Fix compiler error on OS X 10.6 #1465
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This builds on the prior change to extricate builtin command documentation out of the source and into an external text file for each command. Which is the next step into converting the documentation to a modern markup language. Related #507
This is to facilitate extracting the command documentation into text files separate from the source code. Related #507
Move the libcmd builtin documentation into files separate from the source code. Related #507
Move the code for `b_readonly()` into its own module. Refactor it so that the `readonly` and `export` commands only share the `setall()` function.
This also moves `b_unalias()` into the same module. TBD is refactoring the code to move the option parsing out of the `unall()` function and into the respective `b_*()` functions.
Builds on Copr and OBS were failing because `print_scan()` function definition was not prefixed with `static_fn`.
Recently builtins were moved into separate modules. It broke builds in Copr and OBS. `print_scan()` function should be declared non-static as it's used in more than one file. Fix made in previous commit was incorrect.
Move each of the `bg`, `fg`, and `disown` implementations into separate modules.
Move each command implementation into its own module.
Separate the break/continue and return/exit implementations into separate modules to improve clarity. This increases the size of the binary by a few words but that is well worth it for the improved clarity.
Separate the `cd` and `pwd` implementations into separate modules to improve code clarity.
With this change all special builtins are now in their own source module. More importantly this eliminates a lot of cleverness that obfuscates the actual behavior.
Update fallback version to `2020.0.0-beta1`
We want to be able to separate `b_trap()` and `b_kill()` into separate source modules. This is a step towards doing so.
The default `od` output format varies depending on the platform. Force output of octal bytes. Also, eliminate the `head -n1` and tighten up the matching regex.
Recent changes require a `ninja install` step before `meson test`. Update the documentation.
The core of this change is moving this statement after the `err:` label:
if (isfun && fd >= 0) sh_close(fd);
While fixing that I also decided to refactor the function to make the
primary logic of `can_execute()` clearer.
Coverity CID#353337
The default `od` output format varies depending on the platform. Force output of octal bytes. Also, eliminate the `head -n1` and tighten up the matching regex. This is the same fix as commit 031d73a but for a different unit test.
This fixes one, false positive, failure on Cygwin and simplifies and clarifies that test logic.
This change is a hack. It will be moot once the remaining three uses of optget() are modified to use optget_long(). But it's a straightforward way to get the correct version string when running `ksh --version`. Fixes #1456
The `getconf` command hasn't been a builtin for a long time. No known, non AST implementation, supports `getconf LIBPATH`. So remove that pointless test code.
Some platforms (e.g., OpenBSD) allow pretty much any locale string. And it is not clear why it is the responsibility of ksh to issue a diagnostic if an invalid locale is specified by the user. Or why we should be testing for this situation. Even ksh93u+, which uses the AST locale subsystem, doesn't produce a diagnostic message when doing `LC_CTYPE=xyz.UTF-8` on OpenBSD. This is a follow up to commit 6b6146e.
Also, simplify the affected test by removing extraneous items.
The test being removed only passes because of the trailing `exit 0`. Even with ksh93u+ the preceding two commands fail on all platforms I have access to.
I noticed this while reviewing OpenSuse patches that included this, pointless, change: -const char e_crondir[] = "/usr/spool/cron/atjobs"; +const char e_crondir[] = "/var/spool/cron";
This is adapted from the ksh93-optimizeleak.dif included with the OpenSuse ksh source. I confirmed the inner condition is true some of the time and the early return does not break any unit tests.
The old AST team implementation of an improved `cd` command has several problems when enabled for every interactive shell. Not least of which is that the approach of using an alias does not allow it to be, easily, overriden by a user defined `cd` function. It also fails to preserve some builtin `cd` command behavior. There may be other problems this doesn't fix. This is an experiment to see if we can augment commands like `cd` without having to write all behavior in the C language. Something other shells, like fish, have managed for more than a decade. Fixes #1445
Don't use global vars for the private data used by the cd related functions (pushd, nextd, etc.). Move them into the `.sh` namespace. Related #1445
People (and distros) have been building ksh from this project using the default "debug" build type since we switched from Nmake to Meson. That results in a ksh program whose performance is far from optimal. So change the default build type. If a debugging enabled binary is needed, such as in a CI environment, it must be requested explicitly. Related #1449
The typo is harmless since the string is not used for anything but fix it since it is wrong and confusing.
Mac OS X 10.6 (Snow Leopard) causes symbol `ERROR_dictionary` in src/lib/libdll/dlfcn.c to be evaluated. Since that symbol isn't defined it results in a compilation error. Add the identity mapping that is present in the rest of the source code. Fixes #1462
|
@ryandesign I booted my ancient OS X 10.6 VM but it isn't able to install Meson using the Homebrew environment it already had. Sorry, I know that using Homebrew rather than MacPorts is heretical. 😄 Is it possible for you to verify that this change fixes the build for you? If not I'll just merge it and you can then report back if it didn't fix the build for you. |
gordonwoodhull
force-pushed
the
master
branch
from
8cf92b2
to
683bccf
Compare
|
Why haven't the new owners of this project closed this pull-request? Could it be because they aren't competent to manage this project? Nah, that can't possibly be the reason. /snark I also wonder what they intend to do with issue #1462 that resulted in this P.R. |
|
In any case, the PR is only relevant to ksh2020. |
|
@krader1961, we at AT&T are not actively maintaining this repo - everyone should fork from here if they want to continue development. Thanks for pointing out that this PR should be closed as part of #1466. I hope that you will take your fork of ksh to a new repo, or that others who prefer ksh2020 will continue your work if you are no longer interested. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Mac OS X 10.6 (Snow Leopard) causes symbol
ERROR_dictionaryinsrc/lib/libdll/dlfcn.c to be evaluated. Since that symbol isn't defined it
results in a compilation error. Add the identity mapping that is present
in the rest of the source code.
Fixes #1462