AFNI HISTORY

The levels of importance go from 1 to 5, with meanings:
       1 - users would not care
       2 - of little importance, though some users might care
       3 - fairly important
       4 - a big change or new program
       5 - IMPORTANT: we expect users to know

generated via the command : afni_history -html -reverse -min_level 2 -past_months 6 

  ----  log of AFNI updates (most recent first)  ----

21 Jan 2026, RC Reynolds, afni, level 2 (MINOR), type 5 (MODIFY)
    can specify whether to redraw, even on non-macos26

    Have new options -x_needsX11Redraw() : do we redraw on resize?
                     -x_have_MACOS_FORCE_EXPOSE : is the compile flag set?
    Use Y/N env var AFNI_DO_X11_REDRAW to force redrawing.  Else, default
    is still based on macos 26 and compile flag.


20 Jan 2026, RC Reynolds, afni, level 3 (MAJOR), type 4 (BUG_FIX)
    on macos 26, fully redraw any resized windows

    The major updates were from RW Cox.
    If using macos 26, resize events will have top-level windows call
    XClearArea() and XSync(), and recursively on all children.
    RW Cox rewrote AFNI graphs to continue drawing into fd_pxWind, but to
    then copy this into the windows, rather than just having it as the
    background.
    Applying these 2 changes requires building with MACOS_FORCE_EXPOSE set,
    currently done for both official mac systems.
    RW Cox also added FIX_TOPFORM_HEIGHT to control window sizing in Linux.
    These updates were also done with D Glen, who is now working on suma.


08 Jan 2026, RC Reynolds, install_bootcamp_data.tcsh, level 3 (MAJOR), type 1 (NEW_PROG)
    new program to install the AFNI bootcamp data

    Download and install all packages in the checksum file.


07 Jan 2026, RC Reynolds, 1d_tool.py, level 2 (MINOR), type 2 (NEW_OPT)
    add -show_xmat_warnings

    Warn on all-zero regressors or ones with small magnitudes.


24 Sep 2025, DR Glen, afni, level 2 (MINOR), type 4 (BUG_FIX)
    Clusterize plugin wami command fix

    Temporary dataset deleted before getting name for whereami overlap omask.


18 Sep 2025, DR Glen, afni, level 2 (MINOR), type 6 (ENHANCE)
    whereami functionality uses new Julich v3.1 brain atlases

    Julich v3.1 atlases for MNI 2009c and N27 spaces are available.
    Default atlas is now Julich_MNI2009c_v3.1 for Atlas Colors, Goto location.


12 Sep 2025, DR Glen, afni, level 2 (MINOR), type 6 (ENHANCE)
    afni GUI can now sort datasets in chooser alphabetically

    Previously, sessions were sorted by anatomical, func, then NIFTI.
    Now the user has options to set alphabetic, reverse alphabetic,
    mri type (data type including NIFTI) or no re-sorting (original 
    mri type then NIFTI) using AFNI environment variables.


11 Sep 2025, RC Reynolds, afni_util.py, level 2 (MINOR), type 2 (NEW_OPT)
    add get_def_blur_from_dims() to report a hopefully useful blur size

    Set gmean = geometric mean(deltas), scale by 1.6 and take the ceiling,
    truncating to 4 significant bits.  So return:
    
        truncate_to_N_bits(gmean, bits=4, method='ceil', scale=1.6)

11 Sep 2025, RC Reynolds, afni_proc.py, level 2 (MINOR), type 5 (MODIFY)
    minor updates

    - suggest a new default blur (but still use 4.0), if none is given
    - suggest open_apqc.py instead of afni_open
    - allow verb only to show tracked files and create proc script
    - add example class do_21, to match that in AFNI_data7


10 Sep 2025, RC Reynolds, ap_run_simple_rest_me.tcsh, level 2 (MINOR), type 2 (NEW_OPT)
    add -blur_size; have default come from voxel sizes

    The default is now 1.1 times the geometric mean dim, ceiling rounded.

10 Sep 2025, RC Reynolds, ap_run_simple_rest.tcsh, level 2 (MINOR), type 2 (NEW_OPT)
    add -blur_size; have default come from voxel sizes

    The default is now 1.6 times the geometric mean dim, ceiling rounded.


08 Sep 2025, RC Reynolds, afni_proc.py, level 2 (MINOR), type 2 (NEW_OPT)
    add -regress_per_run_ortvec, to process physio_calc.py volbase regs

    The main outputs of physio_calc.py are *slibase.1D (for slice-based
    regressors) and *volbase.1D (for spatially global regressors).
    Keep using the -ricor options for *slibase.1D, but pass the *volbase.1D
    files using -regress_per_run_ortvec, with a label and one file per run.
    E.g.  -regress_per_run_ortvec Vphys resp_r1_volbase.1D resp_r2_volbase.1D


28 Aug 2025, RC Reynolds, gen_group_command.py, level 2 (MINOR), type 2 (NEW_OPT)
    try to detect BIDS; add option -sid_method

    This will also be incorporated in gen_ss_review_table.py.


18 Aug 2025, RC Reynolds, afni_proc.py, level 2 (MINOR), type 2 (NEW_OPT)
    add -volreg_warp_master_box

    This is like -volreg_warp_master, but dxyz will come from EPI.
    If -volreg_warp_dxyz is given, _master is the same as _master_box.


15 Aug 2025, RC Reynolds, build_afni.py, level 2 (MINOR), type 2 (NEW_OPT)
    add -fast_log_commands and -fast_log_messages

15 Aug 2025, RC Reynolds, @compute_OC_weights, level 2 (MINOR), type 2 (NEW_OPT)
    add -prefix_combine, to combine the echoes using computed weights

    Added for P Molfese.


06 Aug 2025, RC Reynolds, afni_proc.py, level 2 (MINOR), type 6 (ENHANCE)
    add basic GTKYD (getting to know your data) and outlier check

    This will grow, including with user control.


29 Jul 2025, RC Reynolds, timing_tool.py, level 2 (MINOR), type 2 (NEW_OPT)
    add option -force_write_type, to force output of simple, AM, DM or AM/DM


25 Jul 2025, RC Reynolds, afni-general, level 2 (MINOR), type 5 (MODIFY)
    gcc-15 : code updates under the SUMA tree

    This hopefully finalizes the June 12, 2025 update, including SUMA.