Fabric Create report API

HeetParkhiya 

Here's a python code you can use to generate the payload. Create a legacy format PBIR file using Power BI Desktop first! The payload will be written to the root directory. 

You can use the "Try It" button and paste the payload in the below link using your Fabric/Power BI workspace GUID.

Items - Create Report - REST API (Report) | Microsoft Learn 

#!/usr/bin/env python3
# PBIR-Legacy payload validator + builder (Fabric Items API - Reports)

import os, sys, json, base64

# --- CONFIG: Point to your PBIR-legacy project root (contains definition.pbir and report.json) ---
PBIP_ROOT = r"C:\Legacy Report\legacy_report.Report"   # <-- UPDATE THIS to your legacy project folder

INCLUDE_PLATFORM = True                     # include .platform if present
INCLUDE_REGISTERED_RESOURCES = True         # include files under RegisteredResources/**
INCLUDE_STATIC_RESOURCES = True             # include files under StaticResources/**

# Optionally constrain resource file extensions to reduce payload bloat
RESOURCE_EXTENSIONS = {".json", ".jpg", ".jpeg", ".png", ".gif", ".pbiviz"}

def die(msg):
    print(f"ERROR: {msg}", file=sys.stderr)
    sys.exit(1)

def is_file(rel):
    return os.path.isfile(os.path.join(PBIP_ROOT, rel))

def must_file(rel):
    p = os.path.join(PBIP_ROOT, rel)
    if not os.path.isfile(p):
        die(f"Missing required file: {rel} at {p}")
    return rel

def b64(rel):
    with open(os.path.join(PBIP_ROOT, rel), "rb") as f:
        return base64.b64encode(f.read()).decode("utf-8")

def add_part(parts, rel):
    # Enforce forward slashes in 'path' for the API
    api_rel = rel.replace("\\", "/")
    parts.append({"path": api_rel, "payload": b64(rel), "payloadType": "InlineBase64"})
    print(f" + Added part: {api_rel}")

def walk_resources(root_rel, parts):
    """Add resource files beneath 'root_rel' (e.g., RegisteredResources or StaticResources)."""
    root_abs = os.path.join(PBIP_ROOT, root_rel)
    if not os.path.isdir(root_abs):
        return
    for dirpath, _, filenames in os.walk(root_abs):
        for name in filenames:
            ext = os.path.splitext(name)[1].lower()
            if RESOURCE_EXTENSIONS and ext not in RESOURCE_EXTENSIONS:
                continue
            # Build relative path that mirrors the on-disk tree from PBIP_ROOT
            rel_path = os.path.relpath(os.path.join(dirpath, name), PBIP_ROOT)
            add_part(parts, rel_path)

# --- Validate PBIR-legacy layout ---
print(f"Using PBIP_ROOT: {PBIP_ROOT}")
must_file("definition.pbir")
must_file("report.json")

# --- Build parts ---
parts = []
print("\nAdding PBIR-legacy core parts:")
add_part(parts, "definition.pbir")
add_part(parts, "report.json")

# Optional .platform
if INCLUDE_PLATFORM and is_file(".platform"):
    add_part(parts, ".platform")

# Optional RegisteredResources/** and StaticResources/**
# (themes, images, private custom visuals, etc.)
if INCLUDE_REGISTERED_RESOURCES:
    print("\nScanning RegisteredResources/**")
    walk_resources("RegisteredResources", parts)

if INCLUDE_STATIC_RESOURCES:
    print("\nScanning StaticResources/**")
    walk_resources("StaticResources", parts)

# --- Print summary and write payload.json ---
print("\nSummary:")
print(f" Total parts: {len(parts)}")
present_paths = [p["path"] for p in parts]
print(" Paths included:")
for p in present_paths:
    print("  -", p)

# Confirm the specific required paths exist in parts
if "definition.pbir" not in present_paths:
    die("CRITICAL: 'definition.pbir' was not included in parts[]")
if "report.json" not in present_paths:
    die("CRITICAL: 'report.json' was not included in parts[]")

payload = {"definition": {"parts": parts}}
out_path = os.path.join(PBIP_ROOT, "payload.json")
with open(out_path, "w", encoding="utf-8") as f:
    json.dump(payload, f, indent=2)
print(f"\nPayload written to: {out_path}")

Hope this helps - please appreciate leaving a Kudos or accepting as a Solution!  

There is no published full schema document listing every property for definition.pbir that you can download from microsoft; what exists is the JSON Schema endpoint URL (use VS Code for validation), and the effective expected content is extremely simple: essentially only these keys matter for REST deployment. What you need is this canonical structure (use this exact JSON with the correct schema URL and semantic model ID):

{
  "$schema": "https://developer.microsoft.com/json-schemas/fabric/item/report/definitionProperties/2.0.0/schema.json",
  "version": "4.0",
  "datasetReference": {
    "byConnection": {
      "connectionString": "semanticmodelid=[SemanticModelId]"
    }
  }
}

replace [SemanticModelId] with your valid semantic model GUID

the schema expects:

• $schema pointing to the published schema JSON (not truncated)

• version (4.0 is typical for reports deployed from modern PBI Desktop/Fabric)

• datasetReference.byConnection.connectionString with semanticmodelid=GUID

Other than those keys, the RESTAPI will reject anything extra or incomplete because it strictly validates against that schema. You cannot include additional keys outside what the schema defines for definition.pbir when creating via REST.  If you need to explore every property the schema defines, you can paste the $schema URL into a code editor with JSON Schema support to see the full fields (this is the source of truth)

HTH!

deborshi_nag, v-achippa, Vinodh247 Thank you above solution has worked and apologies it took longer then you are expecting.