- Add static getParsedFile() helper that caches include-based
file parses per file path
- Replace repeated `include $file` calls in
testLanguageFile_ParsesToArray, key-matching, and key-sorting
tests with calls to the cached helper
- Avoid redundant file includes/parsing when the same language
file is loaded across multiple test methods, speeding up the
test suite
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
Resolve language-file conflicts by taking master's English-removal
cleanup across non-English locales, while stripping the MailChimp
keys master reintroduced (already moved to MailchimpPlugin on this
branch) and reapplying this branch's plugins/integrations rename and
new Module.php plugins keys where master lacked them.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
- Replace untranslated English strings with empty fallback ('') across all non-English catalogs and sort keys alphabetically (closes#4783, closes#4660)
- Restore missing Sales.no_sales_to_display in English while preserving existing translations (closes#4794, references #4637, #4795)
- Register Georgian (ka) and Swahili (sw-KE, sw-TZ) locales, normalize Croatian to hr per BCP 47, and fix datepicker long month keys (partially addresses #4476, references #3891)
- Hardcode invariant brand constants (OSPOS, website, copyrights) in login/footer views, remove from catalogs, and standardize Latin OSPOS branding
- Fix English catalog typos and maintain consistent British English variants in en-GB
- Add LanguageFilesTest with 4,418 automated tests for syntax, key parity, sorting, and bidirectional locale registrations
- Update AGENTS.md with localization guidelines and development best practices
* fix(sales): allow gift card credit for returns + surface errors
- Add incrementGiftcardValue() to Giftcard model for crediting cards
- In Sale::save_value(), use incrementGiftcardValue() for returns
with negative gift card payments (credits the card)
- Add error_message rendering to register view so failed completions
show the error instead of silently reloading
Fixes#4779
---------
Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>
Introduce a CashupAmounts transfer object and break down large Cashups
controller methods into focused private helpers. Includes a breaking
change to constructor parameter order.
- Add CashupAmounts (TransferObjects/Cashups) to bundle open/transfer/
closed cash, due, card, and check amounts; reorder constructor params
to cash-before-due sequence
- Replace positional _calculate_total() with calculateTotal(CashupAmounts),
eliminating argument-order errors
- Decompose getView() into buildEmployeeLookup(), normalizeCashupInfo(),
initializeNewCashup(), isPendingClose(), closeCashup(),
buildReportDateRangeInputs(), applyPaymentTotals(), subtractCashExpenses()
- Decompose postSave() with parseCashupDate() and getCashupAmountsFromPost()
helpers; update both CashupAmounts call sites to new param order
- Rename snake_case vars/params to camelCase throughout Cashups controller
- Drop stale TODO comments in OSPOS.php and Cashups::getIndex filter block
- Scope escape => false to cashup_headers() note column rather than a
broad regex exclusion in transform_headers()
- Add CashupFixtureTrait and ExpenseFixtureTrait under tests/Support;
tighten cashup view test to assertSeeInField on specific form fields
BREAKING CHANGE: CashupAmounts constructor positional argument order
changed — closedAmountCash now precedes closedAmountDue; any callers
passing positional args must swap those two arguments.
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
build/refactor/fix: tighten tooling, rename accessors to camelCase, and correct model bugs
* build: tighten gitignore, lockfile, and script config
- gitignore /build and /.php-cs-fixer.cache
- allow core-js@3.50.0 postinstall script in package.json
* refactor: rename snake_case data accessors to camelCase
Rename get_info, get_multiple_info, get_sale_items, get_sale_payments,
and save_customer across Customer, Employee, Item, Sale, and Supplier
models to PSR-12 camelCase, updating every call site in controllers,
libraries, and views.
* test: update calls to match renamed camelCase accessors
Update HomeTest, ItemsControllerTest, and ItemTest to call
getInfo()/getMultipleInfo() instead of the old snake_case model methods.
* fix: correct independent bugs in Item, Item_taxes, and Attribute models
- Item::delete()/delete_list(): let $this->db->transStatus() alone decide
success instead of accumulating a bitwise-AND over intermediate results,
which could report failure (and trigger error toasts) even when the
delete had actually completed
- Item_taxes: correct $table property from 'item_taxes' to 'items_taxes'
- Attribute::getDefinitionNames(): pass an array to
QueryBuilder->whereNotIn(), which requires one
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Refactor Stock_location model to camelCase naming and introduce
is_default and sort_order fields with full migration, UI, and
item-sync support. Multiple follow-up fixes consolidated.
Database
- Add is_default (TINYINT, backfilled to oldest non-deleted row) and
sort_order (nullable unsigned TINYINT unique, dense 0..N-1) columns
via new migration; MySQL 5.7-compatible backfill using user variables
instead of ROW_NUMBER(); down() drops both columns
Model
- Rename show_locations → showLocations (camelCase convention)
- Add getDefaultLocationId(), setDefaultLocation(), saveSortOrder();
validate location existence before setting as default
- delete() compacts sort_order via two-pass CASE update to avoid
collisions after prior reorders
- Atomic sort_order assignment via explicit SQL COUNT query
Controller / Config
- Modularize save logic into sanitizeSubmittedLocations(),
deleteRemovedLocations(), saveSubmittedLocations(),
saveLocationOrderAndDefault(); wrap in single transaction
- Support stock_location_new inputs for new-row creation
- Resolve new-* ID placeholders to real IDs post-save
- Abort loop on save failure; default missing POST arrays to []
- Validate that at least one location array is non-empty
Views
- Add Default column with center-aligned radio buttons; disable for
unsaved rows; use attr() instead of data() for location IDs
- Escape dynamic location IDs against XSS in all affected partials
- form_count_details.php: fall back to default location when current
is undefined; guard quantity input similarly
- Add aria-label to default radio inputs
Libraries / Items
- Widen Item_lib::get_item_location() and get_stock_destination() to
nullable return types
- Guard null location_id in Items::getRow() and getGenerateBarcodes()
- Item::get_multiple_info(): accept ?int $location_id; short-circuit
with where('item_id', -1) instead of always-false where('1','0')
i18n
- Add stock_location_default, stock_location_default_header,
stock_location_name_header across all locale files
Tests
- Stock_locationTest: getDefaultLocationId, setDefaultLocation,
saveSortOrder, delete compaction, saveValue; rename to
testMethod_Scenario convention
- Add ConfigControllerTest / StockLocationsControllerTest with
AdminAuthTrait covering CRUD, reorder, default assignment, and
malformed/empty/blank payload validation
- DatabaseBootstrapTrait-based setup for ItemTest and Stock_locationTest
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Resolve language file conflicts: take master's translations, strip
MailChimp phrases (MailChimp is now a plugin instead of core), and
retain plugin-system additions (plugins/plugins_configuration keys).
