Package: fontconfig
Version: 2.13.1-2
Severity: wishlist
Today I discovered six years of files, hundreds, several megabytes, in
/var/cache/fontconfig . Nothing is cleaning them up.
Also no owner of /var/cache/fontconfig has been declared, in contrast to
$ dlocate /var/cache$
xfstt: /var/cache
Or just add "any of" in
>> │ system. If this option is chosen, 'jidanni.org', 'localhost' and **ANY OF**
>> │
>> 'jidanni2.jidanni.org;jidanni5.jidanni.org;jidanni7.jidanni.org;jidanni8.jidanni.org'
>> in From,
if semicolon is detected.
retitle 927922 Ask user if really he wants to proceed to a dangerous site.
Don't just quit
thanks
All I know is in such cases
$ w3m some_site.com
should say
some_site.com is using dangerous software.
(ERR: SSL v... v... )
Please ask the site owner to fix it.
[QUIT] [Proceed anyway
retitle 927395 Add a update-exim4.conf.conf.local to avoid edit conflicts
thanks
Well all I know is
/etc/exim4/exim4.conf.localmacros
is the safe place for macros, so there also should be a safe place for
things that would go in
/etc/exim4/update-exim4.conf.conf too.
By safe I mean guaranteed
Package: exim4-config
Version: 4.92-6
Severity: minor
dpkg-reconfigure exim4-config says
Mail Server configuration ├──┐
│
│
│ The headers of outgoing
Package: w3m
Version: 0.5.3-37
Lynx, chromium work fine.
$ w3m
https://branch.taipower.com.tw/content/Messagess/Contents.aspx?SiteID=564732602442427467=564732602561157543=564734220745242356
SSL error: error:1425F102:SSL routines:ssl_choose_client_version:unsupported
protocol
-- System
Package: chromium
Version: 74.0.3729.108-1
Severity: important
$ chromium &
$ sleep 22
$ chromium &
The second one kills the first one, and we see "Restore pages?"
retitle 927395 #include update-exim4.conf.conf-user.pre
thanks
Anyway having a robot edit what the user edits what the robot edits...
often leads to disaster.
A much cleaner design for update-exim4.conf.conf would be something like
#include user.pre #if exists
...guts go here...
#include
Package: exim4-config
Version: 4.92-6
Severity: wishlist
File: /var/lib/dpkg/info/exim4-config.postinst
The 17 line warning at write_header() is only written once and not
updated. Thus user A sees a 2 line warning, user B sees a 12 line
warning, user C sees a 17 line warning, all depending on
X-debbugs-Cc: billpo...@alum.mit.edu
Package: uni2ascii
Version: 4.18-3
Severity: wishlist
We read
NAME
ascii2uni - convert 7-bit ASCII representations to UTF-8 Unicode
and
DESCRIPTION
ascii2uni converts various 7-bit ASCII representations to
X-debbugs-Cc: billpo...@alum.mit.edu
Package: uni2ascii
Version: 4.18-3
Severity: wishlist
$ uni2ascii myfile.txt > /dev/null
Invalid UTF-8 code encountered at byte 4074, character 4074.
OK, but please also mention what line number too.
(Workaround;
$ uni2ascii myfile.txt | nl -b a | tail -n 1
Package: exim4-config
Version: 4.92-6
Severity: wishlist
File: /var/lib/dpkg/info/exim4-config.config
$ set /var/lib/dpkg/info/exim4-config.config
$ iconv -f latin1 -t utf8 $@ | diff $@ - | wc
14 2421770
Emacs says your file is Devangari.
chromium gets it wrong too.
file(1) says it
Package: aptitude
Version: 0.8.11-7
Severity: wishlist
File: /usr/bin/aptitude-curses
Black upon blue text very hard to read:
SM> Does this perhaps depend on settings or something? If you create a new
SM> temporary user account, log in as the new user and run epiphany-browser,
SM> do you get the same warnings?
