This patch adds a function to read and parse a SCPI response which contains a
comma-separated list of floating-point numbers (e.g. "1.0e-5,2.0e-4,3.0e-3").
This is particularly useful if the instrument sends analog measurement
data in this format.
The SCPI standard specifies the "*OPC?" command (Operation complete query) which
queries the instrument for its operative state. When all pending operations are
complete, the instrument responds with a "1".
Some manufacturers block before completing all operations and don't respond
with anything and some of them respond with a "0". This function handles both
cases uniformly.
This patch adds helper functions to read an SCPI response and parse the response
as an integer, boolean, floating-point or double-precision floating-point number.
The Standard Commands for Programmable Instruments (SCPI) defines a standard
for syntax and commands to use in controlling programmable test and measurement
devices.
SCPI documentation:
http://www.ivifoundation.org/docs/scpi-99.pdf
This patch adds helper functions for sending SCPI commands, reading a SCPI
response and reading and parsing a SCPI "*IDN?" response.
This patch adds helper functions for converting a string to different number
formats (double, long, float, int).
These functions are exposed in the public API.
SR_CONF_PATTERN_MODE was not handled in config_list(), yielding
non-working OLS support in PulseView (due to an assert), and a missing
pattern list in sigrok-cli's --show output.
This fixes bug #184.
The fields of this structure should not be used directly by frontends
(and none of the current ones do). Thus, make the struct opaque and hide
its contents from the API.
The text output module keeps buffers for internal state, upon receiving a DF_END
packet it frees the internal context but the buffers are never freed.
This adds a text_cleanup() helper function and registers it as the cleanup
function within all the text output modules.
The scan() function was opening the port in non-blocking mode, the dev_open()
function however was not using the SERIAL_NONBLOCK flag. This led to hangs
in certain situations.
This fixes the OLS e.g. on NetBSD.
A new sp_port is created every time we call serial_open (sp_get_port_by_name
implicitly creates one for us), so free it every time we call serial_close.
Without a SR_DF_END samples could be cached in the internal buffer of an output
module and never flushed, therefore they would be missing in the final output.
By sending a SR_DF_END packet we force the output to be flushed.
Disable drivers that need serial port support if libserialport is not found.
Also, disable building various other serial port related code in that case.
poljar (Damir Jelić)
poljar (Damir Jelić)
Dan Horák
Uwe Hermann
Martin Ling
Bert Vermeulen
Matthias Heidbrink