cocoindex 0.2.7__tar.gz → 0.2.9__tar.gz

{cocoindex-0.2.7 → cocoindex-0.2.9}/Cargo.lock

@@ -1283,7 +1283,7 @@ dependencies = [
 
 [[package]]
 name = "cocoindex"
-version = "0.2.7"
+version = "0.2.9"
 dependencies = [
  "anyhow",
  "async-openai",
@@ -1337,6 +1337,7 @@ dependencies = [
  "schemars 0.8.22",
  "serde",
  "serde_json",
+ "serde_path_to_error",
  "serde_with",
  "sqlx",
  "time",

{cocoindex-0.2.7 → cocoindex-0.2.9}/Cargo.toml

@@ -2,7 +2,7 @@
 name = "cocoindex"
 # Version used for local development is always higher than others to take precedence.
 # Will be overridden for specific release versions.
-version = "0.2.7"
+version = "0.2.9"
 edition = "2024"
 rust-version = "1.88"
 license = "Apache-2.0"
@@ -145,3 +145,4 @@ azure_storage_blobs = { version = "0.21.0", default-features = false, features =
     "enable_reqwest_rustls",
     "hmac_rust",
 ] }
+serde_path_to_error = "0.1.17"

{cocoindex-0.2.7 → cocoindex-0.2.9}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: cocoindex
-Version: 0.2.7
+Version: 0.2.9
 Classifier: Development Status :: 3 - Alpha
 Classifier: License :: OSI Approved :: Apache Software License
 Classifier: Operating System :: OS Independent

{cocoindex-0.2.7 → cocoindex-0.2.9}/THIRD_PARTY_NOTICES.html

@@ -2428,7 +2428,7 @@ Software.
                 <h3 id="Apache-2.0">Apache License 2.0</h3>
                 <h4>Used by:</h4>
                 <ul class="license-used-by">
-                    <li><a href=" https://crates.io/crates/cocoindex ">cocoindex 0.2.7</a></li>
+                    <li><a href=" https://crates.io/crates/cocoindex ">cocoindex 0.2.9</a></li>
                     <li><a href=" https://github.com/awesomized/crc-fast-rust ">crc-fast 1.3.0</a></li>
                     <li><a href=" https://github.com/qdrant/rust-client ">qdrant-client 1.15.0</a></li>
                 </ul>

{cocoindex-0.2.7 → cocoindex-0.2.9}/dev/neo4j.yaml

@@ -1,16 +1,15 @@
-name: cocoindex-neo4j
+name: cocoindex-neo4j-community
 services:
   neo4j:
-    image: neo4j:5-enterprise
+    image: neo4j:2025.08-community-ubi9
     volumes:
-        - /$HOME/neo4j/logs:/logs
-        - /$HOME/neo4j/config:/config
-        - /$HOME/neo4j/data:/data
-        - /$HOME/neo4j/plugins:/plugins
+        - /$HOME/neo4j/community/logs:/logs
+        - /$HOME/neo4j/community/config:/config
+        - /$HOME/neo4j/community/data:/data
+        - /$HOME/neo4j/community/plugins:/plugins
     environment:
         - NEO4J_AUTH=neo4j/cocoindex
         - NEO4J_PLUGINS='["graph-data-science"]'
-        - NEO4J_ACCEPT_LICENSE_AGREEMENT=eval
 
         # Uncomment to enable query logging
         # - NEO4J_db_logs_query_enabled=VERBOSE

{cocoindex-0.2.7 → cocoindex-0.2.9}/docs/docs/ops/targets.md

@@ -442,15 +442,6 @@ docker compose -f <(curl -L https://raw.githubusercontent.com/cocoindex-io/cocoi
 This will bring up a Neo4j instance, which can be accessed by username `neo4j` and password `cocoindex`.
 You can access the Neo4j browser at [http://localhost:7474](http://localhost:7474).
 
