= adjust: Adjust VLBI clocks for correlation E. Himwich, B. Corey, M. Titus Version 0.2 - July 2020

== Installation

. Requires pip3 (alternatively instead of the bash install command, replace the shebang line in adjust.py with what your system needs):

git clone https://github.com/whi-llc/adjust
cd adjust
bash install

. Add the directory where it was installed to your PATH variable

== Usage

For help:

adjust -h

Output: .... Usage: adjust -afhvx: file finds average offset to apply to correlator 'Used' clocks after finding average offset, use -a to print adjusted offsets file: one line per station (three fields minimum, blank delimited): two-letter station, case sensitive, e.g. Ht fmout-gps Used Additional columns ignored. output, unless -a specified: one line per station: station peculiar offset difference from historical "x" excluded or " " included residual to fit for Diff_mean Normalized residual for std Normalized residual for historical RMS station's historical offset station's historical RMS number of data points in history two lines: explanation one line: peculiar offsets version one line: summary number of stations included total number of stations average offset (Diff_mean) std output, with -a specified: one line per station: station Use flag i=included x=excluded n=new Adjusted peculiar offset File name (all the same)

Options: -a print adjusted peculiar offsets -b use broadband peculiar offsets -f sub 8 us from Ft fmout for 512 Mbps -h this text -v print version number -x stations to exclude list of two letter codes, colon separated, case sensitive -z accept zero values ....

== DRAFT:

=== Files

All data files are in the data/ sub-directory

. File types + .. Files +++*++++++_+++po+++_++++++*+++.dat are peculiar offset (PO) data + .. Files +++*++++++_+++sd+++_++++++*+++.dat are sampler delay (SD) data + .. Other files are other information and plots.

. bb files are for broadband (VGOS)

.. Version numbers are of the form vS.P where:

... S is the sampler delay (SD) version ... P is the peculiar offset (PO) version ... PO and SD files are related by the S value + P changes more quickly than S, so there can be multiple P values for a given S

.. SD files: + bb+++_+++sd+++_+++_v_.dat, where v is of the form vS: the S-th SDs + .. PO files: + bb_po_v.dat, where v is of the form vS.P: the P-th POs for the S-th SDs

. sx files are S/X + .. PO files: + sx_po_v.dat, where v is of the form vN

. The symbolic links: bb_po.dat, bb_sd.dat, and sx_po.dat point to the current versions.

=== Plans:

. Modify adjust to read the PO data from the linked file in data/. . Add -d option to select a specific PO version . Add -u option to adjust to print new final values based on adjustment

=== Longer term possibilities:

. Modify -u option to print final values as clock_early statements using clock_early statements as input. . Modify input to read fmout-gps values from clock_early statements (extension to VEX) . Add option to apply a fixed offset final values (adjusted values as well?) . Change PO files to use .py instead of .dat since they are Python code.