Same on AMD64:
$ su - some_vanilla_acct
$ DISPLAY=:0 epiphany
(epiphany:8477): dbind-WARNING **:
https://www.debian.org/doc/debian-policy/ap-pkg-conffiles.html :
"Note that a package should not **modify** a dpkg-handled conffile in its
maintainer scripts. Doing this will lead to dpkg giving the user
confusing and possibly dangerous options for conffile update when the
package is upgraded."
Package: dpkg
Version: 1.19.6
Severity: wishlist
File: /usr/share/man/man1/dpkg.1.gz
Man page says
--clear-selections
Set the requested state of every non-essential package to
deinstall
(since dpkg 1.13.18). This is intended to be used
immediately
Package: aptitude
Version: 0.8.11-7
Severity: minor
One sees a (harmless) // in:
E: Cannot remove aptitude within aptitude
E: Problem parsing '/var/lib/aptitude//pkgstates', is it corrupt or malformed?
You can try to recover from '/var/lib/aptitude//pkgstates.old'.
Reproduce by doing
dpkg
Package: epiphany-browser
Version: 3.32.1.2-1
Users will get the message
Error loading module
'/usr/lib/i386-linux-gnu/epiphany-browser/web-extensions/libephywebextension.so':
libdazzle-1.0.so.0: cannot open shared object file: No such file or
directory
Because you removed the
Package: fonts-noto-core
Version: 20181227-1
Purging configuration files for fonts-noto-core (20181227-1) ...
dpkg: warning: while removing fonts-noto-core, directory
'/usr/share/fonts/truetype/noto' not empty so not removed
-rw-r--r-- 1 root root36 2018-04-12 .uuid
Package: grub2-common
Version: 2.02+dfsg1-16
Severity: wishlist
File: grub2-common
(info "(grub) play")
-- Command: play file | tempo [pitch1 duration1] [pitch2 duration2] ...
The units of duration are not explained. Microseconds?
Package: exim4-config
Version: 4.92-6
During this upgrade,
$ cat /var/log/apt/history.log
Start-Date: 2019-04-18 01:32:49
Upgrade: exim4-base:amd64 (4.92-5, 4.92-6), openssl:amd64 (1.1.1b-1, 1.1.1b-2),
unicode-data:amd64 (12.0.0-1, 12.1.0~pre1-1), exim4-daemon-light:amd64 (4.92-5,
4.92-6),
Package: debconf
Version: 1.5.71
Severity: wishlist
File: /usr/sbin/dpkg-reconfigure
User does
# dpkg-reconfigure some_package
and starts answering some questions,
and some more,
and some more...
Therefore, somewhere on the screen, please add a running tally of which
question we are on, (e.g.,
Package: aptitude
Version: 0.8.11-7
Severity: wishlist
The richer I get, the wider the monitor I buy.
Alas, this just makes the gap,
# aptitude search tile
p crawl-tiles
- Dungeon Crawl, a roguelike game,
Package: python3-qtile
Severity: wishlist
$ aptitude search ...
p node-tilelive-vector - Tilelive store for rendering mapnik vector tiles -
Node.js module
p projectile-doc - project interaction library for Emacs - documentation
p python-mapbox-vector-tile - Mapbox Vector Tile library for Python
Package: libgeo-osm-tiles-perl
Version: 0.04-5
Severity: wishlist
File: /usr/share/man/man3/Geo::OSM::Tiles.3pm.gz
"tile2path($tilex, $tiley, $zoom)" Composes the path to the tile at
$tilex, $tiley, and $zoom at the OSM server. $tilex and $tiley
must be integers in the range
Package: libgeo-osm-tiles-perl
Version: 0.04-5
Severity: wishlist
File: /usr/share/man/man1/downloadosmtiles.1p.gz
These ".pl"'s don't belong here:
NAME
downloadosmtiles.pl - Download map tiles from OpenStreetMap
SYNOPSIS
downloadosmtiles.pl --lat=49.5611:49.6282
Package: net-tools
Version: 1.60+git20180626.aebd88e-1
Severity: wishlist
File: /usr/share/man/man8/ifconfig.8.gz
The man page should mention:
# ifconfig -a
inet 192.168.0.8 netmask 255.255.255.0 broadcast 192.168.0.255
How to zap ^^^ #Yes, use 0.0.0.0, but say so more
GFTG> while this is broken, you could use Alt+/ to complete with filenames.
