Mercurial > hg > gitweb
comparison tests/README @ 2985:4cdb68d7eb92
patch queue: portability.notes
| author | Danek Duvall <danek.duvall@sun.com> |
|---|---|
| date | Mon, 21 Aug 2006 14:13:27 -0700 |
| parents | 1c66aad252f9 |
| children | f63667f694de |
comparison
equal
deleted
inserted
replaced
| 2984:a9d7a43fb3f0 | 2985:4cdb68d7eb92 |
|---|---|
| 29 - diff will show the current time | 29 - diff will show the current time |
| 30 | 30 |
| 31 use hg diff | sed -e "s/\(+++ [a-zA-Z0-9_/.-]*\).*/\1/" \ | 31 use hg diff | sed -e "s/\(+++ [a-zA-Z0-9_/.-]*\).*/\1/" \ |
| 32 -e "s/\(--- [a-zA-Z0-9_/.-]*\).*/\1/" | 32 -e "s/\(--- [a-zA-Z0-9_/.-]*\).*/\1/" |
| 33 to strip dates | 33 to strip dates |
| 34 | |
| 35 You also need to be careful that the tests are portable from one platform | |
| 36 to another. You're probably working on Linux, where the GNU toolchain has | |
| 37 more (or different) functionality than on MacOS, *BSD, Solaris, AIX, etc. | |
| 38 While testing on all platforms is the only sure-fire way to make sure that | |
| 39 you've written portable code, here's a list of problems that have been | |
| 40 found and fixed in the tests. Another, more comprehensive list may be | |
| 41 found in the GNU Autoconf manual, online here: | |
| 42 | |
| 43 http://www.gnu.org/software/autoconf/manual/html_node/Portable-Shell.html | |
| 44 | |
| 45 sh: | |
| 46 | |
| 47 The Bourne shell is a very basic shell. /bin/sh on Linux is typically | |
| 48 bash, which even in Bourne-shell mode has many features that Bourne shells | |
| 49 on other Unix systems don't have (and even on Linux /bin/sh isn't | |
| 50 guaranteed to be bash). You'll need to be careful about constructs that | |
| 51 seem ubiquitous, but are actually not available in the least common | |
| 52 denominator. While using another shell (ksh, bash explicitly, posix shell, | |
| 53 etc.) explicitly may seem like another option, these may not exist in a | |
| 54 portable location, and so are generally probably not a good idea. You may | |
| 55 find that rewriting the test in python will be easier. | |
| 56 | |
| 57 - don't use pushd/popd; save the output of "pwd" and use "cd" in place of | |
| 58 the pushd, and cd back to the saved pwd instead of popd. | |
| 59 | |
| 60 - don't use math expressions like let, (( ... )), or $(( ... )); use "expr" | |
| 61 instead. | |
| 62 | |
| 63 grep: | |
| 64 | |
| 65 - don't use the -q option; redirect stdout to /dev/null instead. | |
| 66 | |
| 67 - don't use extended regular expressions with grep; use egrep instead, and | |
| 68 don't escape any regex operators. | |
| 69 | |
| 70 sed: | |
| 71 | |
| 72 - make sure that the beginning-of-line matcher ("^") is at the very | |
| 73 beginning of the expression -- it may not be supported inside parens. | |
| 74 | |
| 75 echo: | |
| 76 | |
| 77 - echo may interpret "\n" and print a newline; use printf instead if you | |
| 78 want a literal "\n" (backslash + n). | |
| 79 | |
| 80 false: | |
| 81 | |
| 82 - false is guaranteed only to return a non-zero value; you cannot depend on | |
| 83 it being 1. On Solaris in particular, /bin/false returns 255. Rewrite | |
| 84 your test to not depend on a particular return value, or create a | |
| 85 temporary "false" executable, and call that instead. | |
| 86 | |
| 87 diff: | |
| 88 | |
| 89 - don't use the -N option. There's no particularly good workaround short | |
| 90 of writing a reasonably complicated replacement script, but substituting | |
| 91 gdiff for diff if you can't rewrite the test not to need -N will probably | |
| 92 do. |