Changelog — OpenSAK

All notable changes to this project are documented here. Format follows Keep a Changelog.

[1.18.0] — 2026-09-02

First stable release of the 1.18.0 cycle, and the first stable release distributed via the Microsoft Store (product 9P4NBMM84H2D) alongside the existing Windows .exe/ZIP, Linux AppImage, and macOS DMG builds. Replaces the 1.18.0-beta.1 … 1.18.0-beta.3 builds — see git history for the detailed beta-by-beta log if needed. Headline of this cycle: MSIX packaging for the Microsoft Store, plus a round of dialog high-DPI scaling fixes surfaced by Store certification.

Added

MSIX / Microsoft Store packaging (#786) — OpenSAK can now be packaged as an MSIX for Microsoft Store distribution (product 9P4NBMM84H2D). Nothing about the existing Windows .exe/ZIP build changed — this is a side channel wrapping the same PyInstaller output.
MSIX CI build job — a manual-trigger-only GitHub Actions workflow (build-msix.yml) builds and packages the .msix on workflow_dispatch.
scripts/derive_msix_version.py — converts OpenSAK's version string (including any -beta.N suffix) into the strict 4-part numeric version MSIX/the Store requires.

Fixed

Settings and 7 other dialogs could exceed screen height at high DPI scaling (#811, #815) — caught by Microsoft Store certification (10.1.2.10 Functionality) on a 2560x1600 display at 200% scaling. Added a shared clamp_dialog_height_to_screen() helper (widgets.py) and applied it, with QScrollArea wrapping where needed, to: Settings, Trip planner, Column chooser, Welcome wizard, Import, GPS export (#811), and Waypoint/Custom WP, Update Location, and Database Manager (#815). A new test_dialog_height_policy.py tracks the remaining 14 lower-risk dialogs as explicit follow-up work (#816) rather than a silent gap.
"Close" button in the Download Boundary Packs dialog showed the raw translation key btn_close instead of "Close" (#804).
Advanced settings tab controls squashed on small/high-DPI screens (#805) — the Advanced tab now scrolls like the General tab instead of squashing its controls together.
Hint/note text too low-contrast to read, especially in dark mode (#806) — replaced a hardcoded low-contrast grey (duplicated across 42 call sites) with a new theme-aware hint_text_color()/hint_style() helper.
GSAK .db3 import: a genuine 0.0m/sea-level elevation displayed as blank (#794) — the importer now reads GSAK's Resolution column to distinguish "no elevation assigned" from a real sea-level reading.
GSAK .db3 import: plain-text descriptions containing a literal < were misdetected as HTML and lost all line breaks (#803) — the importer now reads GSAK's own ShortHtm/LongHtm columns instead of guessing from a raw < check; plain-text descriptions are also now HTML-escaped before display.
Import dialog's Close button froze the UI mid-import or mid-geocode, and could crash after either finished (#802).

[1.18.0-beta.3] — 2026-08-31

Fixed

Settings and 5 other dialogs could exceed screen height at high DPI scaling (#811) — caught by Microsoft Store certification (10.1.2.10 Functionality) on a 2560x1600 display at 200% scaling: the Settings dialog's "Geocaching profile" group was rendered off-screen with no way to reach it. Root cause: dialogs had no maximum-height constraint, so Qt grew them to fit all content unscrolled — DPI scaling shrinks the effective usable screen area, so on a small enough one that can exceed the available height. Added a shared clamp_dialog_height_to_screen() helper (widgets.py) and applied it, with QScrollArea wrapping where needed, to the 5 highest-content dialogs: Settings, Trip planner, Column chooser, Welcome wizard, Import, and GPS export. A new test_dialog_height_policy.py tracks the remaining 17 dialogs as explicit follow-up work rather than a silent gap — a new dialog (or a stripped protection call) without it now fails CI.

[1.18.0-beta.2] — 2026-08-30

Fixed

"Close" button in the Download Boundary Packs dialog showed the raw translation key btn_close instead of "Close" (#804) — reported by Giles Percival on Facebook after a boundary-pack download completed. BoundaryDownloadDialog._on_done() called tr("btn_close"), a key that didn't exist in any of the 8 language files — tr() falls back to the raw key when a translation is missing. Fixed to reuse the existing "close" key (already used elsewhere, e.g. file_export_dialog.py); no language file changes needed.
Advanced settings tab controls squashed on small/high-DPI screens (#805) — reported by Sean Flook on Facebook (v1.18.0-beta.1). The General tab wraps its content in a QScrollArea to stay usable on small screens/high DPI, but the Advanced tab — which has the most stacked group boxes of any tab (Folders, Search, Location refinement, Updates, Distance) — didn't, so Qt squashed the controls together instead of letting the user scroll when the dialog couldn't grow to fit everything. _build_advanced_tab() now wraps its content in a QScrollArea the same way _build_general_tab() does.
Hint/note text too low-contrast to read, especially in dark mode (#806) — reported by Sean Flook (dark mode) and Robert Long (also hard to read in light mode for older eyes) on Facebook. A hardcoded "color: gray; font-size: 10px;" was duplicated across 42 call sites in 16 files for secondary/hint text — a fixed, theme-independent colour with insufficient contrast in both themes. Replaced with a new theme-aware hint_text_color()/hint_style() helper in src/opensak/gui/theme.py: dark green (#2e7d32, ~4.7:1 contrast) in light mode, light green (#66bb6a, ~6:1 contrast) in dark mode. Centralised in one place so a future user-configurable hint colour only needs to change this one function. (One follow-up fix included: the map popup's cache-type/D/T line is built as a raw JavaScript template embedded in Python and resolved via string-token substitution, not an f-string — the initial colour-substitution accidentally called the Python helper inside the JS source itself, throwing hint_text_color is not defined in the browser console at runtime. Fixed by resolving a HINT_TEXT_COLOR template token in Python before the HTML reaches QtWebEngine, matching the existing INIT_LAT/MAP_TIP_MAXIMIZE-style tokens.)
GSAK .db3 import: a genuine 0.0m/sea-level elevation displayed as blank (#794) — reported by ianwok, root-caused with GSAK maintainer Mike Wood (lignumaqua). GSAK's Elevation column is always numeric (0.0 both before an elevation macro has run and for a real sea-level cache), so the two cases were indistinguishable from the value alone — every 0.0 elevation was nulled on import, including real ones. GSAK actually signals "no elevation assigned" via a blank (not NULL) Resolution column instead — metres of source resolution, or the literal text "USER" for a manually entered value, left blank only when unset. The importer now reads Resolution to make that distinction; when the column is absent entirely (older/partial GSAK exports), it trusts the raw Elevation value rather than guessing.
GSAK .db3 import: plain-text descriptions containing a literal < were misdetected as HTML and lost all line breaks (#803) — reported by nagisml, e.g. GC5K1DY. The importer guessed short_desc_html/ long_desc_html by testing for any < character in the description text, instead of reading GSAK's own authoritative ShortHtm/LongHtm columns (which mirror geocaching.com's real html=true/false flag, the same source the GPX importer already reads). Now reads those columns directly; falls back to a real-tag-matching regex (not the old naive < check) only when they're entirely absent. Also fixed a related issue nagisml flagged in the same report: plain-text descriptions were written into the description view's <pre> block unescaped, so a stray <, > or & in genuinely plain text (e.g. "N < 5 caches", "R&D") could still be misinterpreted as markup — cache_detail.py now HTML-escapes plain-text descriptions before wrapping them.
Import dialog's Close button froze the UI mid-import or mid-geocode, and could crash after either finished (#802) — found by Allan while reviewing the #800/#801 fix. closeEvent() blocked the GUI thread with a synchronous worker.wait(), so clicking Close while an import or the follow-up "Looking up missing location data" step was running froze the window until the background thread stopped — indistinguishable from a hang or crash on a large import. The import step has no cancellation support, so Close is now simply disabled while it runs; the geocode step can cancel responsively (ReverseGeocodeWorker checks a flag between rows), so Close relabels to "Cancel" there and routes to the non-blocking request_cancel() instead of closing. Also added parent=self to the ReverseGeocodeWorker construction, matching #801's fix in update_location_dialog.py. A related crash surfaced during manual testing: self._worker/self._geo_worker were only ever cleared in closeEvent(), never on normal completion, so a finished worker's already-deleteLater()'d C++ object could still be referenced by a later Close click ("RuntimeError: libshiboken: Internal C++ object already deleted"). Fixed by clearing both references on QThread.finished (mirroring update_location_dialog.py's existing pattern), with a defensive try/except RuntimeError as a second safety net.

[1.18.0-beta.1] — 2026-08-27

Added

MSIX / Microsoft Store packaging prototype (#786) — OpenSAK can now be packaged as an MSIX for a private, unlisted Microsoft Store listing (product 9P4NBMM84H2D). A submission passed Store certification and was confirmed installable end-to-end via the Store app itself, clearing the risk that the bundled QtWebEngine/Chromium runtime would be rejected. Nothing about the existing Windows .exe/ZIP build changed — this is a side channel wrapping the same PyInstaller output.
MSIX CI build job — a new manual-trigger-only GitHub Actions workflow (build-msix.yml) builds and packages the .msix on workflow_dispatch, so producing a new build for Partner Center no longer requires the Windows machine for that step. Not tag-triggered and not attached to public GitHub Releases yet, since the Store listing is still private/unlisted.
scripts/derive_msix_version.py — converts OpenSAK's version string (including the -beta.N suffix) into the strict 4-part numeric version MSIX requires.

[1.17.2] — 2026-08-26

Fixed

Detail panel header showed large empty grey padding when the bottom panel was resized taller (#779) — regression from the #755 fix shipped in v1.17.1. Reported by Mike via email with side-by-side screenshots. #755 gave the cache-detail QTabWidget a vertical QSizePolicy of Ignored so Qt would disregard the tab pages' minimum-size contribution when computing the panel's minimum height. Ignored, unlike the tab widget's default Expanding, doesn't carry QSizePolicy.ExpandFlag — so the tabs stopped being the preferred recipient of any leftover vertical space in the panel's QVBoxLayout, and Qt spread that space across every row instead, including the compact header/meta rows. Fixed by replacing the policy change with a small QTabWidget subclass in src/opensak/gui/cache_detail.py that overrides minimumSizeHint() to return QSize(0, 0) — keeping the #755 fix intact while the tabs keep their normal Expanding policy and vertical stretch priority, as in v1.17.0.
GSAK Mark As Found status not detected on GPX/GGZ import when GSAK encodes it via <type>...|Found</type> instead of <sym> or a personal log entry (#766)
Found status from genuine GSAK GPX/GGZ exports not detected on import when GSAK's <sym> doesn't reflect found state (#766)

Known Issues

Boundary updates don't always recognize when they're already current (#781) — the reverse-geocoding boundary refresh can reprocess polygon data that's already up to date. Under investigation.
A small number of caches reverse-geocode to the wrong region (#782) — e.g. GCAJWVJ is assigned country/county values that don't match its position on the map. Not yet clear whether the cause is in the polygon data, the boundary-matching logic, or the map visualization. Under investigation.

