blob: ed27875f6defbbcc972fbfdf329cd5c9a037a418 (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
|
#! /bin/sh
# Repack the boost tarball as xz (much faster to unpack) and remove
# a lot of needless files such as generated html docs.
tarball="$1"
if test -z "$tarball" -o ! -f "$tarball"; then
echo "Usage: $0 <tarball>"
exit 1
fi
tmpdir=$(mktemp -d)
if ! test -d "$tmpdir"; then
echo mktemp failed
exit 1
fi
echo Unpacking "$tarball" ...
tar x -C "$tmpdir" -f "$tarball"
if test $? -ne 0; then
echo tar x failed
rm -rf "$tmpdir"
exit 1
fi
echo Removing unnecessary files ...
find "$tmpdir" \( -name doc -o -name test -o -name example \) -type d -prune -print0 | xargs -0 rm -r
if test $? -ne 0; then
echo file removal failed
rm -rf "$tmpdir"
exit 1
fi
name="$(basename "$tarball" | sed 's/\.tar.*$//').tar.xz"
dir=$(ls "$tmpdir")
echo Creating "$name" ...
# To make the tarball reproducible, use a timestamp of a file inside the tarball (they all seem to have the same mtime).
if ! test -f "$tmpdir/$dir/README.md"; then
echo timestamp retrieval failed, check the script
rm -rf "$tmpdir"
exit 1
fi
# Many of the options are to make the tarball reproducible.
LC_ALL=C tar c -C "$tmpdir" --xz -f "$(pwd)/$name" --format=gnu --sort=name --owner=0 --group=0 --mode=go=rX,u=rwX --mtime "$tmpdir/$dir/README.md" "$dir"
if test $? -ne 0; then
echo tar c failed
rm -rf "$tmpdir"
exit 1
fi
echo Cleaning up ...
rm -rf "$tmpdir"
sha256sum "$name"
echo Done.
exit 0
|