perlos2 - Perl under OS/2, DOS, Win0.3*, Win0.95 and WinNT.
man perlos2 view perl perlos2 explorer perlos2.html info perlos2
to list some (not all may be available simultaneously), or it may be read as is: either as README.os2, or pod/perlos2.pod.
To read the .INF version of documentation (very recommended) outside of OS/2, one needs an IBM's reader (may be available on IBM ftp sites (?) (URL anyone?)) or shipped with PC DOS 7.0 and IBM's Visual Age C++ 3.5.
A copy of a Win* viewer is contained in the ``Just add OS/2 Warp'' package
ftp://ftp.software.ibm.com/ps/products/os2/tools/jaow/jaow.zip
in ?:\JUST_ADD\view.exe. This gives one an access to EMX's .INF docs as well (text form is available in /emx/doc in EMX's distribution). There is also a different viewer named xview.
Note that if you have lynx.exe or netscape.exe installed, you can follow WWW links from this document in .INF format. If you have EMX docs installed correctly, you can follow library links (you need to have "view emxbook" working by setting "EMXBOOK" environment variable as it is described in EMX docs).
The current state is quite close to this target. Known limitations:
While using the standard perl.exe from a text-mode window is possible too, I have seen cases when this causes degradation of the system stability. Using perl__.exe avoids such a degradation.
Please keep this list up-to-date by informing me about other items.
Note that not all features of Perl are available under these environments. This depends on the features the extender - most probably RSX - decided to implement.
Only the latest runtime is supported, currently "0.9d fix 03". Perl may run under earlier versions of EMX, but this is not tested.
One can get different parts of EMX from, say
http://www.leo.org/pub/comp/os/os2/leo/gnu/emx+gcc/ http://powerusersbbs.com/pub/os2/dev/ [EMX+GCC Development] http://hobbes.nmsu.edu/pub/os2/dev/emx/v0.9d/
The runtime component should have the name emxrt.zip.
NOTE. When using emx.exe/rsx.exe, it is enough to have them on your path. One does not need to specify them explicitly (though this
emx perl_.exe -de 0
will work as well.)
Having RSX and the latest sh.exe one gets a fully functional *nix-ish environment under DOS, say, "fork", `` and pipe-"open" work. In fact, MakeMaker works (for static build), so one can have Perl development environment under DOS.
One can get RSX from, say
ftp://ftp.cdrom.com/pub/os2/emx09c/contrib ftp://ftp.uni-bielefeld.de/pub/systems/msdos/misc ftp://ftp.leo.org/pub/comp/os/os2/leo/devtools/emx+gcc/contrib
Contact the author on "rainer@mathematik.uni-bielefeld.de".
The latest sh.exe with DOS hooks is available in
http://www.ilyaz.org/software/os2/
as sh_dos.zip or under similar names starting with "sh", "pdksh" etc.
Note that if you do not plan to build the perl itself, it may be possible to fool EMX to truncate file names. This is not supported, read EMX docs to see how to do it.
For best results use EMX pdksh. The standard binary (5.2.14 or later) runs under DOS (with RSX) as well, see
http://www.ilyaz.org/software/os2/
perl foo.pl arg1 arg2 arg3
If you want to specify perl options "-my_opts" to the perl itself (as opposed to your program), use
perl -my_opts foo.pl arg1 arg2 arg3
Alternately, if you use OS/2-ish shell, like CMD or 4os2, put the following at the start of your perl script:
extproc perl -S -my_opts
rename your program to foo.cmd, and start it by typing
foo arg1 arg2 arg3
Note that because of stupid OS/2 limitations the full path of the perl script is not available when you use "extproc", thus you are forced to use "-S" perl switch, and your script should be on the "PATH". As a plus side, if you know a full path to your script, you may still start it with
perl ../../blah/foo.cmd arg1 arg2 arg3
(note that the argument "-my_opts" is taken care of by the "extproc" line in your script, see ""extproc" on the first line").
To understand what the above magic does, read perl docs about "-S" switch - see perlrun, and cmdref about "extproc":
view perl perlrun man perlrun view cmdref extproc help extproc
or whatever method you prefer.
There are also endless possibilities to use executable extensions of 4os2, associations of WPS and so on... However, if you use *nixish shell (like sh.exe supplied in the binary distribution), you need to follow the syntax specified in ``Switches'' in perlrun.
Note that -S switch supports scripts with additional extensions .cmd, .btm, .bat, .pl as well.
Note however that to use some of these operators you need to have a sh-syntax shell installed (see ``Pdksh'', ``Frequently asked questions''), and perl should be able to find it (see ``PERL_SH_DIR'').
The cases when the shell is used are:
For the sake of speed for a common case, in the above algorithms backslashes in the command name are not considered as shell metacharacters.
Perl starts scripts which begin with cookies "extproc" or "#!" directly, without an intervention of shell. Perl uses the same algorithm to find the executable as pdksh: if the path on "#!" line does not work, and contains "/", then the directory part of the executable is ignored, and the executable is searched in . and on "PATH". To find arguments for these scripts Perl uses a different algorithm than pdksh: up to 3 arguments are recognized, and trailing whitespace is stripped.