-:::warning
-
-The docker compose config above will start a Neo4j Enterprise instance under the [Evaluation License](https://neo4j.com/terms/enterprise_us/),
-with 30 days trial period.
-Please read and agree the license before starting the instance.
-
-:::
-
-
 ### Kuzu
 
 #### Spec

{cocoindex-0.2.7 → cocoindex-0.2.9}/examples/amazon_s3_embedding/pyproject.toml

@@ -4,7 +4,7 @@ version = "0.1.0"
 description = "Simple example for cocoindex: build embedding index based on Amazon S3 files."
 requires-python = ">=3.11"
 dependencies = [
-    "cocoindex[embeddings]>=0.2.4",
+    "cocoindex[embeddings]>=0.2.8",
     "python-dotenv>=1.0.1",
     "psycopg[binary,pool]",
 ]

{cocoindex-0.2.7 → cocoindex-0.2.9}/examples/azure_blob_embedding/pyproject.toml

@@ -4,7 +4,7 @@ version = "0.1.0"
 description = "Simple example for cocoindex: build embedding index based on Azure Blob Storage files."
 requires-python = ">=3.11"
 dependencies = [
-    "cocoindex[embeddings]>=0.2.4",
+    "cocoindex[embeddings]>=0.2.8",
     "python-dotenv>=1.0.1",
     "psycopg[binary,pool]",
 ]

{cocoindex-0.2.7 → cocoindex-0.2.9}/examples/code_embedding/pyproject.toml

@@ -4,7 +4,7 @@ version = "0.1.0"
 description = "Simple example for cocoindex: build embedding index based on source code."
 requires-python = ">=3.11"
 dependencies = [
-    "cocoindex[embeddings]>=0.2.4",
+    "cocoindex[embeddings]>=0.2.8",
     "python-dotenv>=1.0.1",
     "pgvector>=0.4.1",
     "psycopg[binary,pool]",

{cocoindex-0.2.7 → cocoindex-0.2.9}/examples/custom_output_files/pyproject.toml

@@ -3,7 +3,7 @@ name = "custom-output-files"
 version = "0.1.0"
 description = "Simple example for cocoindex: convert markdown files to HTML files and save them to a local directory."
 requires-python = ">=3.11"
-dependencies = ["cocoindex>=0.2.4", "markdown-it-py[linkify,plugins]"]
+dependencies = ["cocoindex>=0.2.8", "markdown-it-py[linkify,plugins]"]
 
 [tool.setuptools]
 packages = []

{cocoindex-0.2.7 → cocoindex-0.2.9}/examples/docs_to_knowledge_graph/pyproject.toml

@@ -3,7 +3,7 @@ name = "manuals-to-kg"
 version = "0.1.0"
 description = "Simple example for cocoindex: extract triples from files and build knowledge graph."
 requires-python = ">=3.11"
-dependencies = ["cocoindex>=0.2.4"]
+dependencies = ["cocoindex>=0.2.8"]
 
 [tool.setuptools]
 packages = []

{cocoindex-0.2.7 → cocoindex-0.2.9}/examples/face_recognition/pyproject.toml

@@ -4,7 +4,7 @@ version = "0.1.0"
 description = "Build index for papers with both metadata and content embeddings"
 requires-python = ">=3.11"
 dependencies = [
-    "cocoindex>=0.2.4",
+    "cocoindex>=0.2.8",
     "face-recognition>=1.3.0",
     "pillow>=10.0.0",
     "numpy>=1.26.0",

{cocoindex-0.2.7 → cocoindex-0.2.9}/examples/fastapi_server_docker/requirements.txt

@@ -1,4 +1,4 @@
-cocoindex[embeddings]>=0.2.4
+cocoindex[embeddings]>=0.2.8
 python-dotenv>=1.0.1
 fastapi==0.115.12
 fastapi-cli==0.0.7

{cocoindex-0.2.7 → cocoindex-0.2.9}/examples/gdrive_text_embedding/pyproject.toml

