tonl parse pretty good

docker/main/requirements.txt

@@ -8,3 +8,4 @@ colorlog>=6.7.0
 favicon>=0.7.0
 markdown>=3.4.4
 Pygments>=2.16.1
+toml>=0.10

gawa/blog/data/articles/de-test.md

@@ -1,16 +1,3 @@
----
-Title:      Bash Arrays
-Subtitle:   sub
-Desc:       Brief intro to Bash Arrays
-Date:       2023-09-29
-Keywords:   bash
-            technology
-            test
-Category:   Test
-Featured:   True
-Public:     True
----
-
 **NOTE**
 
 This is a stolen article from [opensource.com](https://opensource.com/article/18/5/you-dont-know-bash-intro-bash-arrays)

gawa/blog/data/articles/en-empty.md

@@ -0,0 +1,8 @@
+---
+Title:      empty
+Subtitle:   empty
+Desc:       empty
+Date:       2023-09-29
+Featured:   True
+Public:     False
+---

gawa/blog/data/articles/en-test.md

@@ -1,16 +1,3 @@
----
-Title:      Bash Arrays
-Subtitle:   sub
-Desc:       Brief intro to Bash Arrays
-Date:       2023-09-29
-Thumbnail:  img/thumbnails/bash.png
-Keywords:   bash
-            technology
-Category:   Test
-Featured:   True
-Public:     True
----
-
 **NOTE**
 
 This is a stolen article from [opensource.com](https://opensource.com/article/18/5/you-dont-know-bash-intro-bash-arrays)

gawa/blog/data/articles/test.toml

@@ -0,0 +1,24 @@
+date = "2023-09-29"
+keywords = ["test", "bash"]
+category = "Test"
+featured = true
+public = true
+thumbnail = "img/thumbnails/bash.png"
+
+[lang.en]
+title = "title"
+subtitle = "subtitle"
+desc = """
+long
+multiline
+desc
+"""
+
+[lang.de]
+title = "Titel"
+subtitle = "Subtitel"
+desc = """
+Lange,
+mehrzeilige,
+Beschreibung
+"""

gawa/blog/models.py

@@ -1,3 +1,4 @@
+import toml
 import ast
 import re
 import os
@@ -45,6 +46,14 @@ class Category(models.Model):
     def __str__(self):
         return f"{{<{self.__class__.__name__}>\"{self.name}\"}}"
 
+    @staticmethod
+    def get_or_create_uncategorized():
+        try:
+            return Category.objects.get(slug="uncategorized")
+        except Category.DoesNotExist:
+            return Category.objects.create(
+                slug="uncategorized", name="uncategorized")
+
 
 class BlogPost(Searchable):
     """
@@ -53,7 +62,8 @@ class BlogPost(Searchable):
     body_en = models.TextField(blank=True, default="")
     body_de = models.TextField(blank=True, default="")
     category = models.ForeignKey(
-        Category, on_delete=models.SET_NULL, null=True)
+        Category, on_delete=models.SET_DEFAULT, null=False,
+        default=Category.get_or_create_uncategorized)
     thumbnail = models.ImageField(
         blank=True,
         upload_to="img/thumbnails",
@@ -67,12 +77,14 @@ class BlogPost(Searchable):
 
     DATA_DIR = "/app/blog/data/articles"
     DEFAULT_LANGS = {'en': False, 'de': False}
-
+    META_TOP_KEYS = [
-    def has_keywords(self) -> bool:
+        "date",
-        """
+        "keywords",
-        check if the post has keywords
+        "category",
-        """
+        "featured",
-        return self.keywords.first() is not None
+        "public",
+        "lang"]
+    META_LANG_KEYS = ["title", "subtitle", "desc"]
 
     def regenerate(self):
         """
@@ -85,130 +97,80 @@ class BlogPost(Searchable):
         self.suburl = f"/blog/{self.category.name}/{self.slug}"
 
