Mercurial > hg > cc > cirrus_work

annotate bin/do_idx.sh @ 133:3682ef4d2169

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
get 7f (two cases) and %25 working
author Henry S. Thompson <ht@inf.ed.ac.uk>
date Thu, 28 Sep 2023 18:31:23 +0100
parents 5b952d16838c
children 0326805aa6df
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
rev   line source
101
e2e64c3d763e bug4 fixed, but that created a new, earlier bug
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
1 #!/bin/bash
110
a0ea1e4a714d pass in debug flag(s) to merge_date.py
Henry Thompson <ht@markup.co.uk>
parents: 105
diff changeset
2 while [ "$1" = "-d" ]
a0ea1e4a714d pass in debug flag(s) to merge_date.py
Henry Thompson <ht@markup.co.uk>
parents: 105
diff changeset
3 do
a0ea1e4a714d pass in debug flag(s) to merge_date.py
Henry Thompson <ht@markup.co.uk>
parents: 105
diff changeset
4 shift
a0ea1e4a714d pass in debug flag(s) to merge_date.py
Henry Thompson <ht@markup.co.uk>
parents: 105
diff changeset
5 debug="$debug -d"
a0ea1e4a714d pass in debug flag(s) to merge_date.py
Henry Thompson <ht@markup.co.uk>
parents: 105
diff changeset
6 done
101
e2e64c3d763e bug4 fixed, but that created a new, earlier bug
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
7 export res="$1"
e2e64c3d763e bug4 fixed, but that created a new, earlier bug
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
8 orig="$2"
105
9403c02d5034 switch to gzip -7 to get comparable compressed cdx block size
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 103
diff changeset
9 # igzip was faster, but produced bigger files, so went to gzip one step
9403c02d5034 switch to gzip -7 to get comparable compressed cdx block size
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 103
diff changeset
10 # smaller than default (-6), which produces slightly _smaller_ blocks.
116
5b952d16838c a bit more logging
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 110
diff changeset
11 merge_date.py $debug <(LC_ALL=C sort -m -k1,3 -s $res/ks_[0-9]*.tsv) \
5b952d16838c a bit more logging
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 110
diff changeset
12 $orig $res/idx 2>$res/merge.log | \
101
e2e64c3d763e bug4 fixed, but that created a new, earlier bug
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
13 parallel -j 10 'echo {#} {} >$res/merge_{#}.log
e2e64c3d763e bug4 fixed, but that created a new, earlier bug
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
14 echo $(date) {#} {}
e2e64c3d763e bug4 fixed, but that created a new, earlier bug
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
15 export res
105
9403c02d5034 switch to gzip -7 to get comparable compressed cdx block size
Henry S. Thompson <ht@inf.ed.ac.uk>
parents: 103
diff changeset
16 split -l 3000 --filter="gzip -c -7 --keep | \
101
e2e64c3d763e bug4 fixed, but that created a new, earlier bug
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
17 tee >(wc -c >> \
e2e64c3d763e bug4 fixed, but that created a new, earlier bug
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
18 $res/merge_{#}.log)" \
e2e64c3d763e bug4 fixed, but that created a new, earlier bug
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
19 {} > {}.gz && \
e2e64c3d763e bug4 fixed, but that created a new, earlier bug
Henry S. Thompson <ht@inf.ed.ac.uk>
parents:
diff changeset
20 rm {}'