@@ -4,7 +4,7 @@ version = "0.1.0"
 description = "Simple example for cocoindex: build embedding index based on Google Drive files."
 requires-python = ">=3.11"
 dependencies = [
-    "cocoindex[embeddings]>=0.2.4",
+    "cocoindex[embeddings]>=0.2.8",
     "python-dotenv>=1.0.1",
     "psycopg[binary,pool]",
 ]

{cocoindex-0.2.7 → cocoindex-0.2.9}/examples/image_search/pyproject.toml

@@ -4,7 +4,7 @@ version = "0.1.0"
 description = "Image search examples for cocoindex: CLIP and ColPali-based embedding."
 requires-python = ">=3.11"
 dependencies = [
-    "cocoindex[colpali]>=0.2.4",
+    "cocoindex[colpali]>=0.2.8",
     "python-dotenv>=1.0.1",
     "fastapi>=0.100.0",
     "torch>=2.0.0",

{cocoindex-0.2.7 → cocoindex-0.2.9}/examples/live_updates/pyproject.toml

@@ -3,7 +3,7 @@ name = "live-updates-example"
 version = "0.1.0"
 description = "Simple example for cocoindex: perform live updates based on local markdown files."
 requires-python = ">=3.11"
-dependencies = ["cocoindex>=0.2.4", "python-dotenv>=1.1.0"]
+dependencies = ["cocoindex>=0.2.8", "python-dotenv>=1.1.0"]
 
 [tools.setuptools]
 packages = []

{cocoindex-0.2.7 → cocoindex-0.2.9}/examples/manuals_llm_extraction/pyproject.toml

@@ -3,7 +3,7 @@ name = "manuals-llm-extraction"
 version = "0.1.0"
 description = "Simple example for cocoindex: extract structured information from a Markdown file using LLM."
 requires-python = ">=3.11"
-dependencies = ["cocoindex>=0.2.4", "marker-pdf>=1.8.5"]
+dependencies = ["cocoindex>=0.2.8", "marker-pdf>=1.8.5"]
 
 [tool.setuptools]
 packages = []

{cocoindex-0.2.7 → cocoindex-0.2.9}/examples/multi_format_indexing/pyproject.toml

@@ -4,7 +4,7 @@ version = "0.1.0"
 description = "Simple example for cocoindex: build embedding index based on local PDF files."
 requires-python = ">=3.11"
 dependencies = [
-    "cocoindex[colpali]>=0.2.4",
+    "cocoindex[colpali]>=0.2.8",
     "python-dotenv>=1.0.1",
     "pdf2image>=1.17.0",
     "qdrant-client>=1.15.0",

{cocoindex-0.2.7 → cocoindex-0.2.9}/examples/paper_metadata/pyproject.toml

@@ -4,7 +4,7 @@ version = "0.1.0"
 description = "Build index for papers with both metadata and content embeddings"
 requires-python = ">=3.11"
 dependencies = [
-    "cocoindex[embeddings]>=0.2.4",
+    "cocoindex[embeddings]>=0.2.8",
     "pypdf>=5.7.0",
     "marker-pdf>=1.8.5",
 ]

{cocoindex-0.2.7 → cocoindex-0.2.9}/examples/patient_intake_extraction/pyproject.toml

@@ -4,7 +4,7 @@ version = "0.1.0"
 description = "Extract structured information from patient intake forms using LLM."
 requires-python = ">=3.10"
 dependencies = [
-    "cocoindex>=0.2.4",
+    "cocoindex>=0.2.8",
     "python-dotenv>=1.0.1",
     "markitdown>=0.1.2",
     "openai>=1.68.2",

{cocoindex-0.2.7 → cocoindex-0.2.9}/examples/pdf_embedding/pyproject.toml

@@ -4,7 +4,7 @@ version = "0.1.0"
 description = "Simple example for cocoindex: build embedding index based on local PDF files."
 requires-python = ">=3.11"
 dependencies = [
-    "cocoindex[embeddings]>=0.2.4",
+    "cocoindex[embeddings]>=0.2.8",
     "python-dotenv>=1.0.1",
     "marker-pdf>=1.8.5",
     "psycopg[binary,pool]",

