pixelbeat

9 ms·

31.
by pixelbeat 10y ago
Yes exactly. For example GNU coreutils auto enables -Werror for devs when running from a git repo, whereas tarball releases do not add that option. See: http://git.sv.gnu.org/gitweb/?p=coreutils.git;a=blob;f=confi...
32.
by pixelbeat 10y ago
Note mv tries reflink by default (when moving files across BTRFS subvols) since doesn't need a separate copy of the data. Now as storage systems evolve there are less guarantees that one does get multiple copies, with deduplicating at
33.
by pixelbeat 10y ago
The approach that http://www.pixelbeat.org/fslint takes is to md5 and sha1 the files to avoid false positives. The core of the duplicate detection is the shell script here: https://github.com/pixelb/fsl
34.
by pixelbeat 10y ago
Oh cool, this is using my ansi -> html conversion script! http://www.pixelbeat.org/scripts/ansi2html.sh
35.
by pixelbeat 11y ago
GNU coreutils' shuf since v8.22 (Dec 2013) implements this to minimize memory usage when sampling large files or unknown numbers of items read from a pipe. For example this command never finishes, but consumes a constant (small) amount
36.
by pixelbeat 11y ago
I previously wrote about this common double encoding issue at http://www.pixelbeat.org/docs/unicode_utils/ which references tools and techniques to fix up such garbled data
37.
by pixelbeat 11y ago
Note the coreutils-single package in Fedora rawhide which can be used to get coreutils down to about 1MB. Details on that change at http://pkgs.fedoraproject.org/cgit/rpms/coreutils.git/commit...
38.
by pixelbeat 11y ago
That's more awkward and outputs a separate line per interrupt. Also to do that programmatically (to provide progress to a program using dd underneath) is tricky to do robustly due to the default disposition of SIGUSR1. SIGUSR1 is reall
39.
by pixelbeat 11y ago
As mentioned in the online manual at http://gnu.org/software/coreutils/dd that command line syntax is inspired by the DD (data definition) statement of OS/360 JCL
40.
by pixelbeat 11y ago
Yes something like termux might be a sufficient addition to "android". Note a standard GNU coreutils build is about 15M, though one can configure to use a multi-call binary like busybox, reducing the install size to about 1M. See
41.
by pixelbeat 11y ago
s/email/mozilla email/ Thanks for the reference ;)
42.
by pixelbeat 11y ago
Note that's virtual memory. For determining real RAM usage for programs, I find https://github.com/pixelb/ps_mem extremely useful
43.
by pixelbeat 11y ago
Yes GNU spends a lot of time improving performance. A couple of examples from the most recent release, which you might think were too simple to optimize significantly: The yes command (which is generally useful for generating repetitive tex
44.
by pixelbeat 11y ago
xargs basename -a < a.txt
45.
by pixelbeat 11y ago
Discussed at http://lists.gnu.org/archive/html/coreutils/2011-01/msg00080... Summary of that is you can filter using xargs like: get_file_paths | xargs basename -a No point having two ways to do
46.
GNU Coreutils Gotchas(pixelbeat.org)
3 points by pixelbeat 11y ago|0 comments
47.
by pixelbeat 11y ago
There have been a couple of recent changes to the latest GNU version of that related to flexible array members: https://github.com/coreutils/gnulib/blame/master/lib/fts.c#L... https://git
48.
by pixelbeat 11y ago
Personally I use these from the command line: http://www.pixelbeat.org/scripts/urldiff http://www.pixelbeat.org/scripts/idiff See also mergely which supports diffing URLs: http://pixelb
49.
by pixelbeat 11y ago
This uses a few utils and techniques to deal 5 random cards: https://twitter.com/pixelbeat_/status/587703133717057537 paste -d '' <(printf '%s\n' $(seq 2 9) T J Q K A | sed 'p;p;p
50.
by pixelbeat 11y ago
Neat thanks! You can generate HTML (and from that png or whatever) with http://www.pixelbeat.org/scripts/ansi2html.sh like $ COLUMNS=80 timeout 1 htop > t.ansi $ ansi2html.sh --bg=dark < t.ansi > t.h
51.
by pixelbeat 11y ago
It's a joke
52.
by pixelbeat 11y ago
You owe the GNU coreutils project $392 https://github.com/diafygi/gnu-pricing
53.
by pixelbeat 11y ago
It's great to see focus on performance, especially after all the recent stories on web bloat. I've noted a few techniques I've used on my blog to get pages served in a single request to users, including using SSIs and avoid c
54.
by pixelbeat 11y ago
Some practical notes for handling/avoid signed integer overflow http://www.pixelbeat.org/programming/gcc/integer_overflow.ht...
55.
by pixelbeat 11y ago
No, just that sparsification will work better. 1. cp will detect holes < 128KiB 2. cp --sparse=always it will avoid speculative preallocation used on file systems like XFS, which can have a significant impact on sparsification.
56.
by pixelbeat 11y ago
One thing not mentioned in the link are some performance improvement details. For example the yes command (generally useful for generating repetitive text): $ yes-old | pv > /dev/null ^C ... 55.8MiB/s ... $ y
57.
by pixelbeat 11y ago
Note Emacs uses http://git.sv.gnu.org/gitweb/?p=gnulib.git to abstract away platforms differences
58.
by pixelbeat 11y ago
The popup on command completion is a neat feature actually which I've appreciated in gnome-terminal on Fedora 22, nicely integrated into the notification system
59.
by pixelbeat 11y ago
In my experience we can't rely on manual handling of these integer overflow issues, especially with changing compiler behavior over time. I've noted some compile time and run time checking options at: http://www.pixelbe
60.
by pixelbeat 11y ago
Another option for the common case of static output is to render command or script(1) outout directly to static html with something like http://www.pixelbeat.org/scripts/ansi2html.sh

More ›