summaryrefslogtreecommitdiff
path: root/src/lib
AgeCommit message (Collapse)Author
2025-02-16White space: kdm_with_metadata.{cc,h} send_kdm_email_job.{cc,h}Carl Hetherington
2025-02-12pot/merge.Carl Hetherington
2025-02-11Fix/hack build to work again on Ubuntu 1{6,8}.04Carl Hetherington
2025-02-11Further fixes to cs_CZ translation.Carl Hetherington
2025-02-11Fix up i18n after analytics message change.Carl Hetherington
2025-02-11Replace compose with fmt.Carl Hetherington
2025-02-11Remove happy clappy language.Carl Hetherington
2025-02-11Replace compose with fmt.Carl Hetherington
2025-02-11Use a default constructor.Carl Hetherington
2025-02-11Revert "Add stubs for Georgian (ka_KA) language support."Carl Hetherington
This reverts commit b5cecdd0ee7c75388d95e2f987538fe88c23893a. I added this for somebody but no apparently translations were ever forthcoming, so remove it again.
2025-02-10Apply the same fix for DKDM recipients as we previously had for cinemas.Carl Hetherington
Otherwise the same problem happens with dkdm_recipients.xml - it is moved to sqlite3 but the configuration still has the XML path, causing a crash the next time we try to access the DKDM recipients list.
2025-02-09Remove believed pointless pre-storage of file state.Carl Hetherington
I think I previously misunderstood what was going on here and thought the problem was that doing set_cinemas_file() created a sqlite3 file. I now think that the problem is rather that we call Config::instance() early in dcpomatic.cc, then ::drop(). This creates the sqlite3 file, sets its path in the Config object but then deletes the object without saving it. When we create the Config object "for real" we don't reset the path from XML to sqlite3. The part of the previous fix that unconditionally sets the config path seems to be the right one.
2025-02-09Add --audio-channels option to dcpomatic2_create (#2966).Carl Hetherington
2025-02-09Fix out-of-date help.Carl Hetherington
2025-02-03Fix subtitles being written with negative times when trimmed (#2965).Carl Hetherington
2025-02-03Cleanup: rename subtitle_stop -> text_stop.Carl Hetherington
2025-02-02pot/merge.Carl Hetherington
2025-02-02White space: config.{cc,h} player.{cc,h} about_dialog.{cc,h} hints_dialog.{cc,h}Carl Hetherington
2025-01-25White space: film.{cc,h} font.{cc,h}Carl Hetherington
2025-01-25White space: kdm_cli.{cc,h} usl.{cc,h} zipper.{cc,h}Carl Hetherington
2025-01-25White space: audio_delay.{cc,h} audio_point.{cc,h} rng.{cc,h} atmos_test.ccCarl Hetherington
2025-01-25White space: filter.{cc,h}, hints.{cc,h}, gain_calculator_dialog.{cc,h}Carl Hetherington
2025-01-24Updated ru_RU translation from Евгений Дац.Carl Hetherington
2025-01-22Fix typo (#2956).v2.18.6Carl Hetherington
2025-01-22Sort cinemas and DKDM recipients correctly using the collator (#2950).Carl Hetherington
2025-01-22Move sqlite database open/close to a new SQLiteDatabase.Carl Hetherington
2025-01-22Announce changes to cinemas file separately.Carl Hetherington
2025-01-22Only read screen certificates when we need them.Carl Hetherington
2025-01-22Fix updating of screens with new/different trusted devices.Carl Hetherington
2025-01-22Fix adding of trusted devices.Carl Hetherington
2025-01-22Allow KDMRecipient and hence Screen to convert certs to dcp::Certificate lazily.Carl Hetherington
This is pretty slow (as it runs the certificate through OpenSSL) and we don't need to do it for every certificate in a database when we load the database.
2025-01-22Add screens() to get all screens.Carl Hetherington
2025-01-21White space: cross{_osx,_windows,_common}.cc cross.hCarl Hetherington
2025-01-21White space: analyse_audio_job.{cc,h}Carl Hetherington
2025-01-21Cleanup: sorting.Carl Hetherington
2025-01-21White space: verify_dcp_job.{cc,h}Carl Hetherington
2025-01-21Cleanup: default constructor.Carl Hetherington
2025-01-20White space: util.{cc,h}Carl Hetherington
2025-01-20Fix typo in comment.Carl Hetherington
2025-01-20Fix typo accomodate -> accommodate.Carl Hetherington
2025-01-20Remove the strange tolerant flag in Film.Carl Hetherington
Instead, we now pass it in directly to Content::examine() and the Player.
2025-01-20Pass tolerant flag into Player.Carl Hetherington
This is needed because the DCPDecoder ends up needing it when it reads reel information from CPLs. Maybe all that stuff should be in our metadata.
2025-01-20Pass tolerant flag into Content::examine() and the ExamineContentJob.Carl Hetherington
Along the way this helps to fix #2942.
2025-01-19Find missing files better in projects coming from other platforms (#2935).Carl Hetherington
2025-01-19Add search by digest if search by name fails (#2935).Carl Hetherington
2025-01-19Rename search -> search_by_name.Carl Hetherington
2025-01-19const correctness.Carl Hetherington
2025-01-19Add option to force short screen layout (#2946).Carl Hetherington
2025-01-18Fix first frame being shown even when trimmed (#2952).Carl Hetherington
We calculated the DCP time of the first frame (when its content was trimmed) and it was clamped from -8 frames to 0, which meant it was used when it should not have been.
2025-01-18Add comment.Carl Hetherington