If we're deleting a character or operating in overtype
mode, we must account for the potential of 'invisible'
characters. When one follows any character about to be
deleted or replaced both multi-byte sequences must go.
Without this change, there exists the possibility that
top might report some error where no error is apparent
to the user. For example, with 'other filtering' (o/O)
the user could see "unrecognized field name 'COMMAND'"
where the quoted column name appears perfectly normal.
Or maybe a sequences like the 'combining acute accent'
gets applied to an existing character instead of being
deleted as one expects when its parent was eliminated.
So, henceforth whenever any character is being deleted
we will now check for a following 'invisible' sequence
then eliminate it along with that preceding character.
[ admittedly, these scenarios are very rare yet they ]
[ may occur, especially when recalling some previous ]
[ multi-byte strings for editing. and, since we will ]
[ be interacting with a user, performance won't be a ]
[ factor so extra checks for a zero wcwidth is fine. ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
This commit just provides the final protection against
possible screen corruption when processing line input.
[ such corruption was limited to the input line only ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
With the library change calculating 'MEMINFO_MEM_USED'
top must be tweaked in order to retain the distinction
between non-cached used memory and cached used memory.
[ assuming one of the two graphs are being displayed ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
Some kdeinit tasks have a large environment consisting
mostly of nulls which were then followed by one or two
printable characters. Such strange environments should
not be shown with that 'not applicable' (n/a) notation
even though that first string vector is equal to '\0'.
I thought I had covered such a contingency but, due to
a misplaced right parenthesis, that '^N' bottom window
could see 'n/a' + a bunch of spaces + printable stuff.
Well, that won't happen anymore with this tiny change.
Signed-off-by: Jim Warner <james.warner@comcast.net>
This program was well equipped to properly handle utf8
multi-byte characters - except for one important area!
If users typed any unicode character (shift+ctrl+u) or
pasted a utf-8 multi-byte string as a response to some
input prompt, those characters would simply be ignored
since they would not pass the internal 'isprint' test.
Well, now we can handle such data while preserving all
line editing provisions such as insertions, deletions,
destructive backspace, prior line recall (up/down) and
those all important cursor left plus right arrow keys.
[ we even support overtype mode for multi-byte stuff ]
[ even though our gui emulator will not let us alter ]
[ the cursor as confirmation (as we do at a console) ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
Given that we won't always be able to alter the cursor
shapes (from underscore to block) if in input overtype
mode, this commit will at least provide a visual clue.
[ while this libvte quirk will impact gnome-terminal ]
[ and likely others, we're able to change the cursor ]
[ shape from underscore to block at a linux console. ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
In the commit shown below the bottom window was forced
off if a full screen replacement function was invoked.
It did so by setting Frames_signal after calling those
routines from the keys_global function. However, there
was sometimes a possibility such action was premature.
At least two of those full screen replacement routines
may issue an error message & return without corrupting
the screen. As such, forcing off that bottom window is
totally unnecessary. It therefore should be preserved.
So this commit just moves the setting of Frames_signal
to the full screen replacement routines when possible.
Reference(s):
. May, 2022 - bottom window forced off for some
commit d66c1f39b5
Signed-off-by: Jim Warner <james.warner@comcast.net>
[ along the way, we'll fix-up the section 4 commands ]
[ summary which has gotten a little outdated lately. ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
Messages issued by top will be displayed for only 1.25
seconds. And while this length of time would appear to
be acceptable (given the absence of complaints), there
will be times when a specific message might be missed.
So, this commit offers users the opportunity to recall
up to 10 of the most recent messages that were issued.
[ we'll just exploit top's new bottom window feature ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
Now that a potential use of that bottom 'window' for a
menu has been demonstrated, we'll remove such clutter.
[ after all, the prior commit cautioned against text ]
[ made conditional which might also be translatable. ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
While it is acceptable to make text usage conditional,
one must never make the text itself conditional. After
all, the translators must be presented with all of the
text so the opportunity to translate it is never lost.
[ one wonders who the idiot was that did this anyway ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
Fix some errors introduced in the commits shown below.
Reference(s):
. Apr, 2022 - 'keys_global' missed redundancy
commit 27f5904edd
. Apr, 2022 - added extraneous comma in help text
commit fb32021eeb
Signed-off-by: Jim Warner <james.warner@comcast.net>
If the special bottom 'window' routines remain unaware
of potential multi-byte characters, that heading could
be truncated prematurely and unnecessary blank line(s)
added to the reserved rows at the bottom of a display.
So, in both cases, this patch will now account for any
difference between string lengths and display columns.
