Add report_prototype: auto-generate report pages from semantic model by KornAlexander · Pull Request #1155 · microsoft/semantic-link-labs

KornAlexander · 2026-04-09T16:27:18Z

Report Prototype Generator

Auto-generates a visual prototype of a Power BI report as SVG and Excalidraw format. Analyzes the report page layout, visual types, field mappings, and page structure to produce a schematic diagram — useful for report reviews, documentation, and restructuring discussions.

Functions Added

Function	Description
`generate_report_prototype(report, workspace=None, screenshots=False, include_hidden=False, cols=4, thumb_width=480, thumb_height=270, on_progress=None)`	Generates a visual prototype of a Power BI report as SVG + Excalidraw.

Files

src/sempy_labs/report/_report_prototype.py (new file)
src/sempy_labs/report/__init__.py (updated exports)

Usage

import sempy_labs as labs

# Generate a report prototype
labs.report.generate_report_prototype("My Report")

# Include hidden pages and use wider thumbnails
labs.report.generate_report_prototype("My Report", include_hidden=True, thumb_width=640)

PBI Fixer Contribution — Overview

This PR is part of the PBI Fixer contribution to semantic-link-labs — an interactive ipywidgets-based UI for scanning and fixing Power BI reports and semantic models directly in Microsoft Fabric Notebooks.

The PBI Fixer provides a tabbed ipywidgets interface (Semantic Model Explorer, Report Explorer, Perspective Editor, Vertipaq Analyzer) that lets users interactively scan, inspect, and fix Power BI artifacts without leaving the notebook. All underlying fixer functions also work as standalone API calls, so users can integrate them into scripts and pipelines without the UI.

Contribution Structure

The full contribution (~17K lines across 68 files) is split into 22 focused PRs across 6 phases to keep each PR reviewable and self-contained. Only new files are added in Phases 1–4 and 6 — no existing SLL code is modified.

Phase	Focus	PRs	Description
1	Report Fixers	7	Standalone functions that programmatically fix common Power BI report issues: replace pie charts with bar charts, standardize page sizes to Full HD, apply chart formatting best practices, migrate slicers to slicerbars, hide visual-level filters, clean up unused custom visuals, align visuals, migrate report-level measures, and upgrade reports to PBIR format. Each function operates on PBIR-format report definitions.
2	Semantic Model Fixers	4	Functions that fix and enhance semantic models via XMLA/TOM: add calculated calendar and measure tables, add calculation groups for units and time intelligence, discourage implicit measures, and 19 BPA auto-fixers covering formatting conventions, naming standards, data types, column visibility, sort order, and DAX patterns (e.g., use DIVIDE instead of `/`).
3	SM Setup & Analysis	3	Setup utilities: configure cache warming queries, set up incremental refresh policies, and prepare semantic models for AI/Copilot integration (descriptions, metadata enrichment).
4	Report Utilities	3	Report-level utilities: auto-generate report page prototypes from a semantic model's structure, extract and apply report themes, and generate IBCS-compliant variance charts.
5	Upstream Enhancements	3	⚠️ These PRs modify existing SLL code (unlike Phases 1–4 which only add new files). Changes include TOM model `.Find()` fixes and expression capture (`tom/_model.py`), Vertipaq analyzer enhancements with memory/column-level analysis (`_vertipaq.py`, ~1000 lines changed), and various small fixes across `_items.py`, `_item_recovery.py`, `_helper_functions.py`, `_export_report.py`, `_sql.py`, and `admin/_tenant.py`. These carry higher merge conflict risk and may need closer review or discussion.
6	PBI Fixer UI	2	The interactive UI layer: shared UI components (theme, icons, tree builders, layout helpers), BPA scan runners, report helpers, and the main PBI Fixer application with its tabbed interface (SM Explorer, Report Explorer, Perspective Editor, Vertipaq Analyzer). Depends on Phases 1–5 but uses lazy imports to degrade gracefully if individual fixers aren't yet merged.

Dependencies & Review Order

Phases 1–4 are fully independent — they only add new files and can be reviewed/merged in any order.
Phase 5 is also independent but modifies existing code, so it may benefit from early discussion.
Phase 6 (the UI) ties everything together. It depends on the earlier phases but works standalone via lazy imports.
All fixer functions work without the UI — they can be called directly as sempy_labs.report.fix_piecharts(...) or sempy_labs.semantic_model.add_calculated_calendar(...).