{cocoindex-0.2.7 → cocoindex-0.2.9}/examples/postgres_source/pyproject.toml

@@ -4,7 +4,7 @@ version = "0.1.0"
 description = "Demonstrate how to use Postgres tables as the source for CocoIndex."
 requires-python = ">=3.11"
 dependencies = [
-    "cocoindex[embeddings]>=0.2.4",
+    "cocoindex[embeddings]>=0.2.8",
     "python-dotenv>=1.0.1",
     "pgvector>=0.4.1",
     "psycopg[binary,pool]",

cocoindex-0.2.9/examples/product_recommendation/.gitignore

@@ -0,0 +1 @@
+.env

{cocoindex-0.2.7 → cocoindex-0.2.9}/examples/product_recommendation/pyproject.toml

@@ -3,7 +3,7 @@ name = "cocoindex-ecommerce-taxonomy"
 version = "0.1.0"
 description = "Simple example for CocoIndex: extract taxonomy from e-commerce products and build knowledge graph."
 requires-python = ">=3.11"
-dependencies = ["cocoindex>=0.2.4", "jinja2>=3.1.6"]
+dependencies = ["cocoindex>=0.2.8", "jinja2>=3.1.6"]
 
 [tool.setuptools]
 packages = []

{cocoindex-0.2.7 → cocoindex-0.2.9}/examples/text_embedding/pyproject.toml

@@ -4,7 +4,7 @@ version = "0.1.0"
 description = "Simple example for cocoindex: build embedding index based on local text files."
 requires-python = ">=3.11"
 dependencies = [
-    "cocoindex[embeddings]>=0.2.4",
+    "cocoindex[embeddings]>=0.2.8",
     "python-dotenv>=1.0.1",
     "pgvector>=0.4.1",
     "psycopg[binary,pool]",

{cocoindex-0.2.7 → cocoindex-0.2.9}/examples/text_embedding_qdrant/pyproject.toml

@@ -4,7 +4,7 @@ version = "0.1.0"
 description = "Simple example for cocoindex: build embedding index based on local text files."
 requires-python = ">=3.11"
 dependencies = [
-    "cocoindex[embeddings]>=0.2.4",
+    "cocoindex[embeddings]>=0.2.8",
     "python-dotenv>=1.0.1",
     "qdrant-client>=1.6.0",
 ]

{cocoindex-0.2.7 → cocoindex-0.2.9}/python/cocoindex/setting.py

@@ -94,8 +94,7 @@ class Settings:
 
         database_url = os.getenv("COCOINDEX_DATABASE_URL")
         if database_url is not None:
-            db_kwargs: dict[str, Any] = dict()
-            _load_field(db_kwargs, "url", "COCOINDEX_DATABASE_URL", required=True)
+            db_kwargs: dict[str, Any] = {"url": database_url}
             _load_field(db_kwargs, "user", "COCOINDEX_DATABASE_USER")
             _load_field(db_kwargs, "password", "COCOINDEX_DATABASE_PASSWORD")
             _load_field(

{cocoindex-0.2.7 → cocoindex-0.2.9}/src/base/value.rs

@@ -1,13 +1,12 @@
+use crate::prelude::*;
+
 use super::schema::*;
 use crate::base::duration::parse_duration;
-use crate::{api_bail, api_error};
-use anyhow::Result;
 use base64::prelude::*;
 use bytes::Bytes;
 use chrono::Offset;
 use log::warn;
 use serde::{
-    Deserialize, Serialize,
     de::{SeqAccess, Visitor},
     ser::{SerializeMap, SerializeSeq, SerializeTuple},
 };
@@ -1014,7 +1013,8 @@ where
         Ok(Self {
             fields: fields
                 .map(|(s, v)| {
-                    let value = Value::<VS>::from_json(v, &s.value_type.typ)?;
+                    let value = Value::<VS>::from_json(v, &s.value_type.typ)
+                        .with_context(|| format!("while deserializing field `{}`", s.name))?;
                     if value.is_null() && !s.value_type.nullable {
                         api_bail!("expected non-null value for `{}`", s.name);
                     }
@@ -1033,9 +1033,10 @@ where
             fields: fields_schema
                 .map(|field| {
                     let value = match values.get_mut(&field.name) {
-                        Some(v) => {
-                            Value::<VS>::from_json(std::mem::take(v), &field.value_type.typ)?
-                        }
+                        Some(v) => Value::<VS>::from_json(std::mem::take(v), &field.value_type.typ)
+                            .with_context(|| {
+                                format!("while deserializing field `{}`", field.name)
+                            })?,
                         None => Value::<VS>::default(),
                     };
                     if value.is_null() && !field.value_type.nullable {
@@ -1137,7 +1138,7 @@ impl BasicValue {
                 v.as_f64()
                     .ok_or_else(|| anyhow::anyhow!("invalid fp64 value {v}"))?,
             ),
-            (v, BasicValueType::Range) => BasicValue::Range(serde_json::from_value(v)?),
+            (v, BasicValueType::Range) => BasicValue::Range(utils::deser::from_json_value(v)?),
             (serde_json::Value::String(v), BasicValueType::Uuid) => BasicValue::Uuid(v.parse()?),
             (serde_json::Value::String(v), BasicValueType::Date) => BasicValue::Date(v.parse()?),
             (serde_json::Value::String(v), BasicValueType::Time) => BasicValue::Time(v.parse()?),
@@ -1170,7 +1171,11 @@ impl BasicValue {
             ) => {
                 let vec = v
                     .into_iter()
-                    .map(|v| BasicValue::from_json(v, element_type))
+                    .enumerate()
+                    .map(|(i, v)| {
+                        BasicValue::from_json(v, element_type)
+                            .with_context(|| format!("while deserializing Vector element #{i}"))
+                    })
                     .collect::<Result<Vec<_>>>()?;
                 BasicValue::Vector(Arc::from(vec))
             }
@@ -1267,7 +1272,11 @@ where
                     TableKind::UTable => {
                         let rows = v
                             .into_iter()
-                            .map(|v| Ok(FieldValues::from_json(v, &s.row.fields)?.into()))
+                            .map(|v| {
+                                Ok(FieldValues::from_json(v, &s.row.fields)
+                                    .with_context(|| format!("while deserializing UTable row"))?
+                                    .into())
+                            })
                             .collect::<Result<Vec<_>>>()?;
                         Value::LTable(rows)
                     }
@@ -1289,10 +1298,13 @@ where
                                         let mut field_vals_iter = v.into_iter();
                                         let keys: Box<[KeyPart]> = (0..num_key_parts)
                                             .map(|_| {
+                                                let field_schema = fields_iter.next().unwrap();
                                                 Self::from_json(
                                                     field_vals_iter.next().unwrap(),
-                                                    &fields_iter.next().unwrap().value_type.typ,
-                                                )?
+                                                    &field_schema.value_type.typ,
+                                                ).with_context(|| {
+                                                    format!("while deserializing key part `{}`", field_schema.name)
+                                                })?
                                                 .into_key()
                                             })
                                             .collect::<Result<_>>()?;
@@ -1328,7 +1340,14 @@ where
                     TableKind::LTable => {
                         let rows = v
                             .into_iter()
-                            .map(|v| Ok(FieldValues::from_json(v, &s.row.fields)?.into()))
+                            .enumerate()
+                            .map(|(i, v)| {
+                                Ok(FieldValues::from_json(v, &s.row.fields)
+                                    .with_context(|| {
+                                        format!("while deserializing LTable row #{i}")
+                                    })?
+                                    .into())
+                            })
                             .collect::<Result<Vec<_>>>()?;
                         Value::LTable(rows)
                     }

