Skip to content

Releases: Tusharp21/scan_me

Release list

Version-2.3.1

Version-2.3.1 Latest
Latest

Choose a tag to compare

View all tags
@Tusharp21 Tusharp21 released this 12 Jun 06:28
v2.3.1
252f003

Scan Me — Release Notes

All previous releases are on GitHub: https://github.com/Tusharp21/scan_me/releases

v2.3.1

Chromium system-library dependencies declared for Frappe Cloud + clearer failure handling.

🛠 Fixes

  • PDF generation failed on servers missing Chromium's system libraries. The bundled chrome-headless-shell binary launches but exits immediately (code 127) when host libraries such as libatk-1.0.so.0, libcups, libgbm, or libnss3 are absent. Playwright surfaced this only as an opaque TargetClosedError. generator.py now wraps the launch in _launch_chromium(), detects the shared-library failure, writes the full diagnostic to the Error Log, and raises a clean frappe.ValidationError. The outer handler re-raises that message instead of masking it with the generic "PDF generation failed."

📦 Deployment

  • [deploy.dependencies.apt] added to pyproject.toml. Declares the Chromium runtime libraries (libatk1.0-0, libatk-bridge2.0-0, libcups2, libgbm1, libnss3, etc.). Frappe Cloud (Press) reads this table when it builds the deploy image and installs the packages automatically — so a fresh deploy provisions the libraries before the app runs. It is not consumed by bench migrate or bench build (asset build); it is an image-build step. Self-hosted hosts still install the equivalent libraries at the OS level.

🔐 Audit

  • The user-facing failure message contains no shell/install commands — it reads "PDF generation is temporarily unavailable. Please contact your administrator." The actionable detail lives in the Error Log, keeping the Frappe Cloud "install instructions" audit clean.

🔁 No behaviour changes

  • No schema migrations, no settings changes. Existing print flows are unaffected on hosts that already have the libraries.

v2.3.0

Page orientation in Advanced Print + Frappe Cloud audit follow-ups.

✨ New

  • Portrait / Landscape orientation in the Advanced Print page. A new Orientation selector under Print Setup rotates the rendered A4 page; the live preview re-renders on change like every other option. Plumbed end-to-end: orientation is whitelisted in the options schema (api/pdf/options.py), resolved to Playwright's landscape flag in generator.py, and applied to every page.pdf() render path in assembly.py (all-pages, none, reference, and per-segment renders use the same value so page counts and breaks stay consistent).
  • Wider landscape preview. The preview iframe widens from 900px to 1240px (≈ the 1.41× A4 aspect ratio) when Landscape is selected, with a smooth max-width transition, so the rotated page fills the main column instead of staying boxed at portrait width. The mobile breakpoint still wins on narrow screens.

🛠 Fixes

  • Deprecated cur_frm removed. scan_me_settings.js called render_cert_status(cur_frm) from inside the cert-install callback (which had no frm in scope). The form reference is now carried on the dialog (d._frm) and used directly — clears the Semgrep frappe-cur-frm-usage correctness finding.

🔐 Security

  • Semgrep file-traversal findings resolved (false positives). The two new cert-read sites in api/admin.py (_read_pfx_bytes, cert_status_info) read paths that are already realpath-validated against the site's private-files root by _resolve_uploaded_file() / _safe_pfx_path(). Annotated both with # nosemgrep: frappe-security-file-traversal plus rationale, matching the existing directive on the cert-write path.

📦 Marketplace

  • Long Description. The "long description contains install instructions" audit finding is cleared in the listing copy (no bench setup/bench install text). Canonical marketplace copy is now version-controlled in MARKETPLACE_LONG_DESCRIPTION.md — paste it into the Frappe Cloud listing; do not paste DOCUMENTATION.md / RELEASE_NOTES.md (they contain bench install commands by design and would re-trip the check).

🔁 No behaviour changes for existing options

Additive feature only. No schema migrations; orientation defaults to Portrait, preserving current output for every existing print flow.

v2.2.0

Cert management UI + invisible PAdES signatures + audit follow-ups.

✨ New

  • Signing & Certificates tab in Scan Me Settings with:
    • Status panel showing the active signing cert at a glance: 🔴 self-signed (recipients see yellow ⚠), 🟡 CA-issued expiring within 30 days, 🟢 CA-issued healthy, ⚪ no cert installed. Surfaces subject CN, issuer, validity window, days-to-expiry, SHA-256 fingerprint, and a recognised-CA hint for AATL-listed issuers (Sectigo, GlobalSign, eMudhra, DigiCert, etc.).
    • Upload Production Certificate button — Frappe Attach-based dialog with a validate-then-install flow. Server-side preview shows cert metadata plus warnings (expired, expiring soon, missing digital_signature key usage, self-signed) before the admin commits.
    • Regenerate Self-Signed Certificate button — confirms before wiping the auto-generated PFX so a fresh self-signed cert is created on the next PAdES-signed PDF.
  • Signing Certificate Display Name setting — overrides the Subject CN/O baked into auto-generated self-signed certificates. Falls back to Global Defaults → Default Company, then to the site name when blank.
  • In-print signed-PDF indicator — the Advanced Print page shows a blue 🔒 "Final PDF will be cryptographically signed (PAdES). Adobe Reader will show the signature panel — no visual stamp is drawn on the page." hint whenever Apply Signature is on, signature_type involves PAdES, and the master switch is enabled.

🛠 Fixes

  • Invisible PAdES signatures. sign_pdf() previously stamped a PyHanko TextStampStyle at the bottom-right of the last page, which collided with the HTML signature card rendered by signature.py. Now the PAdES signature field has no visual appearance — Adobe Reader, Foxit, and Chrome PDF viewer continue to surface the signature in their Signature Panel UI, and the on-page signature card stands alone. Matches enterprise PAdES practice (DocuSign, Adobe Sign).
  • PLAYWRIGHT_BROWSERS_PATH resolved one directory too high. generator.py:_ensure_browsers_path() computed <bench>/apps/playwright-browsers from a 4-level relative climb out of api/pdf/, but install.py writes Chromium to <bench>/playwright-browsers/. PDFs failed at sync_playwright() with an "executable doesn't exist"–style error. Switched to frappe.utils.get_bench_path() so installer and generator can never drift apart again.
  • Semgrep nosemgrep directive now actually suppresses. The # nosemgrep: frappe-subprocess-exec comment in install.py was followed by a rationale-continuation line, so the immediate preceding line of subprocess.run( was no longer the directive. Reordered.
  • Long-description install-instruction audit finding. The Playwright-missing log at install.py:54 referenced bench setup requirements, which tripped the Frappe Cloud "long description contains install instructions" check (regex bench\s+setup). Rephrased to "Playwright module is missing — reinstall scan_me to refresh dependencies."

🔐 Security

  • New scan_me/api/admin.py — five hardened endpoints, all System Manager only, all POST-only with rate limits (validate 10/hr, install 5/hr, regenerate 5/hr, cancel 20/hr) and audit logging.
  • Realpath-based path validation against sites/<site>/private/files/ for every cert read/write; symlinks pointing outside are refused with ValidationError.
  • Atomic install with rollback — writes to .tmp + fsync + os.replace; backs up old PFX before overwriting; restores backup if update_site_config fails afterwards.
  • Strict input handling — 100 KB binary cap on uploaded PFX, base64 decoded with validate=True, password bytes zeroed after PKCS#12 parse.
  • Audit log — every install / regenerate writes an Error Log row with user, timestamp, fingerprint prefix (first 16 hex chars). Passwords are never logged.
  • Backup retention — keeps the most recent 3 timestamped signing.pfx.*.bak files; older backups are deleted (each contains a private key).
  • Mandatory cleanup of temporary uploads — install_cert_upload deletes the temp Frappe File doc in a finally block. The dialog's hidden.bs.modal handler calls cancel_cert_upload if the admin closes without installing, and a swap-attachment handler cleans up the previous URL before accepting a new one.

🧱 Internal

  • _company_name() in scan_me/utils/pades.py reads the new signing_cert_display_name setting first, falling back to Global Defaults.default_company, then to the site name.
  • get_dialog_settings() in scan_me_settings.py now also returns enable_pades_signing so the print page can decide whether to render the signed-PDF indicator.