…tic model structure

Copilot

Pull request overview

Adds a new report utility module intended to generate a schematic “report prototype” (SVG + Excalidraw JSON) from a Power BI report’s page/visual metadata, optionally embedding exported page screenshots.

Changes:

Introduces generate_report_prototype(...) to analyze pages and navigation links and produce SVG + Excalidraw outputs.
Adds optional parallel screenshot exporting and embedding as base64 PNGs in the Excalidraw output.
Adds save_report_prototype(...) to persist the generated artifacts into the attached lakehouse.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-04-09T16:31:56Z

+                png_bytes = export_report(
+                    report=report,
+                    export_format="PNG",
+                    file_name=f"_prototype_{idx:02d}",
+                    page_name=pg["name"],
+                    workspace=workspace,
+                    _report_id=_resolved_report_id,
+                    _return_bytes=True,
+                )


export_report is invoked with _report_id and _return_bytes, but sempy_labs.report.export_report() does not accept these parameters. As written, screenshots=True will raise TypeError before any export happens. Consider either extending export_report with a supported return_bytes/report_id option (and using public parameters only here), or call the underlying API directly from this module without private kwargs.

Copilot · 2026-04-09T16:31:57Z

+        def _export_one(idx, pg):
+            # Suppress all output inside each thread
+            import sys as _tsys
+            import io as _tio
+            _t_old_stdout = _tsys.stdout
+            _tsys.stdout = _tio.StringIO()
+            try:
+                import IPython.display as _tipd
+                _tipd_orig = _tipd.display
+                _tipd.display = lambda *a, **kw: None
+            except Exception:
+                _tipd = None
+                _tipd_orig = None
+            try:
+                # Win 3: get PNG bytes directly, skip file I/O
+                png_bytes = export_report(
+                    report=report,
+                    export_format="PNG",
+                    file_name=f"_prototype_{idx:02d}",
+                    page_name=pg["name"],
+                    workspace=workspace,
+                    _report_id=_resolved_report_id,
+                    _return_bytes=True,
+                )
+                if png_bytes:
+                    with _lock:
+                        page_images[pg["name"]] = base64.b64encode(png_bytes).decode("ascii")
+                else:
+                    with _lock:
+                        export_errors.append(f"'{pg['display_name']}': empty response")
+            except Exception as e:
+                with _lock:
+                    export_errors.append(f"'{pg['display_name']}': {str(e)[:200]}")
+            finally:
+                _tsys.stdout = _t_old_stdout
+                if _tipd and _tipd_orig:
+                    _tipd.display = _tipd_orig
+                with _lock:
+                    _done_count[0] += 1
+                    done = _done_count[0]
+                if on_progress:
+                    try:
+                        on_progress(done, total_pages, pg["display_name"])
+                    except Exception:
+                        pass
+
+        # Redirect stdout AND suppress IPython.display to prevent notebook output overflow
+        _real_stdout = sys.stdout
+        sys.stdout = _io.StringIO()
+
+        # Monkey-patch IPython display to a no-op during exports
+        _ipd = None
+        _ipd_orig = None
+        _idf = None
+        _idf_orig = None
+        try:
+            import IPython.display as _ipd_mod
+            _ipd = _ipd_mod
+            _ipd_orig = _ipd_mod.display
+            _ipd_mod.display = lambda *a, **kw: None
+        except Exception:
+            pass
+        try:
+            import IPython.core.display_functions as _idf_mod
+            _idf = _idf_mod
+            _idf_orig = _idf_mod.display
+            _idf_mod.display = lambda *a, **kw: None
+        except Exception:
+            pass
+
+        try:
+            if on_progress:
+                on_progress(0, total_pages, "starting exports...")
+            from concurrent.futures import ThreadPoolExecutor
+            with ThreadPoolExecutor(max_workers=min(5, total_pages)) as pool:
+                pool.map(lambda args: _export_one(*args), target_pages)
+        finally:
+            sys.stdout = _real_stdout
+            if _ipd and _ipd_orig:
+                _ipd.display = _ipd_orig
+            if _idf and _idf_orig:
+                _idf.display = _idf_orig
+


The screenshot export path monkey-patches sys.stdout and IPython.display.display globally while running a ThreadPoolExecutor. These globals are shared across threads, so concurrent calls can race and leak side effects to other notebook code (and threads can restore the wrong stdout/display). Prefer a thread-safe approach (e.g., avoid global monkey-patching; add a quiet mode to export_report/API calls, or use contextlib.redirect_stdout + locks around any global state changes).

