[books] Add papers as a data model

vrobbler/apps/books/models.py

@@ -36,6 +36,7 @@ from vrobbler.apps.books.locg import (
     lookup_comic_writer_by_locg_slug,
 )
 from vrobbler.apps.books.sources.google import lookup_book_from_google
+from vrobbler.apps.books.sources.semantic import lookup_paper_from_semantic
 from vrobbler.apps.scrobbles.dataclasses import BookLogData
 
 COMICVINE_API_KEY = getattr(settings, "COMICVINE_API_KEY", "")
@@ -64,22 +65,27 @@ class Author(TimeStampedModel):
     )
     bio = models.TextField(**BNULL)
     wikipedia_url = models.CharField(max_length=255, **BNULL)
-    isni = models.CharField(max_length=255, **BNULL)
-    locg_slug = models.CharField(max_length=255, **BNULL)
     wikidata_id = models.CharField(max_length=255, **BNULL)
+    isni = models.CharField(max_length=255, **BNULL)
     goodreads_id = models.CharField(max_length=255, **BNULL)
-    librarything_id = models.CharField(max_length=255, **BNULL)
     comicvine_data = models.JSONField(**BNULL)
-    amazon_id = models.CharField(max_length=255, **BNULL)
+
+    semantic_id = models.CharField(max_length=50, **BNULL)
 
     def __str__(self):
         return f"{self.name}"
 
-    def fix_metadata(self, data_dict: dict = {}):
-        if not data_dict and self.openlibrary_id:
-            data_dict = lookup_author_from_openlibrary(self.openlibrary_id)
+    def enrich_from_semantic(self, overwrite=False):
+        ...
+
+    def enrich_from_google_books(self, overwrite=False):
+        ...
+
+    def enrich_from_openlibrary(self, overwrite=False):
+        data_dict = lookup_author_from_openlibrary(self.openlibrary_id)
 
         if not data_dict or not data_dict.get("name"):
+            logger.warning("Could not find author on openlibrary")
             return
 
         headshot_url = data_dict.pop("author_headshot_url", "")
@@ -160,26 +166,32 @@ class Book(LongPlayScrobblableMixin):
         if not created and not overwrite:
             return book
 
-        bdict, authors, cover, genres = lookup_book_from_google(
-            title
-        ).as_dict_with_authors_cover_and_genres()
+        book_dict = lookup_book_from_google(title)
 
         if created or overwrite:
             author_list = []
+            authors = book_dict.pop("authors")
+            cover_url = book_dict.pop("cover_url")
+            genres = book_dict.pop("generes")
+
             if authors:
                 for author_str in authors:
                     if author_str:
-                        author_list.append(
-                            Author.objects.get_or_create(name=author_str)[0]
+                        author, a_created = Author.objects.get_or_create(
+                            name=author_str
                         )
+                        author_list.append(author)
+                        if a_created:
+                            # TODO enrich author
+                            ...
 
-            for k, v in bdict.items():
+            for k, v in book_dict.items():
                 setattr(book, k, v)
-                book.save()
+            book.save()
 
-                book.save_image_from_url(cover)
-                book.genre.add(*genres)
-                book.authors.add(*author_list)
+            book.save_image_from_url(cover_url)
+            book.genre.add(*genres)
+            book.authors.add(*author_list)
         return book
 
     def save_image_from_url(self, url: str, force_update: bool = False):
@@ -474,3 +486,68 @@ class Page(TimeStampedModel):
             seconds=self.duration_seconds
         )
         self.save(update_fields=["end_time"])
+
+
+class Paper(LongPlayScrobblableMixin):
+    """Keeps track of Academic Papers"""
+
+    COMPLETION_PERCENT = getattr(settings, "PAPER_COMPLETION_PERCENT", 60)
+    AVG_PAGE_READING_SECONDS = getattr(
+        settings, "AVERAGE_PAGE_READING_SECONDS", 60
+    )
+
+    title = models.CharField(max_length=255)
+    semantic_title = models.CharField(max_length=255, **BNULL)
+    authors = models.ManyToManyField(Author, blank=True)
+    koreader_data_by_hash = models.JSONField(**BNULL)
+    arxiv_id = models.CharField(max_length=50, **BNULL)
+    semantic_id = models.CharField(max_length=50, **BNULL)
+    arxiv_id = models.CharField(max_length=50, **BNULL)
+    corpus_id = models.CharField(max_length=50, **BNULL)
+    doi_id = models.CharField(max_length=50, **BNULL)
+    pages = models.IntegerField(**BNULL)
+    language = models.CharField(max_length=4, **BNULL)
+    first_publish_year = models.IntegerField(**BNULL)
+    publish_date = models.DateField(**BNULL)
+    journal = models.CharField(max_length=255, **BNULL)
+    journal_volume = models.CharField(max_length=50, **BNULL)
+    abstract = models.TextField(**BNULL)
+    tldr = models.CharField(max_length=255, **BNULL)
+    openaccess_pdf_url = models.CharField(max_length=255, **BNULL)
+
+    genre = TaggableManager(through=ObjectWithGenres)
+
+    @classmethod
+    def get_from_semantic(cls, title: str, overwrite: bool = False) -> "Paper":
+        paper, created = cls.objects.get_or_create(title=title)
+        if not created and not overwrite:
+            return paper
+
+        paper_dict = lookup_paper_from_semantic(title)
+
+        if created or overwrite:
+            author_list = []
+            author_dicts = paper_dict.pop("author_dicts")
+            if author_dicts:
+                for author_dict in author_dicts:
+                    if author_dict.get("authorId"):
+                        author, a_created = Author.objects.get_or_create(
+                            semantic_id=author_dict.get("authorId")
+                        )
+                        author_list.append(author)
+                        if a_created:
+                            author.name = author_dict.get("name")
+                            author.save()
+                            # TODO enrich author?
+                            ...
+
+            for k, v in paper_dict.items():
+                setattr(paper, k, v)
+            paper.save()
+
+            if author_list:
+                paper.authors.add(*author_list)
+            genres = paper_dict.pop("genres", [])
+            if genres:
+                paper.genre.add(*genres)
+        return paper