[1.17.1] — 2026-08-19

Fixed

GPX export/import did not preserve "found" status (#766) — reported by Allyn56 via a detailed multi-scenario comparison in #649. Two related bugs in the GPX bridge, both now fixed:
1. Export: _cache_symbol() in src/opensak/gps/garmin.py always wrote <sym>Geocache</sym>, regardless of cache.found. The Groundspeak/GC.com Pocket Query convention — which GSAK and Garmin devices actually read found status from in a GPX — is <sym>Geocache Found</sym>. OpenSAK never wrote this, so found status was invisible to any tool reading an OpenSAK-exported GPX or GGZ, including OpenSAK itself on re-import (the log entries were present and correct, but nothing checks the log list for found status — only <sym>). Found caches now export with <sym>Geocache Found</sym> in both the file-export and send-to-device paths (GGZ reuses the same GPX generator). Lab Caches are unaffected — they have no found-symbol in the Garmin convention, so their distinct icon (Flag, Blue) is unchanged either way.
2. Import: found_by_me in src/opensak/importer/__init__.py was derived exclusively from sym == "Geocache Found", with no fallback — unlike the GSAK database importer, which reads GSAK's own Found column directly and was unaffected. Import now also checks the log list for a found-type log (FOUND_LOG_TYPES) matching the configured gc_finder_id/gc_username, same matching rules already used for found_date/FTF derivation, but only when <sym> is entirely absent from the source wpt — if <sym> is present at all (even a plain "Geocache", not just "Geocache Found"), it's trusted as authoritative. This matters because "Mark as Not Found" deliberately leaves old found-type Log rows in place rather than risk deleting real imported log history — without this guard, re-exporting such a cache (which correctly writes sym="Geocache" after the fix above) and re-importing it would have incorrectly resurrected found status from the stale log.
Vertical splitter couldn't be resized past roughly the middle of the window (#755) — the cache detail panel's QTabWidget (Description, Hint, Logs, Waypoints, Attributes, Trackables, Notes) reported its own minimum size as the largest of all its tab pages' minimums, since any tab could be switched to at any time. That propagated up into the whole bottom panel's minimum height, blocking the main splitter from being dragged much further down — the panel could shrink, but never much past its content's natural minimum. A QSizePolicy.Ignored on the tab widget's vertical component stops that minimum-size propagation without affecting normal sizing or tab switching — each tab's own content already scrolls internally when squeezed, so nothing becomes unreachable. Verified the splitter can now shrink the bottom panel well under its old floor, even with a heavily populated cache (many attributes/logs/waypoints, long description).
Coordinate Converter gave wrong results (#751) — two independent bugs in src/opensak/coords.py, both reported/diagnosed by pjacklam:
1. parse_coords() had no dedicated branch for a plain decimal-degree value written with a hemisphere letter instead of a +/- sign (e.g. "N 59.99999 E 12.99999", no separate minutes component). It fell through to the DMM° regex branch instead, where backtracking let the degrees group give back digits to the minutes group whenever keeping them would leave an unparseable "." right after — silently reinterpreting "59.99999" as degrees=5, minutes=9.99999 (5.16667° instead of 59.99999°), with no error shown. A dedicated branch for this input shape, checked before the DMM° branch, fixes it.
2. _dd_to_dmm()/_dd_to_dms() (and the cache list's format_lat()/ format_lon(), which had the same bug independently) rounded minutes/ seconds directly in an f-string without checking whether the rounded value hit 60 — e.g. 59.999999° displayed as "59° 60.000'" or "59° 59' 60.00"" instead of carrying into the next unit ("60° 00.000'"/"60° 00' 00.00""). Both formatters now share carry-aware helpers that roll 60 minutes into a degree and 60 seconds into a minute (which can itself cascade into a degree). Verified against both of the reporter's exact examples end-to-end.

Added

"Mark as Found" now opens a dialog and creates a real log entry (#649) — right-clicking a cache and choosing "Mark as Found" used to silently set cache.found = True with no found date and no Log row. Since GPX export only ever serializes existing cache.logs, a manually-found cache exported with no groundspeak:finder/"Found it" entry at all — GSAK (or any re-import) then didn't recognize it as found either, even though OpenSAK's own list showed it as found. This is a real workflow gap for Adventure Lab Caches specifically, where geocaching.com auto-logs the parent AD Lab as found with no real per-stage log to import, so manual marking is the only way in. "Mark as Found" now opens a dialog to pick the found date (default: today), and creates a matching Log row (type "Found it", or "Attended" for event-type caches, finder = your configured Geocaching.com username, date = the chosen date) — verified end-to-end that a manually-found cache now round-trips correctly through GPX export. Requires a Geocaching.com username configured under Settings first (used as the log's finder name); you'll be prompted if it isn't set yet. A new "Edit found date…" entry lets you change the date afterwards without unmarking first — it updates your existing found-type log in place rather than adding a duplicate. "Mark as Not Found" now also clears the found date (mirroring GSAK's own found/found-date linkage), so a cache marked not found no longer shows a stale date; it still doesn't touch existing log history.

Fixed

Child waypoints missing from GPX/GGZ export and Send-to-GPS (#753) — generate_gpx() only ever emitted the one <wpt> for a cache's own listing; cache.waypoints (parking areas, trailheads, stages, final locations, etc.) were silently dropped from every export path that uses it — File Export (GPX and GGZ) and Send-to-GPS alike. Each child waypoint now gets its own sibling <wpt> element, as a plain GPX waypoint (no groundspeak:cache block — same convention as custom-waypoint-type caches, #660) with its coordinates, comment, description, sym/type from its waypoint type, and a reconstructed GC-style <name> (prefix + this cache's own code suffix — the exact original geocaching.com-assigned code isn't retained on import, but a short unique code is all a re-import or device needs). Verified end-to-end against pjacklam's original GC1YB0C.gpx / opensak_GC1YB0C.gpx pair — the reconstructed child waypoint now matches the original byte-for-byte on every field except <url> (not retained by GPX import, so intentionally omitted rather than exported empty). Thanks to pjacklam for the report and the side-by-side GPX files, which made the diff straightforward.
Split-screen map: caches with corrected coordinates rendered outside the drawn circle, and stayed unfiltered when a cache was selected (#748, #743) — the split-screen "nearby" map computed radius membership, sort order, and its own circle centre from each cache's raw latitude/longitude, but plots markers at corrected coordinates when set. A cache whose corrected coordinates diverged from its raw ones could therefore pass the radius check yet render outside the visible circle (or the reverse — genuinely nearby via its correction, but excluded). Separately, get_nearby_caches() only ever filtered by distance, ignoring whatever filter was currently active on the main list — selecting a cache while filtered silently reverted the split-screen map to showing every nearby cache. Both are fixed: radius membership/sorting/circle-centre now use each cache's effective (corrected-aware) coordinates, and the currently active filter (advanced
- quick/search-box) is passed through and applied alongside the distance check. Thanks to GeePa67 for both reports.
Info Bar counts wrong when text-searching User Notes (#752) — apply_filters_lightweight()'s fallback check for when it must defer to the full ORM query path never accounted for TextSearchFilter.search_notes (only description/logs/hint were checked). A text search scoped to only User Notes therefore took the fast lightweight path, where the notes exists() subquery — written for, and only ever exercised against, the full ORM query — raised a SQLAlchemy auto-correlation error instead of returning a result. Notes-only text search filters now correctly fall back to the full ORM path, matching how description/logs/hint scoping already behaved. Thanks to pjacklam for the report.
Community Celebration Event caches imported with wrong type (#756) — geocaching.com's GPX export uses the raw type string "Lost and Found Event Caches" for CCE caches, not "Event Cache" as previously assumed (#591). Unrecognized, this string fell through to the generic gray "unknown type" icon on GPX import, and to the Mystery icon on GSAK import (GSAK code F, previously left deliberately unmapped). Both import paths now correctly classify these as Community Celebration Event. Verified end-to-end against a real-world 88-cache CCE export. Thanks to Veé X Péé for the detailed report and test file.

[1.17.0] — 2026-08-17

First stable release of the 1.17.0 cycle. Replaces the run of 1.17.0-beta.1 … 1.17.0-beta.14 builds — see git history for the detailed beta-by-beta log if needed. Headline of this cycle: full-screen and pop-out map support, offline reverse-geocoding enabled by default, and a fix for the long-standing GUI freeze on large-database switch/load.

Summarises net changes since v1.16.3 (the beta.1/4/5 releases that folded the v1.16.1–v1.16.3 hotfixes back into beta are not repeated here, since those fixes already shipped as stable patch releases).

Added

Full-screen / pop-out map (#696, #598) — maximize the map within the main window (F11), or pop it out to its own floating window that can be moved to a second monitor (Ctrl+Shift+M). Thanks @blazerat!
14 new GSAK-compatible columns (#658) — Cache Id, Changed date, Creation date, Elevation, Found count, GC.com note, Guid, Hints, Notes, Owner ID, Owner name, Source, Url, Watch.
Last found / Last GPX update / Last four logs columns (#716) — closing out long-standing GSAK-parity requests #518, #542, #534. Thanks Allyn56 and ianwok.
Wait cursor during cache-list refresh (#647) — including at startup with a large database, so a multi-second refresh is no longer indistinguishable from the app doing nothing.
Previous session's log preserved across restarts (#737) — a new Help menu action, "Open log from last restart", makes it possible to inspect a crash or hang from the previous session.

Changed

Offline reverse-geocoding (Country/State/County) enabled by default (#60) — no longer requires manual opt-in.

Fixed

GUI freeze / force-close on large-database switch, filter, or load (#740) — the cache list and map now refresh on a background thread instead of blocking the GUI thread for up to 85+ seconds on very large databases.
App could hang or silently crash on startup for large or existing databases (#723) — root cause was a missing index on logs.cache_id causing full table scans during startup migrations, plus a separate migration bug that could crash startup entirely on databases that had already run a later schema migration.
PQ import appeared to hang on Windows when county boundary packs weren't cached locally (#722) — missing packs are now pre-fetched in parallel with progress feedback, instead of retried on demand for every affected cache.
Migration and database-switch failures could fail silently (#738) — these now log the full error and show a clear dialog instead of leaving the app in a half-switched or broken state.
Reverse-geocode bulk write could crash on large databases (#710) — fixed a "too many SQL variables" error by chunking the write.
Archived caches hidden by default in the Set Filter dialog, and the setting wasn't remembered (#576) — now defaults to shown, matching GSAK, and the choice persists correctly.
Map didn't reliably pan to the selected cache, and the split-screen map didn't update for caches outside the overview map's display range (#718).
Toolbar title was hardcoded in Danish regardless of UI language (#683) — thanks @urs-beeli.

[1.17.0-beta.14] — 2026-08-16

Fixed

GUI thread blocked 20-90s on large-database switch/filter/load, appearing as a freeze or force-close (#740) — reported by Ron Radix (Facebook community) on a 232,149-cache database: switching to another database could make OpenSAK close itself, with the previous database reopening afterward. A submitted opensak.log ruled out both an uncaught exception and a migration (schema was already current throughout) — the log simply stopped, mid-flow, right after a successful database switch, with no traceback. Root cause: apply_filters_auto() → cache_table.load_caches() → map_widget.load_caches() ran synchronously on the GUI thread and consistently took 16-28s on 39k-232k cache databases, reaching 85+ seconds of continuous blocking when combined with a distance recalculation — long enough that Windows' "Not Responding" detection (or the user via Task Manager) would terminate the process. Confirmed via measured timings across three real databases (98k/198k/348k caches). The DB query now runs on a new background RefreshWorker (QThread); the GUI thread applies the result once ready. A generation counter discards results from a superseded refresh (e.g. two quick database switches) without needing to cancel the older worker. Verified against the same three databases: switching between them no longer freezes or force-closes, at the cost of a measured increase in total completion time on first (cold) access — tracked separately as a performance follow-up in #746, since the query's cost is largely GIL-bound ORM row hydration rather than I/O wait, so backgrounding it doesn't make it faster on its own.

Added

Wait cursor during cache-list refresh (#647) — a multi-second refresh on a large database (see #740 above) previously gave no visual feedback at all, indistinguishable from the app doing nothing. A wait cursor is now shown for the duration of any refresh, including at startup with a large database.

[1.17.0-beta.13] — 2026-08-15

Added

Previous session's log is preserved across restarts (#737) — setup_logging() now renames opensak.log to opensak.log.previous at startup instead of deleting it, so a crash or hang from the last session can still be inspected after restarting. A new Help menu action, "Open log from last restart", opens it directly. Falls back to the old delete-and-recreate behaviour if rotation fails (e.g. a locked file) rather than blocking startup.