Ah, (or ESC /). Good to know.
Package: bash-completion
Version: 1:2.8-6
File: /usr/share/bash-completion/completions/find
$ find /tmp -path /tmp/ssh-dEu does not complete.
X-Debbugs-Cc: dia...@skoll.ca
Package: pppoe
Version: 3.12-1.2
File: /usr/share/man/man8/pppoe.8.gz
# pppoe --help 2>&1 | grep -- -t
-t timeout -- Initial timeout for discovery packets in seconds
# man pppoe | col -b | grep -c -- ' -t'
0
# man pppoe | col -b | grep -c -- ' -T'
2
-t
Package: ppp
Version: 2.4.7-2+4.1
Severity: minor
File: /usr/share/doc/ppp/README.pppoe
Change
http://www.roaringpenguin.com/pppoe to
https://dianne.skoll.ca/projects/rp-pppoe/
Package: usbview
Version: 2.0-21-g6fe2f4f-2
Severity: wishlist
File: /usr/bin/usbview
$ usbview
Shows an error message,
that can be copied with the mouse. Bug 1.
The message mentions many possible problems.
But forgets to say "try again as root". Bug 2.
Package: usbview
Version: 2.0-21-g6fe2f4f-2
Severity: wishlist
File: /usr/bin/usbview
There is no way (e.g., quit with CTRL+Q) to hit any of the buttons with
just keyboard shortcuts. This makes the program less accessible to
certain disabled users.
Package: usbutils
Version: 1:010-3
Severity: wishlist
File: /usr/bin/lsusb
I have an idea.
Real (arboretum) trees also have branches with no leaves. Looking at
# lsusb -t
/: Bus 04.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/8p, 1M
/: Bus 03.Port 1: Dev 1, Class=root_hub,
Package: usbutils
Version: 1:010-3
Severity: minor
File: /usr/bin/lsusb
Here we observe the buses are listed backwards,
# lsusb -t
/: Bus 04.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/8p, 1M
/: Bus 03.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/14p, 480M
|__ Port 10: Dev 2, If
OK, I put my changes in /etc/default/grub.d/jidanni.cfg . You might, in
a README, mention that is the ideal way of "getting the last word", and
also not having ones changes get "polluted back" into a
/etc/default/grub file...
Package: aptitude
Version: 0.8.11-7
Severity: minor
No matter if a package was marked for installation or not,
# aptitude purge firmware-ipw2x00
Package firmware-ipw2x00 is not installed, so it will not be purged
gives the same message.
It should say in addition "no longer marked for
Package: installation-reports
Severity: minor
Just want to let you know after a fresh install, when booting into
"(recovery mode)" from the grub menu, and typing the root passwd,
# cat
^C
doesn't interrupt cat, etc.
One needs ^Z, then "kill %" to stop it.
Yes, these just reported the same as on
Package: pppoe
Version: 3.12-1.2
Severity: minor
File: /usr/share/man/man8/pppoe.8.gz
The pppoe home page is now
https://dianne.skoll.ca/projects/rp-pppoe/
Not the 404 link listed on the man page.
P.S., 3.13 is available.
