Merge branch 'claims' into 'main'

Update claim types

Closes #18 and #34

See merge request peerdb/peerdb!43

Author: mitar

.gitlab-ci.yml

@@ -40,21 +40,23 @@ test_go:
 
   image: golang:1.25-alpine3.22
 
+  tags:
+    - saas-linux-2xlarge-amd64
+
   services:
-    - name: elasticsearch:7.16.3
+    - name: "$CI_REGISTRY_IMAGE/elastic/$ELASTIC_VERSION:latest"
       alias: elasticsearch
       variables:
         network.bind_host: 0.0.0.0
         network.publish_host: elasticsearch
-        ES_JAVA_OPTS: "-Xmx1000m"
         discovery.type: single-node
         xpack.security.enabled: false
         ingest.geoip.downloader.enabled: false
         cluster.routing.allocation.disk.watermark.flood_stage: "100%"
+        ES_LOG_STYLE: file
     - name: registry.gitlab.com/tozd/docker/postgresql:18
       alias: postgres
       variables:
-        LOG_TO_STDOUT: 1
         PGSQL_ROLE_1_USERNAME: test
         PGSQL_ROLE_1_PASSWORD: test
         PGSQL_DB_1_NAME: test
@@ -68,10 +70,16 @@ test_go:
     - (cd /go; go install gotest.tools/gotestsum@v1.13.0)
     - (cd /go; go install github.com/boumenot/gocover-cobertura@v1.4.0)
     - go version
+    - wget --spider "http://elasticsearch:9200/_cluster/health?wait_for_status=yellow&timeout=120s"
 
   script:
     - make test-ci
 
+  after_script:
+    # TODO: This does not really work. We should move services inside a DinD and add docker binary here.
+    - docker cp elasticsearch:/usr/share/elasticsearch/logs elasticsearch-logs
+    - docker cp postgres:/var/log/postgresql postgresql-logs
+
   artifacts:
     when: always
     reports:
@@ -83,6 +91,8 @@ test_go:
       - tests.xml
       - coverage.html
       - coverage.xml
+      - elasticsearch-logs
+      - postgresql-logs
     expire_in: never
 
   coverage: '/coverage: \d+\.\d+% of statements/'
@@ -119,6 +129,9 @@ lint_go:
 
   image: golang:1.25-alpine3.22
 
+  tags:
+    - saas-linux-2xlarge-amd64
+
   before_script:
     - apk --update add make bash gcc musl-dev git
     - wget -O- -nv https://raw.githubusercontent.com/golangci/golangci-lint/master/install.sh | sh -s -- -b $(go env GOPATH)/bin v2.5.0
@@ -273,6 +286,9 @@ docker:
 
   image: docker:28-cli
 
+  tags:
+    - saas-linux-2xlarge-amd64
+
   services:
     - name: docker:28-dind
       command:

.golangci.yml

@@ -20,6 +20,9 @@ linters:
       check-type-assertions: true
     exhaustruct:
       allow-empty: true
+    goconst:
+      ignore-string-values:
+        - "none"
     gocritic:
       disabled-checks:
         - ifElseChain

CLAUDE.md

@@ -13,7 +13,8 @@ Key features:
 
 - **Versioned document store** with full change history
 - **Real-time collaboration** with conflict detection
-- **Claims-based document schema** supporting 11+ claim types (identifiers, text, relations, amounts, time, files, etc.)
+- **Claims-based document schema** supporting 12 claim types (identifiers, strings, HTML, amounts, time, links,
+  references, etc.)
 - **Adaptive search UI** that automatically adjusts to data and provides relevant filters
 - **Multi-site support** with separate schemas and indices per site
 
@@ -162,56 +163,64 @@ Manages append-only logs for real-time collaboration sessions.
 - Pagination (max 5000 operations per page)
 - PostgreSQL-backed with JSONB columns
 
-**PeerDB usage**: Tracks document editing sessions with conflict detection.
+**PeerDB usage**: Tracks document edit sessions with conflict detection.
 
