PWN-Hunter/android_external_toybox
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
50c0130d6c
android_external_toybox/tests/dd.test

129 lines
5.1 KiB
Plaintext
Raw Normal View History

I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
#!/bin/bash
# Copyright 2013 Robin Mittal <robinmittal.it@gmail.com>
# Copyright 2013 Divya Kothari <divya.s.kothari@gmail.com>
Add tests.
2023-07-23 09:18:31 -07:00
# TODO test skip= for seekable and nonseekable input.
# TODO bs overrides ibs= and obs=, and disables block aggregation
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
[ -f testing.sh ] && . testing.sh
# 'dd' command, stderr prints redirecting to /dev/null
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
exec 2>/dev/null
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
#testing "name" "command" "result" "infile" "stdin"
dd improvements. Support hex (requested by users). Support status=noxfer and status=none (requested by users). Fix status output. Clarify large numbers in --help output. Use O_TRUNC rather than ftruncate unless we're also seeking. New tests. Also partial cleanup (reuse of existing code, removal of non-GLOBALS globals, and merge dd_main and do_dd).
2016-07-07 21:48:33 -07:00
# Test suffixed number parsing; `count` is representative.
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "count=2" "if=input count=2 ibs=1" "hi" "high\n" ""
testcmd "count= 2" "if=input 'count= 2' ibs=1" "hi" "high\n" ""
toyonly testcmd "count=0x2" "if=input 'count=0x2' ibs=1" "hi" "high\n" ""
testcmd 'posix 2x3x4' "if=input count=2x3x4 ibs=1" \
Add test for dd val=2x3x4 posix syntax. I didn't add it to dd's help because $((MATH)) exists in modern shells, but coreutils' mailing list got a bug report about having broken it, so somebody's script somewhere still cares...
2023-06-14 14:45:24 -07:00
'abcdefghijklmnopqrstuvwx' 'abcdefghijklmnopqrstuvwxyz' ''
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "count=-2" "if=input 'count=-2' ibs=1 || echo errored" "errored\n" "" ""
dd improvements. Support hex (requested by users). Support status=noxfer and status=none (requested by users). Fix status output. Clarify large numbers in --help output. Use O_TRUNC rather than ftruncate unless we're also seeking. New tests. Also partial cleanup (reuse of existing code, removal of non-GLOBALS globals, and merge dd_main and do_dd).
2016-07-07 21:48:33 -07:00
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "if=(file)" "if=input" "I WANT\n" "I WANT\n" ""
testcmd "of=(file)" "of=file && cat file" "I WANT\n" "" "I WANT\n"
testcmd "if=file of=file" "if=input of=foo && cat foo && rm -f foo" \
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
"I WANT\n" "I WANT\n" ""
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "if=file | dd of=file" "if=input | dd of=foo && cat foo && rm -f foo" \
"I WANT\n" "I WANT\n" ""
testcmd "(stdout)" '' "I WANT\n" "" "I WANT\n"
testcmd "sync,noerror" \
"if=input of=outFile seek=8860 bs=1M conv=sync,noerror &&
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
stat -c \"%s\" outFile && rm -f outFile" "9291431936\n" "I WANT\n" ""
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "if=file of=(null)" "if=input of=/dev/null && echo 'yes'" "yes\n" \
"I WANT\n" ""
testcmd "with if of bs" \
"if=/dev/zero of=sda.txt bs=512 count=1 &&
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
stat -c '%s' sda.txt && rm -f sda.txt" "512\n" "" ""
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "with if of ibs obs" \
"if=/dev/zero of=sda.txt ibs=512 obs=256 count=1 &&
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
stat -c '%s' sda.txt && rm -f sda.txt" "512\n" "" ""
Add tests.
2023-07-23 09:18:31 -07:00
// TODO check block size of transactions
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "with if of ibs obs count" \
"if=/dev/zero of=sda.txt ibs=512 obs=256 count=3 &&
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
stat -c '%s' sda.txt && rm -f sda.txt" "1536\n" "" ""
ln -s input softlink
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "if=softlink" "if=softlink" "I WANT\n" "I WANT\n" ""
rm -f softlink
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
ln -s file softlink
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "if=file of=softlink" "if=input of=softlink &&
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
[ -f file -a -L softlink ] && cat softlink" "I WANT\n" "I WANT\n" ""
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
rm -f softlink file
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "if=file of=file (same file)" "if=input of=input &&
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
[ -f input ] && cat input && echo 'yes'" "yes\n" "I WANT\n" ""
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "same file notrunc" \
"if=input of=input conv=notrunc && cat input" "I WANT\n" "I WANT\n" ""
Add tests.