Fixed

Migration and database-switch failures could still fail silently (#738) — switch_to() now logs the full exception once at the source and re-raises, instead of the error being swallowed somewhere along the way. All three GUI call sites (toolbar database dropdown, Switch button, new-database auto-switch) catch it, show a clear error dialog, and avoid leaving the UI in a half-switched state. Startup (ensure_active_initialised, the original #723 code path) now shows a clear error and exits cleanly instead of continuing into a MainWindow with no valid database engine. This doesn't fix any specific unknown migration bug, but turns any future occurrence — known or unknown — into a visible error with a full traceback in the log, which is now preserved across restart thanks to #737.

[1.17.0-beta.12] — 2026-08-14

Fixed

App could still silently crash (appearing to hang) on startup for some existing databases (#723 follow-up) — migration 2 checked for an outdated index name (uq_waypoint_cache_prefix_name) that migration 22 had long since replaced with uq_waypoint_cache_wp_code when it relaxed the constraint from (cache_id, prefix, name) to (cache_id, wp_code) (issue #536). Any database that had already run migration 22 didn't have the old index name, so migration 2 wrongly believed it had never run and tried to recreate its own old, stricter constraint — which real-world data can legitimately violate (that's exactly why migration 22 replaced it in the first place). The resulting error was never caught anywhere in the startup path, so the app crashed with no visible error message — indistinguishable from a hang. Root-caused with a local repro against a real 405MB/12,600-cache database; fixed by having migration 2 recognise either index name as evidence it can skip.

[1.17.0-beta.11] — 2026-08-13

Fixed

Split-screen map didn't update for caches outside the overview map's display limit (#718) — reported by Mike Wood (lignumaqua): selecting a cache in the table sometimes left the map unchanged, or centred on the wrong area, depending on the table's current sort order. Root cause: the split-screen map reused the overview map's own capped/sorted marker set (the "Map Display" max-caches limit), so a selected cache outside that set simply had no marker to pan to — panToCache() silently did nothing. Selecting a cache now loads that cache's own neighbourhood (within a configurable radius, independent of the overview limit) and draws a circle on the map at that radius so it's always clear where the view ends; a small label appears only when a new safety cap actually limited the result in a dense area. Two new per-database settings (Settings → Map: split-screen map radius and cache limit) control this, defaulting to 2 km / 500 caches.

[1.17.0-beta.10] — 2026-08-12

Added

Startup and migration timing diagnostics (#723 follow-up) — a report surfaced that the app could still appear to hang on startup for some large existing databases even after the beta.9 fix. To pin down exactly where, the full startup path now logs timestamped checkpoints to opensak.log: each phase in app.py (language load, database check, database load, main window build), every one of the 24 schema migrations individually, and — for the migrations most likely to be slow on a large database — each underlying query separately rather than the migration as a whole (the three backfills in migration 24, each of the index creations in migrations 6 and 12, and the waypoints table rebuild in migration 2). The one-off distance/bearing recalculation that can run on first launch after an upgrade is also broken down by phase (fetch / compute / write). No behaviour changes — this release is diagnostics only, so the next report comes with an exact trace instead of a stopwatch guess.

[1.17.0-beta.9] — 2026-08-12

Fixed

PQ import appeared to hang on Windows when county boundary packs aren't cached locally (#722) — a missing county pack needed by the offline reverse-geocoding step was fetched on demand inside the resolve loop, but a failed fetch was never remembered: the same missing pack needed by a later cache in the same import batch retried the full network fetch (up to 60s) every single time. A PQ spanning many counties combined with slow/blocked outbound requests could stall for a very long time with no visible progress. Fixed with three changes: (1) failed fetches are now cached negatively for the rest of the run, so a missing pack is only attempted once; (2) a new pre-fetch phase collects every distinct county pack a batch will need and downloads them all in parallel with real progress before resolving starts, so the on-demand fetch is now a fallback rather than the common path; (3) a short reachability probe runs before the pre-fetch batch, so a fully blocked network fails fast instead of costing up to 60s per pack.
App appeared to hang on startup for large existing databases (#723) — the startup migrations backfill several cached columns on caches (log_count, last_log_date, last_found_date, last_gpx_update, last_four_logs) using correlated subqueries against the logs table. On older databases without an index on logs.cache_id, each of these ran as a full table scan per row in caches, confirmed by benchmark on the issue. Added a migration that creates ix_logs_cache_id and ix_logs_log_date, placed before the three migrations that need it, so the index is guaranteed to exist before any of the heavy queries run.

[1.17.0-beta.8] — 2026-08-12

Fixed

Archived caches hidden by default in the Set Filter dialog (#576) — the Availability tab's Archived checkbox defaulted to unchecked, silently hiding archived caches even with no other filter criteria set at all (GSAK, by contrast, always shows archived caches unless you explicitly filter them out). Also fixed a related persistence bug: explicitly checking Archived was silently forgotten the next time the filter dialog was reopened, because the "all three availability checkboxes checked" state was treated internally as "no filter needed" and nothing was saved to restore it from. Archived now defaults to checked, matching GSAK, and hiding archived caches is a deliberate, persisted choice like any other filter — it now also counts correctly toward the "N active" badge.
Map didn't pan to the selected cache ~60% of the time (#718) — quickly selecting a different cache in the list updated the detail panel text but left the map pin unmoved. Leaflet.markercluster's zoomToShowLayer() reveal animation can silently drop its completion callback if a new cache is selected before the previous call's callback has fired — worse on macOS, where WebEngine's animation timing makes back-to-back selections more likely to overlap. Fixed with a sequence guard (a stale callback can no longer move the map to the wrong cache) and a timeout fallback for when the callback never fires at all.

Thanks to Mike for both reports.

[1.17.0-beta.7] — 2026-08-11

Added

14 new GSAK-compatible columns (#658) — Cache Id, Changed date, Creation date, Elevation, Found count, GC.com note, Guid, Hints, Notes, Owner ID, Owner name, Source, Url, Watch. All backed by existing data — available immediately via the Column Chooser.
Last found / Last GPX update / Last four logs columns (#716) — follow-up to #658 for the 3 remaining GSAK fields that needed new derived data rather than just exposing existing columns:
- Last found — most recent "Found it"-type log by any finder (unlike the existing "Date found by me" column, which is specifically your own found date).
- Last GPX update — local timestamp of the most recent import that touched a given cache.
- Last four logs — shown as 4 GSAK-style colored squares (green = found, red = DNF, yellow = other, blank = no log), matching GSAK's own compact display.
Existing databases backfill automatically on first launch after updating.

Thanks to Allyn56 and ianwok for the underlying GSAK-parity requests this closes out (#518, #542, #534).

[1.17.0-beta.6] — 2026-08-11

Changed

Offline reverse-geocoding (Country/State/County) enabled by default (#60) — the feature is now on out of the box instead of requiring manual opt-in.

Fixed

Reverse-geocode bulk write could crash on large databases (#710) — the bulk write hit SQLite's parameter limit on very large databases, producing a "too many SQL variables" error. The write is now chunked, so it works regardless of database size.

[1.17.0-beta.5] — 2026-08-10

Folds the v1.16.3 stable hotfix (#695 re-fix) back into beta. No other beta-only changes in this release.

[1.17.0-beta.4] — 2026-08-10

Folds the v1.16.2 stable hotfix (7 bugs) back into beta. No other beta-only changes in this release — see v1.16.2 below for details.

[1.16.3] — 2026-08-10

Corrects a mistake in v1.16.2: the #695 fix (GPX/GGZ export IDs) was investigated and tested during that cycle but never actually committed, despite the v1.16.2 release notes below stating it was fixed. Apologies for the confusion — it's properly fixed now, verified end-to-end against the original bug report's GPX file.

Fixed

GPX/GGZ export: wrong/missing cache and finder IDs (#695) — the exported groundspeak:cache id was the internal database row ID instead of the real Geocaching.com numeric cache ID (the GPX importer parsed this value but never stored it). The groundspeak:owner element and each log's groundspeak:finder id attribute were also missing entirely. All three now export correctly. Caches imported before this fix will need a re-import (GPX or Pocket Query) to pick up the correct ID — until then, export falls back to a visible id="0" placeholder.

Thanks to urs-beeli, pjacklam, and Allyn56 for the reports and for re-testing after the first fix didn't actually land.

[1.16.2] — 2026-08-10

Bugfix release — no new features. Seven issues reported by urs-beeli and pjacklam, all fixed and verified.

Fixed

GPX/GGZ export: wrong/missing cache and finder IDs (#695) — the exported groundspeak:cache id was the internal database row ID instead of the real Geocaching.com numeric cache ID (the GPX importer parsed this value but never stored it). The groundspeak:owner element and each log's groundspeak:finder id attribute were also missing entirely, despite the underlying data already being present in the database. All three now export correctly. Caches imported before this fix will need a re-import (GPX or Pocket Query) to pick up the correct ID — until then, export falls back to a visible id="0" placeholder rather than the old misleading value.
Cache selection could take several seconds on large databases (#685) — caches with many logs/attributes/waypoints/trackables (e.g. a well-logged cache with several trackables) triggered a SQLAlchemy query that joined multiple collections at once, multiplying out to tens of thousands of rows internally before deduplication. Fixed in four places: cache selection, corrected-coordinates reload, and bulk database moves. Up to ~500x faster in testing on worst-case data.
Ctrl+F did nothing (#678) — the View-menu and toolbar filter actions both held the same keyboard shortcut, which Qt silently refuses to resolve when two actions on the same window collide. Also fixed a related sync issue in the customizable-shortcuts feature that could re-introduce the conflict after saving a custom shortcut.
Quick Filter dropdown not visible on Windows (#681) — a combo box added directly to the menu bar rendered unreliably on Windows' native menu styling. Moved to the toolbar, next to the saved-filter dropdown, which doesn't have this problem.
Newly saved filter profile didn't appear in the toolbar dropdown until a filter was applied (#682).
Trip Planner could sometimes not be reopened after closing it (#676) — an intermittent timing issue in Qt's deferred window cleanup.
GPS export status log showed Danish text regardless of UI language (#675).

Thanks to urs-beeli and pjacklam for the detailed bug reports that made this release possible.

[v1.17.0-beta.3] - 2026-08-09

Fixed

Fixed layout issue where docking the map back after using fullscreen/popout would leave the map view split incorrectly (#696)

Added

Full screen and popout map support — pop the map out to its own window or monitor, or maximize it within the app (#696, thanks @blazerat!)

[1.17.0-beta.2] — 2026-08-02

Fixed

Toolbar title was hardcoded in Danish ("Værktøjslinje") regardless of the selected UI language, showing incorrectly in the toolbar's right-click context menu for all non-Danish users (#683, thanks @urs-beeli)

[1.17.0-beta.1] — 2026-07-31

Brings the beta line back in sync with main. Beta had drifted 12 commits behind after the v1.15.1, v1.16.0, and v1.16.1 stable releases went out without being merged back — this release folds all of that in, including the #577 splitter fix below. No other beta-only changes.

Fixed

Status bar / bottom panel could disappear with no way back (#577) — ported from the v1.16.1 stable patch. Dragging the main vertical splitter all the way down let the bottom panel (info bar + detail/map) collapse fully to 0px, with the collapsed position then persisted and restored on every subsequent launch. Both the main and the detail/map splitter now refuse to collapse their panels completely (setChildrenCollapsible(False)), and restoring a previously-saved splitter ratio that would still leave either side below a small minimum size now falls back to the default layout instead of reproducing the stuck state.

[1.16.1] — 2026-07-30

Small patch release for a splitter/layout regression reported on Facebook right after 1.16.0.