{cocoindex-0.2.7 → cocoindex-0.2.9}/src/builder/analyzed_flow.rs

@@ -20,7 +20,9 @@ impl AnalyzedFlow {
         flow_instance_ctx: Arc<FlowInstanceContext>,
     ) -> Result<Self> {
         let (data_schema, setup_state, execution_plan_fut) =
-            analyzer::analyze_flow(&flow_instance, flow_instance_ctx.clone()).await?;
+            analyzer::analyze_flow(&flow_instance, flow_instance_ctx.clone())
+                .await
+                .with_context(|| format!("analyzing flow `{}`", flow_instance.name))?;
         let execution_plan = async move {
             shared_ok(Arc::new(
                 execution_plan_fut.await.map_err(SharedError::new)?,

{cocoindex-0.2.7 → cocoindex-0.2.9}/src/builder/analyzer.rs

@@ -674,7 +674,9 @@ impl AnalyzerContext {
         let global_concurrency_controller = self.lib_ctx.global_concurrency_controller.clone();
         let result_fut = async move {
             trace!("Start building executor for source op `{op_name}`");
-            let executor = executor.await?;
+            let executor = executor
+                .await
+                .with_context(|| format!("Preparing for source op: {op_name}"))?;
             trace!("Finished building executor for source op `{op_name}`");
             Ok(AnalyzedImportOp {
                 executor,
@@ -700,10 +702,7 @@ impl AnalyzerContext {
             ReactiveOpSpec::Transform(op) => {
                 let input_field_schemas =
                     analyze_input_fields(&op.inputs, op_scope).with_context(|| {
-                        format!(
-                            "Failed to analyze inputs for transform op: {}",
-                            reactive_op.name
-                        )
+                        format!("Preparing inputs for transform op: {}", reactive_op.name)
                     })?;
                 let spec = serde_json::Value::Object(op.op.spec.clone());
 
@@ -725,7 +724,7 @@ impl AnalyzerContext {
                 async move {
                             trace!("Start building executor for transform op `{op_name}`");
                             let executor = executor.await.with_context(|| {
-                                format!("Failed to build executor for transform op: {op_name}")
+                                format!("Preparing for transform op: {op_name}")
                             })?;
                             let enable_cache = executor.enable_cache();
                             let behavior_version = executor.behavior_version();
@@ -784,7 +783,7 @@ impl AnalyzerContext {
                         local_field_ref,
                         op_scope: analyzed_op_scope_fut
                             .await
-                            .with_context(|| format!("Analyzing foreach op: {op_name}"))?,
+                            .with_context(|| format!("Preparing for foreach op: {op_name}"))?,
                         name: op_name,
                         concurrency_controller: concur_control::ConcurrencyController::new(
                             &concur_control_options,
@@ -920,7 +919,7 @@ impl AnalyzerContext {
                     let export_context = data_coll_output
                         .export_context
                         .await
-                        .with_context(|| format!("Analyzing export op: {op_name}"))?;
+                        .with_context(|| format!("Preparing for export op: {op_name}"))?;
                     trace!("Finished building executor for export op `{op_name}`");
                     Ok(AnalyzedExportOp {
                         name: op_name,
@@ -1006,7 +1005,8 @@ pub async fn analyze_flow(
         import_ops_futs.push(
             analyzer_ctx
                 .analyze_import_op(&root_op_scope, import_op.clone())
-                .await?,
+                .await
+                .with_context(|| format!("Preparing for import op: {}", import_op.name))?,
         );
     }
     let op_scope_fut = analyzer_ctx
@@ -1059,7 +1059,8 @@ pub async fn analyze_flow(
                     &mut targets_analyzed_ss,
                     &mut declarations_analyzed_ss,
                 )
-                .await?,
+                .await
+                .with_context(|| format!("Analyzing export ops for target `{target_kind}`"))?,
         );
         analyzed_target_op_groups.push(analyzed_target_op_group);
     }

{cocoindex-0.2.7 → cocoindex-0.2.9}/src/execution/source_indexer.rs

@@ -382,7 +382,8 @@ impl SourceIndexingContext {
             error!(
                 "{:?}",
                 e.context(format!(
-                    "Error in processing row from source `{source}` with key: {key}",
+                    "Error in processing row from flow `{flow}` source `{source}` with key: {key}",
+                    flow = self.flow.flow_instance.name,
                     source = self.flow.flow_instance.import_ops[self.source_idx].name,
                     key = row_input.key,
                 ))

