Add interactive OPD, IPD, and diagnosis dashboards with historical date ranges, KPI summaries, doctor and department workload, referral breakdowns, and diagnostic workflow statuses.
Route User Management and Role Master action results through fixed top-right notifications, including SPA partial responses, validation failures, session actions, and network errors.
Move per-user permissions into row actions, add contextual module permission guides, show selected-user details, and relocate online sessions to the profile menu.
Add configurable role management, expanded project permission coverage, secured module routes, and single-active-session enforcement with online user administration.
Integrate Final Diagnosis with canonical disease master data and add reusable My/Master template CRUD for discharge narrative sections, including summary investigations, course, other advice, and discharge summary.
Restore surgery and procedure bindings after SPA updates, validate procedure dates without SQL errors, and prevent medicine autocomplete Enter from submitting while preserving layout and autofill.
Load referral master data into patient registration/edit screens, persist patient referby, expose searchable referral inputs in patient and IPD forms, and make referral title editable as free text in referral master CRUD.
- Add checkForUpdates() with 10-minute cached GitHub API check
- Add fetchLatestCommitInfo() to get latest commit SHA + message from GitHub API
- Save deployed commit SHA to writable/deployed_commit.txt after each deploy
- Invalidate update cache on successful deploy so next check is always fresh
- Button turns yellow + red NEW badge when update available
- Show commit message and short SHA diff (deployed → latest) in alert
- Show green 'Up to date' with deployed SHA when already current
- GitHub API rate limit safe: checks once per 10 mins via file cache
- Remove initial 'running' history entry (orphans when PHP times out)
- Add purgeStaleRunning() to mark old 'running' entries as 'timeout' on next deploy
- Add set_time_limit(300) to prevent PHP timeout during large ZIP download
- Add timestamps to log file entries for easier debugging
- Include ZIP size and source URL in success detail
- History now keeps last 20 entries (was 12)
- Show dash instead of blank for entries with no detail
- Style 'timeout' status same as 'failed' (red pill)
- Move detail modal from system_ops_panel.php to system_ops.php
- Modal is now outside #systemOpsPanel so panel refresh never destroys it
- Skip refreshPanel() while modal is open (prevents backdrop desync)
- Add hidden.bs.modal handler to clean up lingering backdrop and body classes
- Fix show.bs.modal handler to correctly read data-detail via jQuery .data()
- Add white-space:pre-wrap to detail content for readable log output
- readNetworkStatus() now returns array of all interfaces instead of single string
- Parses 'ip addr' output (with ifconfig fallback) for all IPv4 interfaces
- Each interface classified as LAN / VPN / Public by name and IP range
- VPN detected by interface prefix: wg*, tun*, tap*, vpn*
- RFC1918 ranges (192.168.x, 10.x, 172.16-31.x) shown as LAN
- Display: icon + IP + interface name badge + type label per row
- Handles NAT servers showing both local LAN and VPN IPs simultaneously
- Remove Restart Web, Restart PHP-FPM, Shutdown, Reboot buttons
- Remove runServerAction() library method and action() controller
- Remove /system-ops/action route (no longer needed)
- Parse /proc/mdstat into structured array (name, level, health, drives)
- Show each RAID array as a card with colored drive icons per-slot
- Green drive = healthy, red = failed, blue = spare
- Show rebuild progress bar when array is rebuilding
- Internet status shown as colored badge with wifi icon
- Services shown as colored badges (green=running, grey=other)
- Remove 'Update HMS (Git)' button and related functionality
- Remove runUpdate() method from SystemOperations
- Remove update() controller method
- Remove /setting/admin/system-ops/update route
- Remove setup_git_config.sh (no longer needed)
- Keep 'Deploy from GitHub' button as the only update method
- Rationale: Direct GitHub deployment is more reliable and requires no git config
Result: Single, simpler deployment method that always works
- Add POST route for /setting/admin/system-ops/updateDirect
- Add GET route for /setting/admin/system-ops/diagnose
- Both endpoints were implemented but routes were missing
This fixes 'Request failed' error on Deploy from GitHub button
- New 'Deploy from GitHub' button downloads latest code directly
- No git configuration required - more reliable for git-unfriendly servers
- Automatically syncs changed files (skips .git, vendor, writable)
- Both methods available: git pull (if configured) or direct GitHub download
- Each has a clear confirmation dialog explaining the method
- Handles temporary extraction and cleanup automatically
This solves HMS deployment when git is not accessible to PHP process
Enhanced error reporting:
- Show detailed diagnostic information: git version, user config, email, remote, branch
- Check .git directory and config file existence
- Detect if repository has uncommitted changes
- Error categorization: PERMISSION, CONFIG, REPO, NETWORK, SSL, GIT errors
- Actionable error messages with command suggestions
- Report raw error output to log file for debugging
Added setup_git_config.sh script:
- Automated configuration for www-data user
- Creates .ssh directory with proper permissions
- Sets git user.name and user.email globally
- Disables SSL verification (for HTTPS repos)
- Configures git to use HTTPS remote URL
- Fixes directory ownership and permissions
- Tests git status to verify configuration
Usage on live server:
sudo bash setup_git_config.sh
This script handles the common git configuration issues
that prevent 'Update HMS' button from working.
- Added findGitCommand() method to check for git in common locations:
* /usr/bin/git (standard Linux)
* /bin/git (alternative location)
* /usr/local/bin/git (compiled from source)
* Falls back to 'which git' and 'command -v git'
- Uses absolute git path in all shell commands
- Fixes 'Git command not found' error on systems where git is not in PATH
- No dependencies added, uses PHP file_exists() + is_executable() checks
Pre-checks before git pull:
- Verify .git directory exists
- Check if git command is available
- Run git status to detect permission/connectivity issues
Better error messages for common failures:
- Permission denied → 'Git directory may not be writable'
- Authentication failed → 'SSH keys may not be configured for www-data'
- Connection refused → 'GitHub server unreachable'
- fatal errors → Parse and return specific error details
Returns actual error message instead of generic 'Check logs for details'
- Method 1: ping to Google DNS (8.8.8.8)
- Method 2: DNS resolution via nslookup
- Method 3: DNS query using dig
- Method 4: curl HTTP HEAD to google.com
- Method 5: wget --spider
- Method 6: fsockopen to Google DNS (8.8.8.8:53)
- Method 7: curl to Cloudflare DNS (1.1.1.1)
- Method 8: PHP gethostbyname() for DNS resolution
Tries multiple DNS providers (Google, Cloudflare, Quad9)
Falls back to 'Offline' only if all methods fail
Handles firewalls, ICMP blocks, and missing commands
- Method 1: \['SERVER_ADDR'] from Apache/Nginx (most reliable)
- Method 2: /sbin/ip (absolute path, works if ip not in PATH)
- Method 3: Parse /proc/net/route for default gateway interface + IP
- Method 4: hostname -I (space-separated IPs)
- Method 5: ifconfig (fallback for older systems)
- Method 6: Socket connection to 8.8.8.8 (pure PHP, no shell commands)
Each method validates IP and skips loopback (127.x, ::1, ::ffff:127.x)
Fixes 'Unavailable' when shell commands fail or aren't available
- Changed primary method to: ip -4 addr show scope global (gets public/private IPs only)
- Added isLoopbackIp() helper to filter out loopback addresses (127.* and ::1)
- Fallback: hostname -I with loopback filtering
- Last resort: gethostbyname() with loopback check
- On remote server with eth0: 159.89.175.190, will now show public IP instead of 127.0.0.1
Three-level fallback chain:
1. gethostbyname(gethostname()) - PHP built-in, works on Linux and Windows
2. ip -4 addr show scope global | grep inet - Linux shell fallback
3. hostname -I - Linux shell secondary fallback
All results validated with FILTER_VALIDATE_IP before display.
- Added "View HMS Features & Capabilities" link below the login card
- Copied HMS_FEATURES_PRESENTATION.html to public/ so it is accessible
via base_url without auth (same pattern as software-use-license.html)
- Link opens in a new tab
- HMS_FEATURES_PRESENTATION.html: full styled presentation with cover,
professional introduction, 11 feature sections, tech stack, compliance,
print/PDF button, and proper @media print A4 layout
- HMS_FEATURES_PRESENTATION.md: plain Markdown version of the same content
- system_ops.php: fix confirm dialog text for Update HMS button
- Added "Details" column to Update History table
- Shows info icon button when error details are available
- Clicking button opens a modal with full error message
- Details displayed in pre-formatted text for easy reading
- Helps diagnose why HMS updates failed or timed out
When an update fails, users can now click the info icon to see the exact
git error output, making troubleshooting much easier.
Update improvements:
- Renamed button from "Update System" to "Update HMS"
- Changed update to run only 'git pull' (removed database migrations)
- Added 60-second timeout to prevent system hangup
- Better error messages for timeouts and failures
- Logs update status (running/success/failed/timeout) to history
- Confirmation dialog now says "Pull latest HMS code from repository?"
This ensures quick updates without long-running migrations that could hang
the admin interface. Database migrations can be run separately as needed.