Fixed

Status bar could disappear with no way to bring it back (#577) — dragging the main vertical splitter all the way down let the bottom panel (info bar + detail/map) collapse fully to 0px, with the collapsed position then persisted and restored on every subsequent launch. Both the main and the detail/map splitter now refuse to collapse their panels completely (setChildrenCollapsible(False)), and restoring a previously-saved splitter ratio that would still leave either side below a small minimum size now falls back to the default layout instead of reproducing the stuck state.

[1.16.0] — 2026-07-29

First stable release of the 1.16.0 cycle. Replaces the run of 1.16.0-beta.2 … 1.16.0-beta.16 builds — see git history for the detailed beta-by-beta log if needed. Headline of this cycle: a deep investigation into why exported caches weren't showing up correctly on Garmin handhelds ("Send to GPS"), plus a large-database performance pass and several data-integrity fixes for GSAK/GPX imports.

Added

Garmin "Send to GPS" — full content and correct recognition (#656, #453, #454, #455, #502) — a multi-stage investigation, working directly from GGZ/GPX byte comparisons against GSAK's own export of the same caches, found and fixed the actual root cause: OpenSAK exported GPX 1.1 with groundspeak:cache wrapped in an <extensions> element, while Garmin's on-device geocache parser is built around GSAK's GPX 1.0 format, where groundspeak:cache is a direct child of <wpt>. Export format now matches GSAK's exactly. Along the way, several content gaps were found and fixed: missing groundspeak:attributes, short_description/long_description, and state; logs hardcoded to the last 5 with text truncated at 500 characters (now full history, no cap); and element order inside groundspeak:cache corrected to match the official cache.xsd sequence. Confirmed fixed by multiple testers across a GPSMAP 64s, 66s, 66sr, and Montana 700 — description, hint, and previous logs all display correctly now, and exported files are correctly recognized as geocache files (not plain waypoints) on-device.
Custom Waypoints (Hotel/POI, Parking Area, Trailhead, etc.) get proper icons, in both the app and on Garmin devices (#593, #660) — in-app, these now show their own distinct icon in the table, map, and detail panel instead of a generic "unknown" icon. For Garmin export, they're now written as plain GPX waypoints with a matching native Garmin icon (e.g. "Parking Area", "Lodging", "Trail Head") instead of showing up as an empty "fake geocache" with blank D/T stars.
"Use database name as filename" for GPS export — new checkbox in the Send to GPS dialog (community suggestion — GSAK has an equivalent), on by default, pre-fills the export filename from the active database's name for both file-mode and device-mode exports. Its on/off state is remembered between exports.
Send to GPS collision handling — device-mode exports ("Send to GPS") now prompt before silently overwriting a same-named file on the device, matching the protection file-mode export already had. The "delete old files before upload" option now also covers GGZ exports (previously GPX-only).
Large-database performance (#627 and its follow-ups) — a new lightweight query path (apply_filters_lightweight()/ apply_filters_auto()) avoids ORM row-hydration cost with an automatic, always-correct fallback to the full path whenever a filter needs it; SQL pushdown extended to every remaining filter type; map loading is dramatically faster via icon caching and bulk marker loading; a "Max caches shown on map" setting and a "disable map entirely" setting both target the same large-database load cost. Measured on a 250,000-cache benchmark database: total time to show all caches dropped by ~19% end-to-end, with individual steps (map load, filtered queries) 2–48x faster depending on the scenario. A new scripts/benchmark_large_db.py harness backs all of these numbers and is available for future performance work too.
Default Column View (#607) — named, saveable column configurations (visible columns, widths, container/type display), with a toolbar quick-switch dropdown and a designated global default for new/ unconfigured databases, replacing the previous implicit "last used" fallback.
Vertical gridlines in the cache table (#463), and a center point picker for the distance filter (#511) — choose Home, a saved home point, the selected cache, or a manual coordinate as the filter's center, plus a matching right-click "Set as center point" action.

Fixed

Cache distance not calculated after adding/editing a waypoint (#662) — a newly added cache (including Custom Waypoints) or one with edited coordinates showed no distance and sorted to the bottom of the list until the center/home point was switched away and back. Distance and bearing are now recalculated immediately in both cases.
Wrong attribute mappings from GSAK import (#615) — 42 of 70 Groundspeak attribute IDs were mapped to the wrong attribute, also silently affecting the attribute filter regardless of import source.
Corrected-coordinate caches lost their original coordinates on GSAK database import (#614) — the original (pre-solve) position is now read from GSAK's Corrected table instead of the already-corrected Latitude/Longitude columns.
Hidden date (and log dates) lost when importing Project-GC-style GPX (#617) — an explicit UTC-offset timestamp format (+00:00) wasn't recognized by the old parser and silently came back as None.
Logs wiped on every re-import instead of accumulating (#618) — a partial GPX/PQ re-import now merges logs (update-in-place for a matching ID, add new ones, keep existing ones not present in the current file) instead of deleting and rebuilding from that file alone.
Community Celebration Event caches imported as generic "Event Cache" (#591) — narrow name-based fallback for this un-typed, time-limited Groundspeak program.
New database didn't inherit column settings (#606) — falls back to the last-used configuration instead of hard-coded factory defaults.
Redundant distance recalculation on every startup (#579) — skipped when nothing about the database or home point has changed since the last run.
Boolean filters bypassed their SQL indexes (#628) — IS true/ IS false changed to the index-usable = true/= false form.
Hint markup incorrectly ROT13-scrambled (#595) — bracketed markup like [br] is now left untouched by the hint cipher.
Several dark-theme/UI fixes: unreadable placeholder text (#624), an unreadable Where-filter SQL error box (#613), and Country/Region/ County columns left-aligned instead of centered (#603).
Database dropdown/lists not sorted alphabetically (#531, #601), and a misleading "database created" confirmation message (#464).

[1.16.0-beta.16] — 2026-07-28

Beta release — further "Send to GPS" polish following tester feedback on beta.15, plus proper Garmin icons for Custom Waypoints and Adventure Lab caches.

Fixed

"Send to GPS" silently overwrote a same-named file on the device with no warning (#656) — reported by CheminerWill after testing beta.15. The existing #501 fix only covered file-mode exports; device-mode ("Send to GPS") now runs the same collision check: if a file with the same name already exists in the device's Garmin/GPX or Garmin/GGZ folder, you're prompted for a new name (with an auto-suggested next-available name), unless "delete old files before upload" is checked, in which case the old files are cleared first. That checkbox now also works for GGZ exports (previously GPX-only).
Custom Waypoints (Hotel/POI, Parking Area, Trailhead, etc.) showed up on Garmin devices as empty "fake geocaches" — with blank D/T stars and Size: (Not Chosen) — instead of looking like the simple waypoints they are. They're now exported as plain GPX waypoints with a proper native Garmin icon (e.g. "Parking Area", "Lodging", "Trail Head"). Confirmed fixed on a GPSMAP 64s via GPX import.

Added

"Use database name as filename" checkbox in the GPS export dialog (community suggestion — GSAK has an equivalent), checked by default, covering both file-mode and device-mode exports. Pre-fills the export filename with the currently active database's name instead of a fixed "opensak" default, reducing how often the collision prompt above gets triggered in normal day-to-day use.

Notes

Adventure Lab stages were also given a distinct icon attempt ("Flag, Blue" instead of the standard geocache icon) — this had no visible effect on the tested GPSMAP 64s (device firmware appears to always use its own icon for anything containing a full groundspeak:cache block, regardless of the sym field). Left in place since it's harmless and may help on other device/firmware combinations; Lab stages keep their full description/D-T/hint content either way, which was the more important fix.

[1.16.0-beta.15] — 2026-07-28

Beta release — a cluster of fixes to GPX/GGZ export for Garmin devices ("Send to GPS"), including the root cause of Garmin firmware not recognizing exported files as geocaches at all.

Fixed

GPX/GGZ export missing attributes, descriptions, state, and full log history (#656) — generate_gpx() (used by both GPX and GGZ export, since GGZ embeds a generated GPX internally) never wrote groundspeak:attributes, groundspeak:short_description / long_description, or groundspeak:state at all, and hardcoded log export to only the last 5 logs with text truncated to 500 characters. Confirmed via a controlled side-by-side comparison against GSAK's export of the same cache (GC.com direct download and GSAK both included full attributes, description, and complete log history for the same cache; OpenSAK's export was missing all of it). All of the above is now exported in full, with no artificial limits.
Garmin devices not recognizing OpenSAK's GPX/GGZ exports as geocaches (#656) — the underlying root cause of the above and of related reports (#453, #454, #455, #502): OpenSAK exported GPX 1.1 (xmlns=".../GPX/1/1") with groundspeak:cache wrapped inside a GPX-1.1-style <extensions> element. GSAK exports GPX 1.0 (xmlns=".../GPX/1/0") with groundspeak:cache as a direct child of <wpt> — no <extensions> wrapper — which is what Garmin's on-device geocache parser is built around. Export format now matches GSAK's exactly, including declaring the groundspeak XML namespace locally on the <groundspeak:cache> element itself rather than at the GPX root. Confirmed fixed on a Garmin GPSMAP 64s: description, logs and hint all now display correctly for an exported cache, where previously only the hint displayed and description/logs did not.

[1.16.0-beta.14] — 2026-07-26

Beta release — vertical gridlines in the database grid (#463).

Added

Vertical gridlines in the cache table (#463) — columns in the main cache grid now have a thin vertical separator line at each column boundary, in addition to the existing alternating row colours. The line colour follows the active theme's palette (light/dark), so no additional theme handling is needed. User-configurable colours, independent show/hide toggles for horizontal vs. vertical lines, and the proposed new Appearance settings tab remain out of scope for now — basic functionality comes first, per the discussion on the issue.

[1.16.0-beta.13] — 2026-07-26

Beta release — a new Default Column View system (named, saveable column configurations with a toolbar quick-switch, replacing #606's implicit "last used" fallback), plus three small UI fixes: centered text columns, dark-theme placeholder text, and a readable Where-filter error box in dark mode.

Fixed

Country/Region/County columns left-aligned instead of centered (#603) — #431 centered "similar short-value columns" (Placed By, dates, etc.) but missed Country, Region (state) and County. Centered here for the same consistency #431 was going for.
Placeholder/hint text invisible in Dark theme (#624) — QPalette's PlaceholderText role wasn't set explicitly for either theme, so Qt fell back to a derived default that was unreadable in dark mode. Both palettes now set it explicitly to a legible, dimmed gray.
Where-filter SQL error box unreadable in Windows dark mode (#613) — the error box used a hardcoded light-theme style (dark-red text on a transparent background); in dark mode that rendered as dark-red-on-dark gray. It now picks an explicit, theme-appropriate style via the existing effective_theme() helper.

Added

Default Column View (#607) — the "Choose columns" dialog now supports named, saveable "Column Views" (visible columns, widths, container/type display), parallel to saved filter profiles. A saved view can be picked from a dropdown, and one view can be marked as the global default via a new "Set as Default" button (shown with a ★ in the dropdown). Any database without its own explicit column configuration — including brand-new databases — now falls back to the designated default view instead of the hard-coded factory defaults. This replaces #606's implicit "last used" fallback, which silently changed on every save regardless of user intent; setting the default is now an explicit, deliberate action and does not retroactively affect databases that already have their own saved configuration. A quick-switch Column View dropdown has also been added to the main toolbar, next to the existing filter-profile dropdown, so a saved view can be applied to the active database with one click, without opening the dialog.

[1.16.0-beta.12] — 2026-07-26

Beta release — a cluster of GPX-import fixes reported by the community: a lost hidden date from Project-GC exports, logs being wiped on every re-import instead of accumulating, Community Celebration Event caches showing the wrong type, and column settings not carrying over to a new database.