Copilot · 2026-04-09T16:31:57Z

+            if on_progress:
+                on_progress(0, total_pages, "starting exports...")
+            from concurrent.futures import ThreadPoolExecutor
+            with ThreadPoolExecutor(max_workers=min(5, total_pages)) as pool:
+                pool.map(lambda args: _export_one(*args), target_pages)


If include_hidden is false and all pages are hidden (or the report has zero pages after filtering), total_pages becomes 0 and ThreadPoolExecutor(max_workers=min(5, total_pages)) will be created with max_workers=0, raising ValueError. Add an early return/skip when total_pages == 0.

Suggested change

if on_progress:

on_progress(0, total_pages, "starting exports...")

from concurrent.futures import ThreadPoolExecutor

with ThreadPoolExecutor(max_workers=min(5, total_pages)) as pool:

pool.map(lambda args: _export_one(*args), target_pages)

if total_pages == 0:

if on_progress:

on_progress(0, total_pages, "no pages to export")

else:

if on_progress:

on_progress(0, total_pages, "starting exports...")

from concurrent.futures import ThreadPoolExecutor

with ThreadPoolExecutor(max_workers=min(5, total_pages)) as pool:

pool.map(lambda args: _export_one(*args), target_pages)

Copilot · 2026-04-09T16:31:58Z

+def generate_report_prototype(
+    report: str,
+    workspace: Optional[str | UUID] = None,
+    screenshots: bool = False,
+    include_hidden: bool = False,
+    cols: int = 4,
+    thumb_width: int = 480,
+    thumb_height: int = 270,
+    on_progress=None,


The public parameters cols, thumb_width, and thumb_height are passed into _build_diagram, but _build_diagram currently uses hard-coded IMG_W/IMG_H/COL_W/SLOT_GAP constants and does not use the provided values. This makes the API misleading (callers cannot actually control the output size/layout). Either wire these parameters into the layout calculations or remove them from the public signature/docstring.

Copilot · 2026-04-09T16:31:58Z

+    )
+
+    safe_name = report.replace(" ", "_").replace("/", "_")
+    from sempy_labs._helper_functions import _mount


save_report_prototype writes into the attached lakehouse via _mount() but does not check lakehouse_attached() first (unlike other report functions). This can surface a less actionable exception when no lakehouse is attached. Add the standard lakehouse_attached() guard and raise a ValueError with the usual user-facing message.

Suggested change

from sempy_labs._helper_functions import _mount

from sempy_labs._helper_functions import _mount, lakehouse_attached

if not lakehouse_attached():

raise ValueError(

"A lakehouse must be attached to the notebook to save the report prototype."

)

Copilot · 2026-04-09T16:31:58Z

+    safe_name = report.replace(" ", "_").replace("/", "_")
+    from sempy_labs._helper_functions import _mount
+    local_path = _mount()
+
+    exc_path = f"{local_path}/Files/{safe_name}_prototype.excalidraw"
+    with open(exc_path, "w", encoding="utf-8") as f:
+        f.write(result["excalidraw"])
+    print(f"\u2713 Saved {safe_name}_prototype.excalidraw")
+
+    svg_path = f"{local_path}/Files/{safe_name}_prototype.svg"
+    with open(svg_path, "w", encoding="utf-8") as f:
+        f.write(result["svg"])


safe_name only replaces spaces and /. A user-supplied report name can still contain other path separators (e.g., .., \\) or characters that produce unintended file paths. Consider sanitizing more robustly (e.g., allowlist [A-Za-z0-9._-], replace everything else, and use os.path.join) to ensure the output is always written under Files/.

Copilot · 2026-04-09T16:31:58Z

+def generate_report_prototype(
+    report: str,
+    workspace: Optional[str | UUID] = None,
+    screenshots: bool = False,
+    include_hidden: bool = False,
+    cols: int = 4,
+    thumb_width: int = 480,
+    thumb_height: int = 270,
+    on_progress=None,
+) -> dict:
+    """


This module introduces public functions but does not use the @log decorator pattern that is consistently applied across other sempy_labs.report top-level APIs. If logging/instrumentation is expected for report functions, decorate these public entrypoints as well for consistency.