📚 Docs

  • DSC_DESKTOP_APP_DESIGN.md (new) — solution design for an optional future desktop signer that integrates Indian Class 3 DSC USB tokens (ePass2003, ProxKey, mToken) with Scan Me. Covers architecture, technology stack rationale, cryptographic flow, server-side changes, security architecture, 4-phase delivery plan, cost summary, alternatives (Aadhaar eSign, emSigner Cloud SDK, organisational Document Signer Certificate).

🧹 Migration

  • Run bench migrate to pick up the new tab and signing_cert_display_name field. Existing settings are preserved — the new field defaults to blank, which keeps current behaviour (Default Company is used in self-signed cert subjects).
  • No data migrations. No backwards-incompatible changes.

v2.1.1

Patch release — public verify page rendered unstyled.

🛠 Fixes

  • Verify page CSS/JS 404. scan_me/www/verify_document/index.html referenced verify_document.css and verify_document.js with relative paths. Frappe serves the page at ...
Read more
Assets 2
Loading

Version-2.3.0

Version-2.3.0

Choose a tag to compare

View all tags
@Tusharp21 Tusharp21 released this 07 Jun 06:40
v2.3.0
05cae95

Scan Me — Release Notes

All previous releases are on GitHub: https://github.com/Tusharp21/scan_me/releases

v2.3.0

Page orientation in Advanced Print + Frappe Cloud audit follow-ups.

✨ New

  • Portrait / Landscape orientation in the Advanced Print page. A new Orientation selector under Print Setup rotates the rendered A4 page; the live preview re-renders on change like every other option. Plumbed end-to-end: orientation is whitelisted in the options schema (api/pdf/options.py), resolved to Playwright's landscape flag in generator.py, and applied to every page.pdf() render path in assembly.py (all-pages, none, reference, and per-segment renders use the same value so page counts and breaks stay consistent).
  • Wider landscape preview. The preview iframe widens from 900px to 1240px (≈ the 1.41× A4 aspect ratio) when Landscape is selected, with a smooth max-width transition, so the rotated page fills the main column instead of staying boxed at portrait width. The mobile breakpoint still wins on narrow screens.

🛠 Fixes

  • Deprecated cur_frm removed. scan_me_settings.js called render_cert_status(cur_frm) from inside the cert-install callback (which had no frm in scope). The form reference is now carried on the dialog (d._frm) and used directly — clears the Semgrep frappe-cur-frm-usage correctness finding.

🔐 Security

  • Semgrep file-traversal findings resolved (false positives). The two new cert-read sites in api/admin.py (_read_pfx_bytes, cert_status_info) read paths that are already realpath-validated against the site's private-files root by _resolve_uploaded_file() / _safe_pfx_path(). Annotated both with # nosemgrep: frappe-security-file-traversal plus rationale, matching the existing directive on the cert-write path.

📦 Marketplace

  • Long Description. The "long description contains install instructions" audit finding is cleared in the listing copy (no bench setup/bench install text). Canonical marketplace copy is now version-controlled in MARKETPLACE_LONG_DESCRIPTION.md — paste it into the Frappe Cloud listing; do not paste DOCUMENTATION.md / RELEASE_NOTES.md (they contain bench install commands by design and would re-trip the check).

🔁 No behaviour changes for existing options

Additive feature only. No schema migrations; orientation defaults to Portrait, preserving current output for every existing print flow.

v2.2.0

Cert management UI + invisible PAdES signatures + audit follow-ups.

✨ New

  • Signing & Certificates tab in Scan Me Settings with:
    • Status panel showing the active signing cert at a glance: 🔴 self-signed (recipients see yellow ⚠), 🟡 CA-issued expiring within 30 days, 🟢 CA-issued healthy, ⚪ no cert installed. Surfaces subject CN, issuer, validity window, days-to-expiry, SHA-256 fingerprint, and a recognised-CA hint for AATL-listed issuers (Sectigo, GlobalSign, eMudhra, DigiCert, etc.).
    • Upload Production Certificate button — Frappe Attach-based dialog with a validate-then-install flow. Server-side preview shows cert metadata plus warnings (expired, expiring soon, missing digital_signature key usage, self-signed) before the admin commits.
    • Regenerate Self-Signed Certificate button — confirms before wiping the auto-generated PFX so a fresh self-signed cert is created on the next PAdES-signed PDF.
  • Signing Certificate Display Name setting — overrides the Subject CN/O baked into auto-generated self-signed certificates. Falls back to Global Defaults → Default Company, then to the site name when blank.
  • In-print signed-PDF indicator — the Advanced Print page shows a blue 🔒 "Final PDF will be cryptographically signed (PAdES). Adobe Reader will show the signature panel — no visual stamp is drawn on the page." hint whenever Apply Signature is on, signature_type involves PAdES, and the master switch is enabled.

🛠 Fixes

  • Invisible PAdES signatures. sign_pdf() previously stamped a PyHanko TextStampStyle at the bottom-right of the last page, which collided with the HTML signature card rendered by signature.py. Now the PAdES signature field has no visual appearance — Adobe Reader, Foxit, and Chrome PDF viewer continue to surface the signature in their Signature Panel UI, and the on-page signature card stands alone. Matches enterprise PAdES practice (DocuSign, Adobe Sign).
  • PLAYWRIGHT_BROWSERS_PATH resolved one directory too high. generator.py:_ensure_browsers_path() computed <bench>/apps/playwright-browsers from a 4-level relative climb out of api/pdf/, but install.py writes Chromium to <bench>/playwright-browsers/. PDFs failed at sync_playwright() with an "executable doesn't exist"–style error. Switched to frappe.utils.get_bench_path() so installer and generator can never drift apart again.
  • Semgrep nosemgrep directive now actually suppresses. The # nosemgrep: frappe-subprocess-exec comment in install.py was followed by a rationale-continuation line, so the immediate preceding line of subprocess.run( was no longer the directive. Reordered.
  • Long-description install-instruction audit finding. The Playwright-missing log at install.py:54 referenced bench setup requirements, which tripped the Frappe Cloud "long description contains install instructions" check (regex bench\s+setup). Rephrased to "Playwright module is missing — reinstall scan_me to refresh dependencies."

🔐 Security

  • New scan_me/api/admin.py — five hardened endpoints, all System Manager only, all POST-only with rate limits (validate 10/hr, install 5/hr, regenerate 5/hr, cancel 20/hr) and audit logging.
  • Realpath-based path validation against sites/<site>/private/files/ for every cert read/write; symlinks pointing outside are refused with ValidationError.
  • Atomic install with rollback — writes to .tmp + fsync + os.replace; backs up old PFX before overwriting; restores backup if update_site_config fails afterwards.
  • Strict input handling — 100 KB binary cap on uploaded PFX, base64 decoded with validate=True, password bytes zeroed after PKCS#12 parse.
  • Audit log — every install / regenerate writes an Error Log row with user, timestamp, fingerprint prefix (first 16 hex chars). Passwords are never logged.
  • Backup retention — keeps the most recent 3 timestamped signing.pfx.*.bak files; older backups are deleted (each contains a private key).
  • Mandatory cleanup of temporary uploads — install_cert_upload deletes the temp Frappe File doc in a finally block. The dialog's hidden.bs.modal handler calls cancel_cert_upload if the admin closes without installing, and a swap-attachment handler cleans up the previous URL before accepting a new one.

🧱 Internal

  • _company_name() in scan_me/utils/pades.py reads the new signing_cert_display_name setting first, falling back to Global Defaults.default_company, then to the site name.
  • get_dialog_settings() in scan_me_settings.py now also returns enable_pades_signing so the print page can decide whether to render the signed-PDF indicator.

📚 Docs

  • DSC_DESKTOP_APP_DESIGN.md (new) — solution design for an optional future desktop signer that integrates Indian Class 3 DSC USB tokens (ePass2003, ProxKey, mToken) with Scan Me. Covers architecture, technology stack rationale, cryptographic flow, server-side changes, security architecture, 4-phase delivery plan, cost summary, alternatives (Aadhaar eSign, emSigner Cloud SDK, organisational Document Signer Certificate).