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Adds DatabaseBootstrapTrait and OsposConfigFixtureTrait to replace
duplicated per-file setUp()/tearDown() bootstrap and OSPOS config mock
boilerplate across test classes, bringing them in line with the
ItemsControllerTest gold-standard shape.
Fixes the real root cause of tabular_helperTest's suite-order-dependent
failure: dropAllForeignKeyConstraints()/foreignKeyExists()/
dropColumnIfExists() temporarily blank the shared connection's DBPrefix
via overridePrefix() but never restored it if the query between the two
calls threw (e.g. migration_helperTest's mocked-failure tests) -
permanently corrupting prefixed table lookups for the rest of the
process. Wrapped each in try/finally. Also widens executeScript() to
accept an explicit connection so migration SQL can't silently run
against the wrong DB group.
Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>
These landed on master by accident via a local branch that was
mistakenly created tracking OpensourcePOS/master directly, rather than
pushed as its own branch. Reverting to restore master to its prior
state; the reverted work will come back through a proper PR instead.
This reverts commit 00bf4cb8af.
This reverts commit 37cc1b15e7.
Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>
Adds DatabaseBootstrapTrait and OsposConfigFixtureTrait to replace
duplicated per-file setUp()/tearDown() bootstrap and OSPOS config mock
boilerplate across test classes, bringing them in line with the
ItemsControllerTest gold-standard shape.
Fixes the real root cause of tabular_helperTest's suite-order-dependent
failure: dropAllForeignKeyConstraints()/foreignKeyExists()/
dropColumnIfExists() temporarily blank the shared connection's DBPrefix
via overridePrefix() but never restored it if the query between the two
calls threw (e.g. migration_helperTest's mocked-failure tests) -
permanently corrupting prefixed table lookups for the rest of the
process. Wrapped each in try/finally. Also widens executeScript() to
accept an explicit connection so migration SQL can't silently run
against the wrong DB group.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
- Normalize all language files to single quotes with aligned => operators;
no translation text or logic changes
- Add Georgian (ka) scaffolding across ~18 modules with placeholder
translations, matching project style conventions
- Add LanguageFilesTest to validate every PHP file under app/Language
parses without syntax errors; new files auto-discovered via glob
Add MySQL 8.4 compatibility, including a utf8mb4 conversion migration,
hardened foreign key handling, and expanded CI database coverage.
Migrations:
- Add Mysql84Utf8mb4Conversion migration using
sqlscripts/3.4.3_mysql84_utf8mb4_conversion.sql via executeScript()
- Drop ospos_people composite index before conversion, since its
prefix length exceeds the 3072-byte key limit under utf8mb4
- Add assertNoCollationCollisions() pre-flight check to fail fast on
unique values that collide under utf8mb4_unicode_520_ci, instead of
hitting "Duplicate entry" partway through ALTER TABLE
- Wrap the conversion in try/catch so foreign keys are recreated on
failure; log restore errors separately so they don't mask the
original exception
- Throw RuntimeException if re-adding the people index fails
- Remove the re-add of the sales_items_taxes FK in down(), which
MySQL 8.4 rejects without a unique key
- Revert edits to historical migrations; fix FK handling and table
names
migration_helper:
- Add shared dropAllForeignKeyConstraints() and
recreateForeignKeyConstraints(); table/column now optional, dropping
all FKs in the schema when omitted
- Match CONSTRAINT_SCHEMA when joining KEY_COLUMN_USAGE to
REFERENTIAL_CONSTRAINTS to avoid cross-schema name collisions
- Rebuild FK lookup with the query builder and bound values instead
of interpolated SQL
CI/tests:
- Add mysql:5.7, mariadb:10.6 and mariadb:10.11 to PHPUnit matrix
- Use master instead of latest image tag; improve CI robustness
- Mock table() in FailedQueryConnectionTrait so query builder chains
work against the failing connection
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
Co-authored-by: aliworksx08 <57456290+aliworksx08@users.noreply.github.com>
Co-authored-by: jekkos <jeroen.peelaerts@gmail.com>
Co-authored-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>
* fix(taxes): reject negative tax rates and percentages
Reject negative tax values across all tax entry points so they can
never be persisted and produce inverted sales totals:
- Items::postSave / postBulkUpdate: validate tax_percents.* as
nonNegativeDecimal (was unvalidated, persisted raw).
- Items::save_tax_data (CSV import): fail the row on a negative
Tax N Percent instead of silently dropping the tax.
- Taxes::postSave: reject negative tax_rate before saving.
- Config::postSaveTax: reject negative default tax rates.
Adds non-negative regression tests for Items, Taxes and Config, plus
boundary tests proving a zero tax rate/percent is still accepted. New
language strings for the rejection messages.
* test: adopt test{Method}_{Purpose} naming convention for new tests
Align the negative-tax test method names with the convention established
in ConfigTest (testPostSave_RejectsNegativeTaxPercent, etc.), per review.
* fix(config): add language entries to all locales, extract TaxFixtureTrait
* fix(i18n): place tax_rate_non_negative in alphabetical order
Move the new tax_rate_non_negative entry to its correct alphabetical
position (after tax_rate_error_adding_updating) in all locale files.
---------
Co-authored-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
* fix(models): handle query() returning false in result-consuming methods
As of CI 4.1.2, Connection::query() returns false when a query fails
(previously an empty result object). Guard the model/helper methods that
consume the result so a failed query degrades to an empty result instead
of throwing "Call to member function getResultArray() on false".
- Tax::get_taxes()
- Sale::get_all_suspended()
- Reports\Inventory_low::getData()
- migration_helper: dropAllForeignKeyConstraints(), indexExists(),
primaryKeyExists()
Adds QueryFalseGuardTest forcing query() to return false. Closes#3634.
* test(query-false): mock concrete driver instead of abstract BaseConnection
onlyMethods() does not stub the 16 abstract methods on BaseConnection, so
mocking it produced an invalid double. Mock the concrete driver class
(resolved from the live test connection) instead.
* fix(3634): fail loudly when migration metadata queries return false
Model read methods degrade to an empty result on a failed query, but the
migration metadata helpers (indexExists, primaryKeyExists,
dropAllForeignKeyConstraints) must not mistake a failed query for 'not
found' and proceed with schema changes. They now throw a DatabaseException
so the migration aborts loudly.
- Add an optional injectable BaseConnection param for testability.
- Add missing docstrings (docstring-coverage check).
- Add regression tests asserting the helpers throw on a failed query.
* fix(3634): use supplied connection for FK drop; stub error() in tests
Address CodeRabbit review: dropAllForeignKeyConstraints now builds its
Forge from the supplied connection (consistent with the lookup) instead of
always the default one. Test mock also stubs error() to avoid a warning on
the constructor-disabled connection.
