[FIX] markdown image, prague specifics

district/jekyll_import.py

@@ -6,7 +6,6 @@ import zipfile
 from datetime import date
 from http.client import InvalidURL
 from io import StringIO
-from shutil import rmtree
 from sys import stdout
 from typing import List
 from urllib.error import HTTPError
@@ -15,6 +14,7 @@ import markdown.serializers
 import yaml
 from django.contrib.messages import ERROR, INFO, SUCCESS, WARNING
 from django.core.files.images import ImageFile
+from django.utils import timezone
 from markdown import Markdown
 from markdown.extensions import Extension
 from markdown.inlinepatterns import InlineProcessor
@@ -59,8 +59,9 @@ class ImgProcessor(InlineProcessor):
         el.attrib["format"] = "left"
         collection = get_collection()
         # TODO FIX REPO_NAME
+        parsed_image_path = get_parsed_file_path(m.group(2))
         image_obj = get_or_create_image(
-            params["path"], m.group(2), collection=collection, repo_name=""
+            params["path"], parsed_image_path, collection=collection, repo_name=""
         )
         if not image_obj:
             message_list.append(
@@ -75,6 +76,17 @@ class ImgProcessor(InlineProcessor):
         return el, m.start(0), m.end(0)
 
 
+def get_parsed_file_path(path: str):
+    if "{{" in path:
+        try:
+            parsed_path = path.split("{{")[1].split("|")[0].split("'")[1]
+        except IndexError:
+            parsed_path = path.split("{{")[1].split("|")[0].split('"')[1]
+        return parsed_path
+    else:
+        return path
+
+
 class ImgExtension(Extension):
     def extendMarkdown(self, md):
         IMG_PATTERN = r"!\[(.*?)\]\((.*?)\)"
@@ -133,11 +145,14 @@ def import_post(path, file_path, parent, title_suffix, dry_run, repo_name):
     article.content = [("text", RichText(html))]
     article.perex = get_perex(md)
 
+    if meta.get("date", None):
         meta_date = meta["date"]
         if isinstance(meta_date, date):
             article.date = meta_date
         else:
             article.date = meta["date"].split()[0]
+    else:
+        article.date = timezone.now()
 
     article.title = meta["title"]
     article.author = meta.get("author", "Česká pirátská strana")
@@ -196,6 +211,15 @@ def get_or_create_image(path, file_path, collection, repo_name) -> Image or None
             image = Image(title=file_path, file=file, collection=collection)
             image.save()
             return image
+        except FileNotFoundError:
+            try:
+                file = ImageFile(
+                    open(os.path.join(path, "assets/img", file_path), "rb"),
+                    name=file_path,
+                )
+                image = Image(title=file_path, file=file, collection=collection)
+                image.save()
+                return image
             except FileNotFoundError:
                 img_name = file_path.split("/")[-1]
                 img_assets_folder = repo_name.split(".")[0]  # TODO make as form field
@@ -239,10 +263,12 @@ def clone_repo(url: str) -> (str, str):
     repo_name = url.split("/")[-1]
     repo_path = os.path.join(path, repo_name)
 
+    os.chdir(path)
     if os.path.exists(repo_path):
-        rmtree(repo_path)
+        os.chdir(repo_path)
+        os.system("git pull --depth 1")
+        return repo_path, repo_name
 
-    os.chdir(path)
     os.system("git clone --depth 1 {}".format(url))
 
     return repo_path, repo_name
@@ -369,6 +395,14 @@ def article_parser(
                 defaults={"is_permanent": True, "redirect_page": article},
             )
 
+            # TODO handle redirects! PRAGUE X CB
+            # Redirect.objects.get_or_create(
+            #     site=site,
+            #     old_path="/%s/%s.html"
+            #              % (articlepath, slug),
+            #     defaults={"is_permanent": True, "redirect_page": article},
+            # )
+
             success_counter += 1
         else:
             msg = "ERROR: Nepodporovaná přípona souboru: %s" % ext