If a script does not contain such a cooky, then to avoid calling sh.exe, Perl uses the same algorithm as pdksh: if $ENV{EXECSHELL} is set, the script is given as the first argument to this command, if not set, then "$ENV{COMSPEC} /c" is used (or a hardwired guess if $ENV{COMSPEC} is not set).
When starting scripts directly, Perl uses exactly the same algorithm as for the search of script given by -S command-line option: it will look in the current directory, then on components of $ENV{PATH} using the following order of appended extensions: no extension, .cmd, .btm, .bat, .pl.
Note that Perl will start to look for scripts only if OS/2 cannot start the specified application, thus "system 'blah'" will not look for a script if there is an executable file blah.exe anywhere on "PATH". In other words, "PATH" is essentially searched twice: once by the OS for an executable, then by Perl for scripts.
Note also that executable files on OS/2 can have an arbitrary extension, but .exe will be automatically appended if no dot is present in the name. The workaround is as simple as that: since blah. and blah denote the same file (at list on FAT and HPFS file systems), to start an executable residing in file n:/bin/blah (no extension) give an argument "n:/bin/blah." (dot appended) to system().
Perl will start PM programs from VIO (=text-mode) Perl process in a separate PM session; the opposite is not true: when you start a non-PM program from a PM Perl process, Perl would not run it in a separate session. If a separate session is desired, either ensure that shell will be used, as in "system 'cmd /c myprog'", or start it using optional arguments to system() documented in "OS2::Process" module. This is considered to be a feature.
If everything else fails, you need to build a stand-alone DLL for perl. Contact me, I did it once. Sockets would not work, as a lot of other stuff.
find "pattern" file find pattern file
are not equivalent; find cannot be started directly using the above API. One needs a way to surround the doublequotes in some other quoting construction, necessarily having an extra non-Unixish shell in between.
Use one of
system 'cmd', '/c', 'find "pattern" file'; `cmd /c 'find "pattern" file'`
This would start find.exe via cmd.exe via "sh.exe" via "perl.exe", but this is a price to pay if you want to use non-conforming program.
Note however, that you need to have unzip.exe on your path, and EMX environment running. The latter means that if you just installed EMX, and made all the needed changes to Config.sys, you may need to reboot in between. Check EMX runtime by running
emxrev
Binary installer also creates a folder on your desktop with some useful objects. If you need to change some aspects of the work of the binary installer, feel free to edit the file Perl.pkg. This may be useful e.g., if you need to run the installer many times and do not want to make many interactive changes in the GUI.
Things not taken care of by automatic binary installation:
perl -MConfig -le "print $INC{'Config.pm'}"
While most important values in this file are updated by the binary installer, some of them may need to be hand-edited. I know no such data, please keep me informed if you find one. Moreover, manual changes to the installed version may need to be accompanied by an edit of this file.
NOTE. Because of a typo the binary installer of 5.00305 would install a variable "PERL_SHPATH" into Config.sys. Please remove this variable and put "PERL_SH_DIR" instead.
Note that the extraction with the stored paths is still necessary (default with unzip, specify "-d" to pkunzip). However, you need to know where to extract the files. You need also to manually change entries in Config.sys to reflect where did you put the files. Note that if you have some primitive unzipper (like "pkunzip"), you may get a lot of warnings/errors during unzipping. Upgrade to "(w)unzip".
Below is the sample of what to do to reproduce the configuration on my machine. In VIEW.EXE you can press "Ctrl-Insert" now, and cut-and-paste from the resulting file - created in the directory you started VIEW.EXE from.
For each component, we mention environment variables related to each installation directory. Either choose directories to match your values of the variables, or create/append-to variables to take into account the directories.
unzip perl_exc.zip *.exe *.ico -d f:/emx.add/bin unzip perl_exc.zip *.dll -d f:/emx.add/dll
(have the directories with "*.exe" on PATH, and "*.dll" on LIBPATH);
unzip perl_aou.zip -d f:/emx.add/bin
(have the directory on PATH);
unzip perl_utl.zip -d f:/emx.add/bin
(have the directory on PATH);
unzip perl_mlb.zip -d f:/perllib/lib
If this directory is exactly the same as the prefix which was compiled into perl.exe, you do not need to change anything. However, for perl to find the library if you use a different path, you need to "set PERLLIB_PREFIX" in Config.sys, see ``PERLLIB_PREFIX''.
unzip perl_ste.zip -d f:/perllib/lib/site_perl/5.8.8/
Same remark as above applies. Additionally, if this directory is not one of directories on @INC (and @INC is influenced by "PERLLIB_PREFIX"), you need to put this directory and subdirectory ./os2 in "PERLLIB" or "PERL5LIB" variable. Do not use "PERL5LIB" unless you have it set already. See ``ENVIRONMENT'' in perl.
[Check whether this extraction directory is still applicable with the new directory structure layout!]
unzip perl_blb.zip -d f:/perllib/lib
Same remark as for perl_ste.zip.
unzip perl_man.zip -d f:/perllib/man
This directory should better be on "MANPATH". You need to have a working man to access these files.
unzip perl_mam.zip -d f:/perllib/man
This directory should better be on "MANPATH". You need to have a working man to access these files.
unzip perl_pod.zip -d f:/perllib/lib
This is used by the "perldoc" program (see perldoc), and may be used to generate HTML documentation usable by WWW browsers, and documentation in zillions of other formats: "info", "LaTeX", "Acrobat", "FrameMaker" and so on. [Use programs such as pod2latex etc.]
unzip perl_inf.zip -d d:/os2/book
This directory should better be on "BOOKSHELF".
unzip perl_sh.zip -d f:/bin
This is used by perl to run external commands which explicitly require shell, like the commands using redirection and shell metacharacters. It is also used instead of explicit /bin/sh.