* fix(3634): revert FK drop to Database::forge() (driver-specific forge)
new Forge($db) creates the base Forge which does not implement
dropForeignKey for the driver ('This feature is not available'), breaking
migrations. Restore Database::forge(), which loads the driver-specific
Forge via the factory. The optional connection param is retained solely
for testing the fail-loud behavior on the lookup path.
* test(3634): add regression test for Inventory_low::getData query-failure guard
Covers the low-stock report's degrade-to-empty behavior on a failed query,
addressing the CodeRabbit note that it lacked a focused test.
* test(3634): adopt test{Method}_{Purpose} naming convention
Align QueryFalseGuardTest method names with the convention from #4730
(testGetTaxes_ReturnsEmptyArrayWhenQueryFails, etc.), per review.
* test: split QueryFalseGuardTest into per-component test files
Apply CodeRabbit review: extract the shared failing-connection helpers
into tests/Support/FailedQueryConnectionTrait.php and move each test to
its component's test file (TaxTest, SaleTest, Inventory_lowTest,
migration_helperTest). Use fresh model instances instead of the shared
model() singleton so the injected failing connection does not leak into
other tests.
* style(tests): remove redundant docblocks from query-failure regression tests
---------
Co-authored-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
* fix(appconfig): refresh cached settings once and avoid stale same-request reads
Addresses issue #3693, which has two related problems:
1. Repeated cache refreshes. Appconfig::batch_save() called save() per key,
and save() unconditionally deleted + reloaded + re-cached the settings after
every single write. A settings page with dozens of keys therefore refreshed
the cache dozens of times in one request. batch_save() now persists all keys
in the transaction and refreshes the cached settings exactly once after it
commits successfully. save() keeps refreshing so direct single saves still
work.
2. Stale same-request reads. Controllers copy config(OSPOS::class)->settings at
construction time, so after a save in the same request the local copy is out
of date. Config::postSaveInvoice() read the stale invoice_enable /
default_register_mode values and sent the user to the wrong screen (e.g.
opening a new sale instead of the invoice form after saving an enabled
invoice). It now reads the freshly refreshed settings.
Adds tests/Models/AppconfigTest.php covering batch_save() and save() refreshing
the cached settings after a successful write.
* test(3693): adopt test{Method}_{Purpose} naming convention
Align AppconfigTest method names with the convention from #4730
(testBatchSave_PersistsAllKeysAndRefreshesCachedSettings, etc.), per review.
* style(config): remove inline comments per review feedback
---------
Co-authored-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
* fix(taxes): reject invalid characters in tax code, category, and jurisdiction names
Apply the same unicode_alpha_numeric_punct validation already used for
item tax names to the three tax save endpoints, so stored-XSS payloads
(<, >) are rejected with a validation error instead of being saved.
Mirrors the Items fix and adds a TaxesControllerTest regression suite.
* fix(taxes): permit slash and CJK in tax names, add regression tests
The unicode_alpha_numeric_punct guard rejected '/' (char type Po), which
blocks legitimate slash-separated tax names such as 'GST/HST' and 'VAT/GST'
from being saved. Add '/' to the allowed punctuation set and document it.
Also add regression tests proving acceptance of a slash name (GST/HST) and a
CJK name (消費税) across the tax save endpoints. Addresses CodeRabbit review
on #4679.
* test(4679): adopt test{Method}_{Purpose} naming convention
Align test method names with the convention from #4730
(testPostSaveTaxCodes_RejectsMaliciousName, etc.), per review.
* fix(taxes): allow parentheses in names, make name optional
Address CodeRabbit review on #4733:
- Permit parentheses in tax names (e.g. "VAT (20%)"); slash was
already allowed, so both "GST/HST" and "VAT (20%)" now pass while
< and > are still rejected.
- Drop the rule from tax_code_name, jurisdiction_name and
tax_category so a code with a blank name can be saved, matching the
form (the form does not require the name).
- Add regression tests for the parentheses and blank-name cases.
---------
Co-authored-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
feat(items): optimize search, attribute filtering, and sort for items view (#4652)
BREAKING CHANGE: none
## Search & Query Optimization
- Split item search into two-phase query: Phase A resolves qualifying IDs,
Phase B joins scoped display tables for better performance and readability
- Sanitize `definition_ids` via `array_map` to prevent injection vulnerabilities
- Introduce subquery for SUM aggregation to prevent over-counting across joins
- Add validation requiring both start and end dates before applying date range filter
## Attribute Search (fixes#2919, #2722)
- Add `SHOW_IN_SEARCH` flag (value 8) to Attribute model to separate
searchability from table visibility
- Add `parse_attribute_search()` to parse syntax like `color:blue AND size:large`
- Add `applyNamedAttributeSearch()` supporting decimal and date types with
locale-aware parsing
- Support AND/OR logic for multi-attribute queries
## Sorting
- Add `get_attribute_sort_definition_id()` to detect attribute column sorting
- Join attribute tables dynamically when sorting by attribute columns
- Use `MAX()` for consistent results when sorting by attribute values
- Replace static sort column list with dynamic headers via `itemSortColumns()`
- Add `sanitizeSortColumnAttribute()` to validate attribute definition IDs as sort columns
## Tax & Data Row
- Streamline tax computation in `getItemDataRow()`
## Low Inventory Filter
- Require valid `stock_location_id` before applying low inventory filter
- Add conditional logic to sort by sum of quantities across all locations
when `stock_location_id` is invalid
## Localization
- Add `show_in_search` / `show_in_search_visibility` strings to all language files
- Translated: de-DE, es-ES, fr, it; English placeholder for remaining locales
- Unify single-quote style across all attribute language files
## Refactoring & Style
- Adopt camelCase naming throughout (variables, helpers, methods)
- Replace `sanitizeSortColumnAttribute` with reusable `sanitizeSortColumn`
from `Secure_Controller`
- Simplify column key extraction using `array_key_first`
- Apply PSR-12 formatting
## Tests
- Add tests for tax computation, quantity aggregation (single- and multi-location),
named attribute search, free-text parsing, and date/decimal type handling
- Add `ensureStockLocation` helper to auto-create missing stock locations in tests
- Refactor tests to handle config cache issues
Co-authored-by: Ollama <ollama@steganos.dev>
* fix(security): report unwritable .env.lock, make throttle limits configurable
envFileIsWritable() previously checked is_writable(.env) (the file), which passes in Docker even when the mutex file is root-owned by a prior env:provision run. It now checks the real write path: the directory (to create .env.tmp.* + .env.lock) and any existing .env.lock must be writable, so the app throws the clear 'run env:provision' error instead of crashing on 'Unable to open .env.lock'.