Resource Usage improvements:
- Added visual progress bars for CPU, RAM, and Disk usage
- Progress bars use Bootstrap styling (success/info/warning colors)
- Display percentage and detailed metrics (e.g., RAM 4 / 128 MB (3%))
- Bars animate smoothly with CSS transitions
CPU detection enhancements:
- Added fallback to /proc/stat parsing when shell commands unavailable
- Added fallback to PHP getrusage() for CPU time estimation
- CPU now displays even when shell_exec is disabled
- Graceful fallback chain ensures data is always available
Progress bar styling:
- Success (green) for CPU
- Info (blue) for RAM
- Warning (orange/yellow) for Disk
- All bars styled with height, rounded corners, and smooth transitions
When shell_exec() is disabled on the server (common in production), the system
operations panel now gracefully falls back to PHP built-in functions:
- gethostname() for hostname instead of shell 'hostname' command
- php_uname() for OS info instead of shell 'uname -a' command
- /proc/uptime parsing or unavailable status for uptime
- disk_free_space() and disk_total_space() for disk usage
- memory_get_usage() and PHP memory limit for memory usage
- Improved parsePhpMemoryLimit() to handle memory limit parsing
This ensures the admin panel shows meaningful information even in restricted
production environments where shell_exec is disabled for security reasons.
- Added System Operations link to Admin dashboard (app/Views/Setting/Admin/index.php)
- Uses load_form_div to load the system operations page inline
- Added SystemOperations library for server status collection and actions
- Tile uses server icon (bi-server) and appears in the admin panel grid
- Link: javascript:load_form_div(base_url('setting/admin/system-ops'), 'maindiv', 'System Operations')
- Added SystemOps controller with index and panel endpoints for dynamic content refresh
- Created system_ops_panel.php fragment for AJAX-based UI updates (no full page reload)
- Implemented 30-second auto-refresh using AJAX instead of location.reload()
- Fixed health badge contrast by adding explicit dark text color (#1f2937)
- Added jQuery availability checks to prevent script errors on page load
- Included server overview, resource usage, services, events, and update history
- Integrated delegated event handlers for action buttons (Update, Restart, Reboot, Shutdown)
- All buttons are protected with confirmation dialogs
- Layout maintains NiceAdmin card styling and responsive design
Routes added:
- GET /admin/system-ops/panel - returns panel fragment for refresh
- Diagnosis.php: resolve doctor mapping via ABDM Report Doctor Mapping,
compile pathology/diagnosis reports into separate Letter Head and
Plain Paper PDFs, add compiled-pdf-links endpoint
- Routes.php: add diagnosis/compiled-pdf-links route
- pathology_detail.php: render/persist dual compiled PDF buttons
- Setting/Template.php: per-modality Letter Head / Plain Paper template
mapping via hospital_setting keys
- Setting/Template/index.php: fix tile links using load_form (which only
replaces #main) instead of load_form_div, which broke every internal
AJAX reload (template type switch, modality switch, reset) on the
diagnosis/discharge/document print template settings pages
- diagnosis_print_settings.php: fix corrupted watermark_alpha number
input (missing step attribute) that silently blocked form submission
for templates with non-integer-step legacy values
- Patient.php: broaden non-fatal-pending detection in abdm_content_auto_flow() to recognize more bridge error wordings (No HIU sessions found / matching the criteria) and HTTP 400, stopping spurious 'Auto flow failed' toasts.
- Opd_prescription.php: remove dead opd.fhir.generated bridge_sync_queue enqueue in storePrescriptionFhirBundle(). This unprefixed event type routed to the CSNOtk terminology bridge which has no consumer for it, always 404ing. Real ABDM sharing already works via DreamsoftConnector::sharePrescriptionBundle() (abdm.fhir.share.requested).
- Fix groupWorkflowRowsIntoSessions() to correlate workflow rows by stable
request_id/abdm_consent_request_id instead of chronological proximity to
the newest CONSENT_REQUEST row. Background polling reconciles all
historical sessions in one batch, so older sessions' rows previously got
silently merged into the newest session, causing 'Show Data' to display
an unrelated older session's single facility instead of all consented
facilities.
- Skip synthetic orphan sessions (stray rows with no owning CONSENT_REQUEST)
in getAbdmConsentRequestsList()/getAbdmConsentArtifactDetail() so they
don't show as phantom history rows or get picked as 'most recent'.
- Fix abdm_documents() consent_request_id filter to match all known sibling
artifact IDs for a session (via abdm_hiu_consent_artifacts), not just an
exact single artifact id, so all facilities under one consent request
appear together.
- Fix CreateAbdmHiuConsentArtifacts migration: reduce VARCHAR(180) to
VARCHAR(100) on consent_request_id/artifact_id to satisfy MySQL's
1000-byte composite key length limit (was silently failing to migrate).
- Add fetchAllArtifactsAfterGrant() cascade in M3HiuWorkflowService to fetch
every known artifact under a granted consent, used by both the cron poll
and a new abdm_check_live_status() endpoint/button.
- getLatestAbdmSyncSnapshot() session-scoping fix to avoid an older granted
session masking a newer in-progress one.
- Relabel Dr. Joshi Clinic documents (previously mislabeled 'Test Hospital'
from the HIP's own FHIR bundle Organization name).
Abha::verifyOtp() and Abha::verifyCommOtp() were overwriting the
user-typed, OTP-verified mobile number with whatever the ABDM bridge's
verify-otp response echoed back for 'mobile'. Many bridge responses
(especially plain mobile-OTP verification, as opposed to Aadhaar eKYC)
do not echo the mobile at all, so it silently became empty and was
never saved to patient_master.mphone1 during ABHA Register (By ABHA
Number / QR / Facility QR / Mobile OTP) or ABHA Create/Verify.
Fix: fall back to the originally submitted/verified mobile when the
gateway response omits it. Also updated Patient_V.php JS (Verify ABHA
tab, ABHA Register Method 1 and Method 4) to resend the mobile number
along with the verify_comm_otp call so the server has it available.
Address/email are unaffected by this bug: those fields are already
mapped via applyAbhaMetaColumns() when present, but the ABDM mobile-OTP
verification API generally doesn't return KYC demographic data (that
only comes from the Aadhaar-eKYC flow) - so they may legitimately
stay blank for phone-based registration methods.
HMS instances behind NAT with no live public IP can hit an expired/unreachable
TLS cert on their own base_url() when doing the internal handoff self-call to
AbdmGateway::share_prescription_bundle, causing 'Gateway handoff failed:
curl error 60: Peer's Certificate has expired'. This call never leaves the
local server, so skip peer verification by default via new
fhirGatewayHandoffSslVerify() helper. Set FHIR_HANDOFF_SSL_VERIFY=true in .env
to re-enable strict verification on deployments with a valid public cert.
Does not affect the actual outbound ABDM bridge submission (EAtriaBridgeConnector),
which keeps its own independent SSL verification/toggle.
Those operations can store the entire decrypted FHIR bundle (incl. base64 scanned-document attachments, sometimes MB per row) in response_json. fetchAbdmWorkflowRows()/getLatestAbdmSyncSnapshot() only ever read small consent-status/hi_types fields out of response_json, which never appear in bulk data_fetch payloads anyway, so cap it at the SQL level instead of buffering it into PHP memory just to build a list view.
Confirmed empirically: the SAME clinical document (identical FHIR Bundle.id) gets re-delivered by the bridge under a DIFFERENT transaction_id/consent_ref every time it is re-fetched (e.g. via the by-ABHA-address historical fetch, which returns every past consent session for the patient). The old dedup hash included transaction_id+consent_ref, so every re-fetch created a brand-new duplicate row instead of updating the existing one -- observed as literal duplicate rows in the OPD ABDM Fetched Data list (same date/title appearing 2-3x). bundle_id is set once by the HIP and is stable across re-fetches/consents, so pair it with care_context_reference to uniquely identify a document instead.
- M3HiuWorkflowService: data_fetch now supports fetch_by_request_id=1, sending request_id (not consent_id) per bridge Method 1 -- targets the current/self consent request's own session, distinct from Method 2 (consent_id, which can resolve to a stale session sharing the same artifact).
- Patient::abdm_content_fetch_only(): tries Method 2 (consent_id) -> Method 1 (request_id, self data) -> Method 3 (abha_address, historical fallback) in order, stopping as soon as real documents are found.
- Fixed a real pre-existing bug this surfaced: getLatestAbdmSyncSnapshot() used the same flawed 'status=SUCCESS implies COMPLETED' check (without requiring workflow_state=DATA_RECEIVED) already fixed elsewhere in computeConsentSessionDetail(). This let broad/historical data_fetch rows (status=success, but decrypted_data=[]) hijack priority-500 'best' selection and silently drift the resolved 'latest granted consent' to an unrelated historical consent_id after Method 3 fallback ran. Now requires workflow_state=DATA_RECEIVED, matching the earlier fix.
Per bridge gateway team guidance, GET /v1/hiu/data/fetch?abha_address=... returns ALL historical decrypted records across every past consent for a patient, unlike the consent/request-scoped Methods 1/2 previously used exclusively. Added M3HiuWorkflowService support for a fetch_by_abha_address flag on the data_fetch operation, and wired abdm_content_fetch_only() to automatically fall back to this broader lookup when the consent-scoped fetch finds nothing new -- this surfaced 14 previously-invisible historical documents (consultation reports, pathology/cytology) for the test patient that were never being fetched before.
abdm_content_fetch_only() always returned phase=COMPLETED and 'Records fetched successfully' whenever the data_fetch API call itself succeeded (fetchResult.ok===1), even when documents_persisted+documents_updated was 0 (bridge/HIP hasn't actually delivered any records yet -- decrypted_data still empty, session status 'requested'). Now check documents_persisted+documents_updated: if 0, report phase=GRANTED with a clear 'not sent any records yet' message instead of falsely claiming success.
computeConsentSessionDetail() treated any data_fetch/hi_data_push_callback row with status=success as proof the session reached COMPLETED. But the DB 'status' column only reflects whether the underlying API call/poll succeeded (no HTTP/transport error) -- NOT whether data was actually received; that's tracked separately via workflow_state (DATA_PENDING vs DATA_RECEIVED). A data_fetch poll that succeeds at the HTTP level but returns decrypted_data=[] (bridge still processing) was incorrectly marked COMPLETED instead of staying at GRANTED. Now also require workflow_state === DATA_RECEIVED for the COMPLETED phase.