🧹 Migration

  • Run bench migrate to pick up the new tab and signing_cert_display_name field. Existing settings are preserved — the new field defaults to blank, which keeps current behaviour (Default Company is used in self-signed cert subjects).
  • No data migrations. No backwards-incompatible changes.

v2.1.1

Patch release — public verify page rendered unstyled.

🛠 Fixes

  • Verify page CSS/JS 404. scan_me/www/verify_document/index.html referenced verify_document.css and verify_document.js with relative paths. Frappe serves the page at /verify_document (no trailing slash), so the browser resolved them to /verify_document.css and /verify_document.js — both 404. Switched to absolute paths (/verify_document/verify_document.css, /verify_document/verify_document.js), which Frappe serves correctly. Page now loads with full styling and scanner JS as intended. CSP (style-src 'self' / script-src 'self') unchanged — same-origin so no policy update needed.

🔁 No behaviour changes

HTML-only fix. No schema changes, no migrations, no dependency bumps.

v2.1.0

Reliability + print-format author tooling.

🛠 Fixes

  • Chromium install pipeline rewrite — root-causes the "Executable doesn't exist" launch failure that hit benches after a Playwright pip upgrade left a stale chromium_headless_shell-* cache dir behind. Five mechanical changes:
    • Cache moved to {bench_path}/playwright-browsers/ (was user-global ~/.cache/ms-playwright/). Survives Playwright pip upgrades, one cache per bench, no ~/.cache pollution.
    • Cache validity now requires Playwright's INSTALLATION_COMPLETE marker file — not just directory existence. Stale and partially-downloaded dirs now trigger a fresh install instead of being silently reused.
    • _cleanup_partial() wipes marker-less Chromium dirs before retry so the next install starts clean.
    • New scan_me/__init__.py exports PLAYWRIGHT_BROWSERS_PATH so every web/worker process picks up the bench-local cache at import time.
    • Switched from chromium to chromium-headless-shell (Playwright's PDF-focused build — ~110 MB vs ~170 MB, same rendering engine).
  • Scan Me Settings allowlist now actually rejects Single + child-table doctypes. The previous row-level...
Read more
Loading

Version-2.2.0

Version-2.2.0

Choose a tag to compare

View all tags
@Tusharp21 Tusharp21 released this 24 May 12:42
v2.2.0
afd4e04

Scan Me — Release Notes

All previous releases are on GitHub: https://github.com/Tusharp21/scan_me/releases

v2.2.0

Cert management UI + invisible PAdES signatures + audit follow-ups.

✨ New

  • Signing & Certificates tab in Scan Me Settings with:
    • Status panel showing the active signing cert at a glance: 🔴 self-signed (recipients see yellow ⚠), 🟡 CA-issued expiring within 30 days, 🟢 CA-issued healthy, ⚪ no cert installed. Surfaces subject CN, issuer, validity window, days-to-expiry, SHA-256 fingerprint, and a recognised-CA hint for AATL-listed issuers (Sectigo, GlobalSign, eMudhra, DigiCert, etc.).
    • Upload Production Certificate button — Frappe Attach-based dialog with a validate-then-install flow. Server-side preview shows cert metadata plus warnings (expired, expiring soon, missing digital_signature key usage, self-signed) before the admin commits.
    • Regenerate Self-Signed Certificate button — confirms before wiping the auto-generated PFX so a fresh self-signed cert is created on the next PAdES-signed PDF.
  • Signing Certificate Display Name setting — overrides the Subject CN/O baked into auto-generated self-signed certificates. Falls back to Global Defaults → Default Company, then to the site name when blank.
  • In-print signed-PDF indicator — the Advanced Print page shows a blue 🔒 "Final PDF will be cryptographically signed (PAdES). Adobe Reader will show the signature panel — no visual stamp is drawn on the page." hint whenever Apply Signature is on, signature_type involves PAdES, and the master switch is enabled.

🛠 Fixes

  • Invisible PAdES signatures. sign_pdf() previously stamped a PyHanko TextStampStyle at the bottom-right of the last page, which collided with the HTML signature card rendered by signature.py. Now the PAdES signature field has no visual appearance — Adobe Reader, Foxit, and Chrome PDF viewer continue to surface the signature in their Signature Panel UI, and the on-page signature card stands alone. Matches enterprise PAdES practice (DocuSign, Adobe Sign).
  • PLAYWRIGHT_BROWSERS_PATH resolved one directory too high. generator.py:_ensure_browsers_path() computed <bench>/apps/playwright-browsers from a 4-level relative climb out of api/pdf/, but install.py writes Chromium to <bench>/playwright-browsers/. PDFs failed at sync_playwright() with an "executable doesn't exist"–style error. Switched to frappe.utils.get_bench_path() so installer and generator can never drift apart again.
  • Semgrep nosemgrep directive now actually suppresses. The # nosemgrep: frappe-subprocess-exec comment in install.py was followed by a rationale-continuation line, so the immediate preceding line of subprocess.run( was no longer the directive. Reordered.
  • Long-description install-instruction audit finding. The Playwright-missing log at install.py:54 referenced bench setup requirements, which tripped the Frappe Cloud "long description contains install instructions" check (regex bench\s+setup). Rephrased to "Playwright module is missing — reinstall scan_me to refresh dependencies."

🔐 Security

  • New scan_me/api/admin.py — five hardened endpoints, all System Manager only, all POST-only with rate limits (validate 10/hr, install 5/hr, regenerate 5/hr, cancel 20/hr) and audit logging.
  • Realpath-based path validation against sites/<site>/private/files/ for every cert read/write; symlinks pointing outside are refused with ValidationError.
  • Atomic install with rollback — writes to .tmp + fsync + os.replace; backs up old PFX before overwriting; restores backup if update_site_config fails afterwards.
  • Strict input handling — 100 KB binary cap on uploaded PFX, base64 decoded with validate=True, password bytes zeroed after PKCS#12 parse.
  • Audit log — every install / regenerate writes an Error Log row with user, timestamp, fingerprint prefix (first 16 hex chars). Passwords are never logged.
  • Backup retention — keeps the most recent 3 timestamped signing.pfx.*.bak files; older backups are deleted (each contains a private key).
  • Mandatory cleanup of temporary uploads — install_cert_upload deletes the temp Frappe File doc in a finally block. The dialog's hidden.bs.modal handler calls cancel_cert_upload if the admin closes without installing, and a swap-attachment handler cleans up the previous URL before accepting a new one.

🧱 Internal

  • _company_name() in scan_me/utils/pades.py reads the new signing_cert_display_name setting first, falling back to Global Defaults.default_company, then to the site name.
  • get_dialog_settings() in scan_me_settings.py now also returns enable_pades_signing so the print page can decide whether to render the signed-PDF indicator.

📚 Docs

  • DSC_DESKTOP_APP_DESIGN.md (new) — solution design for an optional future desktop signer that integrates Indian Class 3 DSC USB tokens (ePass2003, ProxKey, mToken) with Scan Me. Covers architecture, technology stack rationale, cryptographic flow, server-side changes, security architecture, 4-phase delivery plan, cost summary, alternatives (Aadhaar eSign, emSigner Cloud SDK, organisational Document Signer Certificate).

🧹 Migration

  • Run bench migrate to pick up the new tab and signing_cert_display_name field. Existing settings are preserved — the new field defaults to blank, which keeps current behaviour (Default Company is used in self-signed cert subjects).
  • No data migrations. No backwards-incompatible changes.

v2.1.1

Patch release — public verify page rendered unstyled.

🛠 Fixes

  • Verify page CSS/JS 404. scan_me/www/verify_document/index.html referenced verify_document.css and verify_document.js with relative paths. Frappe serves the page at /verify_document (no trailing slash), so the browser resolved them to /verify_document.css and /verify_document.js — both 404. Switched to absolute paths (/verify_document/verify_document.css, /verify_document/verify_document.js), which Frappe serves correctly. Page now loads with full styling and scanner JS as intended. CSP (style-src 'self' / script-src 'self') unchanged — same-origin so no policy update needed.

🔁 No behaviour changes

HTML-only fix. No schema changes, no migrations, no dependency bumps.

v2.1.0

Reliability + print-format author tooling.

