@pdfvector/instance-contract 0.0.47 → 0.0.55

package/.tsc/lib/page-markdown-schema.d.ts

@@ -0,0 +1,5 @@
+import { z } from "zod";
+export declare const pageMarkdownSchema: z.ZodObject<{
+    pageNumber: z.ZodNumber;
+    markdown: z.ZodString;
+}, z.core.$strip>;

package/.tsc/lib/page-markdown-schema.js

@@ -0,0 +1,9 @@
+import { z } from "zod";
+export const pageMarkdownSchema = z.object({
+    pageNumber: z
+        .number()
+        .int()
+        .min(1)
+        .describe("1-based page number in the source document"),
+    markdown: z.string().describe("Extracted markdown content for this page"),
+});

package/.tsc/lib/router/academic/parse.d.ts

@@ -9,6 +9,7 @@ export declare const parse: import("@orpc/contract").ContractProcedureBuilderWit
         max: "max";
         auto: "auto";
     }>>>;
+    includePages: z.ZodDefault<z.ZodOptional<z.ZodBoolean>>;
     callback: z.ZodOptional<z.ZodObject<{
         url: z.ZodURL;
         type: z.ZodOptional<z.ZodString>;
@@ -41,4 +42,8 @@ export declare const parse: import("@orpc/contract").ContractProcedureBuilderWit
     credits: z.ZodNumber;
     requestId: z.ZodNumber;
     html: z.ZodOptional<z.ZodString>;
+    pages: z.ZodOptional<z.ZodArray<z.ZodObject<{
+        pageNumber: z.ZodNumber;
+        markdown: z.ZodString;
+    }, z.core.$strip>>>;
 }, z.core.$strip>, Record<never, never>, Record<never, never>>;

package/.tsc/lib/router/academic/parse.js

@@ -1,6 +1,8 @@
 import { oc } from "@orpc/contract";
+import { academicParseOpenApiCodeSamples } from "@pdfvector/api-docs";
 import { z } from "zod";
 import { fetchableUrlSchema } from "../../fetchable-url-schema.js";
+import { pageMarkdownSchema } from "../../page-markdown-schema.js";
 import { pdfvectorModelSchema } from "../../pdfvector-model-schema.js";
 import { documentParseModelDescription, outputModelDescription, } from "../../supported-mimes.js";
 import { providerSchema } from "./provider.js";
@@ -33,6 +35,11 @@ const parseInputSchema = z
         .optional()
         .default("auto")
         .describe(documentParseModelDescription),
+    includePages: z
+        .boolean()
+        .optional()
+        .default(false)
+        .describe("Set to true to include page-separated markdown in the pages array. Existing markdown output is still returned as the full paper text."),
     callback: callbackSchema,
 })
     .superRefine((input, ctx) => {
@@ -82,6 +89,10 @@ const parseOutputSchema = z.object({
         .string()
         .optional()
         .describe("Full HTML representation of the paper content. Only available when using the 'max' model."),
+    pages: z
+        .array(pageMarkdownSchema)
+        .optional()
+        .describe("Page-separated markdown content. Returned only when includePages is true."),
 });
 const requestExamples = {
     "Parse ArXiv paper by ID": {
@@ -105,6 +116,14 @@ const requestExamples = {
             model: "auto",
         },
     },
+    "Parse paper with page-separated output": {
+        summary: "Parse paper with page-separated output",
+        value: {
+            id: "1706.03762",
+            model: "auto",
+            includePages: true,
+        },
+    },
 };
 export const parse = oc
     .input(parseInputSchema)
@@ -121,6 +140,7 @@ export const parse = oc
                 mediaType.examples = requestExamples;
             }
         }
+        op["x-codeSamples"] = academicParseOpenApiCodeSamples;
         return op;
     },
 });

package/.tsc/lib/router/document/parse.d.ts

@@ -10,6 +10,10 @@ export declare const parse: import("@orpc/contract").ContractProcedureBuilderWit
         max: "max";
         auto: "auto";
     }>>>;
+    includePages: z.ZodDefault<z.ZodOptional<z.ZodUnion<readonly [z.ZodBoolean, z.ZodPipe<z.ZodEnum<{
+        true: "true";
+        false: "false";
+    }>, z.ZodTransform<boolean, "true" | "false">>]>>>;
     callback: z.ZodOptional<z.ZodObject<{
         url: z.ZodURL;
         type: z.ZodOptional<z.ZodString>;
@@ -27,4 +31,8 @@ export declare const parse: import("@orpc/contract").ContractProcedureBuilderWit
     requestId: z.ZodNumber;
     html: z.ZodOptional<z.ZodString>;
     documentId: z.ZodOptional<z.ZodString>;
+    pages: z.ZodOptional<z.ZodArray<z.ZodObject<{
+        pageNumber: z.ZodNumber;
+        markdown: z.ZodString;
+    }, z.core.$strip>>>;
 }, z.core.$strip>, Record<never, never>, Record<never, never>>;