When a consent session's overall phase resolves to GRANTED/COMPLETED via a data_fetch or hi_data_push_callback row (proof data was actually fetched), but no consent_reconcile/consent_status row in that session ever recorded which hi_types were granted (e.g. data arrives via async push before any reconcile poll runs), grantedHiTypes stayed empty and every requested HI type was incorrectly marked DENIED. Fall back to treating the full requested set as granted when we have positive proof of GRANTED/COMPLETED but no negative evidence of per-type denial.
- New Patient::abdm_content_request_custom(int \) endpoint builds a custom
consent{} block (purpose, hiTypes, permission.dateRange, dataEraseAt) from
user-submitted hi_types[]/date_from/date_to/purpose form fields, instead of
always using M3HiuWorkflowService::buildDefaultConsent()'s hardcoded
OPConsultation+DiagnosticReport / 365-day defaults. Reuses the existing
requester/hiu-id resolution and hiTypes normalization already built into
preparePayloadForOperation() for the consent_request operation, so no
M3HiuWorkflowService changes were needed.
- New route (POST) patient/abdm_content_request_custom/(:num).
- New 'Custom Request' button + '#abdmCustomConsentModal' form
(Patient_Profile_Opd_V.php) with HI type checkboxes (OPConsultation,
DiagnosticReport, Prescription, DischargeSummary, HealthDocument,
ImmunizationRecord, Wellness, Invoice), Valid From/To date inputs
(defaulted to today-365d..today), and a Purpose text field. On submit,
reuses the existing auto-flow polling loop (runAutoFlowStep) with the
returned request_id to track status through to data fetch.
- Live-verified end-to-end on patient 11 (Devender Singh): submitted a
request for OPConsultation + DiagnosticReport + Prescription; sandbox
granted only OPConsultation/DiagnosticReport, and the Phase 1 consent
detail modal correctly showed Prescription as DENIED alongside the two
GRANTED types, confirming the partial-grant per-HI-type breakdown works
correctly end-to-end.
- New Patient::getAbdmConsentArtifactDetail() helper builds a requested-vs-granted
HI type breakdown for a patient's current ABDM consent artifact, using the
bridge-team-confirmed 'Option A' model: overall status is single-valued
(GRANTED/REVOKED/EXPIRED/DENIED), per-type status is derived by comparing the
requested hiTypes (from the saved consent_request request_json) against the
granted hi_types array (already flattened onto responses by
M3HiuGatewayClient). No bridge changes or DB migrations required.
- New Patient::abdm_consent_detail(int \) JSON endpoint + route
patient/abdm_consent_detail/(:num).
- New 'View Consent' link + Bootstrap modal on the OPD ABDM tab
(Patient_Profile_Opd_V.php) rendering a lifecycle stepper, consent metadata
(Consent ID, Purpose, Valid From/To, Erase Date), and a per-HI-type table
(Document Type / Permission / Status / Timestamp).
- Fixed a bug found during live testing: once a consent reaches COMPLETED
(after a successful data_fetch), the highest-priority workflow row has no
hi_types field (it's a FHIR data bundle), which was causing all granted
types to show as DENIED. Now searches workflow history for the most recent
row that actually carries the granted hi_types payload.
- Live-verified against patient 15 (Meera Bisht): both requested HI types
(OPConsultation, DiagnosticReport) correctly show GRANTED with timestamps.
Previously, clicking 'Fetch ABDM Records' when a prior consent request was
stuck in REQUESTED/PENDING (no GRANTED/DENIED callback ever received from
the ABDM bridge/PHR - e.g. patient never opened/approved it in their PHR
app) would just resume/re-poll that same dead request indefinitely, since
getLatestAbdmSyncSnapshot() always prioritized the existing REQUESTED
request over starting a new one.
Now:
- getLatestAbdmSyncSnapshot() computes the age of a REQUESTED/PENDING
snapshot's updated_at and marks restart_required = true once it has
been pending for more than ABDM_PENDING_STALE_SECONDS (1 hour).
- abdm_content_auto_flow() honors this flag: when the last sync is stale,
it no longer resumes the old request_id and instead falls through to
create a brand-new consent_request against the bridge.
Verified locally: a 6h-old stale REQUESTED request for a test ABHA was
correctly abandoned and a fresh consent_request was created and granted
on the next 'Fetch ABDM Records' click, instead of re-polling the dead
request.
- M3HiuDocumentRepository::persistFromDataFetch no longer includes the
per-call request_id in the document dedup hash. request_id is unique
every time data_fetch runs, so it was causing a brand-new duplicate
row to be inserted on every re-fetch of the exact same ABDM document.
Now dedup key is transaction_id + consent_ref + care_context_reference
+ bundle_id, which stays stable across repeated fetches -> re-fetches
UPDATE the existing row instead of duplicating it.
- Added Patient::abdm_content_fetch_only() + route
billing/patient/abdm_content_fetch_only/(id): a lightweight action for
when consent is already GRANTED/COMPLETED. It calls data_fetch directly
using the saved consent reference, skipping consent_request + reconcile
polling entirely, avoiding unnecessary bridge calls/noise.
- Patient_Profile_Opd_V.php: added a 'Fetch Records' button that only
appears once consent phase is GRANTED or COMPLETED, wired to the new
endpoint; on success it refreshes the document list in place. The
original 'Fetch ABDM Records' button still owns the full consent
request -> reconcile -> fetch flow and already stops/re-enables
correctly on DENIED/FAILED (verified, no change needed there).
- One-time cleanup: deduped existing abdm_hiu_documents rows in dev DB
(13 -> 3) and recomputed doc_hash with the new stable key.
Adds a 'Verify SSL Certificate' switch on Admin -> ABDM Gateway page,
persisted as hospital_setting.ABDM_BRIDGE_SSL_VERIFY (default ON/secure).
Wired into M3HiuGatewayClient, EAtriaBridgeConnector, and the settings
page's own Test Connection check. Per-machine ABDM_BRIDGE_SSL_VERIFY env
var still overrides this DB setting for deployments needing a quick
local-only workaround. Verification stays on by default everywhere;
disabling it is an explicit, visible opt-in with an on-page warning.
Some local WAMP/XAMPP/Laragon PHP installs ship an outdated cacert.pem,
causing outbound HTTPS calls to the bridge to fail with cURL error 60
(SSL certificate problem). Rather than globally disabling certificate
verification (as bridge support suggested), verification stays ON by
default and can only be disabled per-deployment by explicitly setting
ABDM_BRIDGE_SSL_VERIFY=false in that machine's .env. The real fix on
affected machines is updating curl.cainfo in php.ini.
- add facility QR support and OPD queue bridge fixes
- support care-context push warnings for lab, radiology, IPD and remaining HI types
- improve ABHA patient matching and diagnosis autocomplete performance
- clean up webcam streams on page navigation
Serialize the full repaired body instead of only the temporary wrapper contents so DOMDocument recovery does not drop later discharge sections before mPDF rendering.