         # load from markdown
-        self.sync_file()
+        # self.sync_file()
 
+        # redundand vvvv
         self.save()
 
     def sync_file(self):
         """
         generate an article fromm it's original markdown file
         """
-        logger.info(f"regenerating article from markdown for: {self}")
+        logger = logging.getLogger(__name__)
+        logger.info(f"syncing article to markdown for: {self}")
+        # read metadata
         try:
+            fmeta = open(f"{self.DATA_DIR}/{self.slug}.toml", "r")
+        except Exception as e:
+            logger.error(f"could not find meta file for '{self}'")
+            return
+        data = toml.load(fmeta)
+        for key in self.META_TOP_KEYS:
+            if key not in data:
+                logger.error(f"Key '{key}' missing in meta file for '{self}'")
+                raise ValueError(
+                    f"Key '{key}' missing in meta file for '{self}'")
+        for lang in data['lang']:
+            for key in self.META_LANG_KEYS:
+                if key not in data['lang'][lang]:
+                    logger.error(
+                        f"Key '{key}' ('{lang}') missing in meta file for '{self}'")
+                    raise ValueError(
+                        f"Key '{key}' ('{lang}') missing in meta file for '{self}'")
+            with open(f"{self.DATA_DIR}/{lang}-{self.slug}.md") as f_en:
                 MD.reset()
-            with open(f"{self.DATA_DIR}/en-{self.slug}.md") as f_en:
+                body: str = f_en.read()
-
+                # FIXME:
-                body_en: str = f_en.read()
+                # these fields are loaded but wrong?
-
+                # de is in en, de is empty
-                html_en: str = MD.convert(body_en)
+                match lang:
-                try:
+                    case "en":
-                    # NOTE: MD.Meta is generated after MD.convert() by the meta
+                        self.title_en = data['lang'][lang]["title"]
-                    # extension.
+                        self.subtitle_en = data['lang'][lang]["subtitle"]
-                    # Meta not being shown is a reported issue:
+                        self.desc_en = data['lang'][lang]["desc"]
-                    # https://github.com/Python-Markdown/markdown/issues/1383
+                        self.body_en = MD.convert(body)
-                    if not hasattr(MD, 'Meta'):
+                    case "de":
-                        logger.error("Metadata extension for markdown\
+                        self.title_en = data['lang'][lang]["title"]
-                        not loaded")
+                        self.subtitle_en = data['lang'][lang]["subtitle"]
-                        raise ValueError("Metadata extension for markdown\
+                        self.desc_en = data['lang'][lang]["desc"]
-                        not loaded")
+                        self.body_de = MD.convert(body)
-                    meta_en = MD.Meta
+                    case _:
-                    self.title_en = meta_en["title"][0]
+                        logger.error(
-                    self.subtitle_en = meta_en["subtitle"][0]
+                            f"unknown language '{lang}' in meta file for '{self}'")
-                    self.desc_en = meta_en["desc"][0]
+        self.date = data["date"]
-                    self.date = meta_en["date"][0]
+        self.featured = data["featured"]
-                    self.featured = meta_en["featured"][0] == "True"
+        self.public = data["public"]
-                    self.public = meta_en["public"][0] == "True"
+        # NOTE: thumbnail is optional
-                    if "thumbnail" in meta_en:
+        if "thumbnail" in data:
-                        self.thumbnail = meta_en["thumbnail"][0]
+            self.thumbnail = data["thumbnail"]
-
+        # NOTE: thumbnail is optional
+        if "category" in data:
             try:
                 category: Category = Category.objects.get(
-                            slug=meta_en['category'][0])
+                    slug=data['category'])
             except Category.DoesNotExist:
                 category = Category.objects.create(
-                            name=meta_en['category'], slug=meta_en['category'])
+                    name=data['category'], slug=data['category'])
             self.category = category
-
-                    # NOTE: we need to save before we can use the manytomany
-                    # logic.
         self.save()
-                    for item in meta_en["keywords"]:
+        logger.debug("keywords next")