🛠 Fixes

  • Chromium install pipeline rewrite — root-causes the "Executable doesn't exist" launch failure that hit benches after a Playwright pip upgrade left a stale chromium_headless_shell-* cache dir behind. Five mechanical changes:
    • Cache moved to {bench_path}/playwright-browsers/ (was user-global ~/.cache/ms-playwright/). Survives Playwright pip upgrades, one cache per bench, no ~/.cache pollution.
    • Cache validity now requires Playwright's INSTALLATION_COMPLETE marker file — not just directory existence. Stale and partially-downloaded dirs now trigger a fresh install instead of being silently reused.
    • _cleanup_partial() wipes marker-less Chromium dirs before retry so the next install starts clean.
    • New scan_me/__init__.py exports PLAYWRIGHT_BROWSERS_PATH so every web/worker process picks up the bench-local cache at import time.
    • Switched from chromium to chromium-headless-shell (Playwright's PDF-focused build — ~110 MB vs ~170 MB, same rendering engine).
  • Scan Me Settings allowlist now actually rejects Single + child-table doctypes. The previous row-level validate() was dead code (Frappe doesn't call children's validate() during the parent save flow) — moved the check to ScanMeSettings.validate() so it fires at the real entry point.

✨ New

  • PRINT_FORMAT_GUIDE.md — complete authoring guide for print formats: body-header/footer extraction (id="header-html" / id="footer-html"), utility classes (.page-break, .new-page, .no-split, .no-print, .screen-only, .breakable), edge-case CSS rules already in place, Jinja helpers, worked examples, common pitfalls.
  • Print CSS extended with seven edge-case rules (image overflow protection, orphans/widows on paragraphs and rows, list-item keep-whole, preformatted-block wrapping, caption-stays-with-table, long-string break-anywhere) and six opt-in utility classes — all backwards-compatible additions.

🧱 Internal

  • api/pdf.py refactored into api/pdf/ package (one file per feature: generator / assembly / attach / css / images / letterhead / options / pades / qr / signature / watermark). Public API path scan_me.api.pdf.generate_chrome_pdf preserved via package re-export.
  • Body-defined header/footer (id="header-html" / id="footer-html") now auto-promoted to repeating page header/footer with dynamic height measurement so Chromium reserves the right margin without manual tuning.

v2.0.2

Patch release — Frappe Cloud audit compliance.

🛠 Fixes

  • Replaced raw frappe.cache.get / setex / incrby calls in the per-UUID rate limiter (api/verify_document_qr.py) with multitenancy-safe get_value / set_value. Resolves the Semgrep frappe-cache-breaks-multitenancy finding from the Frappe Cloud audit. Behaviour preserved — UUIDs over 60 hits / hour still get blocked, with the cache entry now correctly site-prefixed on multi-tenant benches.
  • Removed remaining external links from the Marketplace long description per audit guidance.

🔁 No behaviour changes

This release is audit / fix only. No feature additions, no schema changes, no migrations.

v2.0.1

Patch release — Frappe Cloud audit compliance.

🛠 Fixes

  • Added [tool.bench.frappe-dependencies] section to pyproject.toml declaring supported Frappe range (>=14.0.0,<=17.0.0-dev).
  • Removed external links from the long description (README.md) per Frappe Cloud audit guidance.
  • Tightened the ...
Read more
Loading

Version-2.1.1

Version-2.1.1

Choose a tag to compare

View all tags
@Tusharp21 Tusharp21 released this 17 May 15:56
v2.1.1
b6300b4

Scan Me — Release Notes

All previous releases are on GitHub: https://github.com/Tusharp21/scan_me/releases

v2.1.1

Patch release — public verify page rendered unstyled.

🛠 Fixes

  • Verify page CSS/JS 404. scan_me/www/verify_document/index.html referenced verify_document.css and verify_document.js with relative paths. Frappe serves the page at /verify_document (no trailing slash), so the browser resolved them to /verify_document.css and /verify_document.js — both 404. Switched to absolute paths (/verify_document/verify_document.css, /verify_document/verify_document.js), which Frappe serves correctly. Page now loads with full styling and scanner JS as intended. CSP (style-src 'self' / script-src 'self') unchanged — same-origin so no policy update needed.

🔁 No behaviour changes

HTML-only fix. No schema changes, no migrations, no dependency bumps.

v2.1.0

Reliability + print-format author tooling.

🛠 Fixes

  • Chromium install pipeline rewrite — root-causes the "Executable doesn't exist" launch failure that hit benches after a Playwright pip upgrade left a stale chromium_headless_shell-* cache dir behind. Five mechanical changes:
    • Cache moved to {bench_path}/playwright-browsers/ (was user-global ~/.cache/ms-playwright/). Survives Playwright pip upgrades, one cache per bench, no ~/.cache pollution.
    • Cache validity now requires Playwright's INSTALLATION_COMPLETE marker file — not just directory existence. Stale and partially-downloaded dirs now trigger a fresh install instead of being silently reused.
    • _cleanup_partial() wipes marker-less Chromium dirs before retry so the next install starts clean.
    • New scan_me/__init__.py exports PLAYWRIGHT_BROWSERS_PATH so every web/worker process picks up the bench-local cache at import time.
    • Switched from chromium to chromium-headless-shell (Playwright's PDF-focused build — ~110 MB vs ~170 MB, same rendering engine).
  • Scan Me Settings allowlist now actually rejects Single + child-table doctypes. The previous row-level validate() was dead code (Frappe doesn't call children's validate() during the parent save flow) — moved the check to ScanMeSettings.validate() so it fires at the real entry point.

✨ New

  • PRINT_FORMAT_GUIDE.md — complete authoring guide for print formats: body-header/footer extraction (id="header-html" / id="footer-html"), utility classes (.page-break, .new-page, .no-split, .no-print, .screen-only, .breakable), edge-case CSS rules already in place, Jinja helpers, worked examples, common pitfalls.
  • Print CSS extended with seven edge-case rules (image overflow protection, orphans/widows on paragraphs and rows, list-item keep-whole, preformatted-block wrapping, caption-stays-with-table, long-string break-anywhere) and six opt-in utility classes — all backwards-compatible additions.

🧱 Internal

  • api/pdf.py refactored into api/pdf/ package (one file per feature: generator / assembly / attach / css / images / letterhead / options / pades / qr / signature / watermark). Public API path scan_me.api.pdf.generate_chrome_pdf preserved via package re-export.
  • Body-defined header/footer (id="header-html" / id="footer-html") now auto-promoted to repeating page header/footer with dynamic height measurement so Chromium reserves the right margin without manual tuning.

v2.0.2

Patch release — Frappe Cloud audit compliance.

🛠 Fixes

  • Replaced raw frappe.cache.get / setex / incrby calls in the per-UUID rate limiter (api/verify_document_qr.py) with multitenancy-safe get_value / set_value. Resolves the Semgrep frappe-cache-breaks-multitenancy finding from the Frappe Cloud audit. Behaviour preserved — UUIDs over 60 hits / hour still get blocked, with the cache entry now correctly site-prefixed on multi-tenant benches.
  • Removed remaining external links from the Marketplace long description per audit guidance.

🔁 No behaviour changes

This release is audit / fix only. No feature additions, no schema changes, no migrations.

v2.0.1

Patch release — Frappe Cloud audit compliance.

🛠 Fixes

  • Added [tool.bench.frappe-dependencies] section to pyproject.toml declaring supported Frappe range (>=14.0.0,<=17.0.0-dev).
  • Removed external links from the long description (README.md) per Frappe Cloud audit guidance.
  • Tightened the Advanced Print page role from All to System User, blocking Guest / Website User access while keeping all desk users (Sales User, Accountant, Stock User, etc.) functional.
  • Annotated the two intentional file-system call sites (api/pdf.py, utils/pades.py) with inline # nosemgrep: frappe-security-file-traversal markers — both are already gated by allow-list containment / fixed internal paths.

🔁 No behaviour changes

This release is audit / metadata only. No feature additions, no schema changes, no migrations.

v2.0.0

Major release — Chromium PDF pipeline, PAdES digital signatures, Advanced Print page, and a hardened Verified QR with tamper detection.