Package: emacs-common
Version: 1:26.1+1-3.2
Severity: minor
File: /usr/share/man/man1/emacsclient.emacs.1.gz
At
For emacsclient to work, you need an already running Emacs with a
server. Within Emacs, call the functions "server-start" or
"server-mode". (Your ".emacs" file can do
X-debbugs-Cc: util-li...@vger.kernel.org, k...@redhat.com
Package: fdisk
Version: 2.33.1-0.1
Severity: minor
File: /sbin/cfdisk
cfdisk: after a [write], the cursor should move to [quit].
Don't leave it on [write].
(And then the words at the bottom when the cursor is on [quit] should not
be still
Package: aptitude
Version: 0.8.11-7
Severity: minor
Search ~b~i finds something,
# aptitude search ~b~i
iBA libjack-jackd2-0 - JACK Audio Connection Kit (libraries)
But
# aptitude show ~b~i
shows nothing,
and
# aptitude purge ~b~i
purges nothing.
$ cat /etc/apt/apt.conf.d/10jidanni
X-Debbugs-Cc: j...@calhariz.com
Package: at
Version: 3.1.23-1
Severity: wishlist
File: /usr/share/doc/at
As we know computers are usually replaced every few years and we want to
move the at jobs from the old system to the new system.
Please document the proper way to do this.
Yes certainly it
Package: debian-installer
Severity: minor
I recall when the graphical installer got to this message,
Name: partman-crypto/progress/plain_erase_text
Type: text
Description: The installer is now overwriting ${DEVICE} with zeroes to delete
its previous contents. This step may be skipped by
Package: xutils
Version: 1:7.7+19
Severity: minor
Apparently this is a "transitional dummy package that can safely be
removed".
Well then add that statement and some tags to help various scanners
looking for such packages find them.
No I don't know what the standard tags are.
At least you
Package: bash-completion
Version: 1:2.8-6
Severity: minor
File: /usr/share/bash-completion/completions/chown
# chown jidanni. doesn't complete.
It should complete the possible groups.
unmerge 925891
retitle 925891 AMD/ATI Carrizo driver incorrectly detected?
thanks
Now making the incorrect driver detection a separate issue.
All I know is with that minimal install, it somehow assumes the correct
driver is present, or it is, but it uses it wrong. See also
clone 926149 -1
retitle -1 [AMD/ATI] Carrizo driver incorrectly detected
retitle 926149 Add nomodeset kernel parameter for recovery boot grub entry
thanks
> "BH" == Ben Hutchings writes:
BH> However it might be reasonable to add "nomodeset" to the command line
BH> in GRUB's "recovery" boot
X-Debbugs-Cc: adri...@gnu.org
Package: units
Version: 2.18-1
Severity: minor
File: /usr/share/man/man1/units.1.gz
We see AUTHOR and then it is blank, on man page.
> However it might be reasonable to add "nomodeset" to the command line
> in GRUB's "recovery" boot menu items.
OK, great. Now all that is left is for me to help you figure out why
the installer incorrectly detects that the drivers are available, etc.
I am able to workaround all this, I just
Package: grub-pc
Version: 2.02+dfsg1-16
Severity: wishlist
I am trying to get the "final word" in how these grub variables are
appended.
Can you please specify a files, e.g.,
/etc/default/grub.local.pre
/etc/default/grub.local.post
that are initially absent, and guaranteed never to be written by
Package: chromium
Version: 73.0.3683.75-1
Please see
https://bugs.chromium.org/p/chromium/issues/detail?id=753242#c31
And lo and behold, in /etc/grub.d/10_linux :
if [ "$ubuntu_recovery" = 1 ]; then
GRUB_CMDLINE_LINUX_RECOVERY="$GRUB_CMDLINE_LINUX_RECOVERY nomodeset"
fi
> "BH" == Ben Hutchings writes:
BH> The installer normally uses a dumb framebuffer driver (probably efifb
BH> on this system) that is built into the kernel. This is too low-
BH> performance for a proper desktop.
