mirror of
https://git.savannah.gnu.org/git/parallel.git
synced 2024-11-25 23:47:53 +00:00
Fixed bug #44614: --pipepart --header off by one.
This commit is contained in:
parent
be70739bb7
commit
57c8ca0823
|
@ -136,12 +136,13 @@ pushd
|
|||
git diff
|
||||
|
||||
# Recheck OBS https://build.opensuse.org/package/show?package=parallel&project=home%3Atange
|
||||
|
||||
YYYYMMDD=`yyyymmdd`
|
||||
TAG=MyTag
|
||||
echo "Released as $YYYYMMDD ('$TAG')."
|
||||
|
||||
git commit -a
|
||||
|
||||
Released as 20150x22 ('').
|
||||
|
||||
TAG=MyTag
|
||||
YYYYMMDD=`yyyymmdd`
|
||||
git tag -s -m "Released as $YYYYMMDD ('$TAG')" $TAG
|
||||
git tag -s -m "Released as $YYYYMMDD ('$TAG')" $YYYYMMDD
|
||||
|
||||
|
@ -218,12 +219,20 @@ Haiku of the month:
|
|||
|
||||
New in this release:
|
||||
|
||||
* GNU Parallel was cited in: Bayesian inference of protein structure from chemical shift data https://peerj.com/articles/861/
|
||||
|
||||
* <<afventer opdatering>> CIDER: a pipeline for detecting waves of coordinated transcriptional regulation in gene expression time-course data http://biorxiv.org/content/biorxiv/early/2015/03/17/012518.full.pdf
|
||||
|
||||
* <<afventer opdatering>> GNU Parallel was used (unfortunately without citation) in: MUGBAS: a species free gene-based programme suite for post-GWAS analysis http://www.ncbi.nlm.nih.gov/pubmed/25765345
|
||||
|
||||
taxator-tk http://algbio.cs.uni-duesseldorf.de/webapps/wa-download/ (check it)
|
||||
|
||||
* GNU Parallel was used in: landsat-gifworks https://github.com/KAPPS-/landsat-gifworks
|
||||
|
||||
* GNU Parallel (Sebuah Uji Coba) http://kaka.prakasa.my.id/2014/09/04/gnu-parallel-sebuah-uji-coba/
|
||||
|
||||
* Bug fixes and man page updates.
|
||||
|
||||
GNU Parallel - For people who live life in the parallel lane.
|
||||
|
||||
|
||||
|
|
12
src/parallel
12
src/parallel
|
@ -242,20 +242,19 @@ sub find_split_positions {
|
|||
while(read($fh,substr($buf,length $buf,0),$dd_block_size)) {
|
||||
if($opt::regexp) {
|
||||
# If match /$recend$recstart/ => Record position
|
||||
if($buf =~ /(.*$recend)$recstart/os) {
|
||||
my $i = length($1);
|
||||
push(@pos,$pos+$i);
|
||||
if($buf =~ /^(.*$recend)$recstart/os) {
|
||||
# Start looking for next record _after_ this match
|
||||
$pos += $i;
|
||||
$pos += length($1);
|
||||
push(@pos,$pos);
|
||||
last;
|
||||
}
|
||||
} else {
|
||||
# If match $recend$recstart => Record position
|
||||
my $i = index64(\$buf,$recendrecstart);
|
||||
if($i != -1) {
|
||||
push(@pos,$pos+$i);
|
||||
# Start looking for next record _after_ this match
|
||||
$pos += $i;
|
||||
$pos += $i + length($recendrecstart);
|
||||
push(@pos,$pos);
|
||||
last;
|
||||
}
|
||||
}
|
||||
|
@ -3134,6 +3133,7 @@ sub bibtex {
|
|||
" url = {http://www.gnu.org/s/parallel},",
|
||||
" year = {2011},",
|
||||
" pages = {42-47}",
|
||||
" doi = {10.5281/zenodo.16303}",
|
||||
"}",
|
||||
"",
|
||||
"(Feel free to use \\nocite{Tange2011a})",
|
||||
|
|
|
@ -81,6 +81,7 @@
|
|||
<li><a href="#Records">Records</a></li>
|
||||
<li><a href="#Record-separators">Record separators</a></li>