Set "PERL_SH_DIR" (see ``PERL_SH_DIR'') if you move sh.exe from the above location.
Note. It may be possible to use some other sh-compatible shell (untested).
After you installed the components you needed and updated the Config.sys correspondingly, you need to hand-edit Config.pm. This file resides somewhere deep in the location you installed your perl library, find it out by
perl -MConfig -le "print $INC{'Config.pm'}"
You need to correct all the entries which look like file paths (they currently start with "f:/").
view perl view perl perlfunc view perl less view perl ExtUtils::MakeMaker
(currently the last two may hit a wrong location, but this may improve soon). Under Win* see ``SYNOPSIS''.
If you want to build the docs yourself, and have OS/2 toolkit, run
pod2ipf > perl.ipf
in /perllib/lib/pod directory, then
ipfc /inf perl.ipf
(Expect a lot of errors during the both steps.) Now move it on your BOOKSHELF path.
perldoc perlfunc perldoc less perldoc ExtUtils::MakeMaker
to access the perl documentation in the text form (note that you may get better results using perl manpages).
Alternately, try running pod2text on .pod files.
man perlfunc man 3 less man ExtUtils.MakeMaker
to access documentation for different components of Perl. Start with
man perl
Note that dot (.) is used as a package separator for documentation for packages, and as usual, sometimes you need to give the section - 3 above - to avoid shadowing by the less(1) manpage.
Make sure that the directory above the directory with manpages is on our "MANPATH", like this
set MANPATH=c:/man;f:/perllib/man
for Perl manpages in "f:/perllib/man/man1/" etc.
cd f:/perllib/lib/pod pod2html
After this you can direct your browser the file perl.html in this directory, and go ahead with reading docs, like this:
explore file:///f:/perllib/lib/pod/perl.html
Alternatively you may be able to get these docs prebuilt from CPAN.
gnupatch -p0 < os2\diff.configure sh Configure -des -D prefix=f:/perllib make make test make install make aout_test make aout_install
This puts the executables in f:/perllib/bin. Manually move them to the "PATH", manually move the built perl*.dll to "LIBPATH" (here for Perl DLL * is a not-very-meaningful hex checksum), and run
make installcmd INSTALLCMDDIR=d:/ir/on/path
Assuming that the "man"-files were put on an appropriate location, this completes the installation of minimal Perl system. (The binary distribution contains also a lot of additional modules, and the documentation in INF format.)
What follows is a detailed guide through these steps.
find --version sort --version
). You need the latest version of pdksh installed as sh.exe.
Check that you have BSD libraries and headers installed, and - optionally - Berkeley DB headers and libraries, and crypt.
Possible locations to get the files:
ftp://hobbes.nmsu.edu/os2/unix/ ftp://ftp.cdrom.com/pub/os2/unix/ ftp://ftp.cdrom.com/pub/os2/dev32/ ftp://ftp.cdrom.com/pub/os2/emx09c/
It is reported that the following archives contain enough utils to build perl: gnufutil.zip, gnusutil.zip, gnututil.zip, gnused.zip, gnupatch.zip, gnuawk.zip, gnumake.zip, gnugrep.zip, bsddev.zip and ksh527rt.zip (or a later version). Note that all these utilities are known to be available from LEO:
ftp://ftp.leo.org/pub/comp/os/os2/leo/gnu
Note also that the db.lib and db.a from the EMX distribution are not suitable for multi-threaded compile (even single-threaded flavor of Perl uses multi-threaded C RTL, for compatibility with XFree86-OS/2). Get a corrected one from
http://www.ilyaz.org/software/os2/db_mt.zip
If you have exactly the same version of Perl installed already, make sure that no copies or perl are currently running. Later steps of the build may fail since an older version of perl.dll loaded into memory may be found. Running "make test" becomes meaningless, since the test are checking a previous build of perl (this situation is detected and reported by lib/os2_base.t test). Do not forget to unset "PERL_EMXLOAD_SEC" in environment.
Also make sure that you have /tmp directory on the current drive, and . directory in your "LIBPATH". One may try to correct the latter condition by
set BEGINLIBPATH .\.
if you use something like CMD.EXE or latest versions of 4os2.exe. (Setting BEGINLIBPATH to just "." is ignored by the OS/2 kernel.)
Make sure your gcc is good for "-Zomf" linking: run "omflibs" script in /emx/lib directory.
Check that you have link386 installed. It comes standard with OS/2, but may be not installed due to customization. If typing
link386
shows you do not have it, do Selective install, and choose "Link object modules" in Optional system utilities/More. If you get into link386 prompts, press "Ctrl-C" to exit.
http://www.cpan.org/src/5.0 http://www.cpan.org/src/5.0/unsupported
If not, you may need to dig in the indices to find it in the directory of the current maintainer.
Quick cycle of developers release may break the OS/2 build time to time, looking into
http://www.cpan.org/ports/os2/
may indicate the latest release which was publicly released by the maintainer. Note that the release may include some additional patches to apply to the current source of perl.