OK, the installer could first double check that the framebuffer driver
it
Package: procps
Version: 2:3.3.15-2
File: /usr/share/man/man1/ps.1.gz
Severity: minor
--cols option may be used to exactly determine the width in this case. The
w or -w option may be also be
used to adjust width.
Also mention --width in that sentence.
And mention if --width is the same
Package: procps
Version: 2:3.3.15-2
Severity: minor
$ ps -e --columns 33 -o pid,start_time,cputimes:0,comm
error: column widths must be unsigned decimal numbers
^positive
Package: procps
Version: 2:3.3.15-2
File: /usr/share/man/man1/ps.1.gz
Severity: minor
On the man page it says
lstart STARTED time the command started. See also bsdstart,
start, start_time, and stime.
but there is no stime entry.
So the question becomes why does that installer ISO know how to properly deal
with the
BH> 00:01.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc.
[AMD/ATI] Carrizo [1002:9874] (rev e6)
BH> Subsystem: ASUSTeK Computer Inc. Device [1043:8719]
giving nice splash screens, a
X-Debbugs-Cc: b...@decadent.org.uk
Package: cdimage.debian.org
Severity: minor
On e.g.,
http://cdimage.debian.org/cdimage/unofficial/non-free/cd-including-firmware/current/amd64/iso-cd/
At
What is a netinst image?
The netinst CD here is a small CD image that contains just the core
BH> Please send the output of "lspci -vnn" for the system that shows this
BH> problem.
Would lspci -knn be good enough for now?:
https://bugs.debian.org/cgi-bin/bugreport.cgi?att=0;bug=925556;filename=logs.gz;msg=50
OK. The package is actually installed. It is just the message about that
being buffered until the next interaction probably.
All I know is I researched what ISO I could put on a USB stick
and take to my mountain for a minimal install with no network.
It worked fine, after I added that nomodeset line.
It is a shame if officially debian no longer can be installed without a
network.
Actually try this:
When you get to the popularity-contest question, wait a few moments,
then answer "no".
Later examine the logs.
You will see actual installation (and then removal) indeed occurs
*after* the user has answered "no".
So the user's eyes didn't fool him after all.
I'll be very happy to help you test this week.
Next week I wish to proceed with final installation though.
BH> Was the firmware (firmware-amd-graphics package) installed on your test
BH> system?
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=925556#25
mentions what ISO I used, and all logs
retitle 925556 Use UUID, not sdX on grub linux lines
thanks
The problem simply turns out to be:
grub lines like
linux /boot/vmlinuz-4.9.0-8-amd64 root=/dev/sdb2 ro quiet
are simply bound to fail:
When the computer boots, no matter when for the installer, or later, the
Package: installation-reports
With the latest AMD CPUs the kernel will attempt to use the AMDGPU
kernel driver. Alas this will result in a black screen on the minimal
system installed by the installer.
Therefore on all three of these lines in /boot/grub/grub.cfg
linux /boot/vmlinuz-...
I suppose in the logs, of
Mar 27 23:38:24 in-target: Preparing to unpack
.../popularity-contest_1.64_all.deb ...
Mar 27 23:38:24 in-target: Unpacking popularity-contest (1.64) ...
Mar 27 23:38:24 in-target: Setting up popularity-contest (1.64) ...
Mar 27 23:38:24 in-target: Removing
Trying again with expert install and some different choices,
and putting nomodeset in the grub finally booted it. I'll send new logs soon.
Package: debian-installer
Severity: minor
The Installer splash screen should show the version number.
Maybe just have a message: "press TAB to see all available commands."
That would show both built-ins and non builtins. (Hope TAB works.)
Package: debian-installer
Severity: minor
Let's say you are asking the user what time zone he is in for some country:
Eastern Time Zone
Plains Time Zone
Mountain Time Zone
Other
Well I recall in expert mode installation, UTC is also always present:
Eastern Time Zone
Plains Time Zone
Mountain
Package: debian-installer
Severity: minor
In tasksel_first_0.png
the user is totally helpless trying to figure out what you mean by
"Standard system utilities"
So give some examples. Say
"Standard system utilities (cat, ls...)" or
"Standard system utilities (gnome, kde...)" or
something,
Package: debian-installer
Severity: wishlist
Here on partman_choose_partition_0.png
we see both things that are about to be changed, and things that will
stay the same, but we can't tell which are which.