🚀 New features

🔐 Signing & Verification

  • Multi-signer support — same document can be signed by multiple users, each with their own Verified QR
  • Content hash (SHA-256) recorded at sign time — edits auto-invalidate the signature in PDFs and on the verify page
  • Permanent signature lock — Verified QR records can be locked so no one, including Administrator, can modify or delete
  • Stored-signature mode — server reads the user's saved signature instead of accepting client-supplied images
  • Signature Type selector — one setting controls Visual Block / Cryptographic (PAdES) / Both

🖨️ Advanced Print Page (/app/scan-me-print/{DocType}/{DocName})

  • Live PDF preview pane that re-renders on every option change (debounced)
  • Auto-fills Print Format + Letter Head defaults per doctype
  • Multi-copy (1–5) with optional comma-separated labels (Original / Duplicate / Triplicate / custom)
  • Header & footer repeat modes — All pages / First only / Last only / First + Last / None
  • QR / Barcode insertion at any corner of the PDF
  • Acrobat-style signature card with green VERIFIED / red INVALIDATED states
  • Watermark system — Document Status (auto DRAFT / CANCELLED / status) or Custom Text
  • Attach-to-Document option — saves the PDF as a private attachment on download
  • Breadcrumbs: DocType → DocName → Advanced Print

🧠 PAdES Digital Signatures (PyHanko)

  • Optional Adobe-compatible cryptographic PDF signatures
  • Self-signed certificate auto-generated on first use
  • Replace the PKCS#12 bundle with a CA-issued one for Adobe's green "Signed and all signatures are valid" badge

🌐 Public Verify Page (/verify_document/)

  • Redesigned mobile-first UI with dark-mode support
  • Three distinct result states — Verified, Signature Invalidated (tampered), Not Recognised
  • Admin-configurable detail level — Minimal / Standard (default) / Full — controls what guests see
  • Rate-limited to 30 requests per 60 seconds per IP

⚙️ Settings & Form Integration

  • Scan Me Settings is now a tabbed form — Allowed Documents / Verification & Signing / Public Verification / Print Dialog Features
  • New "Advanced Print" icon on document forms, placed immediately right of Frappe's native Print icon
  • Admin toggle to show/hide the Advanced Print icon

📄 Bundled print formats

  • Sales Invoice Scan Me — full-featured invoice with QR footer
  • SO — sample Sales Order format
  • Basic HTML Template by Scan Me — starter skeleton users can duplicate for any doctype

📝 Bundled Letter Head

  • Letter Head By Scan Me ships as a fixture for reference — admin can mark it default or customize as needed

🧰 Jinja helpers (for print format authors)

  • qr_img, qr_link_img, verify_qr, verify_qr_img — added alongside existing qr, barcode, qr_link

🧱 Dependencies added

  • playwright~=1.47 — Chromium PDF rendering
  • pypdf~=6.1.3 — multi-copy PDF merging
  • pyHanko~=0.25 — PAdES signing
  • cryptography>=41 — self-signed cert generation

📦 Installation

bench get-app https://github.com/Tusharp21/scan_me
bench install-app scan_me

Chromium for PDF rendering (~200 MB) downloads automatically via the
after_install hook. On restricted hosts where auto-download fails,
run ./env/bin/python -m playwright install chromium manually — the app will show
a clear message pointing to this fallback.

Existing installs: bench migrate runs 5 patches that backfill sensible
defaults for new settings fields and will download Chromium if it isn't
already cached.

⚠️ Known limitations

  • Self-signed PAdES cert shows as "signer unknown" in Adobe until replaced with a CA-issued PFX
  • Test suite is skeleton-only — real test coverage planned for 2.1

📚 Documentation

v1.0.2 — 06 Jan 2026

  • System-verified QR code generation with unique document IDs
  • Logo integration in QR codes
  • Signature requirement for QR generation
  • Doctype-based configuration via Scan Me Settings
  • Public verification support with ROOT_URL/verify_document/

Generate a unique system ID linked to a specific document, with verification URL and embedded logo.

v1.0.1 — 18 Nov 2025

  • Fixed line-level issue in jinja_functions.py
  • Updated function logic for accurate output rendering
  • Minor code cleanup for readability

This update focuses on improving reliability in template utilities without introducing any breaking changes.

v1.0.0 — 15 Nov 2025

  • Jinja functions for print formats: qr(), barcode(), qr_link()
  • Global JavaScript function add_qr_to_description()
  • QR/barcode ge...
Read more
Loading

Version-2.1.0

Version-2.1.0

Choose a tag to compare

View all tags
@Tusharp21 Tusharp21 released this 13 May 18:12
v2.1.0
53a66ca

Scan Me — Release Notes

All previous releases are on GitHub: https://github.com/Tusharp21/scan_me/releases

v2.1.0

Reliability + print-format author tooling.

🛠 Fixes

  • Chromium install pipeline rewrite — root-causes the "Executable doesn't exist" launch failure that hit benches after a Playwright pip upgrade left a stale chromium_headless_shell-* cache dir behind. Five mechanical changes:
    • Cache moved to {bench_path}/playwright-browsers/ (was user-global ~/.cache/ms-playwright/). Survives Playwright pip upgrades, one cache per bench, no ~/.cache pollution.
    • Cache validity now requires Playwright's INSTALLATION_COMPLETE marker file — not just directory existence. Stale and partially-downloaded dirs now trigger a fresh install instead of being silently reused.
    • _cleanup_partial() wipes marker-less Chromium dirs before retry so the next install starts clean.
    • New scan_me/__init__.py exports PLAYWRIGHT_BROWSERS_PATH so every web/worker process picks up the bench-local cache at import time.
    • Switched from chromium to chromium-headless-shell (Playwright's PDF-focused build — ~110 MB vs ~170 MB, same rendering engine).
  • Scan Me Settings allowlist now actually rejects Single + child-table doctypes. The previous row-level validate() was dead code (Frappe doesn't call children's validate() during the parent save flow) — moved the check to ScanMeSettings.validate() so it fires at the real entry point.

✨ New

  • PRINT_FORMAT_GUIDE.md — complete authoring guide for print formats: body-header/footer extraction (id="header-html" / id="footer-html"), utility classes (.page-break, .new-page, .no-split, .no-print, .screen-only, .breakable), edge-case CSS rules already in place, Jinja helpers, worked examples, common pitfalls.
  • Print CSS extended with seven edge-case rules (image overflow protection, orphans/widows on paragraphs and rows, list-item keep-whole, preformatted-block wrapping, caption-stays-with-table, long-string break-anywhere) and six opt-in utility classes — all backwards-compatible additions.

🧱 Internal

  • api/pdf.py refactored into api/pdf/ package (one file per feature: generator / assembly / attach / css / images / letterhead / options / pades / qr / signature / watermark). Public API path scan_me.api.pdf.generate_chrome_pdf preserved via package re-export.
  • Body-defined header/footer (id="header-html" / id="footer-html") now auto-promoted to repeating page header/footer with dynamic height measurement so Chromium reserves the right margin without manual tuning.

v2.0.2

Patch release — Frappe Cloud audit compliance.

🛠 Fixes

  • Replaced raw frappe.cache.get / setex / incrby calls in the per-UUID rate limiter (api/verify_document_qr.py) with multitenancy-safe get_value / set_value. Resolves the Semgrep frappe-cache-breaks-multitenancy finding from the Frappe Cloud audit. Behaviour preserved — UUIDs over 60 hits / hour still get blocked, with the cache entry now correctly site-prefixed on multi-tenant benches.
  • Removed remaining external links from the Marketplace long description per audit guidance.

🔁 No behaviour changes

This release is audit / fix only. No feature additions, no schema changes, no migrations.

v2.0.1

Patch release — Frappe Cloud audit compliance.

🛠 Fixes

  • Added [tool.bench.frappe-dependencies] section to pyproject.toml declaring supported Frappe range (>=14.0.0,<=17.0.0-dev).
  • Removed external links from the long description (README.md) per Frappe Cloud audit guidance.
  • Tightened the Advanced Print page role from All to System User, blocking Guest / Website User access while keeping all desk users (Sales User, Accountant, Stock User, etc.) functional.
  • Annotated the two intentional file-system call sites (api/pdf.py, utils/pades.py) with inline # nosemgrep: frappe-security-file-traversal markers — both are already gated by allow-list containment / fixed internal paths.