- EAtriaBridgeConnector: fix OPConsultRecord hi_type normalization (was wrongly
mapping to PrescriptionRecord); allow push with either abha_address or abha_id;
add careContextId alias for care_context_reference in pushRecord payload
- AbdmGateway: migrate OPD share to records/push semantics with unique
care_context_reference, ABHA id/address resolution, bridge_record_id capture,
and 409 DUPLICATE_RECORD non-fatal handling
- AbdmGateway: add non-ABHA local-only storage path for OPD/IPD/Lab share
(push_status=local_only, no bridge call, audit logged)
- AbdmGateway: add POST /records/discover endpoint with ABHA-first + fallback
matching by mobile, hospital ID, patient UHID+birthYear (PHR-compatible)
- AbdmGateway: add GET /records/fetch/{careContextId} endpoint returning
decrypted FHIR payload from health_records
- AbdmGateway: dual-write consent artefacts to abdm_consent_records AND
consents table when both present; getActiveConsentRecord falls back to
consents table when abdm_consent_records is absent
- AbdmAuditService: write to audit_logs (generic schema) when present,
fallback to abdm_audit_logs; field-aware insert supports both schemas
- Routes: register /records/discover and /records/fetch/(:any)
- Webhook signature validator: check EKA_WEBHOOK_SECRET then HMS_WEBHOOK_SECRET
for gateway compatibility
- Add ?html=1 parameter to show_discharge to output HTML directly
- Shows debug info: IPD ID, template name, content lengths
- Bypass PDF generation to see rendered HTML content
- Helps diagnose PHP/rendering errors before PDF conversion
Usage: /Ipd_discharge/show_discharge/1007168/1?html=1&tpl=3
- Restore "Discharge Summary" heading for template extraction to work
- Show status as strong text below heading (not as replacement)
- Include discharge summary HTML editor content (instruction_remark)
- Section extraction requires consistent heading names
Previous fix broke section extraction by changing heading text
- Include ipd_discharge_instructions.comp_remark in discharge summary section
- Previously only showed status header, missing actual discharge narrative
- Reuse existing instructionRowForMeta query to avoid duplicate DB calls
- Apply renderRichText() to properly display HTML editor content
Fixes missing discharge summary text for IPD 1007168 and similar cases
- Enhanced normalizeRichText() to detect and preserve HTML from HTML editors
- Added renderRichText() helper to safely render HTML vs plain text content
- Updated 4 rendering locations (narrative sections, systemic exams, investigations)
- Backward compatible: plain textarea fields still work correctly
- Fixes issue where strip_tags() removed all HTML formatting from editor content
Resolves blank discharge summaries in PDF when content saved from TinyMCE/HTML editors
- Replace all patient_master_exten checks with patient_master across Medical.php
- Fixes blank patient info in multiple pharmacy pages:
* /Medical/ipd_print/{id}/credit - IPD print bills (UHID, Patient name)
* Counter sale forms and invoice editing
* Patient search functionality
* IPD group print consolidated views
* Auto UHID lookup endpoints
- patient_master_exten table exists but lacks core patient fields (p_fname, p_rname, p_code)
- Ensures consistent patient data display throughout pharmacy module
- Matches earlier fixes for billing and invoice list pages
- Use patient_master directly in list_med_inv and list_med_orginv methods
- patient_master_exten exists but lacks p_fname, p_rname, p_code fields
- Fixes blank Name/Age/Gender/P Code in /Medical/list_med_inv page
- Consistent with earlier fix for list_org_ipd
- Use patient_master table directly instead of patient_master_exten
- patient_master_exten may exist but lacks p_fname, p_rname fields
- Matches pattern used in IpdBillingModel::baseIpdListQuery()
- Fixes blank patient names/codes in /Medical/list_org_ipd page
- Change getById() to find() - IpdBillingModel doesn't have getById method
- Use ipdEditModel (IpdModel extends Model) which has standard find() method
- Fixes 500 error: Call to undefined method IpdBillingModel::getById
- Rename toggleIpdStatus() to updateDischargeToggleStatus() to match naming pattern
- Add route mapping for /billing/ipd/panel/{id}/discharge/toggle-status endpoint
- Fixes 404 error when clicking toggle button in discharge panel
- Follows same pattern as other discharge operations (discount, charge, update)
- Update isIpdDischargedInvoice to check ONLY ipd_status field (remove fallback checks)
- Add IPD status toggle UI in discharge panel (Active/Discharged control)
- Create toggleIpdStatus controller method with permission check
- Add billing.ipd.status.manage permission for controlling bill edit lock
- Allows editing bills when needed by toggling status independently of discharge_date
- Auto-releases bed when setting status to discharged
- Change colspan from 3 to 4 (normal mode) and 4 to 5 (discount mode)
- Ensures Gross Total, Net Amount, Payment Recd., Balance amounts appear in Amount column
- Fixes blank cells in Rate column by properly spanning Description through Rate columns
- Use COALESCE to check both bed_master.current_ipd_id and bed_assignment_history
- Join with bed_assignment_history where released_date IS NULL for active beds
- Handles cases where bed is assigned in history but current_ipd_id not set
- Fixes bill print showing blank Bed No when assignment exists in history
- Add bed_master join to fetch current bed assignment via current_ipd_id
- Add ipd_master_doc_list and doctor_master joins to fetch assigned doctors
- Use GROUP_CONCAT to handle multiple doctors in single query
- Fixes bill print displaying empty Bed No and Doctor Name fields
- Fix dose columns showing IDs instead of labels (OD, BID, TDS, etc.)
- Add Rx-Group functionality with modal, search, and bulk add
- Add substitute medicine suggestions with one-click use
- Improve medicine autocomplete with debouncing and scope=all
- Load dose master maps in controller and pass to view
- Match OPD Consult UI/UX patterns for consistency
- Implemented automatic bed release on patient discharge
- Added bed transfer functionality in nursing workspace
- Added date/time tracking for bed assignments with duration calculation
- Fixed bed_master.current_ipd_id as source of truth for occupancy
- Created data fix scripts for existing records
- Added bed assignment history with duration display (hours/minutes)
- Fixed display issues in bed assignment tab
- Added route for bed transfer endpoint
- Supports custom assignment datetime for backdated entries
- Automatically marks previous bed as released on transfer
- Clear all output buffers before sending PDF binary to prevent corruption
- Add X-Content-Type-Options: nosniff to force Chrome to treat response as PDF
- Apply fixes to main, fallback, and ultimate fallback PDF response paths
- Fixes issue where Chrome fails to open PDF while Firefox/Edge work
- Critical for PDFs with Hindi/UTF-8 content where output buffering can corrupt binary data
ISSUE 1: {{SIGNATURE_BLOCK}} placeholder not working
- Signature extraction was using wrong end markers
- Start: 'Signature of Consultant'
- End: 'Signature of Medical Officer', 'Signature of Receiver / Date'
- Problem: All three texts are in SAME table row
- Extraction cut off too early (found end marker immediately after start)
FIX: Remove end markers for signature block extraction
- Signature table is LAST element in content
- Extract from 'Signature of Consultant' to end of content
- Now captures entire signature table correctly
ISSUE 2: Other Advice not showing in PDF
- Cached content was generated BEFORE 'Other Advice:' label was added
- Database has data but cached HTML doesn't include label
FIX: Created cache clearing utility
- clear_discharge_cache.php script
- Clears cached content for specified IPD
- Forces regeneration with new labels
DEBUG TOOLS ADDED:
- debug_discharge_content.php - Check what's in cached content
- clear_discharge_cache.php - Clear cache to force regeneration
TESTING:
1. php clear_discharge_cache.php 1
2. Visit /Ipd_discharge/preview_discharge_report/1?regen=1
3. Verify PDF shows Other Advice: and Discharge Summary: labels
4. Verify {{SIGNATURE_BLOCK}} placeholder works in templates
ISSUE 1: Discharge Summary showing in PDF but without label
ISSUE 2: Need separate placeholders for Other Advice and Discharge Summary
FIX 1: Added label to instruction_remark in PDF
- Added Discharge Summary: label before instruction_remark content
FIX 2: Added new placeholders for template flexibility
- OTHER_ADVICE - Shows only Other Advice field
- INSTRUCTION_REMARK or DISCHARGE_ADVICE - Shows only Discharge Summary field
- DISCHARGE_INSTRUCTIONS - Shows full section with all advice
Now templates can control what appears where using specific placeholders.
CHANGES:
1. Line 2248 - Added Discharge Summary: label
2. Line 574-575 - Extract new tokens from content
3. Line 579-581 - Add tokens to vars array
4. Line 520-522 - Add markers to extraction list
TESTING: Go to discharge PDF and verify labels show correctly
ISSUE: instruction_remark (Discharge Summary) data appearing in instruction_other (Other Advice) field after save
ROOT CAUSE: Legacy table sync was writing wrong variable
- Save handler at line 5478 was writing \ to legacy food_text
- Should have been writing \
- When legacy fallback loads (if food_ids empty), wrong data copied to instruction_other
FIX: Changed line 5478 in Ipd_discharge.php
- Before: \['food_text'] = \;
- After: \['food_text'] = \;
FILES MODIFIED:
1. app/Controllers/Ipd_discharge.php - Fixed legacy save handler
2. app/Views/billing/ipd/discharge_create.php - Removed debug comments
TESTING STEPS:
1. Clear corrupt legacy data: php clear_legacy_food.php 1
2. Go to /Ipd_discharge/ipd_select/1
3. Clear both fields
4. Type 'Hello Other' in Other Advice
5. Type 'Hello Discharge' in Discharge Summary
6. Click 'Save Discharge Advice'
7. After reload: Other Advice should show 'Hello Other', NOT 'Hello Discharge'
SCRIPTS ADDED:
- check_instruction_db.php - Check database values
- check_legacy_food.php - Check legacy table
- clear_legacy_food.php - Clean corrupt legacy data
- list_instruction_records.php - List all records
ISSUE 2: After clicking 'Save Discharge Advice', CKEditor turns into plain textarea
ROOT CAUSE: initInstructionTools() was not re-initializing CKEditor after AJAX section reload
- When user clicks button with data-reload-section='section-instructions'
- Form submits via AJAX, returns updated HTML
- Section content replaced with new HTML
- initInstructionTools() called but did NOT restore CKEditor
- Fields remain as plain textareas
SOLUTION: Added CKEditor re-initialization to initInstructionTools()
Changes to discharge_create.php initInstructionTools():
1. Check if CKEDITOR is available
2. Destroy existing instances first (instruction_other, instruction_remark)
3. Re-create CKEditor for both fields with same config
4. Prevents duplicate instances and ensures editors work after reload
TESTING:
- Type in instruction_remark or instruction_other
- Click 'Save Discharge Advice'
- After reload, both fields should still be CKEditor (not textarea)
- Can continue editing with toolbar
NOTE: Issue 1 (data mixing between fields) requires separate investigation
User can test with: php check_instruction_data.php to see actual DB values
Added sanitizeCKEditorSpecialChars() method to handle smart quotes, zero-width spaces, em/en dashes, and other special characters that can break mPDF.