Extract it like this
tar vzxf perl5.00409.tar.gz
You may see a message about errors while extracting Configure. This is because there is a conflict with a similarly-named file configure.
Change to the directory of extraction.
gnupatch -p0 < os2\diff.configure
You may also need to apply the patches supplied with the binary distribution of perl. It also makes sense to look on the perl5-porters mailing list for the latest OS/2-related patches (see <http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/>). Such patches usually contain strings "/os2/" and "patch", so it makes sense looking for these strings.
sh Configure -des -D prefix=f:/perllib
"prefix" means: where to install the resulting perl library. Giving correct prefix you may avoid the need to specify "PERLLIB_PREFIX", see ``PERLLIB_PREFIX''.
Ignore the message about missing "ln", and about "-c" option to tr. The latter is most probably already fixed, if you see it and can trace where the latter spurious warning comes from, please inform me.
Now
make
At some moment the built may die, reporting a version mismatch or unable to run perl. This means that you do not have . in your LIBPATH, so perl.exe cannot find the needed perl67B2.dll (treat these hex digits as line noise). After this is fixed the build should finish without a lot of fuss.
make test
All tests should succeed (with some of them skipped). If you have the same version of Perl installed, it is crucial that you have "." early in your LIBPATH (or in BEGINLIBPATH), otherwise your tests will most probably test the wrong version of Perl.
Some tests may generate extra messages similar to
However the test engine bleeds these message to screen in unexpected moments. Two messages of this kind should be present during testing.
To get finer test reports, call
perl t/harness
The report with io/pipe.t failing may look like this:
Failed Test Status Wstat Total Fail Failed List of failed ------------------------------------------------------------ io/pipe.t 12 1 8.33% 9 7 tests skipped, plus 56 subtests skipped. Failed 1/195 test scripts, 99.49% okay. 1/6542 subtests failed, 99.98% okay.
The reasons for most important skipped tests are:
Run
make install
It would put the generated files into needed locations. Manually put perl.exe, perl__.exe and perl___.exe to a location on your PATH, perl.dll to a location on your LIBPATH.
Run
make installcmd INSTALLCMDDIR=d:/ir/on/path
to convert perl utilities to .cmd files and put them on PATH. You need to put .EXE-utilities on path manually. They are installed in "$prefix/bin", here $prefix is what you gave to Configure, see Making.
If you use "man", either move the installed */man/ directories to your "MANPATH", or modify "MANPATH" to match the location. (One could have avoided this by providing a correct "manpath" option to ./Configure, or editing ./config.sh between configuring and making steps.)
make perl_
test and install by
make aout_test make aout_install
Manually put perl_.exe to a location on your PATH.
Note. The build process for "perl_" does not know about all the dependencies, so you should make sure that anything is up-to-date, say, by doing
make perl_dll
Building should proceed differently depending on whether the version of perl you install is already present and used on your system, or is a new version not yet used. The description below assumes that the version is new, so installing its DLLs and .pm files will not disrupt the operation of your system even if some intermediate steps are not yet fully working.
The other cases require a little bit more convoluted procedures. Below I suppose that the current version of Perl is 5.8.2, so the executables are named accordingly.
perl5.8.2 -MCPAN -e "install Bundle::OS2_default" < nul |& tee 00cpan_i_1
This may take a couple of hours on 1GHz processor (when run the first time). And this should not be necessarily a smooth procedure. Some modules may not specify required dependencies, so one may need to repeat this procedure several times until the results stabilize.
perl5.8.2 -MCPAN -e "install Bundle::OS2_default" < nul |& tee 00cpan_i_2 perl5.8.2 -MCPAN -e "install Bundle::OS2_default" < nul |& tee 00cpan_i_3
Even after they stabilize, some tests may fail.
Fix as many discovered bugs as possible. Document all the bugs which are not fixed, and all the failures with unknown reasons. Inspect the produced logs 00cpan_i_1 to find suspiciously skipped tests, and other fishy events.
Keep in mind that installation of some modules may fail too: for example, the DLLs to update may be already loaded by CPAN.pm. Inspect the "install" logs (in the example above 00cpan_i_1 etc) for errors, and install things manually, as in
cd $CPANHOME/.cpan/build/Digest-MD5-2.31 make install
Some distributions may fail some tests, but you may want to install them anyway (as above, or via "force install" command of "CPAN.pm" shell-mode).
Since this procedure may take quite a long time to complete, it makes sense to ``freeze'' your CPAN configuration by disabling periodic updates of the local copy of CPAN index: set "index_expire" to some big value (I use 365), then save the settings
CPAN> o conf index_expire 365 CPAN> o conf commit
Reset back to the default value 1 when you are finished.
( perl2ipf > perl.ipf ) |& tee 00ipf ipfc /INF perl.ipf |& tee 00inf
This produces the Perl docs online book "perl.INF". Install in on "BOOKSHELF" path.
Here is a possible workaround: create a toplevel Makefile.PL in $CPANHOME/.cpan/build/ with contents being (compare with ``Making executables with a custom collection of statically loaded extensions'')
use ExtUtils::MakeMaker; WriteMakefile NAME => 'dummy';
execute this as
perl_5.8.2.exe Makefile.PL <nul |& tee 00aout_c1 make -k all test <nul |& 00aout_t1
Again, this procedure should not be absolutely smooth. Some "Makefile.PL"'s in subdirectories may be buggy, and would not run as ``child'' scripts. The interdependency of modules can strike you; however, since non-XS modules are already installed, the prerequisites of most modules have a very good chance to be present.