|
||||
<li><a href="#Header">Header</a></li>
|
||||
<li><a href="#pipepart">--pipepart</a></li>
|
||||
</ul>
|
||||
</li>
|
||||
<li><a href="#Shebang">Shebang</a>
|
||||
|
@ -1828,6 +1829,18 @@
|
|||
|
||||
<p>Output: Same as above.</p>
|
||||
|
||||
<h2 id="pipepart">--pipepart</h2>
|
||||
|
||||
<p>--pipe is not very efficient. It maxes out at around 500 MB/s. --pipepart can easily deliver 5 GB/s. But there are a few limitations. The input has to be a normal file (not a pipe) given by -a or :::: and -L/-l/-N do not work.</p>
|
||||
|
||||
<pre><code> parallel --pipepart -a num1000000 --block 3m wc</code></pre>
|
||||
|
||||
<p>Output (the order may be different):</p>
|
||||
|
||||
<pre><code> 444443 444444 3000002
|
||||
428572 428572 3000004
|
||||
126985 126984 888890</code></pre>
|
||||
|
||||
<h1 id="Shebang">Shebang</h1>
|
||||
|
||||
<h2 id="Input-data-and-parallel-command-in-the-same-file">Input data and parallel command in the same file</h2>
|
||||
|
|
|
@ -1832,6 +1832,22 @@ If the header is 2 lines, --header 2 will work:
|
|||
|
||||
Output: Same as above.
|
||||
|
||||
=head2 --pipepart
|
||||
|
||||
--pipe is not very efficient. It maxes out at around 500
|
||||
MB/s. --pipepart can easily deliver 5 GB/s. But there are a few
|
||||
limitations. The input has to be a normal file (not a pipe) given by
|
||||
-a or :::: and -L/-l/-N do not work.
|
||||
|
||||
parallel --pipepart -a num1000000 --block 3m wc
|
||||
|
||||
Output (the order may be different):
|
||||
|
||||
444443 444444 3000002
|
||||
428572 428572 3000004
|
||||
126985 126984 888890
|
||||
|
||||
|
||||
=head1 Shebang
|
||||
|
||||
=head2 Input data and parallel command in the same file
|
||||
|
|
|
@ -84,4 +84,10 @@ echo '### bug #44546: If --compress-program fails: fail'
|
|||
parallel --line-buffer --compress-program false echo \;ls ::: /no-existing; echo $?
|
||||
parallel --compress-program false echo \;ls ::: /no-existing; echo $?
|
||||
|
||||
echo '### bug #44614: --pipepart --header off by one'
|
||||
seq 10 >/tmp/parallel_44616;
|
||||
parallel --pipepart -a /tmp/parallel_44616 -k --block 5 'echo foo; cat';
|
||||
parallel --pipepart -a /tmp/parallel_44616 -k --block 2 --regexp --recend 3'\n' 'echo foo; cat';
|
||||
rm /tmp/parallel_44616
|
||||
|
||||
EOF
|
||||
|
|
|
@ -128,3 +128,32 @@ parallel: Error: false -dc failed
|
|||
1
|
||||
parallel: Error: false -dc failed
|
||||
parallel: Error: false -dc failed
|
||||
echo '### bug #44614: --pipepart --header off by one'
|
||||
### bug #44614: --pipepart --header off by one
|
||||
seq 10 >/tmp/parallel_44616; parallel --pipepart -a /tmp/parallel_44616 -k --block 5 'echo foo; cat'; parallel --pipepart -a /tmp/parallel_44616 -k --block 2 --regexp --recend 3'\n' 'echo foo; cat'; rm /tmp/parallel_44616
|
||||
foo
|
||||
1
|
||||
2
|
||||
3
|
||||
foo
|
||||
4
|
||||
5
|
||||
6
|
||||
foo
|
||||
7
|
||||
8
|
||||
9
|
||||
foo
|
||||
10
|
||||
foo
|
||||
1
|
||||
2
|
||||
3
|
||||
foo
|
||||
4
|
||||
5
|
||||
6
|
||||
7
|
||||
8
|
||||
9
|
||||
10
|
||||
|
|
Loading…
Reference in a new issue