Removes invisible Unicode chars, replaces special punctuation with ASCII equivalents, cleans HTML comments.
ISSUE: Templates auto-regenerate when deleted
User deletes template ID 3, page reload creates template ID 4 automatically
ROOT CAUSE: ensureDefaultDischargeTemplateSeeded() called on EVERY page load
- discharge_templates() method called it on every view
- getDischargeTemplateRows() called it on every discharge PDF generation
- When count=0, it auto-inserted default templates
- Made it impossible to delete all templates
SOLUTION: Changed to manual opt-in template creation
1. Removed auto-seeding calls:
- Setting/Template::discharge_templates() - removed ensureDefaultDischargeTemplateSeeded()
- Ipd_discharge::getDischargeTemplateRows() - removed ensureDefaultDischargeTemplateSeeded()
- Both now only call ensureDischargeTemplateTable() (structure check only)
2. Added manual template creation:
- New 'Create Default Templates' button shown when table is empty
- New endpoint: discharge_templates_seed() - creates defaults on demand
- Route: POST /setting/template/discharge_templates_seed
- Button confirms before creating, shows success message, reloads page
3. Preserved seeding functions:
- ensureDefaultDischargeTemplateSeeded() kept but not auto-called
- Can be manually triggered via new button
- Creates 'Default Discharge Template' and 'NABH Compliant Discharge Summary'
BEHAVIOR NOW:
- Users can delete ALL templates - they stay deleted
- Empty table shows helpful message with creation button
- Templates only created when user explicitly requests them
- No more surprise template regeneration
TESTING:
- Delete all templates
- Page reloads showing empty table with button
- Click 'Create Default Templates'
- 2 templates created (ID 1 and 2)
ISSUE: Template delete returns success but record appears not deleted
ROOT CAUSE: Delete WAS working (verified in database), but page wasn't refreshing properly after success
CHANGES:
1. JavaScript fix (discharge_templates.php):
- Replaced conditional load_form_div / location.assign logic
- Now uses simple location.reload() after 500ms delay
- Ensures page always refreshes after successful delete
2. Controller debug logging (Template.php):
- Added comprehensive debug logs to trace delete flow
- Logs: request ID, row fetch, query compilation, affected rows
- Logs error codes and messages for troubleshooting
- Will help diagnose future issues
VERIFIED:
- Database confirmed template ID 2 was deleted
- Only frontend refresh was failing
- Now forces reload to show updated list
ISSUE: User reported alignment tools not showing and delete not working
ROOT CAUSE: Over-complicated toolbar config with too many additions
- Added Strike, Subscript, Superscript (not in standard build)
- Added Image (might not be in build)
- Added ShowBlocks (might not be in build)
- Added extra config options that might conflict
SOLUTION: Return to ORIGINAL working toolbar + ONLY alignment buttons
- Restored original toolbar config that was working
- Kept ONLY the alignment row: JustifyLeft, Center, Right, Block
- Removed Strike, Subscript, Superscript, Image, ShowBlocks
- Removed extra config: allowedContent, entities, fillEmptyBlocks, etc.
- Delete function unchanged (already fixed in commit d025111)
TOOLBAR NOW:
✅ Original working buttons (Source, Undo, Redo, Bold, Italic, etc.)
✅ Text alignment buttons (JustifyLeft, Center, Right, Block) - USER'S REQUEST
❌ Removed bonus features that caused issues
This is minimal change - only adds what user asked for.
ERROR: Uncaught TypeError: can't access property 'icons', h is null
FILE: ckeditor.js:267
ROOT CAUSE:
Previous commit added 'extraPlugins: justify,pagebreak' to CKEditor config.
The pagebreak plugin is NOT included in the standard CKEditor build being used.
CKEditor tried to load pagebreak plugin, failed, and threw null reference error.
SOLUTION:
1. Removed 'PageBreak' button from insert toolbar
2. Removed 'extraPlugins: justify,pagebreak' config line
3. Justify plugin is CORE plugin - included by default, no need to load explicitly
4. Text alignment buttons (JustifyLeft, Center, Right, Block) still work perfectly
WHAT STILL WORKS:
✅ Text alignment buttons (user's original request)
✅ All other enhanced buttons (Strike, Subscript, Superscript, Image, ShowBlocks)
✅ No console errors
WHAT WAS REMOVED:
❌ PageBreak button (was bonus feature, not in CKEditor build)
TESTING: Reload /setting/template/discharge_templates - no console errors
ISSUE: Delete button executed but did not actually delete templates
ROUTE: /setting/template/discharge_templates/delete/{id}
ROOT CAUSE:
The delete button's fetch() call was sending 'new FormData(form)' which:
1. Included entire edit form data (template_html, header_html, footer_html)
2. These are HUGE LONGTEXT fields that can cause request payload issues
3. CSRF token from form might be stale or missing if no template is being edited
4. Server received malformed request and silently failed
SOLUTION:
- Create fresh FormData with ONLY the CSRF token
- Delete request now lightweight and reliable
- Added success message display after successful delete
- Server-side delete logic already correct (line 2969-3028 in Template.php)
CHANGED:
- app/Views/Setting/Template/discharge_templates.php (line 618-653)
- Replace: body: new FormData(form)
- With: body: deleteFormData (contains only CSRF token)
- Added: setNotice('success', ...) to show success message
TESTING:
1. Go to /setting/template/discharge_templates
2. Click Delete on any non-default template
3. Confirm deletion
4. Template should now be removed from list
5. Success message should display
ISSUE: Medicines were being saved twice - once via Add button AJAX, and again when clicking 'Save Discharge Advice' button.
ROOT CAUSE:
1. Legacy code in save_main handler (lines 5489-5538) read 'discharge_medicine_json' POST field
2. JavaScript serializeDischargeMedicineTable() collected all medicine table rows into JSON
3. On form submit (Save Discharge Advice), JSON was sent and controller re-inserted all medicines
4. Result: duplicate entries in ipd_discharge_prescrption_prescribed table
SOLUTION:
- DISABLED controller code that processes discharge_medicine_json in save_main handler
- DISABLED JavaScript call to serializeDischargeMedicineTable() on form submit
- Medicines now saved ONLY via Add button AJAX (action='add_drug')
- Save Discharge Advice button now only saves dietary advice, instructions, review after, etc.
TESTING:
1. Add medicines via Add button (saves immediately)
2. Click Save Discharge Advice (does NOT save medicines again)
3. No duplicate entries in database
- Add getDoseMasterRows() to load dose masters (opd_dose_shed, opd_dose_when, opd_dose_frequency)
- Add buildDosageDisplay() to format dosage: 'BID (TWO TIME A DAY) / दिन में दो बार लेना'
- Update buildAutoDischargeContent() medication table query to include dosage, dosage_when, dosage_freq fields
- Change table columns from '#|Medicine|Qty|Days|Notes' to 'Medicine Name|Dosage|Qty|Day' matching old CI3 format
- Dosage column shows full text labels from master tables, not numeric IDs
- Add composition support as small text under medicine name for legacy compatibility
- Remove # serial number column and Notes column to match old HMS format
Problem:
- PDF shows old cached content with 'Date of Discharge: 2026-06-05'
- Debug page shows correct new format (status header only)
- Discharge content is cached in ipd_discharge table
- PDF generation uses cached content from database
- Fix only affects newly generated content, not cached content
Root Cause:
When discharge is first created, content is generated and saved to DB.
PDF generation flow (show_discharge method):
1. Load cached content from ipd_discharge table
2. Only regenerate if content is empty
3. Apply template and render PDF
So old discharges still have old format cached.
Solution:
Added 'refresh' parameter to show_discharge() method:
- If refresh=1, ignore cached content and regenerate
- Regenerate using buildAutoDischargeContent() (has fix)
- Save new content to database
- Then render PDF
Code Change (line 5900-5913):
Added:
\ = (int) (\->request->getGet('refresh') ?? 0) === 1;
Changed:
if (\ || trim(strip_tags(\)) === '') {
Usage:
1. Original URL (uses cache):
http://localhost:8080/Ipd_discharge/show_discharge/1/1?tpl=1
2. Force refresh (regenerates):
http://localhost:8080/Ipd_discharge/show_discharge/1/1?tpl=1&refresh=1
3. After first refresh, cache is updated with new format
Subsequent views will use new cached content
Note:
- preview_discharge_report() already has 'regen=1' parameter
- Same functionality, different parameter name
- Both methods now support force regeneration
For existing discharges:
- Add &refresh=1 to URL once to update cache
- Or recreate discharge from edit page
- Or clear ipd_discharge.content for that IPD ID"
Problem:
- {{DISCHARGE_SUMMARY}} placeholder was showing:
1. Discharge status header (Discharge Summary/LAMA/etc.)
2. Department information
3. Treating Doctor(s) information
4. Date of Admission
- User wants ONLY the status header in this placeholder
- Extra information should be removed from {{DISCHARGE_SUMMARY}}
Solution:
Modified buildAutoDischargeContent() method:
Before (lines 1679-1703):
- Built dischargeSummaryParts array with department/doctor/date
- Added all info in one section with header + div containing details
After:
- {{DISCHARGE_SUMMARY}} now contains ONLY the status header
- Removed department, doctor, and admission date from this section
- Status header comes from ipd_discharg_status table via:
ipd_master.ipd_status (or discarge_patient_status) = ipd_discharg_status.id
Returns ipd_discharg_status.status_details field
Status Values:
- ipd_status = 0 → 'Discharge Summary' (hardcoded)
- ipd_status = 1 → From DB: 'Discharge Summary'
- ipd_status = 2 → From DB: 'Leave Against Medical Advice'
- ipd_status = 3 → From DB: 'Refer To Higher Centre'
- ipd_status = 5 → From DB: 'Dead Summary'
- ipd_status = 7 → From DB: 'Discharge on Request'
Now {{DISCHARGE_SUMMARY}} outputs:
<h4 class='discharge-section-heading'>Discharge Summary</h4>
OR
<h4 class='discharge-section-heading'>Leave Against Medical Advice</h4>
That's it - no extra lines below.