🔁 No behaviour changes

This release is audit / metadata only. No feature additions, no schema changes, no migrations.

v2.0.0

Major release — Chromium PDF pipeline, PAdES digital signatures, Advanced Print page, and a hardened Verified QR with tamper detection.

🚀 New features

🔐 Signing & Verification

  • Multi-signer support — same document can be signed by multiple users, each with their own Verified QR
  • Content hash (SHA-256) recorded at sign time — edits auto-invalidate the signature in PDFs and on the verify page
  • Permanent signature lock — Verified QR records can be locked so no one, including Administrator, can modify or delete
  • Stored-signature mode — server reads the user's saved signature instead of accepting client-supplied images
  • Signature Type selector — one setting controls Visual Block / Cryptographic (PAdES) / Both

🖨️ Advanced Print Page (/app/scan-me-print/{DocType}/{DocName})

  • Live PDF preview pane that re-renders on every option change (debounced)
  • Auto-fills Print Format + Letter Head defaults per doctype
  • Multi-copy (1–5) with optional comma-separated labels (Original / Duplicate / Triplicate / custom)
  • Header & footer repeat modes — All pages / First only / Last only / First + Last / None
  • QR / Barcode insertion at any corner of the PDF
  • Acrobat-style signature card with green VERIFIED / red INVALIDATED states
  • Watermark system — Document Status (auto DRAFT / CANCELLED / status) or Custom Text
  • Attach-to-Document option — saves the PDF as a private attachment on download
  • Breadcrumbs: DocType → DocName → Advanced Print

🧠 PAdES Digital Signatures (PyHanko)

  • Optional Adobe-compatible cryptographic PDF signatures
  • Self-signed certificate auto-generated on first use
  • Replace the PKCS#12 bundle with a CA-issued one for Adobe's green "Signed and all signatures are valid" badge

🌐 Public Verify Page (/verify_document/)

  • Redesigned mobile-first UI with dark-mode support
  • Three distinct result states — Verified, Signature Invalidated (tampered), Not Recognised
  • Admin-configurable detail level — Minimal / Standard (default) / Full — controls what guests see
  • Rate-limited to 30 requests per 60 seconds per IP

⚙️ Settings & Form Integration

  • Scan Me Settings is now a tabbed form — Allowed Documents / Verification & Signing / Public Verification / Print Dialog Features
  • New "Advanced Print" icon on document forms, placed immediately right of Frappe's native Print icon
  • Admin toggle to show/hide the Advanced Print icon

📄 Bundled print formats

  • Sales Invoice Scan Me — full-featured invoice with QR footer
  • SO — sample Sales Order format
  • Basic HTML Template by Scan Me — starter skeleton users can duplicate for any doctype

📝 Bundled Letter Head

  • Letter Head By Scan Me ships as a fixture for reference — admin can mark it default or customize as needed

🧰 Jinja helpers (for print format authors)

  • qr_img, qr_link_img, verify_qr, verify_qr_img — added alongside existing qr, barcode, qr_link

🧱 Dependencies added

  • playwright~=1.47 — Chromium PDF rendering
  • pypdf~=6.1.3 — multi-copy PDF merging
  • pyHanko~=0.25 — PAdES signing
  • cryptography>=41 — self-signed cert generation

📦 Installation

bench get-app https://github.com/Tusharp21/scan_me
bench install-app scan_me

Chromium for PDF rendering (~200 MB) downloads automatically via the
after_install hook. On restricted hosts where auto-download fails,
run ./env/bin/python -m playwright install chromium manually — the app will show
a clear message pointing to this fallback.

Existing installs: bench migrate runs 5 patches that backfill sensible
defaults for new settings fields and will download Chromium if it isn't
already cached.

⚠️ Known limitations

  • Self-signed PAdES cert shows as "signer unknown" in Adobe until replaced with a CA-issued PFX
  • Test suite is skeleton-only — real test coverage planned for 2.1

📚 Documentation

v1.0.2 — 06 Jan 2026

  • System-verified QR code generation with unique document IDs
  • Logo integration in QR codes
  • Signature requirement for QR generation
  • Doctype-based configuration via Scan Me Settings
  • Public verification support with ROOT_URL/verify_document/

Generate a unique system ID linked to a specific document, with verification URL and embedded logo.

v1.0.1 — 18 Nov 2025

  • Fixed line-level issue in jinja_functions.py
  • Updated function logic for accurate output rendering
  • Minor code cleanup for readability

This update focuses on improving reliability in template utilities without introducing any breaking changes.

v1.0.0 — 15 Nov 2025

  • Jinja functions for print formats: qr(), barcode(), qr_link()
  • Global JavaScript function add_qr_to_description()
  • QR/barcode generation in desk forms and print formats
  • Auto-loading via hooks.py configuration
Loading

Version-2.0.2

Version-2.0.2

Choose a tag to compare

View all tags
@Tusharp21 Tusharp21 released this 26 Apr 17:36
v2.0.2
27080f3

Scan Me — Release Notes

All previous releases are on GitHub: https://github.com/Tusharp21/scan_me/releases

v2.0.2

Patch release — Frappe Cloud audit compliance.

🛠 Fixes

  • Replaced raw frappe.cache.get / setex / incrby calls in the per-UUID rate limiter (api/verify_document_qr.py) with multitenancy-safe get_value / set_value. Resolves the Semgrep frappe-cache-breaks-multitenancy finding from the Frappe Cloud audit. Behaviour preserved — UUIDs over 60 hits / hour still get blocked, with the cache entry now correctly site-prefixed on multi-tenant benches.

🔁 No behaviour changes

This release is audit / fix only. No feature additions, no schema changes, no migrations.

v2.0.1

Patch release — Frappe Cloud audit compliance.

🛠 Fixes

  • Added [tool.bench.frappe-dependencies] section to pyproject.toml declaring supported Frappe range (>=14.0.0,<=17.0.0-dev).
  • Removed external links from the long description (README.md) per Frappe Cloud audit guidance.
  • Tightened the Advanced Print page role from All to System User, blocking Guest / Website User access while keeping all desk users (Sales User, Accountant, Stock User, etc.) functional.
  • Annotated the two intentional file-system call sites (api/pdf.py, utils/pades.py) with inline # nosemgrep: frappe-security-file-traversal markers — both are already gated by allow-list containment / fixed internal paths.

🔁 No behaviour changes

This release is audit / metadata only. No feature additions, no schema changes, no migrations.

v2.0.0

Major release — Chromium PDF pipeline, PAdES digital signatures, Advanced Print page, and a hardened Verified QR with tamper detection.

🚀 New features

🔐 Signing & Verification

  • Multi-signer support — same document can be signed by multiple users, each with their own Verified QR
  • Content hash (SHA-256) recorded at sign time — edits auto-invalidate the signature in PDFs and on the verify page
  • Permanent signature lock — Verified QR records can be locked so no one, including Administrator, can modify or delete
  • Stored-signature mode — server reads the user's saved signature instead of accepting client-supplied images
  • Signature Type selector — one setting controls Visual Block / Cryptographic (PAdES) / Both

🖨️ Advanced Print Page (/app/scan-me-print/{DocType}/{DocName})

  • Live PDF preview pane that re-renders on every option change (debounced)
  • Auto-fills Print Format + Letter Head defaults per doctype
  • Multi-copy (1–5) with optional comma-separated labels (Original / Duplicate / Triplicate / custom)
  • Header & footer repeat modes — All pages / First only / Last only / First + Last / None
  • QR / Barcode insertion at any corner of the PDF
  • Acrobat-style signature card with green VERIFIED / red INVALIDATED states
  • Watermark system — Document Status (auto DRAFT / CANCELLED / status) or Custom Text
  • Attach-to-Document option — saves the PDF as a private attachment on download
  • Breadcrumbs: DocType → DocName → Advanced Print

🧠 PAdES Digital Signatures (PyHanko)