{cocoindex-0.2.7 → cocoindex-0.2.9}/src/lib_context.rs

@@ -250,7 +250,7 @@ impl LibContext {
     pub fn require_persistence_ctx(&self) -> Result<&PersistenceContext> {
         self.persistence_ctx
             .as_ref()
-            .ok_or_else(|| anyhow!("Database is required for this operation. Please set COCOINDEX_DATABASE_URL environment variable and call cocoindex.init() with database settings."))
+            .ok_or_else(|| anyhow!("Database is required for this operation. Please set COCOINDEX_DATABASE_URL environment variable OR call `cocoindex.init()` with database settings."))
     }
 
     pub fn require_builtin_db_pool(&self) -> Result<&PgPool> {

{cocoindex-0.2.7 → cocoindex-0.2.9}/src/llm/anthropic.rs

@@ -138,7 +138,7 @@ impl LlmGenerationClient for Client {
             match &mut resp_json["content"][0]["text"] {
                 serde_json::Value::String(s) => {
                     // Try strict JSON parsing first
-                    match serde_json::from_str::<serde_json::Value>(s) {
+                    match utils::deser::from_json_str::<serde_json::Value>(s) {
                         Ok(_) => std::mem::take(s),
                         Err(e) => {
                             // Try permissive json5 parsing as fallback

{cocoindex-0.2.7 → cocoindex-0.2.9}/src/llm/gemini.rs

@@ -276,7 +276,7 @@ impl LlmGenerationClient for VertexAiClient {
             generation_config = Some(
                 GenerationConfig::new()
                     .set_response_mime_type("application/json".to_string())
-                    .set_response_schema(serde_json::from_value::<Schema>(schema_json)?),
+                    .set_response_schema(utils::deser::from_json_value::<Schema>(schema_json)?),
             );
         }
 
@@ -358,7 +358,7 @@ impl LlmEmbeddingClient for VertexAiClient {
             .next()
             .and_then(|mut e| e.get_mut("embeddings").map(|v| v.take()))
             .ok_or_else(|| anyhow::anyhow!("No embeddings in response"))?;
-        let embedding: ContentEmbedding = serde_json::from_value(embeddings)?;
+        let embedding: ContentEmbedding = utils::deser::from_json_value(embeddings)?;
         Ok(super::LlmEmbeddingResponse {
             embedding: embedding.values,
         })

{cocoindex-0.2.7 → cocoindex-0.2.9}/src/ops/factory_bases.rs

@@ -245,7 +245,8 @@ impl<T: SourceFactoryBase> SourceFactory for T {
         EnrichedValueType,
         BoxFuture<'static, Result<Box<dyn SourceExecutor>>>,
     )> {
-        let spec: T::Spec = serde_json::from_value(spec)?;
+        let spec: T::Spec = utils::deser::from_json_value(spec)
+            .with_context(|| format!("Failed in parsing spec for source `{source_name}`"))?;
         let output_schema = self.get_output_schema(&spec, &context).await?;
         let source_name = source_name.to_string();
         let executor = async move { self.build_executor(&source_name, spec, context).await };
@@ -323,7 +324,8 @@ impl<T: SimpleFunctionFactoryBase> SimpleFunctionFactory for T {
         EnrichedValueType,
         BoxFuture<'static, Result<Box<dyn SimpleFunctionExecutor>>>,
     )> {
-        let spec: T::Spec = serde_json::from_value(spec)?;
+        let spec: T::Spec = utils::deser::from_json_value(spec)
+            .with_context(|| format!("Failed in parsing spec for function `{}`", self.name()))?;
         let mut nonnull_args_idx = vec![];
         let mut may_nullify_output = false;
         let mut args_resolver = OpArgsResolver::new(
@@ -397,7 +399,7 @@ pub trait TargetFactoryBase: TargetFactory + Send + Sync + 'static {
     /// Deserialize the setup key from a JSON value.
     /// You can override this method to provide a custom deserialization logic, e.g. to perform backward compatible deserialization.
     fn deserialize_setup_key(key: serde_json::Value) -> Result<Self::SetupKey> {
-        Ok(serde_json::from_value(key)?)
+        Ok(utils::deser::from_json_value(key)?)
     }
 