If user needs department/doctor/date info, they can use individual tokens:
- {{DEPARTMENT}}
- {{DOCTOR_NAMES}}
- {{ADMIT_DATE_ONLY}}
- {{DISCHARGE_DATE_ONLY}}
- {{ADMISSION_TIME}}
- {{DISCHARGE_TIME}}
Or use {{PATIENT_INFO_TABLE}} which has all patient demographics.
Testing:
- Visit: /Ipd_discharge/debug_discharge_html/1
- Check {{DISCHARGE_SUMMARY}} section
- Should only show status header, no other info"
Problem:
- GET: Ipd_discharge/debug_discharge_html/1 returning 404
- GET: Ipd_discharge/debug_ipd_fields/1 returning 404
- Debug endpoints were created but routes were not registered
Solution:
Added two new routes in app/Config/Routes.php:
1. Debug IPD Fields Endpoint:
Route: Ipd_discharge/debug_ipd_fields/(:num)
Controller: Ipd_discharge::debug_ipd_fields/\
Method: GET
Returns JSON showing:
- All available field names in ipd_master table
- Values of time fields (reg_time, discharge_time)
- Values of department fields (dept_id)
- Values of doctor fields (r_doc_id, r_doc_name)
- Result of getDischargeDepartmentName() function
- Result of getDischargeDoctorNames() function
- Result of safeTime() for time fields
2. Debug Discharge HTML Endpoint:
Route: Ipd_discharge/debug_discharge_html/(:num)
Controller: Ipd_discharge::debug_discharge_html/\
Method: GET
Returns HTML page showing:
- Complete discharge summary HTML source
- Syntax highlighted for easy reading
- All placeholders replaced with actual values
- Useful for checking spacing and HTML structure
Route Placement:
- Added after search_patient_ajax route
- Before ipd_select route
- Follows existing Ipd_discharge route pattern
- Uses (:num) parameter for IPD ID
Usage:
- http://localhost:8080/Ipd_discharge/debug_ipd_fields/1
- http://localhost:8080/Ipd_discharge/debug_discharge_html/1
- Replace '1' with actual IPD ID
Note: Routes must be defined BEFORE more general routes
Otherwise, the (:num) pattern in ipd_select would catch them first
Testing:
1. Pull latest code
2. Access /Ipd_discharge/debug_ipd_fields/1
3. Should return JSON with field values
4. Access /Ipd_discharge/debug_discharge_html/1
5. Should return HTML source page
CRITICAL FIX - User Requirement:
- If ipd_status = 0: Show 'Discharge Summary' only
- If ipd_status != 0: Show value from ipd_discharge_status.status_details
- Remove 'Date of Discharge' line completely
Problem:
- {{DISCHARGE_SUMMARY}} still showed 'Discharge Summary' header regardless of status
- 'Date of Discharge: YYYY-MM-DD' line was appearing
- Previous fix only worked on saved content via regex, not on auto-generated content
Root Cause:
- buildAutoDischargeContent() method had hardcoded 'Discharge Summary' header
- getDischargeStatusText() was querying database but header wasn't being used
- Date of Discharge was being added to dischargeSummaryParts array
Solution:
1. Updated buildAutoDischargeContent() method (line ~1670):
- Removed 'Date of Discharge' from dischargeSummaryParts array
- Added comment: 'NOTE: Date of Discharge removed per user request'
- Call getDischargeStatusText() to get dynamic header
- Use status-based header instead of hardcoded 'Discharge Summary'
- Now generates: <h4>status_details_value</h4> instead of <h4>Discharge Summary</h4>
2. Updated getDischargeStatusText() method (line ~1047):
- Added early return for ipd_status = 0: returns 'Discharge Summary'
- For ipd_status != 0: queries ipd_discharge_status table
- Returns status_details field (primary)
- Falls back to status_desc if status_details is empty
- Final fallback: 'Discharge Summary'
3. Updated buildAutoDischargeSummaryTable() method (line ~1517):
- Used for {{PATIENT_INFO_TABLE}} token
- Now uses getDischargeStatusText() for header
- Changes <h2>Discharge Summary</h2> to <h2>status_value</h2>
- Keeps patient info table structure unchanged
Status Mapping Logic:
- ipd_status = 0 → 'Discharge Summary' (special case, no DB lookup)
- ipd_status = 1 → Query DB → 'Discharge Summary' (from status_details)
- ipd_status = 2 → Query DB → 'Leave Against Medical Advice'
- ipd_status = 3 → Query DB → 'Refer To Higher Centre'
- ipd_status = 5 → Query DB → 'Dead Summary'
- ipd_status = 7 → Query DB → 'Discharge on Request'
Content Structure Before:
<h4>Discharge Summary</h4>
<div>
<b>Department:</b> General Surgery<br>
<b>Treating Doctor(s):</b> Dr. Name<br>
<b>Date of Admission:</b> 04-06-2026<br>
<b>Date of Discharge:</b> 07-06-2026
</div>
Content Structure After:
<h4>Leave Against Medical Advice</h4>
<div>
<b>Department:</b> General Surgery<br>
<b>Treating Doctor(s):</b> Dr. Name<br>
<b>Date of Admission:</b> 04-06-2026
</div>
Testing:
1. Set ipd_master.ipd_status = 0 → regenerate → should show 'Discharge Summary'
2. Set ipd_master.ipd_status = 1 → regenerate → should show 'Discharge Summary'
3. Set ipd_master.ipd_status = 2 → regenerate → should show 'Leave Against Medical Advice'
4. Set ipd_master.ipd_status = 5 → regenerate → should show 'Dead Summary'
5. 'Date of Discharge' line should NOT appear in any case
Files Modified:
- app/Controllers/Ipd_discharge.php:
- getDischargeStatusText() - Added ipd_status = 0 check
- buildAutoDischargeContent() - Removed Date of Discharge, use dynamic header
- buildAutoDischargeSummaryTable() - Use dynamic header
IMPORTANT: User must REGENERATE discharge summary after pulling code
- Click 'Regenerate' button on discharge preview page
- Or delete existing discharge content and reload form
- Old saved content will still have old header until regenerated
Problem:
- {{DISCHARGE_SUMMARY}} always showed 'Discharge Summary' header
- User wants header to change based on patient discharge status
- 'Date of Discharge' line was unwanted
- Need to see HTML source to check spacing between placeholders
Solution:
1. Added getDischargeStatusText() method:
- Queries ipd_discharge_status table
- Uses ipd_master.ipd_status as lookup key
- Returns status_details or status_desc from table
- Fallback: 'Discharge Summary' if status not found
2. Modified buildDischargeSectionTokenVars():
- Now accepts panelData parameter (has ipd info)
- Gets discharge status text via getDischargeStatusText()
- Replaces 'Discharge Summary' heading with status text
- Removes 'Date of Discharge: YYYY-MM-DD' line
- Uses regex to handle multiple HTML tag formats (h1-h6, b, strong, p)
3. Added debug_discharge_html() endpoint:
- URL: /Ipd_discharge/debug_discharge_html/{ipdId}
- Shows complete HTML source with syntax highlighting
- Displays rendered template with all placeholders replaced
- Useful for checking spacing and HTML structure
- Auto-generates content if not saved yet
Database Schema:
- ipd_master.ipd_status: 0-7 (discharge status ID)
- ipd_discharge_status table:
- id: 0-7
- status_desc: 'Status Pending', 'Improved/Recovered', 'LAMA', 'Referred', etc.
- status_details: 'Discharge Summary', 'Leave Against Medical Advice', etc.