The Throttle filter now reads throttle.capacity / throttle.seconds from env (default 5 per 60s); capacity <= 0 disables throttling, so operators serving sequential HTTP clients (e.g. Zabbix) are not caught by the lockout.
* fix(security): address CodeRabbit review findings
- Throttle: validate throttle.capacity as an integer before treating a non-positive value as 'disabled', so a non-numeric value (e.g. 'five') falls back to the default instead of silently bypassing the lockout. Apply the same validation to throttle.seconds.
- envFileIsWritable(): also reject an existing non-writable .env on Windows (where rename() cannot replace a read-only destination); keep the check Windows-only since POSIX rename() replaces a read-only dest when the directory is writable.
- Tests: restore the prior throttle.capacity env state in ThrottleTest (capture/restore instead of delete); add an invalid-capacity fallback case; skip the not-writable fixtures when running as root (where is_writable() is bypassed).
* fix(migration): guard ConvertToCI4 key-write branches with envFileIsWritable()
The migration's 'no key' and 'CI3 key' branches called rotateEncryptionKey()/rotateEncryptionKeyTransaction() directly, bypassing the envFileIsWritable() guard that checkEncryption() uses. On a fresh Docker/Compose install where the web runtime cannot write /app/.env.lock, this produced a raw 'fopen(/app/.env.lock): Permission denied' error instead of the actionable 'run php spark env:provision' message.
A valid CI4 key now short-circuits to checkEncryption() (no write); every write branch is gated on envFileIsWritable() first.
* fix: read provisioned encryption.key so config sees it
env:provision persists the key as 'encryption.key' in .env (matching the
throttle path and .env.example), but Config\Encryption only read the
ENCRYPTION_KEY env var. On a fresh Docker instance the provisioned key was
invisible to config('Encryption')->key, so the app believed no key existed
and tried to write one -- hitting the .env.lock permission wall.
Read encryption.key (via $_SERVER/$_ENV/getenv) first, then fall back to
ENCRYPTION_KEY for Docker '-e' usage. Mirrors checkThrottleEncryption().
* fix: cascade encryption.key lookup past empty-string sources
* refactor: extract Encryption::resolveKey() and test it without global env mutation
* fix(security): decode fallback-selected encryption key like BaseConfig
A key picked in the constructor fallback (notably ENCRYPTION_KEY, which
BaseConfig never inspects) was assigned verbatim, bypassing the
hex2bin:/base64: decode the parent applies to `encryption.key`. Route the
selected key through a parseKey() helper mirroring BaseConfig's
parseEncryptionKey() so prefixed values decrypt consistently, and add a
pure regression test.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* chore: remove advisory ID from comments and tighten verbose comments
Per review: treat GHSA advisory IDs like secrets (drop from code) and
replace the multi-paragraph comments with concise one-liners that keep
the non-obvious "why". No logic changes.
---------
Co-authored-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Per the project's policy of treating security advisory IDs as
secret-like, remove the identifiers embedded in source/test comments
and CHANGELOG entries. Each keeps its human-readable description (and
PR number where present), so traceability is preserved. No logic changes.
checkThrottleEncryption() now consults the THROTTLE_KEY environment
variable when throttle.key is empty, mirroring the ENCRYPTION_KEY
fallback in Config/Encryption. This lets Docker/Compose deployments
supply the throttle HMAC secret without writing a shared value into a
read-only .env. An explicit throttle.key always takes precedence.
Adds regression tests to the existing security_helperTest suite and
documents THROTTLE_KEY in .env.example.
* fix(locale): validate language_code against known locales to block path traversal
postSaveLocale() stored language_code from user input with no allow-list validation, and it later flows into Language::setLocale()/load() where the locale segment is require()'d. An authenticated config-grant account could store a relative path (e.g. ../../public/uploads) and, combined with a planted file in public/uploads/, achieve unauthenticated RCE on the next request.
Validate the submitted language against array_keys(get_languages()) before storing, and harden languageExists() to reject path separators and dot-dot sequences. Adds regression tests.
* test(locale): give locale fixture valid reference-code min/max defaults
* fix(locale): reject null bytes in languageExists guard
A stored language_code containing a NUL byte passes the existing path-separator and parent-dir checks, then reaches file_exists(). On PHP 8.5+ file_exists() throws a ValueError for NUL-byte paths, which breaks configuration loading. Reject NUL bytes in the guard and add regression tests.
* fix(security): handle special characters in `.env` key values and improve insertion logic
- Escape backslashes and dollar signs in `applyEnvKeyReplacement` to prevent unintended value corruption.
- Ensure new keys are inserted after `encryption.key` for better organization and manageability.
- Add explicit cast to int to prevent wrong concatenation operator warning.
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
* fix(security): handle null return in `applyEnvKeyReplacement` and ensure proper `.env` updates
- Update `applyEnvKeyReplacement` to return `null` on failure, improving error handling.
- Adjust calls to `atomicWriteFile` with updated content to prevent unintended behavior.
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
* fix(security): improve error logging and exception messages in file locking
- Add detailed logging for file open and locking errors in `security_helper`.
- Remove unused `helper` and `checkThrottleEncryption` calls from `Events` for cleanup.
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
* fix(security): improve atomic file write and handle encryption key placement
- Throw `RandomException` for better error reporting in `atomicWriteFile`.
- Simplify Windows-specific `rename()` fallback logic.
- Fix `encryption.key` assignment order to ensure consistency in `.env` updates.
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
* fix(security): improve `.env` file handling and add unit tests for helper functions
- Suppress warnings in `file_get_contents` to prevent unnecessary error logs.
- Update `applyEnvKeyReplacement` to use `preg_replace_callback` for better safety.
- Add comprehensive unit tests for `security_helper` functions to ensure `.env` updates and key management work as expected.
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
* fix(security): enhance `.env` update logic and add robust exception handling
- Add `RandomException` to improve error reporting in encryption key management.
- Introduce environment file locking for safer `.env` updates.
- Ensure `applyEnvKeyReplacement` properly handles and inserts old key comments.
- Replace direct file writes with `atomicWriteFile` for consistency.
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
* fix(security): refactor `.env` file initialization and encryption key handling
- Introduce `initializeEnvFile` for reusable `.env` setup logic.
- Add `backupEnvFile` and `writeNewEncryptionKey` for robust key management with backups.
- Simplify and clean up redundant `.env` handling code paths.
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
* fix(security): clarify `checkEncryption` docblock return value description
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
* fix(security): escape backslashes and dollar signs in `applyEnvKeyReplacement`
- Ensure `applyEnvKeyReplacement` properly escapes special characters when inserting or appending `.env` keys.
- Add new unit tests to validate correct handling of backslashes and dollar signs.