package/.tsc/lib/router/document/parse.js

@@ -1,6 +1,8 @@
 import { oc } from "@orpc/contract";
+import { documentParseOpenApiCodeSamples } from "@pdfvector/api-docs";
 import { z } from "zod";
 import { fetchableUrlSchema } from "../../fetchable-url-schema.js";
+import { pageMarkdownSchema } from "../../page-markdown-schema.js";
 import { pdfvectorModelSchema } from "../../pdfvector-model-schema.js";
 import { documentParseModelDescription, outputModelDescription, supportedFileFormatsDescription, supportedFileMimeErrorMessage, supportedFileMimes, supportedFileTypesLong, } from "../../supported-mimes.js";
 import { getDefaultSpec } from "./get-default-spec.js";
@@ -22,6 +24,14 @@ const parseInputSchema = z.object({
         .optional()
         .default("auto")
         .describe(documentParseModelDescription),
+    includePages: z
+        .union([
+        z.boolean(),
+        z.enum(["true", "false"]).transform((value) => value === "true"),
+    ])
+        .optional()
+        .default(false)
+        .describe("Set to true to include page-separated markdown in the pages array. Existing markdown output is still returned as the full document text."),
     callback: z
         .object({
         url: fetchableUrlSchema.describe("Webhook URL where results will be POSTed when processing completes"),
@@ -57,6 +67,10 @@ const parseOutputSchema = z.object({
         .string()
         .optional()
         .describe("Document ID if provided via x-pdfvector-document-id header"),
+    pages: z
+        .array(pageMarkdownSchema)
+        .optional()
+        .describe("Page-separated markdown content. Returned only when includePages is true."),
 });
 const requestExamples = {
     "Parse from URL": {
@@ -84,6 +98,13 @@ const requestExamples = {
             model: "nano",
         },
     },
+    "Parse with page-separated output": {
+        summary: "Parse with page-separated output",
+        value: {
+            url: "https://drive.google.com/file/d/13T04Yk20OwBNIDyvJJ3XlUg9WfOsmbjm/view?usp=share_link",
+            includePages: true,
+        },
+    },
     "Parse with small complex documents (pro)": {
         summary: "Parse with small complex documents (pro)",
         value: {
@@ -104,7 +125,11 @@ export const parse = oc
     summary: "Parse a document",
     description: `Extract text and page count from a document. Supports ${supportedFileTypesLong}. Provide the document via file upload, a public URL, or a base64-encoded string.`,
     tags: ["Document"],
-    spec: (op) => getDefaultSpec(op, requestExamples),
+    spec: (op) => {
+        const spec = getDefaultSpec(op, requestExamples);
+        spec["x-codeSamples"] = documentParseOpenApiCodeSamples;
+        return spec;
+    },
 })
     .input(parseInputSchema)
     .output(parseOutputSchema);

