94 lines
3.3 KiB
Python
94 lines
3.3 KiB
Python
from django.db import models
|
|
from apps.utils.models import BaseModel
|
|
from django.conf import settings
|
|
import os
|
|
# Create your models here.
|
|
|
|
class Paper(BaseModel):
|
|
# ===== 全局唯一标识 =====
|
|
openalex_id = models.TextField(unique=True, verbose_name="OpenAlex ID", null=True, blank=True)
|
|
doi = models.TextField(unique=True, verbose_name="DOI")
|
|
# ===== 基本信息 =====
|
|
type = models.CharField(max_length=20, db_index=True)
|
|
title = models.TextField()
|
|
publication_date = models.DateField(null=True, blank=True)
|
|
publication_year = models.IntegerField(db_index=True)
|
|
# ===== 作者（最小可用集）=====
|
|
first_author = models.TextField(null=True, blank=True)
|
|
first_author_institution = models.TextField(null=True, blank=True)
|
|
# ===== 期刊 =====
|
|
publication_name = models.TextField(null=True, blank=True)
|
|
# ===== OA 元信息 =====
|
|
is_oa = models.BooleanField(default=False, db_index=True)
|
|
oa_url = models.TextField(null=True, blank=True)
|
|
# ===== 状态位（调度核心）=====
|
|
has_abstract = models.BooleanField(default=False, db_index=True)
|
|
has_abstract_xml = models.BooleanField(default=False, db_index=True)
|
|
has_fulltext = models.BooleanField(default=False, db_index=True)
|
|
has_fulltext_xml = models.BooleanField(default=False, db_index=True)
|
|
has_fulltext_pdf = models.BooleanField(default=False, db_index=True)
|
|
fetch_status = models.CharField(
|
|
max_length=20,
|
|
default="meta_only", # meta_only / abstract_ready / fulltext_ready / parsed / failed
|
|
db_index=True
|
|
)
|
|
fail_reason = models.CharField(
|
|
max_length=50,
|
|
null=True,
|
|
blank=True
|
|
)
|
|
|
|
source = models.CharField(
|
|
max_length=20,
|
|
default="openalex",
|
|
verbose_name="元数据来源"
|
|
)
|
|
o_search = models.TextField(default="cement")
|
|
o_keywords = models.TextField(null=True, blank=True)
|
|
|
|
def init_save_dir(self):
|
|
publication_date = self.publication_date
|
|
paper_dir = os.path.join(
|
|
settings.BASE_DIR,
|
|
"media/papers",
|
|
str(publication_date.year),
|
|
str(publication_date.month),
|
|
str(publication_date.day)
|
|
)
|
|
os.makedirs(paper_dir, exist_ok=True)
|
|
return paper_dir
|
|
|
|
def save_file_xml(self, content):
|
|
safe_doi = self.doi.replace("/", "_")
|
|
paper_file = os.path.join(self.init_save_dir(), f"{safe_doi}.xml")
|
|
with open(paper_file, "wb") as f:
|
|
f.write(content.encode("utf-8"))
|
|
|
|
def save_file_pdf(self, content):
|
|
safe_doi = self.doi.replace("/", "_")
|
|
paper_file = os.path.join(self.init_save_dir(), f"{safe_doi}.pdf")
|
|
with open(paper_file, "wb") as f:
|
|
f.write(content)
|
|
|
|
def save_fail_reason(self, reason):
|
|
if self.fail_reason:
|
|
self.fail_reason += f";{reason}"
|
|
else:
|
|
self.fail_reason = reason
|
|
self.save(update_fields=["fail_reason", "update_time"])
|
|
|
|
|
|
|
|
class PaperAbstract(BaseModel):
|
|
paper = models.OneToOneField(
|
|
Paper,
|
|
on_delete=models.CASCADE,
|
|
related_name="abstract"
|
|
)
|
|
|
|
abstract = models.TextField()
|
|
source = models.CharField(
|
|
max_length=20,
|
|
verbose_name="摘要来源" # openalex / elsevier / crossref
|
|
)
|