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
* fix(i18n): add localized error messages and improve error reporting in `security_helper`
- Add missing translations for error messages across multiple language files.
- Update `security_helper` to use localized exception messages with placeholders.
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
* Redesign encryption/throttle key provisioning as read-only runtime
- checkEncryption()/checkThrottleEncryption() are now read-only guards that
throw when no valid key is provisioned, instead of writing .env at
request time.
- Add rotateEncryptionKey() and provisionThrottleKey() for explicit,
idempotent provisioning.
- Add php spark env:provision (app/Commands/EnvProvision.php) so Docker can
provision keys once at container startup before any request.
- Add app/Libraries/CI3SecretConverter.php shared CI3->CI4 secret converter
(AES-128-CBC decrypt + CI4 re-encrypt/verify/save) used by both the
interactive migration and the docker startup path.
- Refactor convertToCI4 migration to use the shared converter.
- Persist .env in a named volume and run spark env:provision on boot; stop
baking .env into the shipped image.
- Add guard/rotation/throttle + converter tests; clean up orphaned
msg_pwd_required language keys across all locales.
* fix: save CI4 ciphertext in env:provision and bind-mount a .env file
Addresses CodeRabbit review on PR #4656:
- env:provision CI3 branch was persisting *plaintext* secrets (saveAll($plain))
instead of the CI4 ciphertext, unlike the ConvertToCI4 migration. Now
encrypts with encryptAll(), verifies the round trip, and saves the ciphertext.
- The ospos_env named volume mounted at /app/.env made .env a directory, so
atomicWriteFile's rename() failed and spark env:provision could not start apache.
Switch to a bind mount of a host file (./.env) which persists and stays a file.
- Add a regression test asserting the command persists ciphertext (not plaintext).
* chore: trim redundant docblocks in EnvProvision and provision throttle.key in CI
Follow up on @objecttothis review comments:
- app/Commands/EnvProvision.php: remove the boilerplate docblocks the
property names already convey (group/name/usage/description, run()),
the two inline step comments, the anyNonEmpty() param docblock, and the
legacySecretsPresent() docblock. Keeps the class-level docblock since it
is the only place that states the read-only runtime design + the
never-persist-plaintext invariant.
- .github/workflows/phpunit.yml: provision a per-run throttle.key the same
way the encryption key is already provisioned. The PR makes
checkThrottleEncryption() a read-only guard that throws when
env('throttle.key') is unset; CI only started exporting ENCRYPTION_KEY,
so every test that goes through the Throttle filter (7 ThrottleTest
cases + 4 LoginTest cases) failed with
"No throttle key is provisioned. Run `php spark env:provision`".
Writing `throttle.key=<KEY>` into .env matches what
`php spark env:provision` does on a real container start.
* fix(ci): write throttle.key into .env instead of exporting an OS env var
The previous attempt exported throttle.key via GITHUB_ENV, but CodeIgniter's
env() helper resolves in the order $_ENV[$key] ?? $_SERVER[$key] ?? getenv($key),
and DotEnv populates $_ENV['throttle.key'] from the .env file first. Because the
.env (copied from .env.example) ships with the empty placeholder throttle.key='',
that $_ENV entry exists as '' and short-circuits the ?? chain before getenv()
is reached — so the OS env var was never consulted and every Throttle/Login test
still threw 'No throttle key is provisioned'.
Write the per-run key into the .env file itself (sed-replacing the empty
placeholder), which is exactly what `php spark env:provision` does in
production and is the single source env() actually reads from.
Verify the replacement happened (grep -Eq '^throttle\.key=.') so a future change
to the placeholder format fails the run loudly instead of silently breaking
the 11 throttle-dependent tests.
* fix(security): restore CI3->CI4 auto-provisioning gated by .env writability
checkEncryption()/checkThrottleEncryption() again provision the keys
inline when .env is writable (empty key -> generate; short key -> decrypt,
rotate, re-encrypt, verify, persist legacy CI3 secrets). When .env is not
writable they assume the key was provisioned externally (e.g. docker
env:provision) and throw. Update helper tests to match and correct the
EnvProvision docblock that claimed the runtime was strictly read-only.
* test(security): make short-key conversion branch injectable and test it
checkEncryption() now accepts an optional CI3SecretConverter so the
CI3->CI4 conversion branch can be exercised in unit tests without a
database. Adds testCheckEncryptionConvertsCi3ShortKeyWhenEnvWritable
which seeds CI3-era ciphertexts via a fake Appconfig model and asserts
the key is rotated and the payload verifies back to the original
plaintext.
* fix(security): abort on backup/read/saveAll failure to avoid data loss
Three related data-integrity fixes:
- backupEnvFile() now returns true/false based on whether the backup
actually exists and is readable. rotateEncryptionKey() aborts before
destroying the key when the backup could not be written to disk.
- rotateEncryptionKey() and provisionThrottleKey() throw
RuntimeException(Error.unable_to_read_env_file) when the .env read
fails, instead of silently replacing the whole file with an empty
string. This prevents a permission error from wiping all keys.
- checkEncryption() and EnvProvision::run() now both roll back to the
backup with abortEncryptionConversion() when the post-rotation
saveAll() throws, matching the migration path (which already did this).
A failing fake Appconfig is used to exercise this in the new
testCheckEncryptionRollsBackWhenSaveAllFails test.
* fix(ci): skip comment job in deploy-pr.yml when prepare was not run
The comment job had if: always(), so it ran even when the prepare job
was skipped (e.g. review was not approved). With PR_NUMBER empty the gh
api call posted to issues//comments, received a 404, and the entire run
showed up as failure. Guard the job with
needs.prepare.result == 'success' so it only runs when PR_NUMBER is valid.
* address coderabbit open items: placeholder guards, message neutrality, ar-EG alignment
- backupEnvFile(): fail when mkdir() or either chmod() fails, so the
pre-rotation backup is actually persisted before the key is replaced
- email/message config views: only show the 'already set' placeholder when
the secret is actually present (prevented false positives on fresh installs)
- Error.unable_to_create_env_file / .unable_to_read_env_file (en + en-GB):
use key-neutral wording since both keys are provisioned with the same keys
- ar-EG/Error.php: align all => arrows on the longest key
Item 7 (filesystem test isolation) is a larger refactor — the tests are
serial on CI and tearDown() restores state per test. Left for follow-up.
* test(security): isolate helper FS tests via Config\SecurityEnv
Introduce Config\SecurityEnv holding envPath/backupPath/lockPath so the
security helper reads its target paths from shared configuration instead of
hardcoded ROOTPATH/WRITEPATH literals. security_helperTest.php now redirects
all three to a unique per-run sandbox under sys_get_temp_dir() and tears it
down in tearDown(), so the suite no longer reads/writes the repository's real
.env and is safe to run in parallel.