If you discover some glitches, move directories of problematic modules to a different location; if these modules are non-XS modules, you may just ignore them - they are already installed; the remaining, XS, modules you need to install manually one by one.
After each such removal you need to rerun the "Makefile.PL"/"make" process; usually this procedure converges soon. (But be sure to convert all the necessary external C libraries from .lib format to .a format: run one of
emxaout foo.lib emximp -o foo.a foo.lib
whichever is appropriate.) Also, make sure that the DLLs for external libraries are usable with with executables compiled without "-Zmtd" options.
When you are sure that only a few subdirectories lead to failures, you may want to add "-j4" option to "make" to speed up skipping subdirectories with already finished build.
When you are satisfied with the results of tests, install the build C libraries for extensions:
make install |& tee 00aout_i
Now you can rename the file ./perl.exe generated during the last phase to perl_5.8.2.exe; place it on "PATH"; if there is an inter-dependency between some XS modules, you may need to repeat the "test"/"install" loop with this new executable and some excluded modules - until the procedure converges.
Now you have all the necessary .a libraries for these Perl modules in the places where Perl builder can find it. Use the perl builder: change to an empty directory, create a ``dummy'' Makefile.PL again, and run
perl_5.8.2.exe Makefile.PL |& tee 00c make perl |& tee 00p
This should create an executable ./perl.exe with all the statically loaded extensions built in. Compare the generated perlmain.c files to make sure that during the iterations the number of loaded extensions only increases. Rename ./perl.exe to perl_5.8.2.exe on "PATH".
When it converges, you got a functional variant of perl_5.8.2.exe; copy it to "perl_.exe". You are done with generation of the local Perl installation.
Actually, the log output of pod2ipf during the step 6 gives a very detailed info about which modules are loaded from which place; so you may use it as an additional verification tool.
Check that some temporary files did not make into the perl install tree. Run something like this
pfind . -f "!(/\.(pm|pl|ix|al|h|a|lib|txt|pod|imp|bs|dll|ld|bs|inc|xbm|yml|cgi|uu|e2x|skip|packlist|eg|cfg|html|pub|enc|all|ini|po|pot)$/i or /^\w+$/") | less
in the install tree (both top one and sitelib one).
Compress all the DLLs with lxlite. The tiny .exe can be compressed with "/c:max" (the bug only appears when there is a fixup in the last 6 bytes of a page (?); since the tiny executables are much smaller than a page, the bug will not hit). Do not compress "perl_.exe" - it would not work under DOS.
Include perl5.def so that people can relink the perl DLL preserving the binary compatibility, or can create compatibility DLLs. Include the diff files ("diff -pu old new") of fixes you did so that people can rebuild your version. Include perl5.map so that one can use remote debugging.
use ExtUtils::MakeMaker; WriteMakefile NAME => 'dummy';
perl_ Makefile.PL
make perl
(you may need to manually add "PERLTYPE=-DPERL_CORE" to this commandline on some versions of Perl; the symptom is that the command-line globbing does not work from OS/2 shells with the newly-compiled executable; check with
.\perl.exe -wle "print for @ARGV" *
).
make perl
will produce a customized executable.
If you fill comfortable with embedding interface (see perlembed), such things are easy to do repeating the steps outlined in ``Making executables with a custom collection of statically loaded extensions'', and doing more comprehensive edits to main() of perlmain.c. The people with little desire to understand Perl can just rename main(), and do necessary modification in a custom main() which calls the renamed function in appropriate time.