Status Mapping (from user's data):
- 0: Status Pending
- 1: Improved/Recovered -> 'Discharge Summary'
- 2: LAMA -> 'Leave Against Medical Advice'
- 3: Referred -> 'Refer To Higher Centre'
- 4: Satisfactory -> 'Discharge Summary'
- 5: Expired -> 'Dead Summary'
- 6: Admission Cancelled
- 7: Discharge on Request
Regex Replacements:
1. Header replacement: /<h[1-6][^>]*>\s*Discharge\s+Summary\s*<\/h[1-6]>/i
2. Bold replacement: /<(b|strong)>\s*Discharge\s+Summary\s*<\/\1>/i
3. Date removal (br): /Date\s+of\s+Discharge\s*:\s*[0-9\-\/]+\s*<br\s*\/?>/i
4. Date removal (p): /<p[^>]*>\s*Date\s+of\s+Discharge\s*:\s*[0-9\-\/]+\s*<\/p>/i
Usage:
- When ipd_status = 1: Shows 'Discharge Summary' (from status_details)
- When ipd_status = 2: Shows 'Leave Against Medical Advice'
- When ipd_status = 5: Shows 'Dead Summary'
- 'Date of Discharge: 2026-06-07' line is removed from all sections
Debug Endpoints:
1. /Ipd_discharge/debug_ipd_fields/{ipdId} - Shows database field values
2. /Ipd_discharge/debug_discharge_html/{ipdId} - Shows complete HTML source
Testing:
- Change ipd_master.ipd_status value
- Regenerate discharge summary
- Header should change based on status
- Date line should not appear
- Use debug_discharge_html to see full HTML structure
CRITICAL FIXES based on actual database schema analysis:
Problem 1: {{ADMISSION_TIME}} showing 12:00 AM
- Code was looking for: register_time, admission_time (don't exist)
- Actual field: reg_time (stores '14:44' format)
- Fix: Changed to use ipd->reg_time field
Problem 2: {{DISCHARGE_TIME}} showing 12:00 AM
- Code was looking for: discharge_date datetime field
- Actual field: discharge_time (stores '05:30' format)
- Fix: Changed to use ipd->discharge_time field directly
Problem 3: {{DEPARTMENT}} showing blank
- Code was using: hc_department.id field for lookup
- Actual field: hc_department.iId (primary key is iId, not id)
- Fix: Changed WHERE clause from 'id' to 'iId'
Problem 4: {{DOCTOR_NAMES}} showing blank
- Code was looking for: doc_list, doctor_id, doc_id (don't exist)
- Actual field: r_doc_id (referring doctor ID = 12)
- Fix: Changed primary lookup to r_doc_id field
Technical Improvements:
1. Enhanced safeTime() function:
- Added regex parsing for 'HH:MM' time format
- Validates hour (0-23) and minute (0-59) ranges
- Converts 24-hour to 12-hour format with AM/PM
- Example: '14:44' -> '02:44 PM', '05:30' -> '05:30 AM'
- Falls back to strtotime() for datetime formats
- Returns empty string for invalid times
2. Updated getDischargeDoctorNames():
- Primary: ipd->r_doc_id (referring doctor)
- Fallback: ipd->doc_list (comma-separated IDs)
- Fallback: ipd->doctor_id, doc_id, consultant_id
- Queries doctor_master table
- Gets specializations from doc_spec + med_spec
- Format: 'Dr. Name [Specialization]' or 'Title Name [Spec]'
3. Updated getDischargeDepartmentName():
- Changed: WHERE id = dept_id
- To: WHERE iId = dept_id
- Keeps fallback to ipd_department table
- Returns department vName field
4. Updated debug endpoint:
- Now checks: reg_time, reg_time_bak, discharge_time, discharge_time_bak
- Shows: dept_id, r_doc_id, r_doc_name
- Removed non-existent fields from output
Database Schema (from live data):
- ipd_master.reg_time: '14:44' (admission time)
- ipd_master.discharge_time: '05:30' (discharge time)
- ipd_master.dept_id: 2 -> hc_department.iId (vName field)
- ipd_master.r_doc_id: 12 -> doctor_master.id (p_fname field)
Testing:
- Admission Time should show: 02:44 PM (from reg_time '14:44')
- Discharge Time should show: 05:30 AM (from discharge_time '05:30')
- Department should show department name from hc_department
- Doctor Names should show doctor name with specialization
Problem:
- {{ADMISSION_TIME}} showing 12:00 AM instead of actual time
- {{DISCHARGE_TIME}} showing 12:00 AM instead of actual time
- {{DEPARTMENT}} showing blank
- {{DOCTOR_NAMES}} showing blank
Need to identify:
1. Which datetime fields exist in ipd_master table
2. Which department field exists (dept_id, department_id, or department)
3. Which doctor field exists (doc_list, doctor_id, doc_id, consultant_id)
4. What values are actually stored in these fields
Debug Endpoint Added:
- URL: /Ipd_discharge/debug_ipd_fields/{ipdId}
- Example: http://localhost:8080/Ipd_discharge/debug_ipd_fields/123
- Replace 123 with actual IPD ID from your discharge page
What it shows:
- All available field names in ipd_master table
- Values of: register_date, discharge_date, register_time, discharge_time, admission_time
- Values of: dept_id, department_id, department
- Values of: doc_list, doctor_id, doc_id, consultant_id
- Result of getDischargeDepartmentName() function
- Result of getDischargeDoctorNames() function
- Result of safeTime() for register_date and discharge_date
Usage:
1. Go to your discharge form
2. Note the IPD ID from the URL
3. Open: /Ipd_discharge/debug_ipd_fields/{ipdId} in browser
4. Check the JSON output to see which fields exist and their values
This will help identify:
- If datetime fields have time component or just date
- Which field names are actually used in your database
- Why placeholders are showing blank/wrong values
Problem:
- {{DEPARTMENT}} and {{DOCTOR_NAMES}} placeholders not working in discharge templates
- Complex GROUP_CONCAT query with LEFT JOINs causing issues
- Query might return NULL or fail silently when specializations missing
Solution - getDischargeDoctorNames():
- Removed GROUP_CONCAT to avoid SQL aggregation issues
- Split into two separate queries: doctors first, then specializations
- Changed specialization JOIN from LEFT to INNER to avoid NULL values
- Query doctors from doctor_master separately
- For each doctor, query doc_spec + med_spec to get specializations
- Build full name format: 'Dr. Name [Spec1, Spec2]' or 'Title Name [Spec1, Spec2]'
- Return comma-separated list of all doctors
Technical Changes:
- First query: SELECT d.id, d.p_fname, d.p_title FROM doctor_master
- Second query (per doctor): SELECT m.SpecName FROM doc_spec s JOIN med_spec m
- Use INNER JOIN for specializations to only include valid matches
- Handle missing specializations gracefully (show doctor name without brackets)
- Handle missing title (default to 'Dr.')
getDischargeDepartmentName() - Already Working:
- Tries dept_id from ipd_master
- Queries hc_department table first
- Falls back to ipd_department table
- Falls back to direct department field
- Returns empty string if all methods fail
Token Flow:
1. buildDischargeSummaryTokenVars() calls both functions
2. Returns array with 'DEPARTMENT' and 'DOCTOR_NAMES' keys
3. buildDischargeTemplateTokenVars() merges these tokens
4. Creates lowercase, UPPERCASE, and Ucfirst versions of all tokens
5. applyDischargeTemplateTokens() replaces {{TOKEN}} and {TOKEN} in template
Available Token Variants:
- {{DEPARTMENT}}, {{department}}, {{Department}}
- {{DOCTOR_NAMES}}, {{doctor_names}}, {{Doctor_names}}
- {{DOCTOR_NAME}} (singular alias for same value)
Example Output:
- Single doctor: 'Dr. John Smith [Cardiology]'
- Multiple doctors: 'Dr. John Smith [Cardiology], Dr. Jane Doe [Neurology]'
- No specialization: 'Dr. John Smith'
- With title: 'Prof. John Smith [Cardiology]'
Testing:
- Ensure doc_list field in ipd_master has comma-separated doctor IDs
- Falls back to doctor_id, doc_id, or consultant_id if doc_list empty
- Returns empty string if no valid doctor IDs found
Medicine Save Behavior Change:
- Previously: Medicines added to table locally, saved on 'Save Discharge Advice' click
- Now: AJAX auto-save immediately when '+ADD / Update' button is clicked
- Matches OPD Consult Form workflow (instant save on add)
Implementation:
- Modified btn_discharge_med_add click handler to make AJAX POST
- POST parameters: action='add_drug', ajax_mode='json', new_drug_name, new_drug_type,
new_drug_dose, new_drug_when, new_drug_freq, new_drug_day, new_drug_qty, new_drug_remark
- Controller add_drug action already existed - now returns JSON when ajax_mode='json'
- On success: adds row to table, clears form, shows 'Medicine saved successfully'
- On failure: shows error message from server
- CSRF token updated after each AJAX call
Remove Button Behavior:
- Remove button now shows confirmation dialog
- Warns user: 'This will only remove from current view. Reload to see saved medicines.'
- Note: Remove action is NOT yet AJAX - full page reload shows actual DB state
- Future enhancement: make Remove button also AJAX (action='remove_drug')
Technical Notes:
- Kept discharge_medicine_json serialization as backup (unused now, but harmless)
- Medicine data: med_id=0 marks client-added medicines
- Success check: data.update === 1
- Error handling: network timeout 30s, shows xhr.status on fail
- Form fields cleared after successful save
- Focus returns to medicine name field for next entry
User Benefits:
- Instant feedback - no need to scroll down and click 'Save Discharge Advice'
- Medicines immediately persisted to database
- Can add multiple medicines in quick succession
- Consistent with OPD prescription workflow
- Reduced risk of data loss (saves immediately vs. waiting for form submit)
Medicine Saving Fix (MAJOR):
- Add hidden field 'discharge_medicine_json' to capture medicine table data
- Add serializeDischargeMedicineTable() function to convert DOM table to JSON before form submit
- Hook serialization into form submit event (before syncEditorValues)
- Update controller to process discharge_medicine_json in save_main action
- Insert medicines into ipd_discharge_prescrption_prescribed or ipd_discharge_prescription_prescribed table
- Medicine data now persists on page reload
CKEditor Integration:
- Initialize CKEditor for 'instruction_other' (Other Advice) field
- Initialize CKEditor for 'instruction_remark' (Discharge Summary) field
- Add simplified toolbar: Bold, Italic, Underline, Lists, Undo/Redo
- Properly destroy existing instances before re-initialization
- Height: 120px for Other Advice, 150px for Discharge Summary
Review After Quick Select Buttons:
- Add getNextVisitOptions() method to Ipd_discharge controller (copied from Opd_prescription)
- Add resolveFirstField() helper method
- Pass next_visit_options array to view
- Render quick select buttons: '3 Days (09-06-2026)', '1 Week (13-06-2026)', etc.