Signed-off-by: Jim Warner <james.warner@comcast.net>
This patch attempts to reduce the proliferation of the
bottom window support routines by combining the 'misc'
with 'item' functions. Along the way we can now rename
those two 'pick' functions 'menu' while keeping proper
alphabetic order among all the bottom window routines.
Signed-off-by: Jim Warner <james.warner@comcast.net>
Reference(s):
proc/uptime.c:191:9: warning: variable 'uphours' is used uninitialized whenever 'if' condition is false [-Wsometimes-uninitialized]
Signed-off-by: Jim Warner <james.warner@comcast.net>
Really, extend shift+tab navigation to only some linux
consoles (or maybe most consoles). However, there were
some consoles where shift+tab always yields just '\t'.
Signed-off-by: Jim Warner <james.warner@comcast.net>
Since top now uses string vectors for two items in the
new focus window (cmdline & environ), there is no real
use for that 'separator' specified at toggle time. So,
this commit will find a new use for what is the space.
Henceforth, when a user is navigating using the tab or
backtab keys, should a strv element with focus contain
embedded space(s), it will be emphasized & highlighted
using the current window's message color (capclr_msg).
Signed-off-by: Jim Warner <james.warner@comcast.net>
A commit that changes no code but, hopefully, helps to
better document where and how such literals were used.
Signed-off-by: Jim Warner <james.warner@comcast.net>
This commit just brings that 'bot_focus_strv' function
up to the same standard as the vanilla 'bot_focus_str'
guy. Namely, if there is no cmdline or environ for any
process, that bottom window will now display an 'n/a'.
[ with the nature of those string vectors, it wasn't ]
[ the same simple process found for regular strings. ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
Some of those bottom windows grow quite large, as with
the firefox environment. If one is navigating the data
using the tab key, it can be very inconvenient when an
element should be reexamined but that focus has moved.
Therefore, this commit will support the 'back_tab' key
which is a standard combination of the shift+tab keys.
Now a user can easily backup up to any missed element,
assuming a terminal emulator honors that terminfo key.
[ a gui emulator typically will honor the 'back_tab' ]
[ terminfo string, whereas a linux console does not. ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
Using 'mempcpy' was a mistake where plain old 'memcpy'
was appropriate. More importantly, the careless length
specified resulted in a SEGV under some circumstances.
[ namely, it occurred under a multi-threaded top and ]
[ the top program itself as focus + CtrlN 'environ'. ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
After satisfying increased buffer size in the previous
patch, I decided to improve the highlight accuracy for
both the command lines plus the environment variables.
Since they both may contain elements which themselves
include embedded spaces, to accomplish this goal, one
must trade vanilla strings for string vectors instead.
[ at last such library items have now been justified ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
As it turns out, some of the 'environ' strings are big
enough to overrun the initial buffer sizes. And, while
no real harm was done (other than to silently truncate
the output), this patch will increase the buffer size.
While we're at it. we will normalize a few more buffer
declarations along with correcting a couple typos too.
Signed-off-by: Jim Warner <james.warner@comcast.net>
[ and along the way, those 2 newest namespace fields ]
[ are now mentioned under that same 'X' command part ]
[ this commit already changed for improved response. ]
[ they should have been included in the patch below. ]
Reference(s):
commit 6e78355d4d
Signed-off-by: Jim Warner <james.warner@comcast.net>
[ along the way we will restore '^R' to keys summary ]
[ plus correct a leftover reference to 'Ctrl-V' too. ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
This commit adds a prototype for a potential extension
of that tabbed bottom window capability. It introduces
a 'menu' which then provides for users' selections. It
does not actually do anything meaningful and will only
be enabled with the new #define called 'BOT_PICK_YES'.
[ since it's just a proof of concept, no attempt has ]
[ been made to add the usual NLS translation support ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
This commit introduces the 'tab' key which can be used
to highlight individual elements in that bottom 'Ctrl'
window. This can really help when reviewing such data.
[ note, normal ongoing monitoring continues unabated ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
This commit adds support for displaying the namespaces
in top's separate bottom 'Ctrl' window. Along the way,
the plumbing is now in place to extend this capability
to other data (if we don't run out of '^' keystrokes).
Signed-off-by: Jim Warner <james.warner@comcast.net>
This patch will just set the stage for the next commit
where we'll be adding the capability to display things
that sometimes may necessitate multiple library items.
In that effort we will be adding several new routines.
Signed-off-by: Jim Warner <james.warner@comcast.net>
In that commit shown below, the concept of a corrupted
display was introduced so that 'Ctrl' windows could be
redrawn now that they were made more persistent. Well,
it turns out there is another case where redrawing was
necessary. That is when transitioning between task and
thread modes. So this patch addresses those occasions.