     /// Will not be called if it's setup by user.
@@ -466,8 +468,10 @@ impl<T: TargetFactoryBase> TargetFactory for T {
                 .into_iter()
                 .map(|d| {
                     anyhow::Ok(TypedExportDataCollectionSpec {
+                        spec: utils::deser::from_json_value(d.spec).with_context(|| {
+                            format!("Failed in parsing spec for target `{}`", d.name)
+                        })?,
                         name: d.name,
-                        spec: serde_json::from_value(d.spec)?,
                         key_fields_schema: d.key_fields_schema,
                         value_fields_schema: d.value_fields_schema,
                         index_options: d.index_options,
@@ -476,7 +480,7 @@ impl<T: TargetFactoryBase> TargetFactory for T {
                 .collect::<Result<Vec<_>>>()?,
             declarations
                 .into_iter()
-                .map(|d| anyhow::Ok(serde_json::from_value(d)?))
+                .map(|d| anyhow::Ok(utils::deser::from_json_value(d)?))
                 .collect::<Result<Vec<_>>>()?,
             context,
         )
@@ -511,7 +515,7 @@ impl<T: TargetFactoryBase> TargetFactory for T {
     ) -> Result<Box<dyn setup::ResourceSetupChange>> {
         let key: T::SetupKey = Self::deserialize_setup_key(key.clone())?;
         let desired_state: Option<T::SetupState> = desired_state
-            .map(|v| serde_json::from_value(v.clone()))
+            .map(|v| utils::deser::from_json_value(v.clone()))
             .transpose()?;
         let existing_states = from_json_combined_state(existing_states)?;
         let setup_change = TargetFactoryBase::diff_setup_states(
@@ -542,8 +546,8 @@ impl<T: TargetFactoryBase> TargetFactory for T {
     ) -> Result<SetupStateCompatibility> {
         let result = TargetFactoryBase::check_state_compatibility(
             self,
-            &serde_json::from_value(desired_state.clone())?,
-            &serde_json::from_value(existing_state.clone())?,
+            &utils::deser::from_json_value(desired_state.clone())?,
+            &utils::deser::from_json_value(existing_state.clone())?,
         )?;
         Ok(result)
     }
@@ -596,7 +600,7 @@ impl<T: TargetFactoryBase> TargetFactory for T {
                 .into_iter()
                 .map(|item| -> anyhow::Result<_> {
                     Ok(TypedResourceSetupChangeItem {
-                        key: serde_json::from_value(item.key.clone())?,
+                        key: utils::deser::from_json_value(item.key.clone())?,
                         setup_change: (item.setup_change as &dyn Any)
                             .downcast_ref::<T::SetupChange>()
                             .ok_or_else(invariance_violation)?,
@@ -614,7 +618,7 @@ fn from_json_combined_state<T: Debug + Clone + Serialize + DeserializeOwned>(
     Ok(setup::CombinedState {
         current: existing_states
             .current
-            .map(|v| serde_json::from_value(v))
+            .map(|v| utils::deser::from_json_value(v))
             .transpose()?,
         staging: existing_states
             .staging
@@ -622,7 +626,7 @@ fn from_json_combined_state<T: Debug + Clone + Serialize + DeserializeOwned>(
             .map(|v| {
                 anyhow::Ok(match v {
                     setup::StateChange::Upsert(v) => {
-                        setup::StateChange::Upsert(serde_json::from_value(v)?)
+                        setup::StateChange::Upsert(utils::deser::from_json_value(v)?)
                     }
                     setup::StateChange::Delete => setup::StateChange::Delete,
                 })