However, there is a third way: perl DLL exports the main() function and several callbacks to customize the search path. Below is a complete example of a ``Perl loader'' which
For best results compile the C file below with the same options as the Perl DLL. However, a lot of functionality will work even if the executable is not an EMX applications, e.g., if compiled with
gcc -Wall -DDOSISH -DOS2=1 -O2 -s -Zomf -Zsys perl-starter.c -DPERL_DLL_BASENAME=\"perl312F\" -Zstack 8192 -Zlinker /PM:VIO
Here is the sample C file:
#define INCL_DOS #define INCL_NOPM /* These are needed for compile if os2.h includes os2tk.h, not os2emx.h */ #define INCL_DOSPROCESS #include <os2.h>
#include "EXTERN.h" #define PERL_IN_MINIPERLMAIN_C #include "perl.h"
static char *me; HMODULE handle;
static void die_with(char *msg1, char *msg2, char *msg3, char *msg4) { ULONG c; char *s = " error: ";
DosWrite(2, me, strlen(me), &c); DosWrite(2, s, strlen(s), &c); DosWrite(2, msg1, strlen(msg1), &c); DosWrite(2, msg2, strlen(msg2), &c); DosWrite(2, msg3, strlen(msg3), &c); DosWrite(2, msg4, strlen(msg4), &c); DosWrite(2, "\r\n", 2, &c); exit(255); }
typedef ULONG (*fill_extLibpath_t)(int type, char *pre, char *post, int replace, char *msg); typedef int (*main_t)(int type, char *argv[], char *env[]); typedef int (*handler_t)(void* data, int which);
#ifndef PERL_DLL_BASENAME # define PERL_DLL_BASENAME "perl" #endif
static HMODULE load_perl_dll(char *basename) { char buf[300], fail[260]; STRLEN l, dirl; fill_extLibpath_t f; ULONG rc_fullname; HMODULE handle, handle1;
if (_execname(buf, sizeof(buf) - 13) != 0) die_with("Can't find full path: ", strerror(errno), "", ""); /* XXXX Fill `me' with new value */ l = strlen(buf); while (l && buf[l-1] != '/' && buf[l-1] != '\\') l--; dirl = l - 1; strcpy(buf + l, basename); l += strlen(basename); strcpy(buf + l, ".dll"); if ( (rc_fullname = DosLoadModule(fail, sizeof fail, buf, &handle)) != 0 && DosLoadModule(fail, sizeof fail, basename, &handle) != 0 ) die_with("Can't load DLL ", buf, "", ""); if (rc_fullname) return handle; /* was loaded with short name; all is fine */ if (DosQueryProcAddr(handle, 0, "fill_extLibpath", (PFN*)&f)) die_with(buf, ": DLL exports no symbol ", "fill_extLibpath", ""); buf[dirl] = 0; if (f(0 /*BEGINLIBPATH*/, buf /* prepend */, NULL /* append */, 0 /* keep old value */, me)) die_with(me, ": prepending BEGINLIBPATH", "", ""); if (DosLoadModule(fail, sizeof fail, basename, &handle1) != 0) die_with(me, ": finding perl DLL again via BEGINLIBPATH", "", ""); buf[dirl] = '\\'; if (handle1 != handle) { if (DosQueryModuleName(handle1, sizeof(fail), fail)) strcpy(fail, "???"); die_with(buf, ":\n\tperl DLL via BEGINLIBPATH is different: \n\t", fail, "\n\tYou may need to manipulate global BEGINLIBPATH and LIBPATHSTRICT" "\n\tso that the other copy is loaded via BEGINLIBPATH."); } return handle; }
int main(int argc, char **argv, char **env) { main_t f; handler_t h;
me = argv[0]; /**/ handle = load_perl_dll(PERL_DLL_BASENAME);
if (DosQueryProcAddr(handle, 0, "Perl_OS2_handler_install", (PFN*)&h)) die_with(PERL_DLL_BASENAME, ": DLL exports no symbol ", "Perl_OS2_handler_install", ""); if ( !h((void *)"~installprefix", Perlos2_handler_perllib_from) || !h((void *)"~dll", Perlos2_handler_perllib_to) || !h((void *)"~dll/sh/ksh.exe", Perlos2_handler_perl_sh) ) die_with(PERL_DLL_BASENAME, ": Can't install @INC manglers", "", "");
if (DosQueryProcAddr(handle, 0, "dll_perlmain", (PFN*)&f)) die_with(PERL_DLL_BASENAME, ": DLL exports no symbol ", "dll_perlmain", ""); return f(argc, argv, env); }
WARNING. Calling "getpriority" on a non-existing process could lock the system before Warp3 fixpak22. Starting with Warp3, Perl will use a workaround: it aborts getpriority() if the process is not present. This is not possible on older versions "2.*", and has a race condition anyway.
When finding a program to run, Perl first asks the OS to look for executables on "PATH" (OS/2 adds extension .exe if no extension is present). If not found, it looks for a script with possible extensions added in this order: no extension, .cmd, .btm, .bat, .pl. If found, Perl checks the start of the file for magic strings "#!" and "extproc ". If found, Perl uses the rest of the first line as the beginning of the command line to run this script. The only mangling done to the first line is extraction of arguments (currently up to 3), and ignoring of the path-part of the ``interpreter'' name if it can't be found using the full path.
E.g., "system 'foo', 'bar', 'baz'" may lead Perl to finding C:/emx/bin/foo.cmd with the first line being
extproc /bin/bash -x -c
If /bin/bash.exe is not found, then Perl looks for an executable bash.exe on "PATH". If found in C:/emx.add/bin/bash.exe, then the above system() is translated to
system qw(C:/emx.add/bin/bash.exe -x -c C:/emx/bin/foo.cmd bar baz)
One additional translation is performed: instead of /bin/sh Perl uses the hardwired-or-customized shell (see "``PERL_SH_DIR''").
The above search for ``interpreter'' is recursive: if bash executable is not found, but bash.btm is found, Perl will investigate its first line etc. The only hardwired limit on the recursion depth is implicit: there is a limit 4 on the number of additional arguments inserted before the actual arguments given to system(). In particular, if no additional arguments are specified on the ``magic'' first lines, then the limit on the depth is 4.
If Perl finds that the found executable is of PM type when the current session is not, it will start the new process in a separate session of necessary type. Call via "OS2::Process" to disable this magic.
WARNING. Due to the described logic, you need to explicitly specify .com extension if needed. Moreover, if the executable perl5.6.1 is requested, Perl will not look for perl5.6.1.exe. [This may change in the future.]
Two additional extensions by Andreas Kaiser, "OS2::UPM", and "OS2::FTP", are included into "ILYAZ" directory, mirrored on CPAN. Other OS/2-related extensions are available too.