No helper signature changes; production callers unaffected.
Addresses CodeRabbit item 7 (issue #4700).
Co-Authored-By: opencode <bot@opencode.ai>
* fix(security): run key-conversion as one locked transaction
Address CodeRabbit Major findings from the 4th re-review of the env
helper and its callers:
1. Hold .env.lock for the entire CI3 -> CI4 conversion transaction
(backup -> rotate -> re-encrypt -> verify -> persist -> cleanup) so a
concurrent worker cannot interleave a key write between the rotation
and the ciphertext save. Split rotateEncryptionKey into a lock-free
core (rotateEncryptionKeyUnlock) plus the existing lock wrapper and a
new rotateEncryptionKeyTransaction that owns the lock across the full
unit and performs both the in-lock rollback (abortEncryptionConversion)
and the in-lock backup removal on success.
2. Treat the legacy value '0' as non-empty data so key rotation still
persists the re-encrypted ciphertext when '0' is the only stored
secret (array_filter would have dropped it and skipped saveAll).
3. Wrap the post-rotation re-encrypt/verify/saveAll sequence in a
catch (Throwable) across all three call-sites so CI4
EncryptionException, ReflectionException from batch_save, a failed
round-trip verify, and any other failure all roll the .env key back
to the pre-rotation state.
4. In Docker Compose, use long-syntax bind with create_host_path: false
and document in INSTALL.md that the host .env must be a regular file
(a missing one is no longer auto-created as a directory, and the
mount now rejects a missing source on Compose implementations that
support the flag).
Files touched: app/Helpers/security_helper.php, app/Commands/EnvProvision.php,
app/Database/Migrations/20220127000000_convertToCI4.php, docker-compose.yml,
INSTALL.md. All 4 existing helper tests still pass via CI.
* fix(security): make abortEncryptionConversion fail loudly on restore failure
The rollback path restored the .env backup with a suppressed
file_put_contents() and an unchecked file_get_contents(). If the restore
failed after the key had already been rotated, .env was left holding the new
CI4 key while the DB still held CI3-era ciphertext, so the data became
undecryptable after the next restart.
Now the backup read is checked for false and the restore goes through the
existing atomicWriteFile() helper; either failure throws so the error is
surfaced instead of silently corrupting the config. Adds a regression test
that forces an unreadable backup and asserts the throw plus that .env is
left untouched.
* fix(security): guard abortEncryptionConversion backup read before touching it
Validate the backup is a regular readable file (is_file/is_readable) before
reading it, so a missing/malformed backup fails loudly instead of emitting a
file_get_contents() warning. The unreadable-backup regression test now
exercises this guard rather than relying on a promoted warning.
---------
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
Co-authored-by: jekkos <jeroen.peelaerts@gmail.com>
Co-authored-by: jekkos <jekkos@users.noreply.github.com>
Co-authored-by: opencode <bot@opencode.ai>
- Broaden OSPOSRules filesystem path regex to allow space, colon,
and backslash for Windows sendmail paths while still excluding
shell metacharacters used unescaped in popen(); update docblock
accordingly.
- Restore/translate mailpath_invalid message across all language
files so users get a real error string instead of empty text.
- Guard Config controller license rendering against malformed or
non-array npm-prod/npm-dev LICENSES JSON, skipping entries
missing required keys instead of raising warnings.
- Document Node.js 20+ requirement in BUILD.md for license
reporting dependency.
- Remove security advisory IDs/URLs from INSTALL.md and code
comments per new AGENTS.md policy against referencing CVE/GHSA
identifiers anywhere in the repo.
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
fix(validation): allow Windows sendmail paths, tighten shell metachar exclusions
Broaden PLAIN_FILESYSTEM_PATH_STRICT to accept real-world sendmail formats
while blocking command injection characters not needed in valid paths.
- OSPOSRules.php: allow space, colon, backslash for Windows paths
(e.g. C:\wamp64\...) and trailing args (-t -i); still excludes
ampersand, backtick, subshell, redirect, and cmd.exe metacharacters
- OSPOSRulesTest.php: add cases for Windows paths, trailing args, and
injection payloads
- Remove 7 ConfigTest assertions that expected metacharacter rejection;
add acceptance test for sendmail path with trailing args
i18n(lang): expand mailpath_invalid message across all locales
- Fill previously empty mailpath_invalid keys across all locales
- Update existing translations (de-CH, de-DE, es-ES, es-MX, fr, nl-BE,
nl-NL) to reflect newly allowed characters; nl locales corrected from
English loanwords to proper Dutch terms
- Add missing key to ckb/Config.php
docs: remove security advisory IDs from public-facing files
- AGENTS.md: extend no-advisory-ID rule to documentation and URLs
- INSTALL.md: drop GHSA reference and advisory link from Host Header
Injection guidance; rationale and fix instructions remain intact
Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
* fix(tests): resolve all phpunit failures (#4626)
Bring the phpunit suite from 153 failures to 0 (281 tests passing):
- Employee: decouple grants block from save_value success; restructure
save_employee new-employee + disallowed-grants early return
- Sale: unify sales_payments_temp schema (add sale_cash_refund,
reference_code) so both creators produce an identical superset table
- Employees controller: provide placeholder password/hash in testing env
so new-employee insert succeeds and grant logic is testable
- TestDatabaseBootstrapSeeder: reset shared connection table-name cache
after bootstrap reset to avoid stale listTables()/tableExists() results
- Config: fix postSaveLocale validation rule syntax
- Test data: use unique employee usernames to avoid UNIQUE constraint
collisions latching strict-mode transStatus=false on the shared conn
- Various test-file and language-string corrections
* test: consolidate employee fixtures in shared trait
Route test employee creation through a single EmployeeFixtureTrait
that delegates to Employee::save_employee(), so fixtures exercise the
same production code path instead of raw DB inserts. Removes six
near-duplicate helpers across EmployeeTest, SalesControllerTest, and
EmployeesControllerTest while preserving each test's specific grant
set.
Closes a piece of the fixture-scattering flagged in #4626.
Closes#4626
* test: add global DROP/CREATE grant and commit theme fixtures
* fix(ci): remove redundant symlink step, set working encryption key
* fix(ci): run phpunit with --no-coverage to avoid no-driver warning
* fix: address code review findings
- Config: restore strict locale validation (min required|integer|>0) and
fix max cross-field check with a new gte_field rule (CI4's
greater_than_equal_to[field] does not resolve the field value)
- Tests: assert rejection for non-numeric/zero/negative/min>max limits
- .env.example: remove shared hard-coded encryption.key (auto-generates);
document Docker env-var usage
- phpunit.yml: scope CREATE/DROP grant to ospos_test.* and provision a
per-run encryption key as an env var
* feat: support ENCRYPTION_KEY env var for encryption key
Read ENCRYPTION_KEY as a fallback for the encryption key when the
config value is empty. This is a supported, reliable path for Docker /
container deploys and CI, avoiding reliance on the raw dotted
encryption.key env var.
