summaryrefslogtreecommitdiffstats
path: root/src/lockdown.c
AgeCommit message (Collapse)AuthorFilesLines
2013-09-17Refactor userpref logic to use plist format and implement trust dialog handlingGravatar Martin Szulecki1-47/+210
iOS 7 introduced a new pairing workflow which increases security by showing a trust dialog to the user before pairing with the host is allowed. The userpref system was refactored to use the native plist format, too. Configuration files of the native implementations are used on each platform. Former configuration files are no longer in use and can be deleted.
2013-06-30Fixed memory leak in lockdownd_gen_pair_cert() by freeing X509_EXTENSION ↵Gravatar Aaron Burghardt1-0/+1
object when using OpenSSL.
2013-06-30Fixed memory leak in lockdownd_gen_pair_cert() when getting cert data using ↵Gravatar Aaron Burghardt1-2/+5
OpenSSL.
2013-04-25common: Move debug and userpref code into libinternalcommonGravatar Martin Szulecki1-2/+2
2013-03-14lockdown: only return service descriptor if StartService succeedsGravatar Nikias Bassen1-5/+11
2013-02-26Refactor port number use into service descriptor to enable SSL for servicesGravatar Martin Szulecki1-12/+46
This is a major change which breaks API but is required in order to support SSL communication for services as used by network connections.
2012-04-01lockdown: Correct a few wrong comments showing NP_E_* error messagesGravatar Martin Szulecki1-7/+9
2012-03-22Mass replace UUID by UDID, which is the correct term for itGravatar Martin Szulecki1-12/+12
2012-03-19Fix broken debug_info usageGravatar Nikias Bassen1-2/+2
2012-03-19free bio after using itGravatar Nikias Bassen1-1/+3
2012-03-19Add OpenSSL supportGravatar Nikias Bassen1-10/+154
2012-03-19Refined asprintf/vasprintf detection and inclusionGravatar Nikias Bassen1-2/+0
2012-03-19Fix building on OSXGravatar Martin Szulecki1-0/+2
2012-03-19Add asprintf replacement for systems lacking itGravatar Nikias Bassen1-0/+1
2012-03-18Completely remove glib dependency.Gravatar Nikias Bassen1-3/+2
2011-11-21lockdown: iOS 5: handle 'Error' key in lockdown_check_resultGravatar Nikias Bassen1-1/+15
2011-06-09lockdown: move writing of device uuid to client struct inside ↵Gravatar Nikias Bassen1-7/+6
lockdownd_client_new() ... instead of in lockdownd_client_new_with_handshake(). This way the device UUID is also available when only using a non-paired connection (e.g. when using the idevicepair tool).
2011-06-07lockdown: fix support for iOS 5Gravatar Nikias Bassen1-1/+3
2011-03-31debug: Use explicit size for string when debugging non-null terminatedGravatar Martin Szulecki1-1/+1
The debug code to output the device public key was not taking into account that the key data is not null terminated.
2010-11-27Add a safety check in lockdownd_client_new_with_handshake() to fix a ↵Gravatar Martin Szulecki1-1/+4
possible crash
2010-06-04lockdown: correctly handle activation_record parameter in lockdownd_activateGravatar Nikias Bassen1-1/+1
The activation_record was inserted into the request plist and freed at the end. If the user app then uses plist_free to free the activation record, it gets an invalid free. By using plist_copy the activation_record is not touched.
2010-05-28Add special handling for unpair request; probably a bug in the iPhone OSGravatar Martin Szulecki1-4/+13
2010-05-28Refactor lockdownd_client_new() internalsGravatar Martin Szulecki1-11/+3
2010-05-25Document that lockdownd automatically drops the connection when idleGravatar Martin Szulecki1-0/+7
2010-05-15Use a more general implementation to get data classes the device supportsGravatar Martin Szulecki1-45/+14
Some devices appear to not have anything set in com.apple.mobile.tethered_sync. Thus we use the more general com.apple.mobile.iTunes domain to query which seems to be supported on all tested devices.
