From 2291f021479b6b6d64fab0da9cd1c74dff6a59f2 Mon Sep 17 00:00:00 2001
From: Nick Sweeting <github@sweeting.me>
Date: Sat, 16 Nov 2024 02:48:17 -0800
Subject: [PATCH] setup seed model

---
 archivebox/seeds/models.py | 66 +++++++++++++++++++++++++++++++++-----
 1 file changed, 58 insertions(+), 8 deletions(-)

diff --git a/archivebox/seeds/models.py b/archivebox/seeds/models.py
index 7fe49c83..ce96c913 100644
--- a/archivebox/seeds/models.py
+++ b/archivebox/seeds/models.py
@@ -1,11 +1,19 @@
 __package__ = 'archivebox.seeds'
 
+from typing import TYPE_CHECKING
+from pathlib import Path
 
 from django.db import models
+from django.db.models import QuerySet
 from django.conf import settings
+from django.urls import reverse_lazy
 
+from archivebox.config import CONSTANTS
+from abid_utils.models import ABIDModel, ABIDField, AutoDateTimeField, ModelWithHealthStats, get_or_create_system_user_pk
 
-from abid_utils.models import ABIDModel, ABIDField, AutoDateTimeField, ModelWithHealthStats
+if TYPE_CHECKING:
+    from crawls.models import Crawl, CrawlSchedule
+    from core.models import Snapshot
 
 
 class Seed(ABIDModel, ModelWithHealthStats):
@@ -40,15 +48,42 @@ class Seed(ABIDModel, ModelWithHealthStats):
     abid = ABIDField(prefix=abid_prefix)
     
     uri = models.URLField(max_length=2000, blank=False, null=False)                          # unique source location where URLs will be loaded from
+    label = models.CharField(max_length=255, null=False, blank=True, default='', help_text='A human-readable label for this seed')
+    notes = models.TextField(blank=True, null=False, default='', help_text='Any extra notes this seed should have')
     
-    extractor = models.CharField(default='auto', max_length=32)   # suggested extractor to use to load this URL source
-    tags_str = models.CharField(max_length=255, null=False, blank=True, default='')          # tags to attach to any URLs that come from this source
-    config = models.JSONField(default=dict)                                                  # extra config to put in scope when loading URLs from this source
+    extractor = models.CharField(default='auto', max_length=32, help_text='The parser / extractor to use to load URLs from this source (default: auto)')
+    tags_str = models.CharField(max_length=255, null=False, blank=True, default='', help_text='An optional comma-separated list of tags to attach to any URLs that come from this source')
+    config = models.JSONField(default=dict, help_text='An optional JSON object containing extra config to put in scope when loading URLs from this source')
     
     created_at = AutoDateTimeField(default=None, null=False, db_index=True)
     modified_at = models.DateTimeField(auto_now=True)
     created_by = models.ForeignKey(settings.AUTH_USER_MODEL, on_delete=models.CASCADE, default=None, null=False)
 
+
+    crawl_set: models.Manager['Crawl']
+
+    class Meta:
+        verbose_name = 'Seed'
+        verbose_name_plural = 'Seeds'
+        
+        unique_together = (('created_by', 'uri', 'extractor'),)
+
+
+    @classmethod
+    def from_file(cls, source_file: Path, label: str='', parser: str='auto', tag: str='', created_by: int|None=None, config: dict|None=None):
+        source_path = str(source_file.resolve()).replace(str(CONSTANTS.DATA_DIR), '/data')
+        
+        seed, _ = cls.objects.get_or_create(
+            label=label or source_file.name,
+            uri=f'file://{source_path}',
+            created_by_id=getattr(created_by, 'pk', created_by) or get_or_create_system_user_pk(),
+            extractor=parser,
+            tags_str=tag,
+            config=config or {},
+        )
+        seed.save()
+        return seed
+
     @property
     def source_type(self):
         # e.g. http/https://
@@ -58,8 +93,23 @@ class Seed(ABIDModel, ModelWithHealthStats):
         #      etc..
         return self.uri.split('://', 1)[0].lower()
 
-    class Meta:
-        verbose_name = 'Seed'
-        verbose_name_plural = 'Seeds'
+    @property
+    def api_url(self) -> str:
+        # /api/v1/core/seed/{uulid}
+        return reverse_lazy('api-1:get_seed', args=[self.abid])  # + f'?api_key={get_or_create_api_token(request.user)}'
+
+    @property
+    def api_docs_url(self) -> str:
+        return '/api/v1/docs#/Core%20Models/api_v1_core_get_seed'
+
+    @property
+    def scheduled_crawl_set(self) -> QuerySet['CrawlSchedule']:
+        from crawls.models import CrawlSchedule
+        return CrawlSchedule.objects.filter(template__seed_id=self.pk)
+
+    @property
+    def snapshot_set(self) -> QuerySet['Snapshot']:
+        from core.models import Snapshot
         
-        unique_together = (('created_by', 'uri', 'extractor'),)
+        crawl_ids = self.crawl_set.values_list('pk', flat=True)
+        return Snapshot.objects.filter(crawl_id__in=crawl_ids)