summaryrefslogtreecommitdiff
path: root/controller.c
AgeCommit message (Collapse)Author
2012-02-11Rename 'void *' callback parameters to 'user_data'.Uwe Hermann
This is better, both for consistency reasons, and also because 'data' is a bit too generic and might be confusing, especially since we have other 'data' fields in the code in various places.
2012-02-11srd: Every callback should have a 'void *' parameter.Uwe Hermann
This is required in various situations and for various lib users, especially so GUIs which might want to pass specific info via this mechanism.
2012-02-11srd: rename all instance to instBert Vermeulen
2012-02-10srd: s/python/Python/.Uwe Hermann
2012-02-10srd: Add/improve g_malloc() error messages.Uwe Hermann
2012-02-10srd: Add/use SRD_API/SRD_PRIV macros.Uwe Hermann
This is not yet finished, more things should be made private.
2012-02-10srd: Cosmetics, whitespace, coding-style fixes.Uwe Hermann
2012-02-02srd: Add explicit srd_pd_output_callback_t typedef.Uwe Hermann
2012-02-01srd: rename extra_probes to optional_probesBert Vermeulen
2012-01-31srd: free all decoder instances when unloading decodersBert Vermeulen
2012-01-28srd: Drop unneeded check, g_free() handles NULL fine.Uwe Hermann
2012-01-27srd: add more debugging around probe mappingBert Vermeulen
2012-01-26srd: Remove manual log domain ("srd: ") prefixes.Uwe Hermann
2012-01-25srd: properly recurse into PD instances when looking for one.Bert Vermeulen
This fixes the confusion between parallel and linear stacks.
2012-01-24srd: change struct srd_pd_output to have a path to the DI, not the decoder.Bert Vermeulen
2012-01-23srd: Prepend (not append) PD dirs to sys.path.Uwe Hermann
2012-01-23srd: deal with invalid probe specifications betterBert Vermeulen
2012-01-23srd: use new exception system everywhereBert Vermeulen
2012-01-23srd: minor code cleanupBert Vermeulen
2012-01-22srd: make all debugging and error reporting uniformBert Vermeulen
2012-01-21srd: better check for PDs with no defined probesBert Vermeulen
2012-01-21srd: support for mapping probesBert Vermeulen
2012-01-21srd: decoder class structure check belongs in module loaderBert Vermeulen
2012-01-19accept only supported types in decoder class default option valuesBert Vermeulen
2012-01-19srd: Windows/MinGW Python path format fixes.Uwe Hermann
On Windows/MinGW, Python's sys.path needs entries of the form 'C:\\foo\\bar' instead of '/foo/bar'.
2012-01-18srd: always set default options, regardless of overridesBert Vermeulen
2012-01-17SRD: support setting per-PD optionsBert Vermeulen
All decoder class options are automatically copied to a newly created instance with their default value, which can be overridden with the GHashTable passed to srd_instance_new(). Currently, only strings and integers are supported as option value types. The type is set by the default value in the decoder class, and enforced when overriding them. Integers can be specified in any format python allows: 10, 0x0a and so on.
2012-01-15srd: PDs now get a logic feed with sample numbers, not time/durationBert Vermeulen
2012-01-15srd: clean up module loading/unloading, and the decoder structBert Vermeulen
PDs are now checked for a proper Decoder object, with at least the required attributes. The author, long_desc and func attributes in the decoder object are gone.
2012-01-10srd: Finish consistency rename to ANN/PROTO.Uwe Hermann
2012-01-10srd: Minor doxygen cosmetics.Uwe Hermann
2012-01-10Stacked protocol decoders implementation.Bert Vermeulen
The DDC decoder takes input from the I2C PD.
2012-01-09code cleanupBert Vermeulen
2012-01-09srd: change output_new() API call to add()Bert Vermeulen
2012-01-08pass PD output to the calling frontend, simple annotation viewer in CLI.Bert Vermeulen
2012-01-07convert data coming in from a PD to C structsBert Vermeulen
This is in preparation for passing annotation data back to the calling frontend, and python data up to the next protocol in the stack.
2012-01-05new srd_logic type implementation for PDs to iterate over.Bert Vermeulen
2012-01-03sr/srd: Fix left-over #includes.Uwe Hermann
The libs themselves should use #include "sigrok.h" etc., while the frontends must use #include <sigrok.h> and so on.
2011-12-28Fix utf8 encoding error by changing the Python parameter to a bytes() object ↵Kristoffer Sjöberg
(python3 regression).
2011-12-28python 3 portBert Vermeulen
2011-12-28make time/duration work, at least when loading from a session fileBert Vermeulen
PD decode() call now takes 3 arguments: timeoffset, duration, data as per the current API specification.
2011-12-28new dynamic output stream registration code, not finished.Bert Vermeulen
2011-12-22srd: SRD_ERR_ARGS -> SRD_ERR_ARG.Uwe Hermann
This is done to be consistent with libsigrok's SR_ERR_ARG (nicer that way).
2011-12-05srd: Initialise struct members by name. Minor other fixes.Gareth McMullin
Restored some prototypes in sigrokdecode.h. Abort sigrok-cli on error while decoding (includes KeyboardInterrupt). Fixed passing metadata to Decoder.start() method.
2011-12-04refactored PD framework, now using new sigrok.Decoder objectBert Vermeulen
This uses the new python unified type/class object API to construct an object for PDs to subclass. The sigrok.Decoder class has a method put() which is implemented as a C function, and receives the PD's object instance as its first parameter.