  • Optional Adobe-compatible cryptographic PDF signatures
  • Self-signed certificate auto-generated on first use
  • Replace the PKCS#12 bundle with a CA-issued one for Adobe's green "Signed and all signatures are valid" badge

🌐 Public Verify Page (/verify_document/)

  • Redesigned mobile-first UI with dark-mode support
  • Three distinct result states — Verified, Signature Invalidated (tampered), Not Recognised
  • Admin-configurable detail level — Minimal / Standard (default) / Full — controls what guests see
  • Rate-limited to 30 requests per 60 seconds per IP

⚙️ Settings & Form Integration

  • Scan Me Settings is now a tabbed form — Allowed Documents / Verification & Signing / Public Verification / Print Dialog Features
  • New "Advanced Print" icon on document forms, placed immediately right of Frappe's native Print icon
  • Admin toggle to show/hide the Advanced Print icon

📄 Bundled print formats

  • Sales Invoice Scan Me — full-featured invoice with QR footer
  • SO — sample Sales Order format
  • Basic HTML Template by Scan Me — starter skeleton users can duplicate for any doctype

📝 Bundled Letter Head

  • Letter Head By Scan Me ships as a fixture for reference — admin can mark it default or customize as needed

🧰 Jinja helpers (for print format authors)

  • qr_img, qr_link_img, verify_qr, verify_qr_img — added alongside existing qr, barcode, qr_link

🧱 Dependencies added

  • playwright~=1.47 — Chromium PDF rendering
  • pypdf~=6.1.3 — multi-copy PDF merging
  • pyHanko~=0.25 — PAdES signing
  • cryptography>=41 — self-signed cert generation

📦 Installation

bench get-app https://github.com/Tusharp21/scan_me
bench install-app scan_me

Chromium for PDF rendering (~200 MB) downloads automatically via the
after_install hook. On restricted hosts where auto-download fails,
run ./env/bin/python -m playwright install chromium manually — the app will show
a clear message pointing to this fallback.

Existing installs: bench migrate runs 5 patches that backfill sensible
defaults for new settings fields and will download Chromium if it isn't
already cached.

⚠️ Known limitations

  • Self-signed PAdES cert shows as "signer unknown" in Adobe until replaced with a CA-issued PFX
  • Test suite is skeleton-only — real test coverage planned for 2.1

📚 Documentation

v1.0.2 — 06 Jan 2026

  • System-verified QR code generation with unique document IDs
  • Logo integration in QR codes
  • Signature requirement for QR generation
  • Doctype-based configuration via Scan Me Settings
  • Public verification support with ROOT_URL/verify_document/

Generate a unique system ID linked to a specific document, with verification URL and embedded logo.

v1.0.1 — 18 Nov 2025

  • Fixed line-level issue in jinja_functions.py
  • Updated function logic for accurate output rendering
  • Minor code cleanup for readability

This update focuses on improving reliability in template utilities without introducing any breaking changes.

v1.0.0 — 15 Nov 2025

  • Jinja functions for print formats: qr(), barcode(), qr_link()
  • Global JavaScript function add_qr_to_description()
  • QR/barcode generation in desk forms and print formats
  • Auto-loading via hooks.py configuration
Loading

Version-2.0.1

Version-2.0.1

Choose a tag to compare

View all tags
@Tusharp21 Tusharp21 released this 26 Apr 07:28
v2.0.1
6900959

Scan Me — Release Notes

All previous releases are on GitHub: https://github.com/Tusharp21/scan_me/releases

v2.0.1

Patch release — Frappe Cloud audit compliance.

🛠 Fixes

  • Added [tool.bench.frappe-dependencies] section to pyproject.toml declaring supported Frappe range (>=14.0.0,<=17.0.0-dev).
  • Removed external links from the long description (README.md) per Frappe Cloud audit guidance.
  • Tightened the Advanced Print page role from All to System User, blocking Guest / Website User access while keeping all desk users (Sales User, Accountant, Stock User, etc.) functional.
  • Annotated the two intentional file-system call sites (api/pdf.py, utils/pades.py) with inline # nosemgrep: frappe-security-file-traversal markers — both are already gated by allow-list containment / fixed internal paths.

🔁 No behaviour changes

This release is audit / metadata only. No feature additions, no schema changes, no migrations.

v2.0.0

Major release — Chromium PDF pipeline, PAdES digital signatures, Advanced Print page, and a hardened Verified QR with tamper detection.

🚀 New features

🔐 Signing & Verification

  • Multi-signer support — same document can be signed by multiple users, each with their own Verified QR
  • Content hash (SHA-256) recorded at sign time — edits auto-invalidate the signature in PDFs and on the verify page
  • Permanent signature lock — Verified QR records can be locked so no one, including Administrator, can modify or delete
  • Stored-signature mode — server reads the user's saved signature instead of accepting client-supplied images
  • Signature Type selector — one setting controls Visual Block / Cryptographic (PAdES) / Both

🖨️ Advanced Print Page (/app/scan-me-print/{DocType}/{DocName})

  • Live PDF preview pane that re-renders on every option change (debounced)
  • Auto-fills Print Format + Letter Head defaults per doctype
  • Multi-copy (1–5) with optional comma-separated labels (Original / Duplicate / Triplicate / custom)
  • Header & footer repeat modes — All pages / First only / Last only / First + Last / None
  • QR / Barcode insertion at any corner of the PDF
  • Acrobat-style signature card with green VERIFIED / red INVALIDATED states
  • Watermark system — Document Status (auto DRAFT / CANCELLED / status) or Custom Text
  • Attach-to-Document option — saves the PDF as a private attachment on download
  • Breadcrumbs: DocType → DocName → Advanced Print

🧠 PAdES Digital Signatures (PyHanko)

  • Optional Adobe-compatible cryptographic PDF signatures
  • Self-signed certificate auto-generated on first use
  • Replace the PKCS#12 bundle with a CA-issued one for Adobe's green "Signed and all signatures are valid" badge

🌐 Public Verify Page (/verify_document/)

  • Redesigned mobile-first UI with dark-mode support
  • Three distinct result states — Verified, Signature Invalidated (tampered), Not Recognised
  • Admin-configurable detail level — Minimal / Standard (default) / Full — controls what guests see
  • Rate-limited to 30 requests per 60 seconds per IP

⚙️ Settings & Form Integration

  • Scan Me Settings is now a tabbed form — Allowed Documents / Verification & Signing / Public Verification / Print Dialog Features
  • New "Advanced Print" icon on document forms, placed immediately right of Frappe's native Print icon
  • Admin toggle to show/hide the Advanced Print icon

📄 Bundled print formats

  • Sales Invoice Scan Me — full-featured invoice with QR footer
  • SO — sample Sales Order format
  • Basic HTML Template by Scan Me — starter skeleton users can duplicate for any doctype

📝 Bundled Letter Head

  • Letter Head By Scan Me ships as a fixture for reference — admin can mark it default or customize as needed

🧰 Jinja helpers (for print format authors)

  • qr_img, qr_link_img, verify_qr, verify_qr_img — added alongside existing qr, barcode, qr_link

🧱 Dependencies added

  • playwright~=1.47 — Chromium PDF rendering
  • pypdf~=6.1.3 — multi-copy PDF merging
  • pyHanko~=0.25 — PAdES signing
  • cryptography>=41 — self-signed cert generation

📦 Installation

bench get-app https://github.com/Tusharp21/scan_me
bench install-app scan_me

Chromium for PDF rendering (~200 MB) downloads automatically via the
after_install hook. On restricted hosts where auto-download fails,
run ./env/bin/python -m playwright install chromium manually — the app will show
a clear message pointing to this fallback.

Existing installs: bench migrate runs 5 patches that backfill sensible
defaults for new settings fields and will download Chromium if it isn't
already cached.