 #### 3. **Document Schema** (`document/`)
 
-Claims-based document system supporting 11 claim types:
+Claims-based document system supporting 12 claim types:
 
 - **IdentifierClaim**: External IDs (e.g., Wikidata Q-IDs)
-- **TextClaim**: Rich text with HTML in multiple languages
 - **StringClaim**: Plain text strings
-- **RelationClaim**: Relationships to other documents
-- **AmountClaim/AmountRangeClaim**: Numeric values with units
-- **TimeClaim/TimeRangeClaim**: Timestamps and time ranges
-- **FileClaim**: File references
-- **ReferenceClaim**: URL references
-- **NoValueClaim/UnknownValueClaim**: Missing data markers
+- **HTMLClaim**: Rich text with HTML
+- **AmountClaim**: Numeric values with precision
+- **AmountIntervalClaim**: Numeric intervals with bounds
+- **TimeClaim**: Timestamps with precision
+- **TimeIntervalClaim**: Time intervals with bounds
+- **LinkClaim**: URL/IRI links
+- **ReferenceClaim**: Relationships to other documents
+- **HasClaim**: Property-only claim (can hold nested claims via sub-claims)
+- **NoneClaim**: Explicitly states no value exists
+- **UnknownClaim**: Value exists but is unknown
 
 **Core structure**:
 
 ```go
 type D struct {
-    CoreDocument  // ID (22-char identifier), Score, Scores
-    Mnemonic      // Human-readable identifier
-    Claims        // ClaimTypes (collections of claims)
+    CoreDocument  // ID (22-char identifier), Base (base for computing ID)
+    Claims        // *ClaimTypes (collections of claims)
+}
+
+type CoreClaim struct {
+    ID         identifier.Identifier
+    Confidence Confidence  // float64 in [-1, 1]
+    Sub        *ClaimTypes // optional sub-claims
 }
 ```
 
 **Important patterns**:
 
 - Use the Visitor pattern to traverse/manipulate claims
-- Claims reference properties (also documents) via `prop.id`
-- Built-in classes and properties defined in `core/` (NAME, etc.)
+- Claims reference properties (also documents) via `Prop`
+- Each claim embeds `CoreClaim` with ID, Confidence, and optional Sub-claims
+- Built-in classes and properties defined in `core/` (NAME, DESCRIPTION, etc.)
 
 #### 4. **Search** (`search/search.go`)
 
 Elasticsearch query builder with session-based filtering.
 
 **Filter types**:
 
-- `RelFilter`: Filter by relation claims
+- `RefFilter`: Filter by reference claims
 - `AmountFilter`: Filter by numeric ranges
 - `TimeFilter`: Filter by time ranges
-- `StringFilter`: Text search on string/text claims
 
 **Limits**: Max 1000 search results per query
 
 #### 5. **Storage** (`storage/storage.go`)
 
 Chunked file upload management with begin/append/end lifecycle.
 
-#### 6. **ES Bridge** (`internal/es/bridge.go`)
+#### 6. **ES Bridge** (`internal/search/bridge.go`)
 
 Listens to Store changesets and synchronizes to Elasticsearch using bulk indexing.
 
@@ -293,6 +302,6 @@ Single PeerDB instance can serve multiple sites:
 
 ### ElasticSearch Index
 
-- Index configuration generated in internal/mapping package
+- Index configuration generated in internal/search package
 - Auto-created on first run if missing
 - Run `./peerdb populate` to initialize with core PeerDB properties

Makefile

@@ -39,8 +39,11 @@ package-lock.json: package.json
 test:
 	gotestsum --format pkgname --packages ./... -- -race -timeout 10m -cover -covermode atomic
 
+# We use -p 1 to run only one package test and thus test process at a time. This allows us to control the number of
+# connections made to a PostgreSQL service to not hit the limit of connections to it. If multiple test processes run
+# in parallel, then our current logic where we track connections and pools inside one process is not enough.
 test-ci:
-	gotestsum --format pkgname --packages ./... --junitfile tests.xml -- -race -timeout 10m -coverprofile=coverage.txt -covermode atomic
+	gotestsum --format pkgname --packages ./... --junitfile tests.xml -- -p 1 -race -timeout 10m -coverprofile=coverage.txt -covermode atomic
 	gocover-cobertura < coverage.txt > coverage.xml
 	go tool cover -html=coverage.txt -o coverage.html