This function enables/disables error popups associated with hardware errors (Disk not ready etc.) and software exceptions.
I know of no way to find out the state of popups before the first call to this function.
This function may redirect error popups associated with hardware errors (Disk not ready etc.) and software exceptions to the file POPUPLOG.OS2 at the root directory of the specified drive. Overrides OS2::Error() specified by individual programs. Given argument undef will disable redirection.
Has global effect, persists after the application exits.
I know of no way to find out the state of redirection of popups to the disk before the first call to this function.
MAX_PATH_LENGTH, MAX_TEXT_SESSIONS, MAX_PM_SESSIONS, MAX_VDM_SESSIONS, BOOT_DRIVE, DYN_PRI_VARIATION, MAX_WAIT, MIN_SLICE, MAX_SLICE, PAGE_SIZE, VERSION_MAJOR, VERSION_MINOR, VERSION_REVISION, MS_COUNT, TIME_LOW, TIME_HIGH, TOTPHYSMEM, TOTRESMEM, TOTAVAILMEM, MAXPRMEM, MAXSHMEM, TIMER_INTERVAL, MAX_COMP_LENGTH, FOREGROUND_FS_SESSION, FOREGROUND_PROCESS
See ``Centralized management of resources'' for additional details.
Dies with ``QUITing...'' if WM_QUIT message is obtained.
Returns change in number of windows. If "cnt" is given, it is incremented by the number of messages retrieved.
Dies with ``QUITing...'' if WM_QUIT message is obtained.
See ``Misfeatures'' for details.
(Note that some of these may be moved to different libraries - eventually).
WUNTRACED Not implemented. waitpid() is not implemented for negative values of PID.
Note that "kill -9" does not work with the current version of EMX.
This may lead to problems later in case the socket is accessed via the ``usual'' file-system calls using the ``initial'' name.
What is much worse, some DLLs change the floating point flags when in _DLLInitTerm() (e.g., TCP32IP). This means that even if you do not call any function in the DLL, just the act of loading this DLL will reset your flags. What is worse, the same compiler was used to compile some HOOK DLLs. Given that HOOK dlls are executed in the context of all the applications in the system, this means a complete unpredictablity of floating point flags on systems using such HOOK DLLs. E.g., GAMESRVR.DLL of DIVE origin changes the floating point flags on each write to the TTY of a VIO (windowed text-mode) applications.
Some other (not completely debugged) situations when FP flags change include some video drivers (?), and some operations related to creation of the windows. People who code OpenGL may have more experience on this.
Perl is generally used in the situation when all the floating-point exceptions are ignored, as is the default under EMX. If they are not ignored, some benign Perl programs would get a "SIGFPE" and would die a horrible death.
To circumvent this, Perl uses two hacks. They help against one type of damage only: FP flags changed when loading a DLL.
One of the hacks is to disable floating point exceptions on Perl startup (as is the default with EMX). This helps only with compile-time-linked DLLs changing the flags before main() had a chance to be called.
The other hack is to restore FP flags after a call to dlopen(). This helps against similar damage done by DLLs _DLLInitTerm() at runtime. Currently no way to switch these hacks off is provided.
Perl provides a centralized management of these resources:
If by some reasons perl.h cannot be included, use
extern int Perl_hab_GET(void);
instead.
To get an "HMQ", the extension should call "hmq = perl_hmq_GET(serve)" in C. After this call is performed, "hmq" may be accessed as "Perl_hmq".
To signal to Perl that HMQ is not needed any more, call "perl_hmq_UNSET(serve)". Perl process will automatically morph/unmorph itself into/from a PM process if HMQ is needed/not-needed. Perl will automatically enable/disable "WM_QUIT" message during shutdown if the message queue is served/not-served.
NOTE. If during a shutdown there is a message queue which did not disable WM_QUIT, and which did not process the received WM_QUIT message, the shutdown will be automatically cancelled. Do not call perl_hmq_GET(1) unless you are going to process messages on an orderly basis.
Some "Win*" entry points also overload a ``meaningful'' return value with the error indicator; having a 0 return value indicates an error. Yet some other "Win*" entry points overload things even more, and 0 return value may mean a successful call returning a valid value 0, as well as an error condition; in the case of a 0 return value one should call WinGetLastError() API to distinguish a successful call from a failing one.
By convention, all the calls to OS/2 API should indicate their failures by resetting $^E. All the Perl-accessible functions which call OS/2 API may be broken into two classes: some die()s when an API error is encountered, the other report the error via a false return value (of course, this does not concern Perl-accessible functions which expect a failure of the OS/2 API call, having some workarounds coded).
Obviously, in the situation of the last type of the signature of an OS/2 API, it is must more convenient for the users if the failure is indicated by die()ing: one does not need to check $^E to know that something went wrong. If, however, this solution is not desirable by some reason, the code in question should reset $^E to 0 before making this OS/2 API call, so that the caller of this Perl-accessible function has a chance to distinguish a success-but-0-return value from a failure. (One may return undef as an alternative way of reporting an error.)
The macros to simplify this type of error propagation are
For example, many newer useful APIs are not present in OS/2 v2; many PM-related APIs require DLLs not available on floppy-boot setup.