Fixed

Hidden date lost when importing a GPX file from Project-GC (#617) — _parse_datetime() only handled a bare or Z-suffixed ISO 8601 timestamp (geocaching.com's own format). Project-GC instead exports an explicit UTC offset (2026-03-16T00:00:00+00:00), which matched none of the old strptime patterns and silently came back as None, dropping the hidden date entirely. Now tries datetime.fromisoformat() first and converts properly to UTC (a +02:00 offset is converted to the correct UTC instant, not just relabelled), falling back to the previous patterns for anything else. The same function also parses log dates, so this fixes any log-date loss from Project-GC-style exports too, not just hidden dates.
Logs removed from one GPX import to another (#618) — every re-import (e.g. loading a new Pocket Query) deleted all of a cache's existing logs and rebuilt them from that file alone. Since a single GPX/PQ typically only carries a cache's most recent handful of logs, this meant older logs not present in that particular file were permanently lost on the next import — unlike GSAK, which lets logs accumulate over time. Logs are now merged instead: a matching log ID is updated in place (e.g. an edited log text), a new one is added, and any existing log absent from the current file is left untouched. log_count, last_log_date, and dnf_date are now derived from the full merged set rather than just the current file's logs, so a partial re-import can no longer move them backwards or silently clear them.
Community Celebration Event caches imported as generic "Event Cache" (#591) — geocaching.com's own machine-readable <groundspeak:type> field has no distinct value for Community Celebration Events (a limited-run program, May 2020 – Dec 2021) — it's always exported as plain "Event Cache". The actual event type only survives in the free-text cache name (e.g. "Karlínská kasárna - Community Celebration Event"). Added a narrow fallback: when groundspeak:type is exactly "Event Cache" and the cache name contains the literal phrase "Community Celebration Event", the cache is now classified as such — using the dedicated type/icon that already existed but was never being reached. Verified against a real-world GPX export for the reported cache (GC8T83E); a plain Event Cache, or another type merely mentioning the phrase in its name, is correctly left unchanged.
Displayed field when importing .gpx not consistent with previous settings (#606) — column visibility and widths are saved per database name, so a brand-new database had no saved key yet and fell straight back to the hard-coded defaults, silently reverting any customisation (e.g. an added Country/State/County column) already made in another database. Added a global "last used" fallback key, updated on every save, that a database with no settings of its own now falls back to before the hard-coded defaults — so a new database inherits whatever was last configured, anywhere.

[1.16.0-beta.11] — 2026-07-22

Beta release — a setting to disable the map panel entirely (and a visible "disabled" placeholder instead of an empty-looking map), plus a setting to cap the map to the nearest N caches from your home coordinate. Both target the same thing #627 already identified: map load is the largest remaining cost in "show me my caches" on a large database.

Added

Limit map to nearest N caches from home coordinate (#639) — combines @nagisml's "max caches shown on the map" suggestion (on #638) with sorting by distance from the active home coordinate, since a map with hundreds of thousands of pins isn't very readable at normal zoom anyway. New "Max caches shown on map" spinbox in the Map settings tab (0 = unlimited), default 2000. The map now gets its own fetch, independent of the table's — same active filterset, but sorted by distance and limited — so the table's own result set and sort order are completely unaffected either way.

Added push_limit to apply_filters_lightweight()/apply_filters_auto() (default False, no behavior change for existing callers): when the whole filterset is SQL-pushed and the sort field is SQL-sortable, the limit is pushed into the SQL query itself (LIMIT after ORDER BY) instead of fetching every filtered row and slicing in Python. Measured directly (100,000-cache database, distance-sorted, no filter): the existing Python-slice limit took ~3.0s regardless of requested size (500 through 5000 all fetch and construct every row before slicing); pushing a real SQL LIMIT took 0.31-0.56s, correctly scaling with the requested size. Falls back to the existing (always-correct) Python-slice behavior whenever those two conditions aren't both met — a SQL LIMIT applied before a Python-only sort or Python-only filter pass would silently return the wrong N rows, so this is deliberately conservative; 13 dedicated tests cover both the cases where it activates and the ones where it correctly must not.

End-to-end confirmed (100,000-cache database, fetch + JSON payload build): 4.65s (unlimited) → 0.46s (default limit of 2000) — ~10x faster, with the JSON payload itself shrinking from ~195MB to ~3.9MB.
Setting to disable the map panel (#638) — new "Map" tab in the Settings dialog (split out from General, so future map-specific settings have a natural shared home) with a "Show map" checkbox, defaulting to on (opt-out, zero behavior change for anyone who doesn't touch it). When off, all three mainwindow.py refresh paths skip building and loading the map's marker data entirely — since #627, map load has been the single largest remaining cost in "show me my caches" on a large database, bigger than the database query itself. Measured directly: the map-load step (Python-side payload build) drops from ~1.8s to effectively 0s on a 100,000-cache database when disabled. Toggling the setting back on mid-session needs no special handling — the existing Settings-dialog-close flow already calls _refresh_cache_list() unconditionally, which now correctly re-populates the map on the next call since the guard re-checks the setting fresh every time. Table contents and sort order are completely unaffected either way.

Follow-up: the map's own base tiles/zoom controls rendered regardless of the setting — only the cache markers were actually skipped — which looked like a stuck or empty map rather than an intentional off state (reported after testing the initial version). A QStackedWidget now swaps in a plain "Map disabled" placeholder page instead, so the off state is visually unambiguous. Also skips the map's own page reload (reload_map()) while disabled, since there's no point refreshing tiles nobody's looking at — it reloads normally the next time the setting is re-enabled. Translated to all 8 languages.

[1.16.0-beta.10] — 2026-07-22

Beta release — SQL pushdown for the last group of filters that had none at all, following up on #627's lightweight query path.

Added

SQL pushdown for remaining scalar-column filters (#633) — UserFlagFilter, LockedFilter, DnfFilter, FtfFilter, FavoritePointsFilter, HasCorrectedFilter/NoCorrectedFilter, FoundByMeDateFilter, DnfDateFilter, and LastLogDateFilter previously had no apply_to_query() at all, always falling back to a full Python matches() scan. Under the ORM path this barely mattered (#631 found the Python pass was only ~2% of apply_filters()'s time — ORM hydration dominated regardless of whether a filter narrowed the SQL query or not). Under the lightweight query path (beta.9) the picture is different: without SQL pushdown, apply_filters_lightweight() must still construct a LightweightCache for every single row before Python-filtering it down, so a highly selective filter with no pushdown costs almost as much as fetching the whole table. Measured directly on a 100,000-cache database: FtfFilter (0.6% selectivity) went from 2.42s to 0.05s (~48x faster); UserFlagFilter (~5%) from 2.49s to 0.16s (~15x); DnfFilter (~7%) from 2.39s to 0.20s (~12x).

Each apply_to_query() mirrors its matches() counterpart exactly, including NULL handling — FoundByMeDateFilter/DnfDateFilter treat a NULL date as "include" (found/DNF but undated), while LastLogDateFilter treats NULL as "exclude", and both are preserved precisely in SQL. HasCorrectedFilter/NoCorrectedFilter use a correlated EXISTS/NOT EXISTS against user_notes, which needed an explicit .correlate(Cache) — without it, apply_filters_lightweight() raised InvalidRequestError because its select() already outerjoins user_notes for corrected-coordinate display, confusing SQLAlchemy's auto-correlation. Only broke on the lightweight path, not the full ORM path — caught by testing both, not just one.

25 new parity tests (including NULL edge cases for every field involved) confirm every filter's SQL and Python forms agree exactly; full unit-test suite (2136 tests) green, mypy clean.

[1.16.0-beta.9] — 2026-07-22

Beta release — the lightweight query path (#627): large databases load dramatically faster in both the cache table and the map, on top of #628-#631's smaller fixes from the last two betas. This is a default-behavior change for every install, not opt-in — see below for why that's safe.

Added

Lightweight query path (#627) — apply_filters_lightweight(), a new function in filters/engine.py alongside apply_filters(), fetches cache rows via a SQLAlchemy Core select() instead of session.query(Cache), avoiding the ORM row-hydration cost already identified as apply_filters()'s dominant expense (#631). Results come back as LightweightCache objects — duck-typed to expose the same attribute names as a real Cache for every column the table and map actually use — with an automatic, transparent fallback to the existing apply_filters() ORM path whenever a filter needs a relationship or one of the three heavy/deferred text fields (short_description, long_description, encoded_hints). That fallback means this is always correct, never returning wrong or incomplete results — only sometimes slower than it could be.

mainwindow.py's table and map refresh now go through a single apply_filters_auto() entry point that always attempts the lightweight path. A thorough compatibility audit (CacheTableModel's every column, sort key, and tooltip; map_widget.py's _do_load_caches(), _effective_coords(), and pin-icon generation) found zero source changes were needed in either consumer — both already only touch scalar fields, cached count columns, or .user_note's three attributes, never a relationship collection directly. Row selection already reloads a full Cache via the established _load_full_cache(gc_code) pattern regardless of what's currently in the table.

Confirmed final numbers (250,000-cache database, via the real apply_filters_auto() wiring):

Scenario	`apply_filters`	`apply_filters_auto`	Speedup
No filter	8.05s	3.10s	~2.6x faster
Exclude archived	11.02s	3.81s	~2.9x faster
Within 50km	1.52s	0.97s	~1.6x faster
`CacheTableModel.load()`	0.198s	0.027s	~7.3x faster

Two real bugs were found and fixed during testing, before release: a LightweightCache design that eagerly copied every one of its ~52 fields at construction time fixed a table-load regression (delegating every attribute through __getattr__ was costing more than the fetch-side win it was meant to complement) but overcorrected, nearly erasing the fetch-side win in the process — the final design only eagerly copies the three fields CacheTableModel touches unconditionally on every row (id, distance, bearing), leaving everything else lazy. Separately, reload_caches_full() — the helper GPX/LOC/GGZ export, KML export, GPS-device export, and the trip planner all use to reload full cache data before generating output — checked isinstance(c, Cache), which silently excluded every LightweightCache row from its reload and would have crashed all four export paths the moment they touched a deferred field; fixed by recognizing both types as reloadable. Both were caught by the project's own test suite (the second one by the e2e suite specifically) before ever reaching a tagged release.

Confidence for shipping this as default (not opt-in) behavior comes from: the lightweight path's own automatic per-filterset fallback to the exact same full-ORM code path used today; full parity test coverage (test_filter_sql_parity.py, test_filter_lightweight.py) proving apply_filters_lightweight() never diverges from apply_filters()'s result set across every filter type, NULL edge case, and composition; dedicated compatibility audits and test suites for both the table and the map with zero source changes needed in either; and a full pass of the unit suite (2111 tests), the e2e suite (244 tests), and a 250,000-cache benchmark, all green.

Benchmark harness measures the lightweight query path (#628) — scripts/benchmark_large_db.py now also runs its three apply_filters scenarios through apply_filters_auto(), and runs the map/table-load steps against both result sets, so a single report shows the full before/after picture instead of requiring a separate isolated A/B script. Fixed a measurement-fairness bug found while adding this: the icon HTML @lru_cache (#629) meant whichever "Map load" measurement ran first in the script paid the one-time cache-warming cost and the second one benefited "for free" — fixed with an explicit warmup pass before either timed measurement.

[1.16.0-beta.8] — 2026-07-22

Beta release — a small, safe correctness fix in the filter engine's SQL pushdown, spun off from the #627 large-database investigation.

Fixed

Boolean filters silently bypassed their indexes (#628, part of #627) — FoundFilter, ArchivedFilter, AvailableFilter, AvailabilityFilter, PremiumFilter, and NonPremiumFilter used Cache.<col>.is_(True) / .is_(False) in their SQL pushdown, which compiles to <col> IS true / IS false. SQLite's query planner cannot use an index for that form — verified directly against SQLite 3.45 with EXPLAIN QUERY PLAN — even though the functionally identical <col> = true / = false (what == True/== False compiles to) is index-usable and the relevant indexes have existed since #214. .is_(None) (NULL checks, e.g. DifficultyFilter's unknown-difficulty handling) was never affected and is unchanged. Real-world impact is small at current database sizes — isolated A/B testing showed ~0.11s either way for a selective filter on 100,000 caches, since raw SQL execution is dwarfed by ORM row hydration (same finding as #631) — but this restores the indexing intent from #214 at zero cost and zero risk.