* fix: align Summary_report temp tables with Sale temp table schema
Summary_report created sales_items_taxes_temp and sales_payments_temp with fewer columns than the canonical create_temp_table() in Sale.php. A later reader expecting those columns hit a schema-mismatch SQL error on the shared temp tables. Add internal_tax/sales_tax (sales_items_taxes_temp) and reference_code (sales_payments_temp) so all creators emit the identical column set.
- Require reports_sales grant on postUnsuspend; return 403 on denial
- Reject unsuspend of non-SUSPENDED sales; skip silently on invalid state
- Move clear_all() after validation so an invalid sale_id no longer wipes
the active in-progress cart
- Null-guard get_sale_status() on missing row instead of fatal property
access; widen return type to ?int
- Fix getSaleType null-coalescing — CI4 session default only fires when
key is unset, not when value is null
- Rename get_sale_type → getSaleType, sale_id → saleId (PSR-12 camelCase)
- Extract SaleFixtureTrait with createSale()/createSuspendedSale(); add
regression coverage for auth denial, status gating, and cart preservation
* Validate gift-card payment amounts (GHSA-9847)
Close the negative gift-card amount minting vector: when a forged
payment_type like 'Gift Card:<number>' reaches the catch-all validation
branch, a negative amount_tendered previously passed decimal_locale and was
then routed into Giftcard::decrementGiftcardValue, where value - (-N)
increased the balance (store credit minted at will).
- Add nonNegativeDecimal rule + 'Sales.negative_amount_tendered' message to
the catch-all amount_tendered rules in Sales::postAddPayment(); add the
language key to all 46 locale files (populated in en, empty elsewhere).
- Guard Giftcard::decrementGiftcardValue() against non-positive amounts so
the sink itself can no longer add balance from an inverted subtraction.
- Regression tests: controller-level rejection of negative amount_tendered
and model-level rejection of negative/zero decrements.
* Address PR review: align locale keys, drop advisory refs, add decimal_locale message
- Align negative_amount_tendered '=> with all other keys (46 locale files)
- Remove docblock + inline comment above decrementGiftcardValue()
- Remove GHSA ID and attack-detail description from test; scrub redundant comment
- Add decimal_locale message override + focused malformed-amount test
* Fix formatting and spacing in SalesControllerTest
* fix(lang): remove duplicate negative amount tendered key
Consolidate 'negative_amount_invalid' and 'negative_amount_tendered'
translation keys in Sales.php across all locale files. Both keys held
identical messages, causing redundant translation maintenance.
- Drop 'negative_amount_invalid' key, keep 'negative_amount_tendered'
- Move existing translated text into 'negative_amount_tendered' where
it was previously empty
- Applied across all app/Language/*/Sales.php locale files
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
* fix(sales): allow negative amount_tendered in return mode
Return transactions legitimately produce negative amount_due and
prefilled amount_tendered values, but validation rules previously
enforced nonNegativeDecimal unconditionally, blocking valid returns.
- Detect return mode via sale_lib->get_mode() in Sales::process
- Build amount_tendered rule conditionally: skip nonNegativeDecimal
check when in return mode, keep it for sale/giftcard flows
- Apply the conditional rule to both giftcard and standard payment
branches
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
* test: update expected error message in negative payment test
Sales controller now returns generic numeric-validation message
instead of specific negative-amount message for negative tendered
amounts. Update test assertion to match new lang key.
- tests/Controllers/SalesControllerTest.php: assert
Sales.must_enter_numeric instead of
Sales.negative_amount_tendered
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
* test: remove regression tests for GHSA-9847 negative amount fix
Drop testDecrementGiftcardValueRejectsNegativeAmount and
testDecrementGiftcardValueRejectsZeroAmount from GiftcardTest.
- Remove coverage for decrementGiftcardValue() rejecting
non-positive amounts (negative/zero) in tests/Models/GiftcardTest.php
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
---------
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
Co-authored-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
Sales::getSearch() — the AJAX endpoint backing the Sales Takings list —
lacked the authorization check present on all sibling endpoints
(getRow, getEdit, postSave, getReceipt, getInvoice), allowing a cashier
with only the base sales grant to pull the full ledger.
- Add reports_sales guard with 403 JSON response on denial
- Add regression tests: cashier without grant → 403; employee with
grant → search payload returned
- Clarify getSearch() coverage in SalesControllerTest comments
- Remove duplicate test methods introduced during initial commit
- Validate paymentType is a non-empty string before processing
- Reject negative or zero amounts for all payment types
- Enforce full payment coverage before completing a sale
- Bypass coverage check for invoice and quote mode sales
- Require a valid gift card number before decrementing value;
rollback and return insufficient balance error on missing input
- Add "amount_due_not_covered" and "negative_amount_invalid"
translations across 40+ locales
- Add test coverage for gift card validation, negative amounts,
and quote/invoice zero-payment completion
- Rename snake_case locals to camelCase in postComplete (no behavior change)
- Introduce `valid_path_strict` rule in `OSPOSRules` to enforce stricter path validation, preventing security issues like injection attempts with newline or special characters.
- Update mail configuration validation in `Config` controller to use the new rule for the `mailpath` field.
- Add unit tests in `OSPOSRulesTest` to cover edge cases for `valid_path_strict`.