⚠️ Known limitations

  • Self-signed PAdES cert shows as "signer unknown" in Adobe until replaced with a CA-issued PFX
  • Test suite is skeleton-only — real test coverage planned for 2.1

📚 Documentation

v1.0.2 — 06 Jan 2026

  • System-verified QR code generation with unique document IDs
  • Logo integration in QR codes
  • Signature requirement for QR generation
  • Doctype-based configuration via Scan Me Settings
  • Public verification support with ROOT_URL/verify_document/

Generate a unique system ID linked to a specific document, with verification URL and embedded logo.

v1.0.1 — 18 Nov 2025

  • Fixed line-level issue in jinja_functions.py
  • Updated function logic for accurate output rendering
  • Minor code cleanup for readability

This update focuses on improving reliability in template utilities without introducing any breaking changes.

v1.0.0 — 15 Nov 2025

  • Jinja functions for print formats: qr(), barcode(), qr_link()
  • Global JavaScript function add_qr_to_description()
  • QR/barcode generation in desk forms and print formats
  • Auto-loading via hooks.py configuration
Loading

Version-2.0.0

Version-2.0.0

Choose a tag to compare

View all tags
@Tusharp21 Tusharp21 released this 19 Apr 11:35
v2.0.0
fb533a4

Scan Me — Release Notes

All previous releases are on GitHub: https://github.com/Tusharp21/scan_me/releases

v2.0.0

Major release — Chromium PDF pipeline, PAdES digital signatures, Advanced Print page, and a hardened Verified QR with tamper detection.

🚀 New features

🔐 Signing & Verification

  • Multi-signer support — same document can be signed by multiple users, each with their own Verified QR
  • Content hash (SHA-256) recorded at sign time — edits auto-invalidate the signature in PDFs and on the verify page
  • Permanent signature lock — Verified QR records can be locked so no one, including Administrator, can modify or delete
  • Stored-signature mode — server reads the user's saved signature instead of accepting client-supplied images
  • Signature Type selector — one setting controls Visual Block / Cryptographic (PAdES) / Both

🖨️ Advanced Print Page (/app/scan-me-print/{DocType}/{DocName})

  • Live PDF preview pane that re-renders on every option change (debounced)
  • Auto-fills Print Format + Letter Head defaults per doctype
  • Multi-copy (1–5) with optional comma-separated labels (Original / Duplicate / Triplicate / custom)
  • Header & footer repeat modes — All pages / First only / Last only / First + Last / None
  • QR / Barcode insertion at any corner of the PDF
  • Acrobat-style signature card with green VERIFIED / red INVALIDATED states
  • Watermark system — Document Status (auto DRAFT / CANCELLED / status) or Custom Text
  • Attach-to-Document option — saves the PDF as a private attachment on download
  • Breadcrumbs: DocType → DocName → Advanced Print

🧠 PAdES Digital Signatures (PyHanko)

  • Optional Adobe-compatible cryptographic PDF signatures
  • Self-signed certificate auto-generated on first use
  • Replace the PKCS#12 bundle with a CA-issued one for Adobe's green "Signed and all signatures are valid" badge

🌐 Public Verify Page (/verify_document/)

  • Redesigned mobile-first UI with dark-mode support
  • Three distinct result states — Verified, Signature Invalidated (tampered), Not Recognised
  • Admin-configurable detail level — Minimal / Standard (default) / Full — controls what guests see
  • Rate-limited to 30 requests per 60 seconds per IP

⚙️ Settings & Form Integration

  • Scan Me Settings is now a tabbed form — Allowed Documents / Verification & Signing / Public Verification / Print Dialog Features
  • New "Advanced Print" icon on document forms, placed immediately right of Frappe's native Print icon
  • Admin toggle to show/hide the Advanced Print icon

📄 Bundled print formats

  • Sales Invoice Scan Me — full-featured invoice with QR footer
  • SO — sample Sales Order format
  • Basic HTML Template by Scan Me — starter skeleton users can duplicate for any doctype

📝 Bundled Letter Head

  • Letter Head By Scan Me ships as a fixture — customize and use

🧰 Jinja helpers (for print format authors)

  • qr_img, qr_link_img, verify_qr, verify_qr_img — added alongside existing qr, barcode, qr_link

🧱 Dependencies added

  • playwright~=1.47 — Chromium PDF rendering
  • pypdf~=6.1.3 — multi-copy PDF merging
  • pyHanko~=0.25 — PAdES signing
  • cryptography>=41 — self-signed cert generation

📦 Installation

bench get-app https://github.com/Tusharp21/scan_me
bench install-app scan_me

Chromium for PDF rendering (~200 MB) downloads automatically via the
after_install hook. On restricted hosts where auto-download fails,
run ./env/bin/python -m playwright install chromium manually — the app will show
a clear message pointing to this fallback.

Existing installs: bench migrate runs 5 patches that backfill sensible
defaults for new settings fields and will download Chromium if it isn't
already cached.

⚠️ Known limitations

  • Self-signed PAdES cert shows as "signer unknown" in Adobe until replaced with a CA-issued PFX
  • Test suite is skeleton-only — real test coverage planned for 2.1

📚 Documentation

v1.0.2 — 06 Jan 2026

  • System-verified QR code generation with unique document IDs
  • Logo integration in QR codes
  • Signature requirement for QR generation
  • Doctype-based configuration via Scan Me Settings
  • Public verification support with ROOT_URL/verify_document/

Generate a unique system ID linked to a specific document, with verification URL and embedded logo.

v1.0.1 — 18 Nov 2025

  • Fixed line-level issue in jinja_functions.py
  • Updated function logic for accurate output rendering
  • Minor code cleanup for readability

This update focuses on improving reliability in template utilities without introducing any breaking changes.

v1.0.0 — 15 Nov 2025

  • Jinja functions for print formats: qr(), barcode(), qr_link()
  • Global JavaScript function add_qr_to_description()
  • QR/barcode generation in desk forms and print formats
  • Auto-loading via hooks.py configuration
Loading

Version-1.0.2

Version-1.0.2

Choose a tag to compare

View all tags
@Tusharp21 Tusharp21 released this 06 Jan 14:12
v1.0.2
d2679a1

Full Changelog: v1.0.1...v1.0.2

Scan Me – Release Notes

Version: v1.0.2
App Name: Scan Me
Category: Frappe Utility App (QR Code and Barcode Generator)

Overview

Scan Me is a utility app for Frappe that provides Jinja helper methods for generating QR codes and barcodes inside Print Formats and documents.

Previous versions supported only QR and barcode generation using Jinja methods.
This release introduces system-verified QR code generation with logo integration and public verification support.

What’s New in This Release

System Verified QR Code Generation

  • Generate a unique system ID linked to a specific document.
  • QR code includes a verification URL and embedded logo.
  • Ensures document authenticity and prevents duplicate or fake QR usage.

Signature Required for QR Generation

  • QR code generation now requires a signature from the user.
  • The signature is stored and validated during verification.
  • Improves security, traceability, and accountability.

Doctype-Based Configuration

  • Scan Me Settings child table to configure which doctypes can generate verified QR codes.
  • Only selected doctypes show the Generate QR action button.

Print Format Integration

  • Generate QR code in Print Formats using Jinja methods.
  • Example usage: 
    <h4>QR for ID</h4>
<img src="{{ qr(doc.name, clarity=10, fill_color='black', include_logo=True) }}" width="120">

Existing Features

  • QR code and barcode generation using Jinja methods.
  • Easy integration in Print Formats.
  • Lightweight and developer-friendly.

Verification Flow

  1. Configure allowed doctypes in Scan Me Settings.
  2. Open a configured document.
  3. Click Generate QR from the Action menu.
  4. System generates a unique ID and QR code with logo.
  5. QR appears in the Print Format and PDF.
  6. External users scan the QR and verify the document using the public URL. ROOT_URL/verify_document/

Installation / Upgrade

Available on Frappe Cloud Marketplace:
https://cloud.frappe.io/marketplace/apps/scan_me

Upgrade the app using standard Frappe update process.

Feedback

Issues, feature requests, and contributions are welcome via GitHub.

Loading

Version-1.0.1

Version-1.0.1

Choose a tag to compare

View all tags
@Tusharp21 Tusharp21 released this 18 Nov 16:43
v1.0.1
6f44ca4
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

Release Notes

Fixes

  • Corrected a line-level issue in jinja_functions.py that affected Jinja template processing.
  • Updated the function logic to ensure accurate output rendering.
  • Performed minor code cleanup for improved readability and consistency.

Notes

This update focuses on improving reliability in template utilities without introducing any breaking changes.

Loading