[1.16.0-beta.7] — 2026-07-22

Beta release — large-database performance work (see #627): map load is dramatically faster on big databases thanks to icon caching and bulk marker loading, plus a small, safe win in the filter engine. Includes a new benchmark harness so every step here — and future ones — can be measured instead of guessed at.

Measured on a 250,000-cache synthetic benchmark database (scripts/benchmark_large_db.py): map load dropped from 11.34s to 4.45s (-61%), total time to show all caches dropped from 37.58s to 30.42s (-19%). Full before/after table in #628.

Added

Large-database benchmark harness (#628, part of #627) — scripts/benchmark_large_db.py generates a synthetic database at a configurable scale (default 250,000 caches) and measures distance recalculation, apply_filters(), map load, table load, and info-bar update, printing a table (optionally markdown) for pasting into GitHub issues. Every performance change in #627 is now measured against this harness rather than eyeballed.

Improved

Cache map pin HTML generation (#629, part of #627) — get_map_pin_html() now caches its output with @lru_cache(maxsize=256). The HTML (including base64-encoded SVG) only depends on (cache_type, found, dnf), a small bounded set of combinations, but was previously rebuilt from scratch for every visible cache on every map load. On a 100,000-cache benchmark database (see #628's scripts/benchmark_large_db.py), map load time dropped from ~10.1s to ~3.2s (~68%).
Bulk-load map markers with chunked clustering (#630, part of #627) — the map's loadCaches() called Leaflet.markercluster's addLayer() once per cache, which rebuilds the library's spatial index on every single call. It now builds all markers first and adds them in one addLayers() bulk call, with chunkedLoading: true so the browser's UI thread stays responsive while a large marker set loads. The post-load pan/fit-bounds step is deferred until every chunk has actually been added (via chunkProgress), so it still reflects the complete marker set instead of a partially-loaded one.
Skip redundant Python filter pass when fully SQL-pushed (#631, part of #627) — apply_filters() now skips its Python-level filterset.matches() re-scan when every filter in the filterset was already pushed into the SQL WHERE clause, since every row query.all() returns already satisfies it. Measured impact is modest — the Python pass itself is only ~2% of apply_filters()'s time even on a 100,000-cache database with a large result set (~6.8s total, ~0.13s of which was the redundant pass); ORM hydration dominates and is unaffected by this change. Still a safe, zero-cost win, and it required introducing a new BaseFilter.sql_exact flag: while implementing this, testing surfaced that DistanceFilter's SQL pushdown is a bounding-box pre-narrowing only (not an exact translation — it ignores min_km entirely and doesn't have the true circle shape), so the naive "non-None apply_to_query() == fully handled" assumption would have silently dropped the min_km check for distance-filtered results. sql_exact lets a filter opt out of counting toward the skip decision while still contributing its SQL pre-narrowing; DistanceFilter is the only filter that needs it.

[1.16.0-beta.6] — 2026-07-21

Beta release — two data-integrity fixes for GSAK-database imports: attribute names and the attribute filter were often wrong, and corrected (solved-puzzle) caches lost their original coordinates on import.

Fixed

Wrong attribute settings from GSAK database import (#615) — 42 of the 70 Groundspeak attribute IDs in OpenSAK's internal attribute table were mapped to the wrong attribute (e.g. id 31 resolved to "Food nearby" instead of "Camping available"). Beyond GSAK-database imports, this also affected the attribute filter, which built its checkbox labels and underlying filter values from the same table — so filtering by attribute could silently return the wrong caches regardless of import source. Rebuilt and verified against real GPX exports from geocaching.com.
Caches with corrected coordinates lose the original coordinates when importing GSAK database (#614) — GSAK's own Latitude/Longitude columns reflect the corrected position once a cache has been solved, not the original/posted coordinates. OpenSAK imported these directly as the cache's primary position, silently discarding the true original location on every GSAK-database import of a solved cache. The original position is now read from GSAK's Corrected table instead.

[1.16.0-beta.5] — 2026-07-16

Beta release — startup no longer recalculates every cache's distance unnecessarily, which should noticeably speed up launch on large databases.

Fixed

Redundant distance recalculation on every startup (#579) — the app recalculated distance/bearing for every cache on every launch, even though nothing about the database or home point had changed since the last session. On large databases (100k+ caches) this made startup noticeably slow with no visual indication of what was happening. recalculate_distances() now persists the centre point and distance method it was run with, and on startup the app checks this — plus a cheap single-row spot-check against the database — before deciding whether a full recalculation is actually needed. Normal startup now skips it entirely; a database synced from another machine with a different home point (or otherwise modified outside this OpenSAK install) still triggers a full recalculation as before.

[1.16.0-beta.4] — 2026-07-15

Beta release — the database list/dropdown is now alphabetically sorted, plus a small message cleanup.

Fixed

Database list/dropdown was not sorted alphabetically (#531, #601) — the toolbar database dropdown, the Manage Databases dialog, and the database picker in Move Caches, GSAK import, and GPX/PQ import all listed databases in the order they were added/imported instead of alphabetically. All of these now show databases sorted alphabetically (case-insensitive) by name, matching GSAK's behaviour.
"Database created" message told the user to manually activate it (#464) — creating a new database already switches to it automatically, but the confirmation dialog still said to click "Switch to this" to activate it. The message now simply confirms the database was created and is active.

[1.16.0-beta.3] — 2026-07-15

Beta release — pick any cache, saved home point, or coordinate as the distance filter's center (#511), plus two small bugfixes.

Added

Center point picker for the distance filter (#511) — the "Afstand" filter no longer always centers on Home. Choose Home, any saved home point, the currently selected cache, or a manually entered coordinate as the center, and set an optional minimum distance alongside the existing maximum (both were already supported by the filter engine; only the maximum was previously exposed in the dialog). Built as a standalone, reusable widget for future reuse (planned for #558).
"Set as center point" (right-click) (#511) — right-click any cache or custom waypoint (e.g. a hotel added via Waypoint → Custom Waypoint) and choose "Sæt som centerpunkt" to recompute the Distance column for every cache from that point, exactly like switching Home. The chosen point's GC code/name is shown in the info bar's "Centerpunkt" field and in the Home dropdown until you pick a saved home point or another cache.

Fixed

Hint markup was being ROT13-scrambled (#595) — geocaching.com's own hint markup ([br] for a line break, place-name tags like [Étape] in French hints) was incorrectly rotated along with the rest of the hint text, so [br] showed up as its ROT13'd form [oe] instead of a line break. Bracketed markup is now left untouched by the ROT13 encode/decode, and [br] renders as an actual line break in both the cache detail hint tab and KML export.
Website: corrected GSAK's freeware date (#589) — the landing page's comparison table said GSAK became freeware in 2021; per research from a long-time GSAK user (French GSAK user since 2011), the free v9.0.0 shipped in 2019, with the last forum-provided patch dating from 2022.

[1.16.0-beta.2] — 2026-07-15

Beta release — custom waypoint types get their own icons, and the found-smiley icon set is simplified (#593).

Added

Custom waypoint types now have their own icons — Parking Area, Trailhead, Stage, Final Location, Reference Point, Waypoint, Hotel/POI and Custom each get a distinct icon in the table, map and detail panel, instead of all sharing the generic "unknown" (?) icon. Overridable via the same icons/cache_types/ user-icon mechanism as #519.

Changed

Simplified the found-smiley icon set (#593) — removed the 12 unused colour variants and the per-type colour-selection code behind them. Only gold (Found overlay + "Found" column) and dark_blue (DNF overlay) were ever actually shown in the app; the rest was dead code/assets. Reported by a community member in the OpenSAK Facebook group.

[1.15.0] — 2026-07-14

First stable release of the 1.15.0 cycle. Replaces the run of 1.15.0-beta.1 … 1.15.0-beta.16 builds — see git history for the detailed beta-by-beta log if needed.

Added

Direct GSAK database import (#469) — import an entire GSAK sqlite.db3 file straight into an OpenSAK database, without going via GPX first. Reads caches, waypoints, attributes, logs (full history, not capped like GPX/PQ exports), corrected coordinates, personal notes and trackables directly from the GSAK schema. Confirmed against several independent real-world GSAK databases during development, including a 1.1M-log-row one. GSAK custom fields, the Ignore list are out of scope for this first pass (tracked separately in #473).
Export to Garmin GGZ format (#348) — the GPS export dialog now offers a GPX/GGZ format choice. GGZ packs the exported caches (unlimited count, unlike GPX-based transfers) directly into a ZIP structure Garmin devices read natively, matching GSAK's GGZ layout byte-for-byte.
User-replaceable icon packs (#519) — custom cache-type and found- smiley icons can now be dropped into a new icons/ folder (Settings → Advanced → "Open icons folder") without touching any code or rebuilding the app. The folder lives alongside opensak.json, so it survives app updates/reinstalls. Also covers the fixed, single-instance UI icons (Corrected coordinates, Premium, Fav. points, Trackables) via an icons/ui/ subfolder. A bundled, offline "View icon naming guide" button lists every file name and recommended canvas size.
Trackables (travel bugs / geocoins) column and tab (#489, #538) — an opt-in column showing how many trackables are logged in each cache, and a new Trackables tab on the cache detail panel listing each one with a clickable coord.info link.
GSAK-style icons for Found, Premium and Fav. points (#489) — icons instead of plain text/numbers in the cache list, matching GSAK's own look.
Double-click a cache row to open it on geocaching.com (#471) — matches GSAK's behaviour.
Option to show hints decoded by default (#499) — new checkbox under Settings → Display. Off by default.
"Support OpenSAK" — now that the project is fiscally hosted by Open Source Collective, a Help menu entry, README/website badges, and a button right on the update-available dialog all link to opencollective.com/opensak.

