Retrieve Full-Texts from Sinequa Dev Servers

.envs/.local/.django

@@ -33,9 +33,17 @@ SINEQUA_CONFIGS_REPO_WEBAPP_PR_BRANCH='dummy_branch'
 # Slack Webhook
 # ------------------------------------------------------------------------------
 SLACK_WEBHOOK_URL=''
-LRM_USER=''
-LRM_PASSWORD=''
+
+#Server Credentials
+#--------------------------------------------------------------------------------
+LRM_DEV_USER=''
+LRM_DEV_PASSWORD=''
 XLI_USER=''
 XLI_PASSWORD=''
 LRM_QA_USER=''
 LRM_QA_PASSWORD=''
+
+#Server Tokens
+#--------------------------------------------------------------------------------
+LRM_DEV_TOKEN=''
+XLI_TOKEN=''

config/settings/base.py

@@ -343,7 +343,9 @@
 SLACK_WEBHOOK_URL = env("SLACK_WEBHOOK_URL")
 XLI_USER = env("XLI_USER")
 XLI_PASSWORD = env("XLI_PASSWORD")
-LRM_USER = env("LRM_USER")
-LRM_PASSWORD = env("LRM_PASSWORD")
+LRM_DEV_USER = env("LRM_DEV_USER")
+LRM_DEV_PASSWORD = env("LRM_DEV_PASSWORD")
 LRM_QA_USER = env("LRM_QA_USER")
 LRM_QA_PASSWORD = env("LRM_QA_PASSWORD")
+LRM_DEV_TOKEN = env("LRM_DEV_TOKEN")
+XLI_TOKEN = env("XLI_TOKEN")

sde_collections/admin.py

@@ -6,7 +6,23 @@
 from .models.candidate_url import CandidateURL, ResolvedTitle
 from .models.collection import Collection, WorkflowHistory
 from .models.pattern import DivisionPattern, IncludePattern, TitlePattern
-from .tasks import import_candidate_urls_from_api
+from .tasks import fetch_and_update_full_text, import_candidate_urls_from_api
+
+
+def fetch_and_update_text_for_server(modeladmin, request, queryset, server_name):
+    for collection in queryset:
+        fetch_and_update_full_text.delay(collection.id, server_name)
+    modeladmin.message_user(request, f"Started importing URLs from {server_name.upper()} Server")
+
+
+@admin.action(description="Import candidate URLs from LRM Dev Server with Full Text")
+def fetch_full_text_lrm_dev_action(modeladmin, request, queryset):
+    fetch_and_update_text_for_server(modeladmin, request, queryset, "lrm_dev")
+
+
+@admin.action(description="Import candidate URLs from XLI Server with Full Text")
+def fetch_full_text_lis_action(modeladmin, request, queryset):
+    fetch_and_update_text_for_server(modeladmin, request, queryset, "xli")
 
 
 @admin.action(description="Generate deployment message")
@@ -105,11 +121,10 @@ def import_candidate_urls_from_api_caller(modeladmin, request, queryset, server_
         collection_ids=list(queryset.values_list("id", flat=True)),
         server_name=server_name,
     )
-    collection_names = ", ".join(queryset.values_list("name", flat=True))
     messages.add_message(
         request,
         messages.INFO,
-        f"Started importing URLs from the API for: {collection_names} from {server_name.title()}",
+        f"Started importing URLs from {server_name.upper()} Server",
     )
 
 
@@ -133,19 +148,19 @@ def import_candidate_urls_secret_production(modeladmin, request, queryset):
     import_candidate_urls_from_api_caller(modeladmin, request, queryset, "secret_production")
 
 
-@admin.action(description="Import candidate URLs from Li's Server")
-def import_candidate_urls_lis_server(modeladmin, request, queryset):
-    import_candidate_urls_from_api_caller(modeladmin, request, queryset, "lis_server")
+@admin.action(description="Import candidate URLs from XLI Server")
+def import_candidate_urls_xli_server(modeladmin, request, queryset):
+    import_candidate_urls_from_api_caller(modeladmin, request, queryset, "xli")
 
 
 @admin.action(description="Import candidate URLs from LRM Dev Server")
 def import_candidate_urls_lrm_dev_server(modeladmin, request, queryset):
-    import_candidate_urls_from_api_caller(modeladmin, request, queryset, "lrm_dev_server")
+    import_candidate_urls_from_api_caller(modeladmin, request, queryset, "lrm_dev")
 
 
 @admin.action(description="Import candidate URLs from LRM QA Server")
 def import_candidate_urls_lrm_qa_server(modeladmin, request, queryset):
-    import_candidate_urls_from_api_caller(modeladmin, request, queryset, "lrm_qa_server")
+    import_candidate_urls_from_api_caller(modeladmin, request, queryset, "lrm_qa")
 
 
 class ExportCsvMixin:
@@ -236,9 +251,11 @@ class CollectionAdmin(admin.ModelAdmin, ExportCsvMixin, UpdateConfigMixin):
         import_candidate_urls_production,
         import_candidate_urls_secret_test,
         import_candidate_urls_secret_production,
-        import_candidate_urls_lis_server,
+        import_candidate_urls_xli_server,
         import_candidate_urls_lrm_dev_server,
         import_candidate_urls_lrm_qa_server,
+        fetch_full_text_lrm_dev_action,
+        fetch_full_text_lis_action,
     ]
     ordering = ("cleaning_order",)
 

sde_collections/migrations/0059_candidateurl_scraped_text.py

@@ -0,0 +1,18 @@
+# Generated by Django 4.2.9 on 2024-10-21 23:10
+
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("sde_collections", "0058_candidateurl_division_collection_is_multi_division_and_more"),
+    ]
+
+    operations = [
+        migrations.AddField(
+            model_name="candidateurl",
+            name="scraped_text",
+            field=models.TextField(blank=True, null=True),
+        ),
+    ]

sde_collections/migrations/0060_alter_candidateurl_scraped_text.py

@@ -0,0 +1,24 @@
+# Generated by Django 4.2.9 on 2024-11-07 17:34
+
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("sde_collections", "0059_candidateurl_scraped_text"),
+    ]
+
+    operations = [
+        migrations.AlterField(
+            model_name="candidateurl",
+            name="scraped_text",
+            field=models.TextField(
+                blank=True,
+                default="",
+                help_text="This is the text scraped by Sinequa",
+                null=True,
+                verbose_name="Scraped Text",
+            ),
+        ),
+    ]

sde_collections/models/candidate_url.py

@@ -35,6 +35,13 @@ class CandidateURL(models.Model):
         blank=True,
         help_text="This is the original title scraped by Sinequa",
     )
+    scraped_text = models.TextField(
+        "Scraped Text",
+        default="",
+        null=True,
+        blank=True,
+        help_text="This is the text scraped by Sinequa",
+    )
     generated_title = models.CharField(
         "Generated Title",
         default="",

sde_collections/sinequa_api.py

@@ -1,3 +1,4 @@
+import json
 from typing import Any
 
 import requests
@@ -32,17 +33,17 @@
         "query_name": "query-sde-primary",
         "base_url": "https://sciencediscoveryengine.nasa.gov",
     },
-    "lis_server": {
+    "xli": {
         "app_name": "nasa-sba-smd",
         "query_name": "query-smd-primary",
         "base_url": "http://sde-xli.nasa-impact.net",
     },
-    "lrm_dev_server": {
+    "lrm_dev": {
         "app_name": "sde-init-check",
         "query_name": "query-init-check",
         "base_url": "https://sde-lrm.nasa-impact.net",
     },
-    "lrm_qa_server": {
+    "lrm_qa": {
         "app_name": "sde-init-check",
         "query_name": "query-init-check",
         "base_url": "https://sde-qa.nasa-impact.net",
@@ -53,15 +54,13 @@
 class Api:
     def __init__(self, server_name: str) -> None:
         self.server_name = server_name
-        self.app_name: str = server_configs[server_name]["app_name"]
-        self.query_name: str = server_configs[server_name]["query_name"]
-        self.base_url: str = server_configs[server_name]["base_url"]
-        self.xli_user = settings.XLI_USER
-        self.xli_password = settings.XLI_PASSWORD
-        self.lrm_user = settings.LRM_USER
-        self.lrm_password = settings.LRM_PASSWORD
-        self.lrm_qa_user = settings.LRM_QA_USER
-        self.lrm_qa_password = settings.LRM_QA_PASSWORD
+        config = server_configs[server_name]
+        self.app_name: str = config["app_name"]
+        self.query_name: str = config["query_name"]
+        self.base_url: str = config["base_url"]
+        self.user = getattr(settings, f"{server_name}_USER".upper(), None)
+        self.password = getattr(settings, f"{server_name}_PASSWORD".upper(), None)
+        self.token = getattr(settings, f"{server_name}_TOKEN".upper(), None)
 
     def process_response(self, url: str, payload: dict[str, Any]) -> Any:
         response = requests.post(url, headers={}, json=payload, verify=False)
@@ -74,14 +73,7 @@ def process_response(self, url: str, payload: dict[str, Any]) -> Any:
         return meaningful_response
 
     def query(self, page: int, collection_config_folder: str = "") -> Any:
-        if self.server_name == "lis_server":
-            url = f"{self.base_url}/api/v1/search.query?Password={self.xli_password}&User={self.xli_user}"
-        elif self.server_name == "lrm_dev_server":
-            url = f"{self.base_url}/api/v1/search.query?Password={self.lrm_password}&User={self.lrm_user}"
-        elif self.server_name == "lrm_qa_server":
-            url = f"{self.base_url}/api/v1/search.query?Password={self.lrm_qa_password}&User={self.lrm_qa_user}"
-        else:
-            url = f"{self.base_url}/api/v1/search.query"
+        url = f"{self.base_url}/api/v1/search.query?Password={self.password}&User={self.user}"
         payload = {
             "app": self.app_name,
             "query": {
@@ -94,11 +86,40 @@ def query(self, page: int, collection_config_folder: str = "") -> Any:
         }
 
         if collection_config_folder:
-            if self.server_name in ["lis_server", "lrm_dev_server", "lrm_qa_server"]:
+            if self.server_name in ["xli", "lrm_dev", "lrm_qa"]:
                 payload["query"]["advanced"]["collection"] = f"/scrapers/{collection_config_folder}/"
             else:
                 payload["query"]["advanced"]["collection"] = f"/SDE/{collection_config_folder}/"
 
         response = self.process_response(url, payload)
 
         return response
+
+    def sql_query(self, sql: str) -> Any:
+        """Executes an SQL query on the configured server using token-based authentication."""
+        if not self.token:
+            raise ValueError("You must have a token to use the SQL endpoint")
+
+        url = f"{self.base_url}/api/v1/engine.sql"
+        headers = {"Content-Type": "application/json", "Authorization": f"Bearer {self.token}"}
+        payload = json.dumps(
+            {
+                "method": "engine.sql",
+                "sql": sql,
+                "pretty": True,
+                "log": False,
+                "output": "json",
+                "resolveIndexList": "false",
+                "engines": "default",
+            }
+        )
+        try:
+            response = requests.post(url, headers=headers, data=payload, timeout=10)
+            response.raise_for_status()
+            return response.json()
+        except requests.exceptions.RequestException as e:
+            raise Exception(f"API request failed: {str(e)}")
+
+    def get_full_texts(self, collection_config_folder: str) -> Any:
+        sql = f"SELECT url1, text, title FROM sde_index WHERE collection = '/SDE/{collection_config_folder}/'"
+        return self.sql_query(sql)

sde_collections/tasks.py

@@ -9,6 +9,7 @@
 from django.core.management.commands import loaddata
 
 from config import celery_app
+from sde_collections.models.candidate_url import CandidateURL
 
 from .models.collection import Collection, WorkflowStatusChoices
 from .sinequa_api import Api
@@ -141,3 +142,36 @@ def resolve_title_pattern(title_pattern_id):
     TitlePattern = apps.get_model("sde_collections", "TitlePattern")
     title_pattern = TitlePattern.objects.get(id=title_pattern_id)
     title_pattern.apply()
+
+
+@celery_app.task
+def fetch_and_update_full_text(collection_id, server_name):
+    """
+    Task to fetch and update full text and metadata for all URLs associated with a specified collection
+    from a given server.
+
+    Args:
+        collection_id (int): The identifier for the collection in the database.
+        server_name (str): The name of the server.
+
+    Returns:
+        str: A message indicating the result of the operation, including the number of URLs processed
+             or a message if no records were found.
+    """
+    collection = Collection.objects.get(id=collection_id)
+    api = Api(server_name)
+    full_texts = api.get_full_texts(collection.config_folder)
+
+    records = full_texts.get("Rows", [])
+    if not records:
+        return "No records found in the response."
+
+    for record in records:
+        url, full_text, title = record
+        if not (url and full_text and title):
+            continue
+
+        CandidateURL.objects.update_or_create(
+            url=url, collection=collection, defaults={"scraped_text": full_text, "scraped_title": title}
+        )
+    return f"Successfully processed {len(records)} records and updated the database."