libimobiledevice - A protocol library to access iOS devices on Linux

Age	Commit message (Collapse)	Author	Files	Lines
2014-03-21	idevice: Fix misleading warning in GnuTLS SSL handshake	Martin Szulecki	1	-2/+2

2014-03-21	idevice: Plug memory leak by freeing pair_record	Martin Szulecki	1	-0/+6

2014-03-21	userpref: merge pair_record_generate_from_device_public_key() into ↵	Nikias Bassen	1	-15/+1
	pair_record_generate_keys_and_certs()
2014-03-21	idevice: Use correct import helper for keys and certs	Martin Szulecki	1	-3/+3

2014-03-21	lockdown: Make sure failed pairing validation works with correct error code	Martin Szulecki	1	-1/+1

2014-03-21	lockdown: Plug memory leak due to missing free of public_key	Martin Szulecki	1	-0/+3

2014-03-21	lockdown: Make sure to set SystemBUID in generated pair records	Martin Szulecki	1	-1/+2

2014-03-21	lockdown: Fix wrong pointer dereference of pair_record plist	Martin Szulecki	1	-3/+3

2014-03-21	lockdown: Remove obsolete debug output	Martin Szulecki	1	-6/+0

2014-03-21	Refactor pair record handling to use new usbmuxd pair record interface	Martin Szulecki	3	-436/+130
	This refactoring is mandatory as libimobiledevice should not interact with the pair record configuration directory which is owned by the usbmuxd user. This change also adds compatibility for the native usbmuxd and thus pair records saved by iTunes.
2014-03-20	Bump dependency to libplist 1.11 and remove use of "plist_dict_insert_item()"	Martin Szulecki	15	-127/+127

2014-03-18	lockdown: make sure type is always free'd in lockdownd_client_new_with_handshake	Nikias Bassen	1	-2/+2

2014-03-13	afc: use static buffers where possible	Nikias Bassen	2	-33/+30

2014-03-13	afc: refactor afc_dispatch_packet and improve afc_file_write performance	Nikias Bassen	1	-252/+109

2014-03-13	idevice: call SSL_shutdown a second time if required	Nikias Bassen	1	-1/+4
	This will fix that: lockdownd[25] <Notice>: 00484000 _receive_message: Could not receive size of message, expected 4 bytes, got -1 bytes: (54, Connection reset by peer)
2014-03-12	file_relay: Added file_relay_request_sources_timeout()	Aaron Burghardt	1	-2/+39
	Signed-off-by: Martin Szulecki <m.szulecki@libimobiledevice.org>
2014-03-10	lockdown: disable SSL in lockdownd_stop_session()	Nikias Bassen	1	-0/+5

2014-03-10	lockdown: removed outdated call to lockdownd_goodbye() in lockdown_client_free()	Nikias Bassen	1	-4/+0

2014-02-18	win32: link against ws2_32.dll	Nikias Bassen	1	-1/+1

2014-02-09	mobile_image_mounter: implemented new image file uploading method	Nikias Bassen	1	-0/+122

2014-01-25	Revert "lockdown: do not send Goodbye when SSL session was active"	Nikias Bassen	1	-2/+2
	This reverts commit 56392428a49543bab4b12431451ec04bb799c411. Actually I committed this because of usbmuxd complaining about a TCP RST. But lockdownd is now complaining since it waits for the Goodbye packet (which I apparently didn't see during testing).
2014-01-15	lockdown: do not send Goodbye when SSL session was active	Nikias Bassen	1	-2/+2

2014-01-09	idevice: free ssl object when SSL handshake fails	Nikias Bassen	1	-0/+1

2014-01-09	idevice: don't free ssl bio after SSL_set_bio has been called	Nikias Bassen	2	-3/+0

2013-11-30	lockdown: Make sure to free client->session_id for sure	Martin Szulecki	1	-0/+4

2013-11-30	Fix possible memory corruption by ensuring client pointers NULL'd after free	Martin Szulecki	4	-0/+9

2013-11-30	lockdown: Do not free session_id again after stop_session already did	Martin Szulecki	1	-1/+0

2013-11-20	idevice: properly handle partial SSL_read()s	Nikias Bassen	1	-1/+9

2013-11-10	installation_proxy: Fix operations exiting before being finished due to timeout	Martin Szulecki	1	-4/+5
	The internal status callback used a timeout of 30 seconds to receive status messages about the progress of an operation. However, slow devices or large app archives trigger this timeout causing the handler to return before the operation was actually complete. This fixes it by removing the internal timeout by waiting forever and only returning early in case of a real error.
2013-11-10	property_list_service: Rename RECEIVE_TIMEOUT error for more clarity	Martin Szulecki	4	-3/+5

2013-11-08	lockdown: fix possible invalid free in GnuTLS cert generation code	Nikias Bassen	1	-3/+3

2013-11-04	installation_proxy: Remove dead code from uninstall method	Martin Szulecki	1	-5/+0

2013-11-04	Convert stray spaces to corresponding tabs	Martin Szulecki	6	-96/+97

2013-10-09	Remove duplicate newline from debug messages as one is added automatically	Martin Szulecki	6	-8/+8

2013-10-09	lockdown: Plug few possible memory leaks in generate_pair_record_plist()	Martin Szulecki	1	-16/+6

2013-10-09	property_list_service: improve error checking on receive	Nikias Bassen	1	-2/+19

2013-09-24	restore: improve restored_query_type result checking	Nikias Bassen	1	-11/+14

2013-09-24	lockdown: improve lockdownd_query_type result checking	Nikias Bassen	1	-4/+11

2013-09-23	lockdown: Remove EscrowBag feature for now as it caused services to not start	Martin Szulecki	1	-7/+0

2013-09-23	Avoid versioning shared library for Windows target	Martin Szulecki	1	-0/+1

2013-09-23	Correctly refer to libpthread_LIBS	Martin Szulecki	1	-1/+1

2013-09-21	lockdown: Only sent EscrowBag if we are running in a secure session	Martin Szulecki	1	-2/+6

2013-09-21	idevicebackup2: Add experimental support for using backups in the cloud	Martin Szulecki	1	-3/+9

2013-09-21	lockdown: Send EscrowBag when starting service for automatic unlocking	Martin Szulecki	1	-0/+3

2013-09-19	lockdown: Remove pairing code which is obsoleted by usbmuxd's preflight handler	Martin Szulecki	1	-53/+0

2013-09-19	lockdown: Don't shutdown SSL during StopSession already or device complains	Martin Szulecki	1	-7/+6

2013-09-19	notification_proxy: handle error conditions in notification polling thread	Nikias Bassen	1	-4/+14

2013-09-19	property_list_service: handle timeout condition on receive	Nikias Bassen	2	-1/+5
	this adds the new error code PROPERTY_LIST_SERVICE_E_TIMEOUT that occurs if the given timeout is reached when trying to receive data.
2013-09-17	lockdownd: remove unused udid variable in reconnect function	Nikias Bassen	1	-4/+0

2013-09-17	lockdown: Refactor pair record generation/loading logic a bit for efficiency	Martin Szulecki	1	-42/+47
	The current code always requests the public key and set's the SystemBUID, even if a pair record already exists which has these values ready. The change moves logic to only do that during generation of new certificates.