Fixed

Changing the install/database folder via the setup wizard didn't move anything (#562) — re-running the setup wizard with a different install and/or database folder only updated the stored pointers, never the actual files, which could silently reset all settings or leave existing databases behind. Settings, custom icon packs, and the Geocaching.com OAuth token now move with the install folder (with a clear warning on collision instead of failing silently); changing the database folder now offers to move existing databases along; moving/ deleting the active database no longer crashes with "Database not initialised"; the "New Database" dialog now defaults to the right folder; and old, now-empty folders (including nested ones) are cleaned up automatically.
"Access is denied" crash saving settings on Windows (#574) — happened right after a reboot or update, when antivirus/indexing/ roaming-profile sync briefly held opensak.json open during the atomic save. The write now retries a few times before giving up.
Filter window always opened on the primary monitor (#580) on multi-monitor setups, regardless of which screen OpenSAK itself was running on. Now opens on the same monitor as the main window.
A cleared filter silently came back when returning to a database — clicking the red ✕, choosing "None" from the filter dropdown, pressing Escape, or clicking "All" in the status bar reset the filter in the current view but never persisted that per-database, so switching away and back reapplied the filter you'd just cleared.
Beta users never discovered a newer stable release — the update checker only ever compared a running beta against other betas, so beta.16 users wouldn't have been notified that this stable release existed.
Critical: severe UI freeze switching to or filtering a large database (#540) — the icon-override folder was being resolved from scratch (file check + JSON read + several mkdir() calls) for every single icon lookup, per row — commonly intercepted synchronously by antivirus on Windows, compounding into 45-60 second freezes on large databases. Now resolved once per session.
Dynamic map zoomed out to show the whole world for caches with hidden-coordinate (0/0) waypoints (#546), e.g. a finale left hidden after a GSAK import.
Clear-filter button (✕) had no hover highlight (#559), looking non-interactive compared to the rest of the toolbar.
A batch of GSAK Database Import fixes found during real-world testing: waypoints sharing a name but not a code were silently dropped (#536); renaming a database didn't move the underlying file, so a new database under the freed name silently reopened the old one (#539); a leftover favorite_point column crashed inserts on some databases (#530); non-UTF-8 text fields aborted the entire import instead of falling back gracefully; Adventure Lab and five other cache types imported as "Unknown Cache" (#532); county wasn't imported from GSAK-exported GPX (#521); trackables and premium status weren't imported (#538, #541); and the "New Database" default folder pointed at the install folder instead of the configured database folder.
Distance column could show stale values after editing a home point (#522) — only switching center points via the toolbar recalculated distances; editing a home point's coordinates in Settings didn't.
"Has trackables" filter crashed on any database created before v1.14.0 (#491) — a missing table-creation migration.
Map didn't update when correcting coordinates via the cache list's right-click menu (#474), unlike the same action from the detail panel.
SMALL row-height setting silently ignored on some systems (#490).
Potential crash on databases with a "Favourite" (★) column enabled (#488) — removed; GSAK only tracks community Fav. points.
Found date missing for webcam caches and events on import (#457) — found_date was derived only from "Found it" logs.
FTF detection flagged logs that only mentioned "first to find" in passing (#458) — now matched exclusively against ProjectGC's official tags.
Several GGZ export bugs (#348): a crash on databases with mixed dated/ undated logs; files written to the wrong device folder; wrong dates inside the exported ZIP; and a severe slowdown on large exports (#466), 200-1000× faster after fixing an accidentally-quadratic offset calculation.
Corrected Coordinates icon inconsistency (follow-up to #354) — a consistent SVG warning-triangle icon everywhere, replacing a hard-to-see emoji.
Found count under the grid counted found caches, not found logs (#552) — a relocatable/multi-visit cache found several times only ever contributed 1 to the total.
Filter couldn't be cleared via the toolbar "None" dropdown, plus a new configurable Escape shortcut to clear the active filter (#553).
Large Text setting not applied consistently — the GC Code column stood out at the wrong size (#547).
Deleting the active saved filter left it applied until the next unrelated action (#491).
Flag and locked column icons distorted on found-cache rows (#509) — emoji glyphs don't have a real italic form on some platforms.
File-mode GPS export silently overwrote existing files (#501) — now prompts for a new filename if the target already exists.
Filters with zero matches emptied the cache list (#444) — now rejected with a warning instead, matching GSAK's behaviour.

[1.14.0] — 2026-06-29

First stable release of the 1.14.0 cycle. Replaces the run of 1.14.0-beta.1 … 1.14.0-beta.20 builds — see git history for the detailed beta-by-beta log if needed.

Added

Lock a cache against import overwrites (closes #202) — a long-requested GSAK feature. Locking a cache freezes its scalar fields (name, type, container, coordinates, D/T, owner, status, descriptions, hint, country/state/county) so a later PQ/GPX re-import can't silently change data your stats depend on. Logs, attributes and waypoints still refresh normally. Filterable and sortable like any other column.
Personal notes, round-trippable with GSAK (closes #389, #390, #391, #392) — a new "Notes" tab on the cache detail panel for free-text notes per cache, separate from the geocaching.com description and logs. Imported from and exported back to GSAK's gsak:UserNote extension, so a note survives an export → GSAK → re-import round trip.
Child waypoints are now visible in the UI (closes #376, #377, #378, #393) — cache names with waypoints show in bold in the list, a new "Waypoints" tab lists each one's prefix, type, name, coordinates and description, and selecting it shows the markers on the map.
Attributes tab in the cache detail panel (closes #417) — lists every cache attribute with a green ✓ or red ✗ marker.
Keyboard Shortcuts dialog (closes #205) — Help → "Keyboard Shortcuts…" opens a searchable reference of every shortcut. Shortcuts are managed through a central registry; user overrides persist across restarts.
Full-text search filter (closes #294) — a new "Text Search" tab in the filter dialog searches cache descriptions, logs, and personal notes (hint text off by default), pushed down to SQL so it stays fast on large databases.
Cache type icon in the detail panel (closes #286) — shown next to the cache title, scaling with the text-size setting. Found/DNF map-pin smileys now correctly use gold/dark-blue regardless of cache type, matching GSAK.
Type column display options (closes #413, #414, #415, #416) — show icon only (default), name as text, or both, via a new column-dialog setting.
Distance calculation reworked (closes #60) — now computed once per centre-point change instead of on every refresh, which kept large databases noticeably faster. A new Vincenty (WGS84) method is available alongside the existing Haversine default in Settings → Advanced.
Active filter count in the info bar (closes #373) — shows e.g. "3 filters active" instead of a generic label.
Welcome wizard for first-run setup (closes #210) — walks new installations through language, installation folder, database folder, optional Geocaching.com profile, and a confirmation screen. A new "Run setup wizard again" button in Settings → Advanced (fixes #358) lets you re-run it later, e.g. to change folders.
JSON-based settings store (closes #209) — replaces QSettings and the old preferences.json with a single opensak.json file. Existing installations migrate automatically and transparently on first launch of this version.
Database and installation folders manageable from Settings → Advanced — view both folders, and move existing databases to a new folder (with the option to keep or delete the originals) without going through the setup wizard again.
Per-database column views with drag-to-reorder (closes #199) — visible columns and widths are remembered separately per database; drag column headers to reorder them.
UI text and icon size is now adjustable (closes #286, #287, #290) — a new Settings → Display option offers Small, Medium (default), and Large, affecting the cache list, detail panel, and tab labels.
GC Code colors and clickable status counts now match GSAK (issue #270) — found caches show yellow, your own caches show green, and clicking a colored count in the info bar (Found / My caches / Inactive / All) filters the list to that status.
GSAK personal/user fields are now imported (closes #269) — UserFlag, IsPremium, UserSort, UserData/User2/User3/User4 and FavPoints from GSAK-exported GPX are imported without overwriting data on a later plain Pocket Query re-import.
Full log text shown without truncation (fixes #218), and links in logs are now clickable (fixes #219), matching the existing behaviour of the cache description tab.
User Guide link in the Help menu — opens the online User Guide directly in your default browser.
Debug logging system — writes to opensak.log in the install directory (resets on startup, rotates at 1 MB). "Open log file" was added to the Help menu, making it easy to attach when reporting issues.
New "no corrected coordinates" filter (fixes #274) — mirrors the existing Premium/Non-Premium filter pair; previously unchecking "has corrected coordinates" alone produced no filter at all.

Changed

Owned-cache counting and coloring now use the owner field instead of placed_by (issue #270) — an adopted cache is now attributed to its current owner, matching GSAK.

Fixed

Hint encoding detection was reversed (fixes #329) — geocaching.com PQ exports deliver hints as plaintext, not ROT13 ciphertext as previously assumed; OpenSAK was showing plaintext hints as gibberish and vice versa. Display defaults to obscured either way; "Decode hint" reveals it.
Google Maps link in the cache detail pane didn't open (fixes #321).
GSAK GPX logs were capped at 20 entries (fixes #266) — all logs are now shown.
A companion -wpts.gpx file could import as a duplicate set of caches (fixes #410) — detection now inspects file content instead of filename.
Container/size column sorted alphabetically instead of by actual size (fixes #412).
Favorites column showed on new databases despite always being empty (fixes #418) — off by default now, since populating it requires the Geocaching.com Live API, which OpenSAK doesn't have yet.
Adventure Lab stages with non-GC/LC prefixes were silently dropped on import (fixes #359).
Newly imported caches showed no distance or bearing until restart (fixes #359).
GC Code text could be unreadable in dark mode (fixes #366).
Unset flag column had no visual indicator (fixes #290).
Locale-aware dates weren't zero-padded consistently (fixes #369).
Enter key in the filter dialog triggered "Save profile" instead of Apply (fixes #370).
Text/icon size setting didn't take effect until reselecting a cache (fixes #371).
Import progress bar was indeterminate (fixes #372) — now shows real progress based on a waypoint pre-scan.
Small/Large text size options looked almost identical to Medium (fixes #374, #375) — range widened, and the setting now also applies to the cache grid's font and row height.
Cache detail panel could crash when sorting logs with some entries missing a date (fixes #429).
Several cache-table columns weren't center-aligned like their neighbours (fixes #431).
Update checker failed with SSL certificate errors on Windows — the bundled .exe now explicitly uses certifi's certificate bundle.
Setup wizard's database-folder step defaulted to the install folder instead of the actual database folder on re-run.
Boolean settings could silently corrupt to base64 strings in the new JSON settings store — existing corrupted values repair automatically on startup.

For planned features and known issues see the GitHub Issues list.

[1.13.12] — 2026-06-15

Added

Export progress shows how far it has reached (closes #207) — the GPS, file (GPX/LOC/GGZ) and KML export dialogs now display a determinate progress bar with the number of caches processed and the percentage (e.g. 320 / 500 (64%)) instead of an indeterminate "running" bar, giving a sense of how long the export will take. Suggested in issue #207.

Fixed

Export no longer crashes with DetachedInstanceError — the cache table loads rows with the description/hint text and logs/waypoints left out for speed, so exporting them straight from the table raised DetachedInstanceError (and would otherwise have dropped hints and logs from the output). Exports now reload the full cache data first, so GPX/LOC/GGZ/KML files always include hints, logs and waypoints.
Re-importing an exported GPX no longer imports 0 caches — OpenSAK exports GPX 1.1 (with the Groundspeak data wrapped in <extensions>), but the importer only recognised GPX 1.0 with the Groundspeak block as a direct child, so importing an OpenSAK-exported file (or any GPX 1.1 file) found nothing. The importer now reads both GPX 1.0 and 1.1.
Reverse geocoding no longer crashes in released builds (#215) — reverse_geocoder and pycountry were declared in pyproject.toml but missing from requirements.txt, which CI and the PyInstaller builds installed from; because they are imported lazily the app started fine but the Country/State/County lookup crashed in every shipped binary, undetected by CI. pyproject.toml is now the single source of truth — CI and builds install the project (pip install -e ".[dev]"), requirements.txt is removed, the bundles ship the libraries' data files (GeoNames CSV, ISO tables), and a smoke test exercises the real lookup so a missing dependency fails CI.

For planned features and known issues see the GitHub Issues list.

[1.13.11] — 2026-05-29

Fixed

Adventure Lab caches from lab2gpx can now be imported — GPX files generated by lab2gpx use LC-prefixed codes (e.g. LC378B-2) instead of the standard GC prefix. These were previously silently skipped during import. OpenSAK now accepts both GC and LC codes, so lab2gpx files import correctly. Cache type, name, coordinates and description are all parsed as expected, and Lab Cache entries are shown with the L label in the container column.

[1.13.10] — 2026-05-09

Added

Drag & drop to import GPX / ZIP files (closes #181) — GPX, ZIP and LOC files can now be dragged from a file manager and dropped anywhere on the OpenSAK window. The import dialog opens immediately with the dropped files pre-loaded and ready to import. Multiple files can be dropped at once. Suggested by Fabio-A-Sa.
Target database selector in import dialog — The import dialog now shows a database dropdown pre-filled with the currently active database. Any known database can be selected as the import target, making it possible to import a PQ directly into a specific database without switching the active database first. Works with both drag & drop and the normal Browse button.

[1.13.9] — 2026-05-09

Added

File → Export menu with GPX, LOC and GGZ support (closes #203) — A new Export submenu has been added under the File menu with three file format options:
- GPX — full Groundspeak GPX 1.1 with cache details, logs and attributes
- LOC — lightweight waypoint format supported by most GPS apps and devices
- GGZ — Garmin's ZIP-based container format that lifts the 10,000-cache limit on supported devices (e.g. GPSMAP 64/66, Oregon 700+). The GGZ file contains a full GPX file plus a Garmin index, identical in structure to GSAK's GGZ export.
All three formats use corrected coordinates automatically when available. Export runs in a background thread so the UI stays responsive for large databases.
Export to Google Maps (KML) moved to File → Export — The Export to Google Maps (KML)… item has been moved from the GPS menu to the new File → Export submenu, where it fits better alongside the other file export formats.