2023-07-23 09:18:31 -07:00
testcmd "seek truncate" \
'of=input bs=3 count=2 seek=7 oflag=seek_bytes && cat input' \
'1234567ABCDEF' '1234567890abcdefghijklmnopqrstuvwxyz' 'ABCDEFGHIJK'
testcmd "seek notrunc" \
'of=input bs=3 count=2 seek=7 conv=notrunc oflag=seek_bytes && cat input' \
'1234567ABCDEFdefghijklmnopqrstuvwxyz' '1234567890abcdefghijklmnopqrstuvwxyz'\
'ABCDEFGHIJK'
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "with ibs obs bs" "ibs=2 obs=5 bs=9" "I WANT\n" "" "I WANT\n"
testcmd "with ibs obs bs if" "ibs=2 obs=5 bs=9 if=input" "I WANT\n" "I WANT\n"\
""
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "with ibs obs count" "ibs=1 obs=1 count=1" "I" "" "I WANT\n"
testcmd "with ibs obs count if" "ibs=1 obs=1 count=3 if=input" "I W" "I WANT\n"\
""
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "with count" "count=1" "I WANT\n" "" "I WANT\n"
testcmd "with count if" "count=1 if=input" "I WANT\n" "I WANT\n" ""
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "with skip" "skip=0" "I WANT\n" "" "I WANT\n"
testcmd "with skip if" "skip=0 if=input" "I WANT\n" "I WANT\n" ""
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "with seek" "seek=0" "I WANT\n" "" "I WANT\n"
testcmd "with seek if" "seek=0 if=input" "I WANT\n" "I WANT\n" ""
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
# Testing only 'notrunc', 'noerror', 'fsync', 'sync'
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "conv=notrunc" "conv=notrunc" "I WANT\n" "" "I WANT\n"
testcmd "conv=notrunc with IF" "conv=notrunc if=input" "I WANT\n" \
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
"I WANT\n" ""
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "conv=noerror" "conv=noerror" "I WANT\n" "" "I WANT\n"
testcmd "conv=noerror with IF" "conv=noerror if=input" "I WANT\n" \
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
"I WANT\n" ""
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "conv=fsync" "conv=fsync" "I WANT\n" "" "I WANT\n"
testcmd "conv=fsync with IF" "conv=fsync if=input" "I WANT\n" \
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
"I WANT\n" ""
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "conv=sync" "conv=sync | head -n 1" "I WANT\n" "" "I WANT\n"
testcmd "conv=sync with IF" "conv=sync if=input | head -n 1" "I WANT\n" \
I have developed few testsuite for toybox commands - ls, ln, rm, mv, printf, dd, renice.
2014-06-26 05:25:20 -07:00
"I WANT\n" ""
dd improvements. Support hex (requested by users). Support status=noxfer and status=none (requested by users). Fix status output. Clarify large numbers in --help output. Use O_TRUNC rather than ftruncate unless we're also seeking. New tests. Also partial cleanup (reuse of existing code, removal of non-GLOBALS globals, and merge dd_main and do_dd).
2016-07-07 21:48:33 -07:00
# status=noxfer|none
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "status=noxfer" "if=input status=noxfer ibs=1 2>&1" \
"input\n6+0 records in\n0+1 records out\n" "input\n" ""
testcmd "status=none" "if=input status=none ibs=1 2>&1" \
"input\n" "input\n" ""
dd: fix seek= on stdout.
2019-07-09 09:49:12 -07:00
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testing "seek stdout" "yes | dd bs=8 seek=2 count=1 > out && xxd -p out" \
dd: fix seek= on stdout.
2019-07-09 09:49:12 -07:00
"00000000000000000000000000000000790a790a790a790a\n" "" ""
dd: iflags, oflags, fix ^C, fix the fundamental loop. Investigating why the toybox tar tests fail on Android with toybox dd, I realized I was rewriting a part of dd I'd rewritten before! This is a re-send of my 2019-02-22 patch, rebased against the current ToT... This patch was originally motivated because after suggesting to the author of https://stackoverflow.com/questions/17157820/access-vdsolinux/54797221#54797221 that he could tell dd to work in bytes rather than blocks, I realized that our dd doesn't actually support that. But the rewrite of the main loop is necessary to fix the incorrect output from the dd calls in the tar test. Without this patch, `yes | dd bs=65536 count=1 > fweep` basically gives random output, based on how many bytes the pipe feels like giving you in your first read. (As far as I know, dd *without* bs= was fine, but I can't guarantee that that's true, just that I haven't seen it fail.) Also switch to TAGGED_ARRAY and comma_* for conv rather than add two more copies of an undesired idiom. It turned out -- contrary to the belief of cp(1) -- that comma_scan isn't suitable for this because of its magic handling of "no" prefixes. (It's actually harmless in cp because none of the --preserve options begin with "no", but some dd options do.) To this end, comma_remove is a less-magic comma_scan. I've also changed an `if` to a `while` because other implementations allow things like `--preserve=mode,mode` or `conv=sync,sync`. (If we decide this is a bug rather than a feature, we should at least fix the error message to be clear that we're rejecting the *duplication*, not the option itself.) I've also fixed the ^C behavior by simply adding a direct SIGINT handler rather than trying to be clever inside the read loop (which is why we weren't handling the SIGINT until the read returned). I've also removed `strstarteq` and just added the '=' to each literal when calling regular `strstart`. Plus basic tests.