+        for keyword in data["keywords"]:
             try:
-                            self.keywords.add(Keyword.objects.get(slug=item))
+                self.keywords.add(Keyword.objects.get(slug=keyword))
             except Keyword.DoesNotExist:
                 self.keywords.create(
-                                slug=item, text_en=item, text_de=item)
+                    slug=keyword, text_en=keyword, text_de=keyword)
         logger.debug(f"keywords of '{self}': {self.keywords}")
-
-                except Exception as e:
-                    logger.warning(
-                        f"could not generate metadata {self.slug} from markdown: {e}")
-
-                self.body_en = ""
-                self.body_en = html_en
-        except FileNotFoundError as e:
-            # TODO: mark as untranslated
-            pass
-        except Exception as e:
-            logger.warning(
-                f"could not generate article {self.slug} from markdown: {e}")
-        try:
-            MD.reset()
-            with open(f"{self.DATA_DIR}/de-{self.slug}.md") as f_de:
-
-                body_de: str = f_de.read()
-
-                html_de: str = MD.convert(body_de)
-                try:
-                    # NOTE: MD.Meta is generated after MD.convert() by the meta
-                    # extension.
-                    # Meta not being shown is a reported issue:
-                    # https://github.com/Python-Markdown/markdown/issues/1383
-                    if not hasattr(MD, 'Meta'):
-                        logger.error("Metadata extension for markdown\
-                        not loaded")
-                        raise ValueError("Metadata extension for markdown\
-                        not loaded")
-                    meta_de = MD.Meta
-                    self.title_de = meta_de["title"][0]
-                    self.subtitle_de = meta_de["subtitle"][0]
-                    self.desc_de = meta_de["desc"][0]
-                    # TODO: parse date from markdown
-                    self.featured = meta_de["featured"][0] == "True"
-                    self.public = meta_de["public"][0] == "True"
-                    if "thumbnail" in meta_de:
-                        self.thumbnail = meta_de["thumbnail"][0]
-
-                    try:
-                        category: Category = Category.objects.get(
-                            slug=meta_de['category'][0])
-                    except Category.DoesNotExist:
-                        category = Category.objects.create(
-                            name=meta_de['category'], slug=meta_de['category'])
-                    self.category = category
-
-                    # NOTE: we need to save before we can use the manytomany
-                    # logic.
-                    self.save()
-                    for item in meta_de["keywords"]:
-                        try:
-                            self.keywords.add(Keyword.objects.get(slug=item))
-                        except Keyword.DoesNotExist:
-                            self.keywords.create(
-                                slug=item, text_en=item, text_de=item)
-                    logger.debug(f"keywords of '{self}': {self.keywords}")
-                except Exception as e:
-                    logger.warning(
-                        f"could not generate metadata {self.slug} from markdown: {e}")
-
-                self.body_de = ""
-                self.body_de = html_de
-        except FileNotFoundError as e:
-            # TODO: mark as untranslated
-            pass
-        except Exception as e:
-            logger.warning(
-                f"could not generate article {self.slug} from markdown: {e}")
         self.save()
 
     def get_langs(self) -> dict[str, bool]:

gawa/blog/templates/blog/featured.html

@@ -34,7 +34,6 @@
                             <li class="list-group-item">
                                 {% translate "category" %}: <b>{{ post.category.name }}</b>
                             </li>
-                           {% if post.has_keywords %}
                             {% for keyword in post.keywords.all %}
                                 {% if LANGUAGE_CODE == "de" %}
                                     <li class="list-group-item">{{ keyword.text_de }}</li>
@@ -44,7 +43,6 @@
                                     <li class="list-group-item">{{ keyword.text_en }}</li>
                                 {% endif %}
                             {% endfor %}
-                       {% endif %}
                         </ul>
                     </div>
                 </a>