Copilot · 2026-04-09T16:31:59Z

+import base64
+import json
+import os
+import uuid
+from typing import Optional
+from uuid import UUID
+
+
+def generate_report_prototype(
+    report: str,
+    workspace: Optional[str | UUID] = None,
+    screenshots: bool = False,
+    include_hidden: bool = False,
+    cols: int = 4,
+    thumb_width: int = 480,
+    thumb_height: int = 270,
+    on_progress=None,
+) -> dict:
+    """
+    Generates a visual prototype of a Power BI report as SVG + Excalidraw.
+
+    Parameters
+    ----------
+    report : str
+        Name of the Power BI report.
+    workspace : str | uuid.UUID, default=None
+        The Fabric workspace name or ID.
+    screenshots : bool, default=False
+        If True, exports each page as PNG via the Export API and embeds as images.
+    include_hidden : bool, default=False
+        If True, includes hidden pages in the diagram.
+    cols : int, default=4
+        Number of columns in the page grid layout.
+    thumb_width : int, default=480
+        Width of each page thumbnail in the diagram.
+    thumb_height : int, default=270
+        Height of each page thumbnail in the diagram.
+
+    Returns
+    -------
+    dict
+        A dictionary with keys:
+        - "svg": str — the SVG diagram as a string
+        - "excalidraw": str — the Excalidraw JSON as a string
+        - "pages": list — page metadata
+        - "screenshots": int — number of screenshots captured
+        - "errors": list — export error messages
+    """
+    from sempy_labs.report import connect_report
+
+    # Layout constants
+    pad_x = 40
+    pad_y = 60
+    header_h = 30
+    footer_h = 25
+


Unused imports/variables add noise: os is imported but not used, and layout constants pad_x/pad_y/header_h/footer_h are defined/passed but not used by _build_diagram. Removing these (or actually using them) will reduce confusion about which knobs affect the output.

Copilot · 2026-04-09T16:31:59Z

+def generate_report_prototype(
+    report: str,
+    workspace: Optional[str | UUID] = None,
+    screenshots: bool = False,
+    include_hidden: bool = False,
+    cols: int = 4,
+    thumb_width: int = 480,
+    thumb_height: int = 270,
+    on_progress=None,
+) -> dict:


PR description mentions updating src/sempy_labs/report/__init__.py to export generate_report_prototype, but the PR diff provided only adds this new module. As-is, users won’t be able to call labs.report.generate_report_prototype(...) via package exports; please ensure the report package exports are updated in this PR (or adjust the documented usage accordingly).

Add report_prototype: auto-generate report page prototypes from seman…

0dfd93f

…tic model structure

Copilot AI review requested due to automatic review settings April 9, 2026 16:27

Copilot started reviewing on behalf of KornAlexander April 9, 2026 16:28 View session

Copilot AI reviewed Apr 9, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add report_prototype: auto-generate report pages from semantic model#1155

Add report_prototype: auto-generate report pages from semantic model#1155
KornAlexander wants to merge 1 commit intomicrosoft:mainfrom
KornAlexander:feature/report-prototype

KornAlexander commented Apr 9, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Apr 9, 2026

Uh oh!

Copilot AI Apr 9, 2026

Uh oh!

Copilot AI Apr 9, 2026

Uh oh!

Copilot AI Apr 9, 2026

Uh oh!

Copilot AI Apr 9, 2026

Uh oh!

Copilot AI Apr 9, 2026

Uh oh!

Copilot AI Apr 9, 2026

Uh oh!

Copilot AI Apr 9, 2026

Uh oh!

Copilot AI Apr 9, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

-    from sempy_labs._helper_functions import _mount
+    from sempy_labs._helper_functions import _mount, lakehouse_attached
+    if not lakehouse_attached():
+        raise ValueError(
+            "A lakehouse must be attached to the notebook to save the report prototype."
+        )

Conversation

KornAlexander commented Apr 9, 2026

Report Prototype Generator

Functions Added

Files

Usage

PBI Fixer Contribution — Overview

Contribution Structure

Dependencies & Review Order

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI Apr 9, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 9, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 9, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 9, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 9, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 9, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 9, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 9, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 9, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants