improve documentation of the histedit 'mesg' command prompted by a question by op@

tog: only request commits when child hsplit increases Fix bug introduced in 3c1dfe12b3 that fails to properly populate child log views due to incorrect request_log_commits() calls: (1) when increasing the bottom hsplit in a ref/log splitscreen; and (2) when reopening a child log view after closing a resized child log view: $ TOG_VIEW_SPLIT_MODE=h tog ref return # open log view in bottom split 4+ # increase log (child/bottom) split *new log lines are not populated* q # close log view return *commits are not loaded* ok stsp@

tog: adjust view line offset when resizing hsplit Squish bug that can move the selection cursor offscreen when resizing horizontal splits due to bogus offset: $ TOG_VIEW_SPLIT_MODE=h tog # 80x24 22j return # open diff view in a hsplit tab # focus log (top) split 10+ # increase top split by 10 lines 22j return # open diff view in a hsplit F # toggle fullscreen diff view tab # focus log (parent) view in fullscreen *selection cursor will be off the bottom of the screen* ok stsp@

tog: enable moving to prev/next blame line in diff view Blame/diff view counterpart of log/diff <,.> key maps to traverse commits corresponding to each line of the annotated file from the diff view. Includes fix from stsp@ to work while the file is still annotating. ok stsp@

tog: make 'S' switch split mode irrespective of the view Always switch internal split state and redraw view according to the constraints of the new mode. suggested by and ok stsp@

make 'tog log' error out in shallow Git repositories instead of hanging ok op@

tog: implement +/- keymaps to resize the focussed split Increase/decrease the currently active split. New dimensions persist for the tog session, irrespective of subsequent closures and new child views. suggested by and ok stsp@

tog: implement global 'S' key map to switch split mode If in a vertical (horizontal) split, switch to horizontal (vertical), and make new split mode persist. ok op@

use got_privsep_flush_imsg instead of reinventing it ok stsp@

get rid of unneeded recv_imsg_error calls got_privsep_recv_imsg users don't need to bother to look whether the imsg type is GOT_IMSG_ERROR because got_privsep_recv_imsg already takes care of that and turns GOT_IMSG_ERROR into `struct got_error's automatically. ok stsp@

rework got_privesp_recv_tree to use got_privsep_recv_imsg ok stsp@

fix buf_alloca error handling Left-over from how OpenRCS/OpenCVS do error handling. OK stsp@

regress: use bs=1M for dd, not bs=1m; fixes -portable test failure on ubuntu GNU dd only accepts upper-case letters for the size units.

delta_cache.c needs to include <stdint.h>; unbreaks -portable build on ubuntu

add signer_id option to got.conf(5) Setting this option will cause 'got tag' to sign all created tags using the SSH key, unless overridden by the -s flag. ok stsp@

mention Austin Appleby's public domain code for completeness

mention additional copyright holders in got(1)

add two more authors how I missed during my sweep of commit history

Xr gotweb(8) from got(1)

typo

long overdue update of AUTHORS lists in man pages; based on commit history

regress test SSH key revocations ok stsp@

move 'got init' command to 'gotadmin init' This functionality is better suited for gotadmin because it is technically a server-side repository operation when we consider the 'got' tool as the client-side tool. I have plans to move 'got import' into gotadmin as well. ok op@

add missing revoked_signers grammar to got.conf(5) ok stsp@

fix potential type mismatch between format specifier and argument Cast printf argument of type time_t to long long to match the %lld format specifier on platforms where this might not be the case.