[1.13.8] — 2026-05-08

Added

Edit cache in right-click menu (fixes #124) — A new ✏️ Edit cache… item has been added to the right-click context menu in the cache list. It opens the same edit dialog as Waypoint → Edit cache in the menu bar, making it faster to edit a cache without leaving the list.
FTF checkbox in Edit Cache dialog (fixes #123) — The Status tab in the Edit Cache dialog now includes a FTF (First to Find) checkbox, making it possible to set or clear the FTF flag manually directly from the dialog.
FTF toggle by clicking the FTF column — Clicking directly on a cell in the FTF column toggles the First to Find flag on or off, the same way the User Flag column works.
FTF filter in filter dialog — A new FTF (First to Find) 🥇 filter group has been added to the Other tab in the filter dialog, allowing you to filter caches by their FTF status.
Double-click corrected coordinates cell (fixes #200) — Double-clicking a cell in the Corrected column now opens the corrected coordinates dialog directly, without needing to use the right-click menu.
Enhanced corrected coordinates dialog — The corrected coordinates dialog now shows the cache's original coordinates and the entered corrected coordinates in all three formats (DMM, DMS, DD), each with a copy-to-clipboard button for easy use in other applications.

Fixed

Clear filter button is now red when active (fixes #201) — The ✕ clear filter button in the toolbar is now displayed in red when a filter is active, making it immediately obvious that the cache list is filtered. The button turns gray and is disabled when no filter is applied.
Crash on exit during update check — OpenSAK could crash with a core dump when closing the window while a background update check was still running. The update worker is now stopped cleanly when the main window closes.

[1.13.7] — 2026-05-08

Added

Filter profile dropdown in toolbar — A new dropdown next to the 🔍 filter button lets you switch between saved filter profiles instantly without opening the filter dialog. Selecting a profile applies it immediately; selecting None clears the active filter. The active profile is remembered per database and restored automatically on startup and when switching databases.
New filter tab: Other — A fifth tab has been added to the filter dialog with additional filter options:
- Country / State / County — text contains search (case-insensitive)
- User Flag — filter on whether the user flag is set or not
- DNF — filter on Did Not Find status
- Favorite points — filter by a minimum/maximum favorite point count
Extended Dates tab — Two new date range filters have been added alongside the existing Hidden date and Last log date filters:
- Found by me date — filter on when you personally found the cache
- DNF date — filter on when a DNF was recorded

Fixed

Filter profile not persisted across restarts — Selecting a filter profile from the toolbar dropdown was not remembered when OpenSAK restarted. The active profile is now saved to QSettings per database alongside the sort order and restored on next launch.
Selecting "None" in filter dropdown did not update cache list — Switching back to no filter via the toolbar dropdown now immediately refreshes the cache list.
Country / State / County filters returned no results — These filters previously required an exact match against a list. They now use case-insensitive contains search, consistent with the Name and GC code filters.

[1.13.6] — 2026-05-07

Added

Export to Google Maps (KML) — New menu item under GPS → Export to Google Maps (KML)… exports the currently filtered caches to a .kml file that can be imported directly into Google My Maps. The file contains two layers: one for geocaches (colour-coded by cache type with paddle icons) and one for custom waypoints. Corrected coordinates are used automatically when available. Options: include/exclude custom waypoints and already-found caches.

Fixed

Corrected coordinates crash — Setting corrected coordinates via right-click now saves correctly without crashing. The cache list updates immediately to show the 📍 indicator without requiring a manual refresh.

[1.13.5] - 2026-05-07

Update notification improvements

Update popup now includes a "See changelog" link opening the full changelog on GitHub
Added "Skip this version" button — suppresses the popup for that release until a newer version is available
Manual update check (Help → Check for updates) always shows the popup, regardless of skipped version
Added automatic update check toggle in Settings → Advanced

[1.13.4] — 2026-05-07

Added

Light / Dark / Automatic theme — A new Appearance section in Settings lets you choose between a light theme, a dark theme, or Automatic which follows the operating system setting. The change takes effect immediately without restarting. Dark mode is detected natively on macOS (System Preferences), Windows 10/11 (registry) and modern Linux desktops (freedesktop portal / GTK theme).

Fixed

Consistent look across Linux, Windows and macOS — OpenSAK now forces Qt's Fusion style on all platforms, giving a uniform baseline appearance regardless of the desktop environment or OS theme. A platform-appropriate default font is applied automatically (Segoe UI on Windows, SF Pro on macOS, Ubuntu on Linux).
Cache list text invisible in dark mode — The GC code column delegate used hardcoded black text in all cases. Rows without a status colour (archived / found / placed) now use palette.text() so the text is readable in both light and dark themes. Status-coloured rows (red / yellow / green pastels) keep black text since the pastel backgrounds are always light.
Strikethrough and colour confined to GC code column (fixes #196) — Strikethrough for archived caches and the orange disabled colour were previously applied to the cache name and type icon columns as well. They are now shown exclusively in the GC code column, making the status easier to read at a glance without affecting the other columns.
Theme change did not update all open windows — Switching theme in Settings left already- visible widgets (including the cache list) unchanged until restart. The theme engine now explicitly propagates the new palette to every open window and its child widgets, so the entire UI updates in one go when you click OK.

[1.13.3] — 2026-05-06

Added

Colour-coded GC codes (fixes #117) — Cache type colours are now applied to the GC code column in the cache list, making it easy to spot cache types at a glance. The colours in the Count: summary bar have been updated to match.

Fixed

Strikethrough for archived and disabled caches (fixes #118) — Cache entries that are archived or temporarily disabled are now shown with strikethrough text in the cache list, giving a clear visual indication that the cache is not currently active.
Delete database — empty folder cleanup (fixes #146) — After deleting a database, OpenSAK now checks whether the containing folder is empty. If it is, a prompt is shown offering to delete the folder as well, so no orphaned folders are left behind.

[1.13.2] — 2026-05-05

Added

Found status and date set automatically on PQ import — When importing a standard Pocket Query, caches you have found are now automatically marked as found and given the correct found date. OpenSAK reads the <sym>Geocache Found</sym> flag that Geocaching.com sets in PQ files for the requesting user's own finds, then locates your log entry to extract the exact date. Your Geocaching username (configured in Settings) is used to match the log; the numeric finder ID is learned automatically on first import and stored for faster matching in future imports.

Fixed

FTF false positives on PQ import — The First To Find flag was incorrectly set on all found caches when importing a Pocket Query. The previous detection logic checked whether the user's log was the earliest of the five logs shown in the PQ — but Geocaching.com only includes the five most recent logs, so an old find would often appear first among those five even if hundreds of people had found the cache earlier. FTF is now detected exclusively from keywords in the user's own log text (FTF, First to find, First finder, Første til at finde), which is the only reliable signal available from a standard PQ.

[1.13.1] — 2026-05-05

Added

Home location in Geocaching profile (fixes #183) — A dedicated Home location field has been added to the Geocaching profile section in Settings. This sets a permanent home coordinate that is used as the default center point for all new databases and as the ★ Home entry in the location dropdown.
User locations renamed (fixes #183) — The Home coordinates group in Settings has been renamed to User locations to better reflect its purpose. The ★ Home entry (from Geocaching profile) always appears at the top and cannot be edited or deleted from this list — it is managed exclusively via the Geocaching profile section.
Welcome dialog on first launch (fixes #183) — If username or home location is not configured, a welcome dialog is shown a few seconds after startup prompting the user to open Settings and complete the setup.

Fixed

Map centers on correct location at startup (fixes #183) — The map now starts at the active location for the current database instead of a hardcoded position in Denmark. The starting coordinates are injected directly into the Leaflet HTML before the page loads, so the correct location is visible from the very first render.
Location saved per database (fixes #183) — Switching the active location via the toolbar dropdown now correctly saves the chosen location for that specific database. Switching to a different database and back restores each database's own last-used location.
Toolbar dropdown reflects active location after DB switch (fixes #183) — The location dropdown in the toolbar now correctly updates to show the active location for the newly selected database when switching databases.
New database uses Home location as default center (fixes #183) — When creating a new database, the center point is automatically set to the Home location from the Geocaching profile. If no Home location is configured, the last active location is used as a fallback.
First cache no longer auto-selected on load — After loading or refreshing caches, the first entry in the list was automatically selected and shown on the map without any user action. The list now loads with no selection, so the map is not unintentionally panned.
test_db_manager match patterns — Four unit tests used raw translation keys as match patterns in pytest.raises(). Since tr() returns translated text, the patterns never matched and the tests always failed. Updated to match on stable substrings present in the translated messages.

[1.13.0] — 2026-05-05

Added

Dutch translation — OpenSAK is now available in Nederlands (Dutch). The translation was generated by Claude AI and has not yet been reviewed by a native speaker — feedback and corrections are welcome via GitHub issues or the Facebook group.
Last log date column (fixes #186) — A new Last log column shows the date of the most recent log entry for each cache. The column can be sorted and is populated automatically for existing databases via a migration.
Enable / disable all cache types (fixes #159) — The cache type filter now has an Enable all / Disable all toggle so you can quickly select or deselect every type at once.

Improved

Search performance (fixes #127) — Name and GC code searches are now pushed to SQL LIKE queries that exploit the existing B-tree index, making live search significantly faster on large databases. An adaptive debounce and minimum-character threshold reduce unnecessary queries while typing. Search settings (debounce delay and minimum characters) are available in the new Advanced tab in the Settings dialog.

Uh oh!

FilesExpand file tree

CHANGELOG.md

Latest commit

History

CHANGELOG.md

File metadata and controls

Changelog — OpenSAK

[1.18.0] — 2026-09-02

Added

Fixed

[1.18.0-beta.3] — 2026-08-31

Fixed

[1.18.0-beta.2] — 2026-08-30

Fixed

[1.18.0-beta.1] — 2026-08-27

Added

[1.17.2] — 2026-08-26

Fixed

Known Issues

[1.17.1] — 2026-08-19

Fixed

Added

Fixed

[1.17.0] — 2026-08-17

Added

Changed

Fixed

[1.17.0-beta.14] — 2026-08-16

Fixed

Added

[1.17.0-beta.13] — 2026-08-15

Added

Fixed

[1.17.0-beta.12] — 2026-08-14

Fixed

[1.17.0-beta.11] — 2026-08-13

Fixed

[1.17.0-beta.10] — 2026-08-12

Added

[1.17.0-beta.9] — 2026-08-12

Fixed

[1.17.0-beta.8] — 2026-08-12

Fixed

[1.17.0-beta.7] — 2026-08-11

Added

[1.17.0-beta.6] — 2026-08-11

Changed

Fixed

[1.17.0-beta.5] — 2026-08-10

[1.17.0-beta.4] — 2026-08-10

[1.16.3] — 2026-08-10

Fixed

[1.16.2] — 2026-08-10

Fixed

[v1.17.0-beta.3] - 2026-08-09

Fixed

Added

[1.17.0-beta.2] — 2026-08-02

Fixed

[1.17.0-beta.1] — 2026-07-31

Fixed

[1.16.1] — 2026-07-30

Fixed

[1.16.0] — 2026-07-29

Added

Fixed

[1.16.0-beta.16] — 2026-07-28

Fixed

Added

Notes

[1.16.0-beta.15] — 2026-07-28

Fixed

[1.16.0-beta.14] — 2026-07-26

Added

[1.16.0-beta.13] — 2026-07-26

Fixed

Added

[1.16.0-beta.12] — 2026-07-26

Fixed