package/CHANGELOG.md

@@ -1,5 +1,77 @@
 # @pdfvector/instance-contract
 
+## 0.0.55
+### Patch Changes
+
+
+
+- [#273](https://github.com/phuctm97/pdfvector/pull/273) [`905025f`](https://github.com/phuctm97/pdfvector/commit/905025f926b760b5b1b7f6cfc876fc8557c70d46) Thanks [@khanhduyvt0101](https://github.com/khanhduyvt0101)! - Release npm packages with trusted publishing
+
+- Updated dependencies [[`905025f`](https://github.com/phuctm97/pdfvector/commit/905025f926b760b5b1b7f6cfc876fc8557c70d46)]:
+  - @pdfvector/api-docs@0.0.5
+  - @pdfvector/util@0.0.23
+
+## 0.0.54
+### Patch Changes
+
+
+
+- [#271](https://github.com/phuctm97/pdfvector/pull/271) [`054e4db`](https://github.com/phuctm97/pdfvector/commit/054e4dbeb712d5fe482c80f08082c33e09480942) Thanks [@khanhduyvt0101](https://github.com/khanhduyvt0101)! - Authenticate npm package release workflows
+
+- Updated dependencies [[`054e4db`](https://github.com/phuctm97/pdfvector/commit/054e4dbeb712d5fe482c80f08082c33e09480942)]:
+  - @pdfvector/api-docs@0.0.4
+
+## 0.0.53
+### Patch Changes
+
+
+
+- [#270](https://github.com/phuctm97/pdfvector/pull/270) [`e3503c4`](https://github.com/phuctm97/pdfvector/commit/e3503c4922384cad889e92cc2a6f5979b05544fe) Thanks [@khanhduyvt0101](https://github.com/khanhduyvt0101)! - Use npm publish for npm package release workflows
+
+- Updated dependencies [[`e3503c4`](https://github.com/phuctm97/pdfvector/commit/e3503c4922384cad889e92cc2a6f5979b05544fe)]:
+  - @pdfvector/api-docs@0.0.3
+
+## 0.0.52
+### Patch Changes
+
+
+
+- [#269](https://github.com/phuctm97/pdfvector/pull/269) [`11c2f38`](https://github.com/phuctm97/pdfvector/commit/11c2f3891ceaa9f051664598624a3e75aa5f096d) Thanks [@khanhduyvt0101](https://github.com/khanhduyvt0101)! - Fix npm package release publishing
+
+- Updated dependencies [[`11c2f38`](https://github.com/phuctm97/pdfvector/commit/11c2f3891ceaa9f051664598624a3e75aa5f096d)]:
+  - @pdfvector/api-docs@0.0.2
+
+## 0.0.51
+### Patch Changes
+
+
+
+- [#268](https://github.com/phuctm97/pdfvector/pull/268) [`37c6f35`](https://github.com/phuctm97/pdfvector/commit/37c6f35a6d2129e95a79fd44f7834a6e12df04ee) Thanks [@khanhduyvt0101](https://github.com/khanhduyvt0101)! - Add page-separated parse output
+
+- Updated dependencies [[`37c6f35`](https://github.com/phuctm97/pdfvector/commit/37c6f35a6d2129e95a79fd44f7834a6e12df04ee)]:
+  - @pdfvector/api-docs@0.0.1
+
+## 0.0.50
+### Patch Changes
+
+
+
+- [#267](https://github.com/phuctm97/pdfvector/pull/267) [`d6f79b3`](https://github.com/phuctm97/pdfvector/commit/d6f79b3d9a8bf5554e3f8dc0912468031ddb0c7a) Thanks [@khanhduyvt0101](https://github.com/khanhduyvt0101)! - Suppress image empty parse Sentry noise
+
+## 0.0.49
+### Patch Changes
+
+
+
+- [#260](https://github.com/phuctm97/pdfvector/pull/260) [`e24aa70`](https://github.com/phuctm97/pdfvector/commit/e24aa706cdd48a10efa13d3d613d63cebf921d16) Thanks [@khanhduyvt0101](https://github.com/khanhduyvt0101)! - Update workspace dependencies
+
+## 0.0.48
+### Patch Changes
+
+
+
+- [#250](https://github.com/phuctm97/pdfvector/pull/250) [`a2e6883`](https://github.com/phuctm97/pdfvector/commit/a2e68833d9f0dd6b38ea5b4b2a91aefb9f13aaf8) Thanks [@khanhduyvt0101](https://github.com/khanhduyvt0101)! - Upgrade workspace dependencies and adapt SPA navigation state typing for React Router 7.15.1.
+
 ## 0.0.47
 ### Patch Changes
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pdfvector/instance-contract",
-  "version": "0.0.47",
+  "version": "0.0.55",
   "type": "module",
   "description": "API contract definitions for PDF Vector instance server",
   "license": "MIT",
@@ -20,9 +20,10 @@
   },
   "main": ".tsc/lib/index.js",
   "dependencies": {
-    "@orpc/client": "^1.14.2",
-    "@orpc/contract": "^1.14.2",
-    "@pdfvector/util": "0.0.22"
+    "@orpc/client": "^1.14.5",
+    "@orpc/contract": "^1.14.5",
+    "@pdfvector/api-docs": "^0.0.5",
+    "@pdfvector/util": "^0.0.23"
   },
   "peerDependencies": {
     "zod": "^4.4.3"