So right after the words "write changes to disk" you need to add "(You
will be first shown a
Package: debian-installer
Severity: minor
Please also ask "Do not configure the network at this time" *before* trying
to configure the network. Not only after, and failing.
That way users installing offline wouldn't need to answer several futile
questions and have futile DHCP packets sent to
Package: debian-installer
Users often hear "Stretch", "Buster", but on the installer, they see
just "Debian 9".
Therefore please make it say "Debian 9 (Stretch)" or "Debian 9 Stretch".
In fact much wiser would be to do what iOS does, and keep them a top secret.
Package: debian-installer
Severity: wishlist
In bug #925556, I showed you what a failed install looks like to a user
who has booted via grub's "(recovery mode)" entries.
If the user hadn't touched the keyboard, a normal boot would instead
proceed, whereupon the user would just end up on a black
Package: debian-installer
Severity: wishlist
The menus are great,
But you forgot one item.
"Reboot".
You see, no matter how the user scours the menus,
he cannot find the proper way to exit (reboot).
"Well he is supposed to know how to do that himself!"
Yes, but he wants the "proper" way to
Package: debian-installer
Today let's discuss Rescue Mode.
I think you should offer a second rescue mode choice.
Call it "Quick Rescue Mode".
List it right after Rescue Mode.
Quick Rescue Mode would drop the user into a shell right away.
No questions about languages. Yup, unfair. Sorry.
No
Package: debian-installer
Severity: minor
Here there is a high chance the user will chose the wrong item,
because he does not recall which of his disks are which.
Therefore you need to present more details, which yes, can fit on the
same line. Add sizes and partition types and disk label stuff
logs.gz
Description: application/gzip
I tried installing again. (This time with no network.) The normal
graphical install. All goes fine, but in the logs there is one "Failed
to get":
$ grep -A 2 'Failed to get' syslog |cut -c 17-
base-installer: warning: Failed to get debconf answer
'base-installer/kernel/linux/initrd'.
> "SM" == Steve McIntyre writes:
SM> Please attach the installer syslog so we have a chance to see what
SM> happened. On the installed system, that's in /var/log/installer/syslog
What should I give at that
initramfs prompt you see
that would put it on the screen?
Then I could take a
Package: debian-installer
User has a USB drive, loaded with
$ mount
/dev/sdg2 on /var/cache/apt/archives type ext4
/dev/sdg1 on /var/lib/apt/lists type ext4
He takes this drive, along with another drive,
containing a Debian installation ISO, to a remote mountain (offline)
computer, intending to
LS> Clearly the initramfs was able to mount /dev and run fsck, but mounting
LS> it to /root/dev to transfer to the real rootfs failed due to a missing
LS> directory.
So it booted, meaning it is not a
https://www.debian.org/releases/stretch/amd64/ch03s06.html.en#UEFI
problem...
Package: debian-installer
Here the user is left staring at an empty screen. (Many look that way
for a while at the beginning. Even with the latest fastest hardware.)
Not for long you might say.
But long enough for the user to scratch his head, and then casually get
out his cellphone and take a
Package: debian-installer
In https://wiki.debian.org/UEFI
We read
"debian-installer's support for UEFI is mostly contained in two modules.
First comes the partman-efi module, and this will be loaded
automatically if d-i recognises it has been booted in UEFI mode."
The problem here is that on
Package: apt
Version: 1.8.0
Severity: minor
Here we observe parallel download:
# aptitude full-upgrade
...
35% [5 firefox 33.9 MB/44.5 MB 76%] [1 google-earth-pro-stable 2,215 kB/56.5 MB
4%]
What if we don't want parallel download?