2010-05-15Add documentation for lockdownd_get_sync_data_classes/data_classes_free()Gravatar Martin Szulecki1-0/+22
2010-05-13Plug some memory leaks in lockdownd due to missing free() callsGravatar Bryan Forbes1-1/+4
2010-05-13Add lockdownd_get_data_sync_classes() helper for mobilesync logicGravatar Bryan Forbes1-0/+111
2010-04-11Fix memory leak and actually set the label supplied for new lockdown clientsGravatar Martin Szulecki1-1/+1
2010-04-11plug memory leak in lockdownd_gen_pair_cert()Gravatar Petr Uzel1-0/+3
2010-04-11plug memory leak in lockdownd_gen_pair_cert()Gravatar Petr Uzel1-0/+10
2010-04-11plug memory leaks in generate_pair_record_plist()Gravatar Petr Uzel1-0/+7
2010-03-24Remove unused and fix some includesGravatar Nikias Bassen1-2/+0
2010-03-16Complete documentation of public interface and fix a lot of bogus commentsGravatar Martin Szulecki1-79/+168
This change unifies the documentation comment syntax, fixes a few bad documentation comments and completes documentation where it was missing.
2010-03-16Rename the *_int to *_private for better semantic and fix public typedefsGravatar Martin Szulecki1-1/+1
This changes the typedef logic slightly for the public headers in order for doxygen to correctly pickup and enable the client struct documentation.
2010-01-29Add LOCKDOWN_E_INVALID_ACTIVATION_RECORD if activation errors outGravatar Martin Szulecki1-0/+12
2010-01-29Global renames due to project rename to libimobiledeviceGravatar Nikias Bassen1-14/+14
2010-01-26Auto-pair devices within lockdownd_client_new_with_handshake()Gravatar Martin Szulecki1-3/+16
This brings back the automatic pairing feature if not yet paired. [#111 state:resolved]
2010-01-22Rename all *_recv functions to *_receive to match underlying APIGravatar Martin Szulecki1-13/+13
2010-01-21Add LOCKDOWN_E_INVALID_SERVICE error and functionalityGravatar Martin Szulecki1-2/+11
This error is returned if the service does not exist on the device. It comes handy to detect if for instance the AFC2 service is actually setup on a device or not.
2010-01-13lockdown/tools/python: use correct type for port parameterGravatar Nikias Bassen1-2/+2
2010-01-13Refactor pairing to allow implementations to fully handle pairingGravatar Martin Szulecki1-56/+108
Implementations can now supply a full pair record and thus use their own preferred method of managing pairing records if needed.
2010-01-13lockdown: Use more correct error code when SSL startup failsGravatar Nikias Bassen1-0/+3
2010-01-13lockdown/property_list_service: use new SSL codeGravatar Nikias Bassen1-196/+14
2010-01-12Refactor and unify internal debug system for ease of use and verbosityGravatar Martin Szulecki1-53/+53
This introduces a new debug_info macro which automatically prints the calling function, file and line number information instead of having that information passed to every old log_debug_msg call.
2010-01-12Remove support for debug masks due to inconsistent useGravatar Martin Szulecki1-41/+41
2010-01-12Rename utils to debug as it does exactly that, contain debug codeGravatar Martin Szulecki1-1/+1
2010-01-12Refactor lockdown session handling and expose it in public APIGravatar Martin Szulecki1-216/+218
This splits out SSL code and allows implementations to manually handle session start and stop if needed. Also brings the API closer to the lockdown request protocol.
2010-01-12Introduce error code for lockdownd requests which require a running sessionGravatar Martin Szulecki1-2/+8
2010-01-12Allow lockdown client creation without performing full handshakeGravatar Martin Szulecki1-4/+36
The lockdown constructor was doing more than needed. Pairing and session negotiation is now handled by lockdownd_client_new_with_handshake().