- Add JavaScript handler: click button → populate #discharge_review_after input
- Change Review After field from col-md-4 to col-md-6 for better layout
- Add col-md-12 wrapper for button row with flex-wrap gap-2 layout
Technical Details:
- Medicine JSON structure: {med_type, med_name, dosage, dosage_when, dosage_freq, no_of_days, qty, remark}
- JSON processing includes column existence checks for flexibility
- Quick select buttons use data-value attribute for value insertion
- Falls back to hardcoded date options if opd_nextvisit table missing
- Medicine insert uses med_id=0 to mark client-added medicines
User Workflow:
1. Add medicines → shown in table (Type, Medicine, Dose labels, Duration formatted)
2. Click 'Save Discharge Advice' → medicines serialize to JSON → POST to controller
3. Controller parses JSON → inserts to database → medicines persist on reload
4. Other Advice/Discharge Summary now have rich text editing (bold, lists, etc.)
5. Review After: click '1 Week (13-06-2026)' button → auto-fills input
Medicine Table Improvements:
- Swap column order: Type column now appears before Medicine Name
- Display dose labels (text) instead of numeric IDs for Dose, When, Freq
- Auto-format duration: numeric values get 'days' suffix (e.g., '5' -> '5 days')
- Single day shows 'day' (singular), multiple days show 'days' (plural)
- Preserve custom duration formats like '1 month' or '3 doses'
Section Rename:
- Change 'Discharge Instructions / Advice' to 'Discharge Summary'
- Updated in card header, navigation menu, and form label
- Better reflects comprehensive discharge information
Technical:
- Use doseMasterCache to resolve ID->label for Dose/When/Freq/Route
- Regex check for numeric-only duration values before formatting
- Maintain backward compatibility with existing data
- Change from rx_group_medicine_suggest to Opd_prescription/medicine_search endpoint
- Match OPD Consult Form data source and behavior
- Auto-fill Type, Dose, When, Frequency, Route, Days, Qty, and Remark from medicine master
- Add ensureDoseOption helper to handle current values not in dropdown list
- Store medicine rows for auto-fill on selection
- Improved user experience with comprehensive auto-complete
- Replace simple text inputs with advanced prescription form
- Add medicine autocomplete with datalist suggestions
- Add dropdown selects for Dose, When, Frequency, and Route from dose masters
- Integrate with existing Opd_prescription/rx_group_dose_masters endpoint
- Add medicine type auto-fill from medicine database
- Add quick-fill buttons for common values (BF/AF/WF, OD/BD/TDS/HS, 3/5/7 Days)
- Improve UI with two-column layout: medicine list table (left) + add form (right)
- Add real-time medicine suggestions via rx_group_medicine_suggest endpoint
- Support Remove All functionality
- Dynamic table row add/remove with proper event binding
- Better UX matching OPD Consult Form prescription interface
- Updated Patient::search() to only render view with search query
- Created Patient::search_ajax() method for DataTables AJAX endpoint
- Implemented server-side processing with pagination, sorting, and filtering
- Results ordered by Last Visit DESC by default
- Retrieves all patient data from patient_master table
- Added proper SQL escaping to prevent injection
- Updated Patient_Search_V.php to use DataTables with AJAX remote data
- Added DataTables configuration with 25 rows per page default
- Supports dynamic length menu (10, 25, 50, 100, All)
- Added route for billing/patient/search_ajax endpoint
- Improved user experience with loading spinner and customized messages
- Added 'Discharge' button in sidebar under In-Patient & Nursing Care section
- Created search_patient() and search_patient_ajax() controller methods
- Implemented AJAX-based patient search by IPD Code, UHID, Name, or Mobile
- Added ipd_discharge/search_patient.php view with dynamic results table
- Search results show patient details, bed info, doctor assignments, and discharge status
- Added 'Create Discharge' and 'Print Discharge' action buttons for each patient
- Integrated DataTables for sorting and pagination
- Fixed database column names (patient_master.gender, age, age_in_month)
- Added routes for search page and AJAX endpoint
- Add 'Discharge Summary' section to buildAutoDischargeContent with department, doctors, and dates so {{DISCHARGE_SUMMARY}} token extracts properly
- Improve getDischargeDepartmentName to support dept_id, department_id fields and check both hc_department and ipd_department tables
- Improve getDischargeDoctorNames to fallback to doctor_id, doc_id, consultant_id if doc_list is empty
- Update DISCHARGE_TEMPLATE_TOKENS.md to clarify what {{DISCHARGE_SUMMARY}} contains
- Clear cached content for IPD 1 to regenerate with new logic
Fixes issue where {{DISCHARGE_SUMMARY}}, {{DOCTOR_NAMES}}, {{DEPARTMENT}} tokens were not working (replaced with empty values)
Fixed the error: Uncaught TypeError: can't access property 'getElementsByTag', b is null
Root Cause:
- CKEditor was not properly destroyed before reinitialization
- No error handling for editor state transitions
- Missing timing delays for DOM readiness when loaded via AJAX
- Inconsistent checking for CKEditor availability
Improvements:
1. Safe destroy pattern - wrap destroy() in try-catch to handle state issues
2. Added typeof check for CKEDITOR instead of window.CKEDITOR
3. Added 100ms setTimeout delay before CKEDITOR.replace() to ensure cleanup
4. Added 150ms delay for initial editor load (especially for AJAX loads)
5. Wrapped all setData() calls in try-catch with textarea fallback
6. Added console warnings for debugging instead of silent failures
7. Check document.readyState before initialization
8. Re-initialize editor check after AJAX template load
9. Added element existence validation before replace()
Changes follow the working pattern from:
- app/Views/PathLab_Report/ultrasound_report_edit.php (safeReplace pattern)
- app/Views/diagnosis/lab_final_report_editor.php (typeof checks)
This ensures the editor works properly when:
- Page loads normally
- Page loads via AJAX (load_form_div)
- User switches between templates using Edit button
- User cancels edit and resets form
Tested scenarios:
- Initial page load
- Edit template -> load data -> CKEditor shows content
- Cancel -> CKEditor resets
- Multiple edit/cancel cycles
New placeholders added:
1. PATIENT_PHONE - Patient's phone number from patient_master.mphone1 or ipd_master.P_mobile1
Example: 9876543210
2. INSURANCE_COMPANY - Insurance company name from ipd_master.ins_short_name or ins_company_name
Example: Star Health Insurance
Default: 'Direct' if no insurance
3. DOCTOR_NAMES / DOCTOR_NAME - Comma-separated list of treating doctors with specialties
Example: Dr. Rajesh Kumar [General Medicine], Dr. Priya Sharma [Cardiology]
Reads from ipd_master.doc_list field and joins with doctor_master and med_spec tables
Changes:
- Added getDischargeDoctorNames() helper function
- Updated buildDischargeSummaryTokenVars() to include three new tokens
- Updated DISCHARGE_TEMPLATE_GUIDE.md with new token documentation
- Added example template showing usage of new tokens
- Updated CI3 to CI4 migration mapping table
These tokens match the data shown in the auto-generated patient summary table
and allow templates to display this information in custom layouts.
OLD CI3 BEHAVIOR (correct):
- Patient demographic table was IN the template file
- Content variables contained ONLY clinical sections
- Template controlled entire layout
NEW CI4 FIX (matching CI3):
- Patient table removed from buildAutoDischargeContent completely
- Added {{PATIENT_INFO_TABLE}} token for templates to use if needed
- Auto-generated content now contains ONLY clinical sections
- Blank template = blank output (no patient table forced)
- Templates can add patient table using token or individual tokens
Changes:
- Removed includeSummaryTable parameter from buildAutoDischargeContent
- Removed templateHasDemographicTokens helper (no longer needed)
- Added PATIENT_INFO_TABLE token in buildDischargeTemplateTokenVars
- Simplified show_discharge and preview_discharge_report (no template detection)
- Added comment explaining CI3 compatibility
This matches the old CI3 chamunda_discharge_template_html.php pattern where:
- Template had hardcoded patient table structure
- Variables like \, \ were injected
- Content did NOT include patient demographics by default
- Modified buildAutoDischargeContent to accept includeSummaryTable parameter
- Added templateHasDemographicTokens helper to detect if template has patient info tokens
- Updated show_discharge and preview_discharge_report to check template before auto-generation
- When template has {{PATIENT_NAME}}, {{UHID}}, etc., skip auto-summary table
- Added clear_discharge_content.php script to clear cached content for regeneration
- Solves issue where blank templates still showed patient information table
- Updated all controllers: Opd, Diagnosis, Billing\Charges, Billing\Ipd, Storestock, AbdmGateway
- freeserif provides better Unicode coverage and more readable font weight
- Ipd_discharge and DoctorDocument already used good fonts (freeserif/freesans)
- Added comprehensive font configuration documentation
IMPROVEMENT: Increased modal size for better readability
Changed modal-dialog from default to modal-lg
- Provides more space for SNOMED CT and ICD code fields
- Better readability for field labels and help text
- More comfortable data entry experience
- Consistent with other modals in the system