2019-07-09 16:08:16 -07:00
# Duplicated options are fine.
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "conv=sync,sync" "conv=sync,sync | head -n 1" "I WANT\n" "" "I WANT\n"
dd: iflags, oflags, fix ^C, fix the fundamental loop. Investigating why the toybox tar tests fail on Android with toybox dd, I realized I was rewriting a part of dd I'd rewritten before! This is a re-send of my 2019-02-22 patch, rebased against the current ToT... This patch was originally motivated because after suggesting to the author of https://stackoverflow.com/questions/17157820/access-vdsolinux/54797221#54797221 that he could tell dd to work in bytes rather than blocks, I realized that our dd doesn't actually support that. But the rewrite of the main loop is necessary to fix the incorrect output from the dd calls in the tar test. Without this patch, `yes | dd bs=65536 count=1 > fweep` basically gives random output, based on how many bytes the pipe feels like giving you in your first read. (As far as I know, dd *without* bs= was fine, but I can't guarantee that that's true, just that I haven't seen it fail.) Also switch to TAGGED_ARRAY and comma_* for conv rather than add two more copies of an undesired idiom. It turned out -- contrary to the belief of cp(1) -- that comma_scan isn't suitable for this because of its magic handling of "no" prefixes. (It's actually harmless in cp because none of the --preserve options begin with "no", but some dd options do.) To this end, comma_remove is a less-magic comma_scan. I've also changed an `if` to a `while` because other implementations allow things like `--preserve=mode,mode` or `conv=sync,sync`. (If we decide this is a bug rather than a feature, we should at least fix the error message to be clear that we're rejecting the *duplication*, not the option itself.) I've also fixed the ^C behavior by simply adding a direct SIGINT handler rather than trying to be clever inside the read loop (which is why we weren't handling the SIGINT until the read returned). I've also removed `strstarteq` and just added the '=' to each literal when calling regular `strstart`. Plus basic tests.
2019-07-09 16:08:16 -07:00
# _bytes options
Convert dd tests to testcmd and replace $obs with an exec redirect. (Each command.test runs as a separate child process now.)
2023-06-14 16:38:05 -07:00
testcmd "iflag=count_bytes" "if=input count=2 ibs=4096 iflag=count_bytes" "hi"\
"high" ""
testcmd "iflag=skip_bytes" "if=input skip=2 ibs=4096 iflag=skip_bytes" "gh" \
"high" ""
testcmd "oflag=seek_bytes" \
"if=input of=output seek=2 obs=4096 oflag=seek_bytes status=none && \
dd: iflags, oflags, fix ^C, fix the fundamental loop. Investigating why the toybox tar tests fail on Android with toybox dd, I realized I was rewriting a part of dd I'd rewritten before! This is a re-send of my 2019-02-22 patch, rebased against the current ToT... This patch was originally motivated because after suggesting to the author of https://stackoverflow.com/questions/17157820/access-vdsolinux/54797221#54797221 that he could tell dd to work in bytes rather than blocks, I realized that our dd doesn't actually support that. But the rewrite of the main loop is necessary to fix the incorrect output from the dd calls in the tar test. Without this patch, `yes | dd bs=65536 count=1 > fweep` basically gives random output, based on how many bytes the pipe feels like giving you in your first read. (As far as I know, dd *without* bs= was fine, but I can't guarantee that that's true, just that I haven't seen it fail.) Also switch to TAGGED_ARRAY and comma_* for conv rather than add two more copies of an undesired idiom. It turned out -- contrary to the belief of cp(1) -- that comma_scan isn't suitable for this because of its magic handling of "no" prefixes. (It's actually harmless in cp because none of the --preserve options begin with "no", but some dd options do.) To this end, comma_remove is a less-magic comma_scan. I've also changed an `if` to a `while` because other implementations allow things like `--preserve=mode,mode` or `conv=sync,sync`. (If we decide this is a bug rather than a feature, we should at least fix the error message to be clear that we're rejecting the *duplication*, not the option itself.) I've also fixed the ^C behavior by simply adding a direct SIGINT handler rather than trying to be clever inside the read loop (which is why we weren't handling the SIGINT until the read returned). I've also removed `strstarteq` and just added the '=' to each literal when calling regular `strstart`. Plus basic tests.
2019-07-09 16:08:16 -07:00
xxd -p output" "000030313233\n" "0123" ""
Add tests.
2023-07-23 09:18:31 -07:00
# The sleep decouples input blocks to force short reads for conv=sync to pad
testing 'sync padding ticks down count' \
'for i in one two three four five "$(seq 1 100)"
do echo "$i"; sleep .1; done | dd bs=1024 count=5 conv=sync | sha1sum' \
'02dcf1f497ccbe940f57818dfc34f2d0def8b3f9 -\n' '' ''
Reference in New Issue Copy Permalink