blob: 4d53e80ab3d7a58a957872a9562160610c749998 (
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
|
#!/bin/sh
set -e
# These are text files.
text=`{ find . -name CVS -prune -o \
-name .cvsignore -prune -o \
-name .svn -prune -o \
-name .git -prune -o \
-name LATEST.VER -prune -o \
-name CHECKLST.txt -prune -o \
-name mksrcarc.sh -prune -o \
-name '*.chm' -prune -o \
-name '*.cur' -prune -o \
-type f -print | sed 's/^\.\///'; } | \
grep -ivE 'test/.*\.txt|MODULE|website.url' | grep -vF .ico | grep -vF .icns`
# These are files which I'm _sure_ should be treated as text, but
# which zip might complain about, so we direct its moans to
# /dev/null! Apparently its heuristics are doubtful of UTF-8 text
# files.
bintext=test/*.txt
# These are actual binary files which we don't want transforming.
bin=`{ ls -1 windows/*.ico windows/website.url; \
find . -name '*.chm' -print -o -name '*.cur' -print; }`
verbosely() {
echo "$@"
"$@"
}
verbosely zip -l putty-src.zip $text
verbosely zip -l putty-src.zip $bintext
verbosely zip putty-src.zip $bin
|