Reference(s):
. May, 2022 - introduced Frames_signal BREAK_screen
commit d66c1f39b5
Signed-off-by: Jim Warner <james.warner@comcast.net>
In the patch referenced below the bottom 'Ctrl' window
was made to remain until it was dismissed by the user.
As such, it required abandoning a 'Cap_nl_clreos' used
when all the terminal rows had not been drawn. Instead
each line had to be cleared individually until the row
reserved for that 'Ctrl' window area had been reached.
Unfortunately, while a 'Cap_nl_clreos' didn't create a
problem for 'batch' mode, a 'Cap_clr_eol' used instead
yielded an unending loop if those rows were limitless.
So, now we will only clear that area when top isn't in
'batch' mode (which should have always been the case).
Reference(s):
. May, 2022 - bottom window made sticky
commit 0f2a755b0b
Signed-off-by: Jim Warner <james.warner@comcast.net>
Well darn it, whoever wrote that new library caught me
with my pants down (again?). Shoot, they were not just
down but somehow missing altogether. Here's the story.
Any item from that library supported by dynamic memory
can only be represented in user's stacks exactly once.
Should any string based enumerator be duplicated among
the items array, for any instance beyond the first the
library will return '[ duplicate ENUM ]' for a result.
That's where I lost my pants. While command lines were
given special handling (and never duplicated) I failed
to turn on CGROUPS, SUPGRPS & ENVIRON when testing the
Ctrl-G, Ctrl-U & Ctrl-N keys. If any of those 3 are on
that's when a Ctrl window sees a 'duplicate' notation.
[ and who runs top with such fields displayed anyway ]
In responding to this oops, the internals were changed
quite dramatically & vastly simplified in the process.
More importantly, the 'duplicate' results are no more.
Signed-off-by: Jim Warner <james.warner@comcast.net>
To complete the implementation of this new feature, we
must address the problem created by any commands which
offer their own full replacement displays. Should that
'Ctrl' bottom window be active, a return from any such
command leaves that lower portion of a terminal blank.
While we could issue a 'TAGGED_TOSS' macro right after
invoking any of those 4 (currently) affected commands,
it is wrong to make them aware of this 'tagged' stuff.
So, I'll opt for a more generalized solution whereby a
notice regarding a corrupted screen is broadcast. Then
the logic already aware of 'tagging' can deal with it.
Signed-off-by: Jim Warner <james.warner@comcast.net>
After making that new bottom window more persistent, I
decided to offer a build choice for somebody who might
prefer to have the rug pulled out from under him while
reviewing window contents should some target task end.
If 'TAGGED_BRIEF' is defined old behavior is restored.
[ it should be noted, however, that when this define ]
[ is active we'll lose any benefit of onetime update ]
[ for the bottom window. just like with the original ]
[ behavior, it is repainted with every update cycle. ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
I grew tired of inadvertently causing a premature exit
from top when all I wanted was to review some process'
environment variables. That 'V' key was just too close
to the 'C' key so the '^V' attempt became one of '^C'.
Well, not any more. Should a user wish to see a task's
environment, it will now require the '^N' combination.
[ it also now places G/U & N/K on a similar diagonal ]
Along the way, lower case Ctrl key combinations on the
help screen were all changed into upper case versions.
Signed-off-by: Jim Warner <james.warner@comcast.net>
After using this new feature for awhile, I encountered
numerous times when that 'Ctrl' window would disappear
as I was studying the contents. This was caused by the
death of the target task but it made me think that the
window could/should persist until dismissed by a user.
Well, now it will persist until that same Ctrl key has
been struck a 2nd time or a different Ctrl key used to
target another process after positioning it on row #1.
So now, instead of that former 'moving target' type of
approach, this window is more of a 'snapshot' in time.
[ even more significant, perhaps, is the fact that a ]
[ Ctrl window will now be drawn exactly 1 time only, ]
[ instead of being repainted after each delay cycle. ]
[ this will remain true no matter how many intervals ]
[ might elapse while its contents were under review. ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
When support for the 'LIBPROC_HIDE_KERNEL' environment
variable was introduced, a deficiency was present that
allowed any rejected proc_t (i.e. a kworker thread) to
preserve the strdup'd 'cmd' value. That residual value
would prevent stat2proc or status2proc from updating a
cmd field with the proper program name for some tasks.
This patch just ensures a proc_t is freshened whenever
it has been rejected due to an active PT->hide_kernel.
[ again thanks to Björn for initiating the extension ]
Reference(s):
. original hide_kernel implementation
commit 2a7ec67ac8
. original hide_kernel proposal
https://gitlab.com/procps-ng/procps/-/merge_requests/147
Signed-off-by: Jim Warner <james.warner@comcast.net>
This minor addition to that new 'Ctrl' feature is more
important than it seems. It's responsible for alerting
yours truly to the 'LIBPROC_HIDE_KERNEL' bug addressed
in the very next commit. All this while, program names
for some tasks would be mysteriously replaced with the
kernel 'kworker' names. However, customary clutter for
the top process display tended to obscure such a fact.
[ and, of course, to notice the problem in the first ]
[ place required one to have the command line toggle ]
[ 'off' or perhaps to toggle its state 'on' & 'off'. ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
There are times when one might want to see some task's
particular variable width data. However, prior to this
commit, the only way was to first turn on a field then
scroll through it via repeated right arrow keystrokes.
[ this also required that field to be displayed last ]
Needless to say, given the potential length of some of
that variable data this could be extremely cumbersome.
Now with this patch, a Ctrl keystroke combination will
create a separate window at the bottom of the terminal
screen where such variable width data is seen in full.
[ the targeted task is the 1st task displayed, which ]
[ is a convention employed in some existing commands ]
[ the targeted data was determined by these Ctrl key ]
[ combinations: CtrlG = ctrl group; CtrlK = cmdline; ]
[ CtrlU = supplementary groups; plus CtrlV = environ ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
When autogroups are active programs such as renice and
nice are basically useless because the nice value will
only affect the target task priority relative to other
processes in the same autogroup. So to accomplish what
we thought of as renice, /proc/<pid>/autogroup must be
changed. Altering a single member in an autogroup will
also affect every other member of that same autogroup.
Since top's renice provision ('r') suffers constraints
like those of the stand alone nice/renice programs, we
will now provide a means to manipulate that nice value
found within some process' /proc/<pid>/autogroup file.
[ to alter this file for other user's tasks requires ]
[ root privileges, as does setting a negative value. ]
[ however, unlike that 'r' command, this new command ]
[ allows raising *and* lowering all positive values. ]
Reference(s):
. Aug 2021, autogroups added to librady
commit 631e5d91f3
. Aug 2021, autogroups added to top
commit b789b46f84
Signed-off-by: Jim Warner <james.warner@comcast.net>
To avoid potential user confusion, like that reflected
in the previous commit message, a short narrative will
be included in the header file as programmer comments,
Signed-off-by: Jim Warner <james.warner@comcast.net>
With openSUSE's tumbleweed release of procps-ng-4.0.0,
their source rpm contained a questionable patch. It is
ostensibly devoted to increasing the accuracy for %cpu
calculations in the ps program. However, in my opinion
it goes too far and is quite flawed for these reasons:
1. Six separate files were impacted instead of just 1.
2. While ps was the object, libproc-2 was changed too.
3. A header file's alphabetic item order was violated.
4. The library API and ABI were altered unnecessarily.
It should be noted that all the <pids> TIME items were
made 'real' when representing seconds so that they can
be converted into jiffies/tics if multiplied by hertz.
Thus, there was absolutely no justification for adding
a new PIDS_TICS_ELAPSED ull_int item to the interface.
As an example, assuming a hertz value of 100, there is
enough capacity in all of those 'TIME' seconds to hold
nearly 3 million years worth of tics without a loss of
precision whenever they are changed back into jiffies.
[ that's nine quadrillion seven trillion one hundred ]
[ ninety-nine billion two hundred fifty-four million ]
[ seven hundred forty thousand nine hundred ninety + ]
[ one! or more concisely, as: 9,007,199,254,740,991. ]
So, any need for increased accuracy in that ps program
can be realized within the single output.c file alone.
That's what will be accomplished with this new commit.
And for any doubters here's a 'double' capacity proof:
beg ------ 9007199254740991.000000000000000000000 tic|
d /= 100, 90071992547409.90625000000000000000000 sec
d /= 60, 1501199875790.165039062500000000000000 min
d /= 60, 25019997929.83608245849609375000000000 hrs
d /= 24, 1042499913.743170142173767089843750000 day
d /= 7, 148928559.1061671674251556396484375000 wks
d /= 52, 2864010.752041676081717014312744140625 yrs
d *= 52, 148928559.1061671674251556396484375000 wks
d *= 7, 1042499913.743170142173767089843750000 day
d *= 24, 25019997929.83608245849609375000000000 hrs
d *= 60, 1501199875790.165039062500000000000000 min
d *= 60, 90071992547409.90625000000000000000000 sec
d *= 100, 9007199254740991.000000000000000000000 tic
end ------ 9007199254740991.000000000000000000000 " |
[ who knows, maybe we'll even convince openSUSE that ]
[ the original sledgehammer 'dif' should be dropped. ]
Reference(s):
. openSUSE equivalent: 'procps-ng-4.0.0-accuracy.dif'
Signed-off-by: Jim Warner <james.warner@comcast.net>
This patch trades some recurring per-task calculations
for calculations performed once at get, select or reap
time. It was prompted by the openSUSE dif named below.
[ my next commit will deal more thoroughly with that ]
Reference(s)
. openSUSE patch named: 'procps-ng-4.0.0-accuracy.dif'
Signed-off-by: Jim Warner <james.warner@comcast.net>
With the memset to zero of a 'new' meminfo_data struct
near the beginning of the meminfo_read_failed function
there's never a need to later set anything in it to 0.
[ who knows, our patch might even coax opensuse into ]
[ reevaluating that 4.0.0 'overflow' guy named below ]
Reference(s):
. overflow: 'procps-ng-4.0.0-integer-overflow.patch'
Signed-off-by: Jim Warner <james.warner@comcast.net>
Now that this 'c' variable initial assignment has been
moved outside the looping code, there is no longer any
need to specifically check for space/comma when 'c' is
not '%'. So, let us eliminate those two lines of code.
[ i promise not to change this algorithm ever again! ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
On occasion, even as the top author, I wonder why that
'x' toggle has stopped working. Of course, it actually
was working but a locate request ('L') or other filter
('O') operation was active and thus temporarily turned
if off. Such behavior is documented in top's man page.
Well, with this patch that 'x' suppression is no more.
[ the original justification, however, remains true. ]
[ but there's really only one character which causes ]
[ any potential trouble & i'm gonna' keep it secret. ]
[ besides, if a display is corrupted, there's always ]
[ that '=' key which restores things back to normal. ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
As a result of the issue referenced below, we'll trade
our homegrown offset generator for an 'offsetof' macro
found in the stddef.h header file. This pleases clang.
[ and thanks to Daniel Kolesa for the report and fix ]
Reference(s):
. bug report & recommended solution
https://gitlab.com/procps-ng/procps/-/issues/235
. clang error message
proc/readproc.c:673:9: error: initializer element is not a compile-time constant
mkENT(Rss),
^~~~~~~~~~
proc/readproc.c:661:34: note: expanded from macro 'mkENT'
#define mkENT(F) { #F ":", -1, (int)((void*)&q->smap_ ## F - (void*)&q->fZERO) }
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Jim Warner <james.warner@comcast.net>
It would appear that openSUSE was the first to release
procps-ng version 4.0.0 (in the tumbleweed distro). As
a result I checked their source rpm and found a couple
of patches I'm porting to newlib for the next release.
This particulate patch, with some changes, is openSUSE
patch 'procps-ng-3.3.10-large_pcpu.patch'. The changes
were strictly cosmetic. They involved whitespace only.
[ it should be noted that the net effect for raising ]
[ the maximum is to produce a '?' in the %CPU field. ]
[ that's because its width is fixed at five columns. ]
[ but at least the '?' won't distort actual results. ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
It would appear that openSUSE was the first to release
procps-ng version 4.0.0 (in the tumbleweed distro). As
a result I checked their source rpm and found a couple
of patches I'm porting to newlib for the next release.
This particulate commit was a refactor of the openSUSE
patch 'procps-ng-3.3.8-bnc634840.patch'. Unfortunately
their original patch did not have the intended effect.
That was because the amended signal handling logic was
performed well before the command line parameters were
parsed. So the global 'Batch' flag was in its 0 state.
. what follows is the original openSUSE commit message
------------------------------------------------------
Do not setup SIGHUP signal handler if we are in the batch mode
Top enables a signal handler for the SIGHUP signal (loss of terminal). While
this makes sense for top's default interactive mode, it doesn't make any sense
for batch mode. If you run top in nohup just to collect data over time and
disconnect top finishes which is not what one would expect.
------------------------------------------------------
Signed-off-by: Jim Warner <james.warner@comcast.net>
With the commit referenced below a nasty bug affecting
aix parsing was swatted. Beyond the bug, the logic was
enhanced to disallow commas in the format string since
they would otherwise be shown with their field's data.
However, there remained many characters other than ','
that could survive the edits to then be shown adjacent
to the data. Well, with this patch they won't anymore!
[ along the way we will no longer try to forgive the ]
[ use of a double '%%' prefix since that resulted in ]
[ display of one '%' and field code instead of data. ]
Reference(s):
. March, 2022 - restored aix bahavior
commit 8cb646bdfc
Signed-off-by: Jim Warner <james.warner@comcast.net>
The commit shown below broke the aix behavior that Dr.
Fink recently reported. However, in the proposed patch
the old behavior, showing garbage when '%cpu' was used
with an invalid formatting option, would appear again.
So this patch, based on Werner's patch, goes the extra
distance to prevent that. Along the way we'll disallow
commas in the aix format str to prevent their display.
Reference(s):
https://www.freelists.org/post/procps/Procpsng-400-released-with-newlib,2
. Mar, 2022 - where aix bug was introduced
commit 81df85a1b5
Prototyped-by: Dr. Werner Fink <werner@suse.de>
Signed-off-by: Jim Warner <james.warner@comcast.net>
When top's rcfile was reorganized, trading that 'char'
based field representation for the 'int' scheme, there
was a function prologue that should have been changed.
Thus, this patch will correct that original oversight.
Reference(s):
. Mar, 2022 - rcfile reorganized
commit 46aa96e438
Signed-off-by: Jim Warner <james.warner@comcast.net>
When we transitioned from 'char' based fields to 'int'
proper initial cursor positioning (highlight) was lost
in the 'fields_utility' function. The highlight should
appear on the current sort field. Instead, however, it
was often positioned on the first field listed or some
completely different field other than that sort field.
So this patch will restore the master branch behavior.
[ and on the assumption that a sort field is visible ]
[ we will search first for a field with the 'on' bit ]
Reference(s):
. Mar, 2022 - trade 'char' for 'int'
commit 46aa96e438
Signed-off-by: Jim Warner <james.warner@comcast.net>
At this point I don't know if the Rc.id ever carried a
value of 'b', 'c', 'd' or 'e' in a public release. But
I do know that those values ought not to be recognized
in any configuration file. Now they won't be for sure.
[ the net result is that now a user with such a file ]
[ will see the 'incompatible rcfile' message instead ]
[ of that misleading 'window entry corrupt' message. ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
When the configuration file is first written, there is
just the normal single newline which ends that rcfile.
However, when any existing rcfile was rewritten, there
was one extra newline '\n' character added at the end.
This will happen just once. The file does NOT continue
to grow with the extra blank lines always being added.
Anyway, this patch will put an end to such a practice.
[ along the way i fixed an unrelated whitespace oops ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
When %CUC was added in the commit shown below I failed
to update the man document. This fixes that oversight.
Reference(s):
commit cfa5538832
Signed-off-by: Jim Warner <james.warner@comcast.net>
This new integer based version of 'fieldscur' deviated
a bit too much from that character based logic when it
comes to ancient 3.2.8 rcfile support. This results in
an inappropriate extra field shown as the last column.
If we're truly dealing with 20 year old 3.2.8 rcfiles,
it's important the 3.3.9 extension (RCF_PLUS_H) not be
concatenated to what was read. That's because a search
for the special SUSE characters will always find a '\'
in the 26th position (after normal 3.2.8 'fieldscur').
[ for symmetry, we'll also avoid RCF_PLUS_J (3.3.17) ]
All other 'old' rcfiles are transformed without error.
Reference(s):
. Mar, 2022 - 3.2.8 support restores
Signed-off-by: Jim Warner <james.warner@comcast.net>
This patch will restore the previously abandoned older
rcfile support. To be honest, I wasn't sure this would
be possible given the vast difference between formats.
But after some very intense hacking our newlib top can
again fully support seamless migration from some older
'char' based file to this new 'int' based file format.
[ top even supports a 20+ year old 3.2.8 config file ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
As I speculated in a prior commit, this patch will now
demonstrate how simple it is to extend the upper limit
of supportable fields. With these changes we'll now go
from facing the old limit of 86 to a new limit of 100.
No other changes are needed nor is any logic impacted.
Signed-off-by: Jim Warner <james.warner@comcast.net>
With each of those references shown below, the maximum
number of fields was increased. However, with a 'char'
based implementation we're nearing the upper limits of
total displayable fields. We currently use 76 of a max
of 86 fields. With extra effort, 94 might be possible.
But 94 is the absolute upper limit of possible fields!
Moreover, the current implementation yields characters
that were unprintable in the rcfile. This could become
an issue with that 'inspect' feature when/if an rcfile
is edited to add entries (as opposed to using 'echo').
So, with this commit the internals of field management
has been completely reimagined. It is now based on the
integer type, not a character. And whereas that former
design used the high order bit to show the 'on' state,
thus yielding an unprintable character, the new design
uses the low order bit for the state. As such, numbers
will be kept small and an even number will be an 'off'
field whereas an odd number will become an 'on' state.
The bottom line is that this new design will afford an
unlimited number of new fields while keeping an rcfile
completely free of that potential unprintable garbage.
And it is embarrassingly easy to extend the maximum of
supportable fields from the currently implemented 100.
Who knows, maybe a future patch will prove this point.
[ unless a subsequent commit proves otherwise, given ]
[ the dramatic differences in rcfile contents, i had ]
[ to abandon the practice of supporting old rcfiles. ]
Reference(s):
. Nov, 2013 - RCF_PLUS_H introduced
commit af4e6533ba
. Jul, 2016 - RCF_PLUS_J introduced
commit d5c5051fb3
Signed-off-by: Jim Warner <james.warner@comcast.net>
This warning was added after discovering openSUSE uses
the option with (at least) their 'tumbleweed' version.
The man document sections impacted are: 4e, 5d and 5e.
That is where it was asserted sort column highlighting
would be temporarily turned off if search ('L') and/or
filtering ('O') is active. This option makes it false.
[ shame on suse for not also correcting the man page ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
This commit just compliments a change referenced below
by providing the value which includes reaped children.
[ as an aside, it looks like ps only includes reaped ]
[ children values under an obscure BSD or GNU option ]
[ so, this change is possibly of even more relevance ]
Reference(s):
. Mar, 2022 - added %CUU
commit 2ac72e2e80
Signed-off-by: Jim Warner <james.warner@comcast.net>
This commit just compliments a change referenced below
by providing the value which includes reaped children.
Reference(s):
. Feb, 2022 - added %CUU field
commit 7647e96b0a
Signed-off-by: Jim Warner <james.warner@comcast.net>
Some items in the <pids> API carry a '_C' suffix which
implies such values include reaped children. So, we'll
now extend this practice to that new UTILIZATION item.
Reference(s):
. Mar, 2022 - tweak PIDS_UTILIZATION
commit 9c0e8e9429
. Feb, 2022 - added PIDS_UTILIZATION
commit c69104b2b8
Signed-off-by: Jim Warner <james.warner@comcast.net>
When %CUU was added, in the commit referenced below, I
stated that 4 similar specifiers already existed. Well
I misspoke since there is actually 5 including 'util'.
Each of those fields had no sort capability. Since the
values are dynamically calculated, they were forced to
use 'PIDS_extra' as the format_array 'sr' designation.
Now each will use 'PIDS_UTILIZATION' and be sort-able.
[ yes, sometimes the calculated values could contain ]
[ reaped children while the sort field does not. but ]
[ such disparity depends on obscure bsd/gnu options. ]
Reference(s):
. Mar, 2022 - added %CUU
commit 2ac72e2e80
Signed-off-by: Jim Warner <james.warner@comcast.net>
When scale_tics was refactored and then Ctrl-E support
added to top, the complete range of scaling values was
not visible. Namely, a single 'd' (days) & 'w' (weeks)
was never seen with ^E. With this commit they will be.
Reference(s):
. Mar, 2022 - introduced ^E tics scaling
commit 402bf1903b
. Mar, 2022 - refactored scale_tics
commit 71eb90c1b2
Signed-off-by: Jim Warner <james.warner@comcast.net>
Whoa, my head really hurts but this commit should help
with a speedy recovery hopefully, after it is applied.
If the '%cpu' field is used as a format specifier with
that 'o' option, you will encounter a SIGSEGV if there
is also an invalid argument on that same command line.
For example, try 'ps/pscommand -o %cpu,x' with newlib.
With any format specifier other than the '%cpu', there
is an error message, as would happen with '-o pcpu,x'.
For a 3.3.17 version of ps, there's no abend. Instead,
the program will just display a bunch of gobbledygook.
This boo-boo was found to exist as far back as v3.3.0.
[ ok, i am starting to feel very much better already ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
This just parallels the top program by adding that new
library PIDS_UTILIZATION item to the ps repertoire. It
should be noted, however, that the new %CUU field is a
little redundant. I mean, ps already has 4 such fields
implemented identified as: '%cpu', 'c', 'cp' & 'pcpu'.
Oh well, at least the newest one offers a little value
added in the form of extra precision. We'll follow the
top lead and display results in the form of: '##.###'.
Signed-off-by: Jim Warner <james.warner@comcast.net>
When preparing the ps program for a %CUU field, it was
revealed the PIDS_UTILIZATION item may sometimes yield
a result of 'nan' or 'inf' for that ps program itself.
Apparently, it was caused by the short lived nature of
such a one-shot program. And, while this anomaly could
be handled in ps, the solution belongs in the library.
So, I reworked an item's algorithm to avoid this oops.
Signed-off-by: Jim Warner <james.warner@comcast.net>
[ i also tweaked that 'STARTED' narrative just a bit ]
[ since its original wording implied the value could ]
[ change, whereas it's fixed when a task is started. ]
Signed-off-by: Jim Warner <james.warner@comcast.net>
That normalization of the 'scale_tics' function in the
prior commit convinced me that I won't please everyone
with my arbitrary choices for the scaling transitions.
So, this patch will provide the users with a means for
setting their own scaling transition points with a new
toggle. Ctrl-E was chosen since the 'e/E' toggles were
already present as a means of scaling (albeit memory).
[ this toggle will also serve an educational purpose ]
[ by allowing one to see all the scaling conventions ]
The scaling a user establishes is saved in the rcfile.
Signed-off-by: Jim Warner <james.warner@comcast.net>
This patch refactors the 'scale_tics' function to more
closely parallel uptime shown on the first line of the
summary area. The old logic has been preserved through
the header file's new #define SCALE_FORMER provision.
However, the former logic was actually a big disaster.
These are some potential problems with that old logic:
1. With respect to our time fields top no longer deals
solely with cpu time. So, the old limits of '68 weeks'
could possibly be insufficient to reflect those times.
2. Given the widths of top's new time fields, the code
never got beyond scaling to hours. For example, with a
ridiculously large span of 19 years, the scaled result
would then be shown as '167832h'. We never reached the
days ('6993d') or even the weeks ('999w') equivalents.
3. Similarly, with that 'TIME+' field and a large tics
value, results would then appear as 'MMMMMM:SS' rather
than the more meaningful 'HH:MM:SS' or days and hours.
So henceforth we will adopt these scaling conventions:
MMM:SS.hh ... minutes:seconds.hundredths
MMM:SS ...... minutes:seconds
HH,MM ....... hours,minutes
D+H ......... days+hours (with 'd' & 'h' suffixes)
D ........... days (with 'd' suffix)
W+D ......... weeks+days (with 'w' & 'd' suffixes)
W ........... weeks (with 'w' suffix)
Note that, unlike our former scaling logic, that 'MMM'
portion won't be allowed to grow unconditionally. It's
limited (arbitrarily?) to 360 total minutes (6 hours).
Additionally, the 'HH' guy will be limited to 96 hours
(4 days) while that 'D' limit was set at 14 (2 weeks).
Whenever a limit is hit, scaling will advance a level.
Reference(s):
. Feb, 2022 - added 'ELAPSED'
commit 9348d3b008
. Feb, 2022 - added 'STARTED'
commit 7647e96b0a
Signed-off-by: Jim Warner <james.warner@comcast.net>
When the 'STARTED' field was added, in the message for
the commit referenced below, I explained why 'ELAPSED'
shouldn't be implemented though it might be preferred.
Well, after climbing out of my box to do a little more
thinking, I came up with the way to add that 'ELAPSED'
field while avoiding the possible performance penalty.
Just do not show what would change with every refresh!
If we do not show the seconds portion of a scaled tics
amount then the problem goes away. And this comes with
an additional benefit. The HH,MM (hours,minutes) style
then is readily compared with that system uptime shown
as HH:MM. The only difference is just the comma/colon.
[ assuming the top uptime/load average toggle was on ]
Reference(s):
. introduced 'start time' field
commit 7647e96b0a
Signed-off-by: Jim Warner <james.warner@comcast.net>
When the 'STARTED' field was added, the width was made
the same as the 'TIME+' field. Thus, a full time could
be shown (which then included hundredths of a second).
That kind of granularity is totally unnecessary. After
all, this column is potentially confusing enough since
it is so counterintuitive. So, this commit will reduce
the width of the field with some help from scale_tics.
Henceforth it will not include those ol' centiseconds.
[ along the way let's expand the man document with a ]
[ a remainder about content representation & scaling ]
Reference(s):
. introduced 'start time' field
commit 7647e96b0a
Signed-off-by: Jim Warner <james.warner@comcast.net>
While we really do not need the extra precision, we'll
trade that recently introduced float type for a double
for efficiency on a 64 bit platform. Additionally, the
UTILIZATION algorithm was tweaked producing less code.
The net result is four fewer machine instructions with
a reduction of 14 total bytes in that function's size.
Reference(s):
. introduced 'real' type & 'utilization' item
commit c69104b2b8
Signed-off-by: Jim Warner <james.warner@comcast.net>
The library refactored 'TIME' items for consistency so
we must adapt to some new data types and calculations.
Signed-off-by: Jim Warner <james.warner@comcast.net>
Jim Warner