I.e., just download one server at a time?
Well the closest
Package: procps
Version: 2:3.3.15-2
Severity: minor
File: /usr/share/man/man1/ps.1.gz
Man page says
the normal default output can be produced with this: ps -eo "%p %y %x %c"
# ps|wc
5 20 140
# ps -eo "%p %y %x %c"|wc
134 5364783
# ps -o "%p %y %x %c"|wc
5
Package: procps
Version: 2:3.3.15-2
Severity: minor
File: /usr/share/man/man1/ps.1.gz
All of these,
OUTPUT FORMAT CONTROL
-c Show different scheduler information for the -l option.
--context
Display security context format (for SELinux).
-f Do
Package: lshw
Version: 02.18.85-0.1
File: /usr/share/doc/lshw/copyright
Author address listed is no longer valid.
: host mail1.ec-lyon.fr[156.18.19.129] said: 550 5.1.1
: Recipient address rejected: ec-lyon.fr (in reply to
RCPT TO command)
Alas, as https://ezix.org/src/issues account activation mail never
comes, there is no way, just looking at
http://ezix.org/project/wiki/HardwareLiSter to contact the author.
As you can see on
https://sourceforge.net/p/hdparm/bugs/82/
# fdisk -l
Disk /dev/nvme0n1: 119.2 GiB, ...
Disk model: INTEL SSDPEKKW128G7
I attempted to report this at
https://ezix.org/src/issues
alas, account activation mail never comes.
Package: systemd
Version: 241-1
Severity: minor
File: /usr/share/bash-completion/completions/journalctl
It's not happy about me wanting to do
# journalctl > file
See:
# journalctl >
AVAILABLE=
AVAILABLE_PRETTY=
CODE_FILE=
...
# journalctl > /
/bin/su
/dev/
/lib/systemd/systemd
Package: ffmpeg-doc
Version: 7:4.1.1-1
Severity: minor
$ cat /usr/share/doc-base/ffmpeg-doc
Document: ffmpeg-doc
Title: FFmpeg API Documentation
Author: FFmpeg Developers
Abstract: This is the API documentation for FFmpeg.
Section: Programming
Format: HTML
Index:
Package: ffmpeg-doc
Version: 7:4.1.1-1
Kindly at the top of each of
ffmpeg: /usr/share/man/man1/ffmpeg-all.1.gz
ffmpeg: /usr/share/man/man1/ffplay-all.1.gz
ffmpeg: /usr/share/man/man1/ffprobe-all.1.gz
ffmpeg-doc: /usr/share/doc/ffmpeg/manual/ffmpeg-all.html
ffmpeg-doc:
Well it might as well write it to both STDOUT and STDERR then.
> "GFTG" == Gabriel F T Gomes writes:
GFTG> You could be right (specially since bash completes any filename),
GFTG> however, where did you get this statistic from?
OK, maybe even less,
$ file /usr/bin/*|grep -c shell\ script
194
$ ls /usr/bin/*.sh
/usr/bin/gettext.sh
But that's just my
OK I let it run, and indeed now it is getting packages from
snapshot.debian.org/archive/ ... at rate 24.3KB/s...
So nevermind, instead of
$ jigdo-lite
http://cdimage.debian.org/cdimage/weekly-builds/amd64/jigdo-cd/debian-testing-amd64-netinst.jigdo
and selecting a local Taiwan mirror,
I'll just
Also when jigdo-lite is asking us questions, it can first mention "don't
worry about 404s"
SM> You don't show to the end of the log. Jigdo will switch to grabbing
SM> missing files from snapshot.d.o once it has finished with all the
SM> files it can find on the mirrors you've told it about.
GREAT: BUT:
The user will start seeing these 404s interspersed with the 200s and
think:
"Yup,
601 - 700 of 2728 matches
Mail list logo