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
bugfix(items, validation): reject unsafe tax names and fix payments temp table collision
- Add unicode_alpha_numeric_punct rule (OSPOSRules) to allow accented/CJK
chars in text fields while blocking HTML-unsafe chars (<, >) as
defense-in-depth against injection
- Items controller: extract validateItemFields/validateBulkUpdateFields,
validate tax_names on save and bulk update using new rule; add shared
validateFields helper in Secure_Controller to DRY up validation +
JSON error response
- Escape tax_group output in sales/quote.php and receipt_email.php views
to harden output encoding at render time
- Rename sales_payments_temp -> sales_report_payments_temp (Summary_report)
and -> sales_search_payments_temp (Sale model) to avoid name collision
between concurrently-created temp tables
- AGENTS.md: document alignment rule for => columns when inserting new
language keys
Tests:
- Add ItemsControllerTest covering postSave/bulkupdate tax_names validation
- Reject <, > in tax_names on /items/save and /items/bulkupdate
- Verify unicode and apostrophe-containing tax names are accepted
- Cover CSV import helpers: header generation (basic, multiple locations,
attributes), stock-location/attribute header builders, get_csv_file
parsing (plain, BOM-prefixed, multi-row)
- Validate required-header detection for import templates
- Remove outdated tax name test from SalesControllerTest
- Simplify Database class references in SalesControllerTest
i18n:
- Add tax_name_invalid translation to Items.php for 20+ locales, inserted
alphabetically after tax_category in each file
- Normalize quote style in ar-EG/Items.php to single quotes
- Add ka/Items.php Georgian locale scaffold
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
fix(reports, home): strengthen method grant validation and URI decoding (#4660)
- Reports, Home: fix URI decoding in hasGrant() method checks to use
urldecode consistently, preventing malformed URI segments from
bypassing access controls
- Reports: rename snake_case variables to camelCase for PSR-12 compliance
- Reports: adjust access checks to accurately handle null submodule IDs
Tests:
- Add grant check tests for encoded URI inputs across Reports and Home
- Add test case for employee access with base reports grant
- Add secondary grant check for reports_customers in relevant test cases
- Confirm logout bypass remains functional and properly controlled
- Refactor TestDatabaseBootstrapSeeder to expose static reset() for
per-class DB re-initialization instead of only via seeder run()
- Standardize session handling, setup logic, and boolean declarations
- Use unique data in test helpers to avoid collisions
- Add docblocks to ReportsControllerTest and HomeTest for PSR-5 compliance
- Add exception handling for failed employee creation in test setup
- Wrap password validation test in try-finally to guarantee state cleanup
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
- Sales::getSearch now enforces reports_sales grant before running
search, returns 403 with lang message when missing
- Rename snake_case helpers/methods to camelCase across
Sales controller, Sale model, and tabular_helper
(get_sale_data_row -> getSaleDataRow, get_payments_summary ->
getPaymentsSummary, sales_headers -> salesHeaders, etc.)
- Config/OSPOS: reset DB data cache before checking app_config
table existence to avoid stale schema cache in tests
- TestDatabaseBootstrapSeeder: expose static reset() so tests can
rebuild schema once per class instead of only via seeder run()
- SalesControllerTest: bootstrap DB once per class, seed once,
refresh app settings each setUp, add tests for search endpoint
authorization (cashier denied, supervisor allowed), move
createTestItem into shared ItemFixtureTrait
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
fix(barcode): escape payload fields to prevent XSS; PSR-12 refactor
- Apply `esc()` to name, ID, item number, category, and company name in `Barcode_lib` payloads
- Remove redundant `urldecode()` in `Item_kitsController` to prevent triple decoding
- Rename variables and methods to camelCase across barcode, item_kits, and tests
- Add type hint for `$layoutType` parameter in `manageDisplayLayout`
- Add/update unit tests covering escaping and HTTP response assertions
Signed-off-by: objecttothis <17935339+objecttothis@users.noreply.github.com>
fix(sales): gate per-record endpoints behind reports_sales grant (REDACTED)
Cashiers holding only the base `sales` grant could reach per-sale endpoints
(getRow, getEdit, postSave, getReceipt, getInvoice, getSendPdf, getSendReceipt)
that require `reports_sales`. getManage() enforced this at the list level, but
individual endpoints did not re-check. Regression tests added.
Auth:
- Introduce `IsLoggedIn` filter to centralize login checks across controllers
- Replace custom `AccessDeniedRedirectException` with built-in `RedirectException`
Employees:
- Add `DISALLOW_PASSWORD_CHANGE` and `DISALLOW_GRANT_CHANGE` env vars to restrict
credential and permission changes in locked-down environments
- Extract `hasGrantsChanged()` to streamline `postSave`
Refactor:
- Rename snake_case variables to camelCase in Sales, Items, and Employees
controllers for PSR-12 compliance
- Use explicit `db_connect()` for transaction clarity in Items controller
Fixes:
- SMTP config entries fall back to defaults via null coalescing
- Migration uses `DROP FOREIGN KEY` instead of `DROP CONSTRAINT`
- Password hash upgrade only sets session on successful `hash_version` update
- Correct lang key for unknown error in Module model
Language:
- Translate `error_grant_change_disallowed` / `error_password_change_disallowed`
across all 44 supported locales with => alignment matching en reference
- Fix "cannot be deleted" messages and misc typos across ~15 language files
Tests:
- Bootstrap seeder only once in ItemsCsvImportTest; close connection after
- Restore `DISALLOW_GRANT_CHANGE` in teardown to prevent side effects
- Use `uniqid()` for test user data to avoid collisions
Signed-off-by: 17935339+objecttothis@users.noreply.github.com
* fix(items): add explicit sentinel value for clearing supplier in bulk edit
This fixes a regression introduced in the fix for [REDACTED]
Introduce `Item::CLEAR_SUPPLIER_OPTION = 'NONE'` to distinguish between
\"leave supplier_id unchanged\" (empty string) and \"clear supplier_id\"
(sentinel). Previously, empty string was ambiguous.
- Add `CLEAR_SUPPLIER_OPTION` constant with doc comment explaining intent
- Update supplier dropdown to include sentinel as first real option
- Shift empty string to mean \"do nothing\" across all bulk edit fields
* test(items): add regression tests for mass assignment in bulk edit
Cover [REDACTED]: Item::update_multiple() bypasses model
$allowedFields via Query Builder, allowing unintended field writes
during bulk edit operations.
* fix(items): add type validation to bulk-edit field filter
filterBulkEditFields now validates field values before accepting them:
- Non-scalar values are rejected (array injection guard)
- Price/quantity fields are locale-parsed to floats, invalid strings skipped
- Boolean fields must be 0 or 1, other values skipped
- supplier_id must be numeric; CLEAR_SUPPLIER_OPTION still nulls it
Update tests to assert parsed types (float for prices, int for
supplier_id) and replace the fill-all-fields fixture with a realistic
input that only covers fields a form would actually submit.
* test(items): add supplier cleanup and helper methods to bulk update tests
- Track created supplier person IDs for teardown cleanup
- Delete supplier records in tearDown to prevent test pollution
- Extract item/supplier creation into reusable helper methods
* style(tests): rename variables to camelCase in ItemBulkUpdateTest
* refactor(items): rename snake_case variables to camelCase
Convert Item model, Items controller, and bulk update tests to
PSR-compliant camelCase naming per project conventions.
- Rename update_multiple to updateMultiple in Item model
- Rename local variables (item_data, items_to_update, tax_names, etc.)
to camelCase across Items controller and Item model
- Update ItemBulkUpdateTest to use new updateMultiple method name
- Reorder and update AGENTS.md naming conventions
* style(tests): convert snake_case variables to camelCase in ItemBulkUpdateTest
Rename local variables and property names to camelCase for PSR-12
consistency, matching convention used elsewhere in new test code.
objecttothis