "") + name_type[0],
+                max_length=100,
+            )
             + name_type[1]
         )
 
@@ -161,7 +164,7 @@ class SpipObject(SpipWritable):
     def dir_slug(self, include_date: bool = False, end_slash: bool = True) -> str:
         date: str = self.date + "-" if include_date else ""
         slash: str = "/" if end_slash else ""
-        return slugify(date + self.titre) + slash
+        return slugify(date + self.titre, max_length=100) + slash
 
     # Get filename of this object
     def filename(self) -> str: