Merge pull request #863 from UN-OCHA/feature/RW-1035

feat: Index entities

PATCHES/remove-tcpdf-footer.patch

@@ -0,0 +1,39 @@
+diff --git a/src/MyPdf.php b/src/MyPdf.php
+index 082131a..bb3cb96 100644
+--- a/src/MyPdf.php
++++ b/src/MyPdf.php
+@@ -21,6 +21,7 @@ class MyPdf extends TCPDF
+     protected $_transf      = array();
+     protected $_myLastPageGroup = null;
+     protected $_myLastPageGroupNb = 0;
++    protected $tcpdflink = false;
+
+     // used to make a radius with bezier : (4/3 * (sqrt(2) - 1))
+     const MY_ARC = 0.5522847498;
+@@ -267,7 +268,7 @@ class MyPdf extends TCPDF
+         $cornerBL = null,
+         $cornerBR = null
+     ) {
+-    
++
+         // init the path
+         $path = '';
+
+@@ -1087,7 +1088,7 @@ class MyPdf extends TCPDF
+         $drawFirst = true,
+         $trans = false
+     ) {
+-    
++
+         // if we want the no trigo direction : add 2PI to the begin angle, to invert the direction
+         if (!$direction) {
+             $angleBegin+= M_PI*2.;
+@@ -1387,7 +1388,7 @@ class MyPdf extends TCPDF
+         $page = null,
+         $fontName = 'helvetica'
+     ) {
+-    
++
+         // bookmark the Title if wanted
+         if ($bookmarkTitle) {
+             $this->Bookmark($titre, 0, -1);

README.md

@@ -96,3 +96,25 @@ XDEBUG_MODE=coverage vendor/bin/phpunit
 # without coverage
 vendor/bin/phpunit
 ```
+
+## For dev
+
+```bash
+drush rapi-i --alias job --verbose
+drush rw-job:index --verbose
+
+cget ocha_ai.settings --include-overridden
+cget ocha_ai_tag.settings --include-overridden
+cget ocha_ai_chat.settings --include-overridden
+cget reliefweb_api.settings --include-overridden
+
+cset ocha_ai.settings plugins.source.reliefweb.api_url https://dev.api-reliefweb-int.ahconu.org/v1
+cset ocha_ai.settings plugins.source.reliefweb.converter_url https://xxx:[email protected]/search/converter/json
+cset reliefweb_api.settings api_url https://dev.api-reliefweb-int.ahconu.org/v1
+cset reliefweb_api.settings api_url_external https://dev.api-reliefweb-int.ahconu.org/v1
+cset reliefweb_api.settings website: https://dev.reliefweb-int.ahconu.org
+
+queue:list
+queue:run --verbose reliefweb_job_tagger
+
+```

composer.json

@@ -78,6 +78,7 @@
         "reliefweb/api-indexer": "^v2.8",
         "reliefweb/simple-autocomplete": "^v1.3",
         "reliefweb/simple-datepicker": "^v1.3",
+        "spipu/html2pdf": "^5.2",
         "symfony/uid": "^6.2",
         "unocha/common_design": "^9.4",
         "unocha/gtm_barebones": "^1.1",

composer.patches.json

@@ -36,6 +36,9 @@
         "drush/drush": {
             "https://humanitarian.atlassian.net/browse/OPS-8026": "PATCHES/drush--timeout-override.patch"
         },
+        "spipu/html2pdf": {
+            "Remove promo link": "PATCHES/remove-tcpdf-footer.patch"
+        },
         "unocha/common_design": {
             "https://humanitarian.atlassian.net/browse/CD-519": "PATCHES/common_design--default-logo-remove.patch"
         }

html/modules/custom/reliefweb_semantic/README.md

@@ -0,0 +1,76 @@
+# ReliefWeb - Semantic module
+
+This module provides integration with the ReliefWeb Semantic API.
+
+## To do
+
+- [x] Add service to query API
+
+## AWS
+
+Dashboards:
+
+- https://us-east-1.console.aws.amazon.com/s3/lens/dashboard/RW-KB?region=us-east-1&bucketType=general
+
+Config:
+
+- Role: `BedrockRoleKbRw`
+- Collection: `arn:aws:aoss:us-east-1:694216630861:collection/b2h8ajgjb3x87ur892hc`
+- Vector field: `embedding`
+- Text field name: `AMAZON_BEDROCK_TEXT_CHUNK`
+- Metadata field name: `AMAZON_BEDROCK_METADATA`
+
+All content is using a single KB, current Id is `VIEPSPYNSS`
+
+## Drush
+
+```bash
+drush reliefweb-semantic:index            Index content in the ReliefWeb API.
+drush reliefweb-semantic:list-kbs         List kbs.
+drush reliefweb-semantic:list-datasources List datasources.
+drush reliefweb-semantic:list-jobs        List ingestion jobs.
+drush reliefweb-semantic:trigger-sync     Trigger sync..
+drush reliefweb-semantic:query-kb --id=WYBGQOFQLN --q="Any jobs in Europe"
+drush reliefweb-semantic:list-apikeys
+```
+
+## Openseach
+
+```json
+GET /bedrock-knowledge-base-default-index/_search
+{
+  "query": {
+    "match_all": {}
+  }
+}
+```
+
+```json
+GET /bedrock-knowledge-base-default-index/_search
+{
+  "query": {
+    "match": {
+      "country": "*"
+    }
+
+  }
+}
+```
+
+```json
+GET /bedrock-knowledge-base-default-index/_search
+{
+  "query": {
+    "match": {
+      "title": {
+        "query": "china",
+        "fuzziness": "AUTO"
+      }
+    }
+  }
+}
+```
+
+## Questions
+
+- Describe the weather in china in june 2024

html/modules/custom/reliefweb_semantic/drush.services.yml

@@ -0,0 +1,12 @@
+services:
+  reliefweb_semantic.commands:
+    class: \Drupal\reliefweb_semantic\Commands\ReliefWebSemanticCommands
+    arguments: ['@config.factory', '@entity_field.manager', '@entity_type.manager', '@module_handler', '@state', '@http_client', '@file_system', '@renderer']
+    tags:
+      - { name: drush.command }
+  reliefweb_semantic.aws_commands:
+    class: \Drupal\reliefweb_semantic\Commands\ReliefWebSemanticAwsCommands
+    arguments: ['@config.factory', '@state', '@reliefweb_semantic.search_service']
+    tags:
+      - { name: drush.command }
+

html/modules/custom/reliefweb_semantic/reliefweb_semantic.info.yml

@@ -0,0 +1,5 @@
+type: module
+name: ReliefWeb Semantic
+description: 'Provides integration with the ReliefWeb Semantic API.'
+package: reliefweb
+core_version_requirement: ^9 || ^10