To make these calls fail only when the calls are executed, one should call these API via a dynamic linking API. There is a subsystem in Perl to simplify such type of calls. A large number of entry points available for such linking is provided (see "entries_ordinals" - and also "PMWIN_entries" - in os2ish.h). These ordinals can be accessed via the APIs:
CallORD(), DeclFuncByORD(), DeclVoidFuncByORD(), DeclOSFuncByORD(), DeclWinFuncByORD(), AssignFuncPByORD(), DeclWinFuncByORD_CACHE(), DeclWinFuncByORD_CACHE_survive(), DeclWinFuncByORD_CACHE_resetError_survive(), DeclWinFunc_CACHE(), DeclWinFunc_CACHE_resetError(), DeclWinFunc_CACHE_survive(), DeclWinFunc_CACHE_resetError_survive()
See the header files and the C code in the supplied OS/2-related modules for the details on usage of these functions.
Some of these functions also combine dynaloading semantic with the error-propagation semantic discussed above.
It can load perl dynamic extensions, and it can fork().
Note. Keep in mind that fork() is needed to open a pipe to yourself.
This is the only executable with does not require OS/2. The friends locked into "M$" world would appreciate the fact that this executable runs under DOS, Win0.3*, Win0.95 and WinNT with an appropriate extender. See ``Other OSes''.
Note. Usually (unless explicitly redirected during the startup) STDIN, STDERR, and STDOUT of a PM application are redirected to nul. However, it is possible to see them if you start "perl__.exe" from a PM program which emulates a console window, like Shell mode of Emacs or EPM. Thus it is possible to use Perl debugger (see perldebug) to debug your PM application (but beware of the message loop lockups - this will not work if you have a message queue to serve, unless you hook the serving into the getc() function of the debugger).
Another way to see the output of a PM program is to run it as
pm_prog args 2>&1 | cat -
with a shell different from cmd.exe, so that it does not create a link between a VIO session and the session of "pm_porg". (Such a link closes the VIO window.) E.g., this works with sh.exe - or with Perl!
open P, 'pm_prog args 2>&1 |' or die; print while <P>;
The flavor perl__.exe is required if you want to start your program without a VIO window present, but not "detach"ed (run "help detach" for more info). Very useful for extensions which use PM, like "Perl/Tk" or "OpenGL".
Note also that the differences between PM and VIO executables are only in the default behaviour. One can start any executable in any kind of session by using the arguments "/fs", "/pm" or "/win" switches of the command "start" (of CMD.EXE or a similar shell). Alternatively, one can use the numeric first argument of the "system" Perl function (see OS2::Process).
There are two distinctive features of the dyna-linking model of OS/2: first, all the references to external functions are resolved at the compile time; second, there is no runtime fixup of the DLLs after they are loaded into memory. The first feature is an enormous advantage over other models: it avoids conflicts when several DLLs used by an application export entries with the same name. In such cases ``other'' models of dyna-linking just choose between these two entry points using some random criterion - with predictable disasters as results. But it is the second feature which requires the build of perl.dll.
The address tables of DLLs are patched only once, when they are loaded. The addresses of the entry points into DLLs are guaranteed to be the same for all the programs which use the same DLL. This removes the runtime fixup - once DLL is loaded, its code is read-only.
While this allows some (significant?) performance advantages, this makes life much harder for developers, since the above scheme makes it impossible for a DLL to be ``linked'' to a symbol in the .EXE file. Indeed, this would need a DLL to have different relocations tables for the (different) executables which use this DLL.
However, a dynamically loaded Perl extension is forced to use some symbols from the perl executable, e.g., to know how to find the arguments to the functions: the arguments live on the perl internal evaluation stack. The solution is to put the main code of the interpreter into a DLL, and make the .EXE file which just loads this DLL into memory and supplies command-arguments. The extension DLL cannot link to symbols in .EXE, but it has no problem linking to symbols in the .DLL.
This greatly increases the load time for the application (as well as complexity of the compilation). Since interpreter is in a DLL, the C RTL is basically forced to reside in a DLL as well (otherwise extensions would not be able to use CRT). There are some advantages if you use different flavors of perl, such as running perl.exe and perl__.exe simultaneously: they share the memory of perl.dll.
NOTE. There is one additional effect which makes DLLs more wasteful: DLLs are loaded in the shared memory region, which is a scarse resource given the 512M barrier of the ``standard'' OS/2 virtual memory. The code of .EXE files is also shared by all the processes which use the particular .EXE, but they are ``shared in the private address space of the process''; this is possible because the address at which different sections of the .EXE file are loaded is decided at compile-time, thus all the processes have these sections loaded at same addresses, and no fixup of internal links inside the .EXE is needed.
Since DLLs may be loaded at run time, to have the same mechanism for DLLs one needs to have the address range of any of the loaded DLLs in the system to be available in all the processes which did not load a particular DLL yet. This is why the DLLs are mapped to the shared memory region.
Current EMX environment does not allow .EXE files compiled in "omf" format to fork(). fork() is needed for exactly three Perl operations:
While these operations are not questions of life and death, they are needed for a lot of useful scripts. This forces "a.out"-style compile of perl.exe.
path1;path2
or
path1 path2
If the beginning of some prebuilt path matches path1, it is substituted with path2.
Should be used if the perl library is moved from the default location in preference to "PERL(5)LIB", since this would not leave wrong entries in @INC. For example, if the compiled version of perl looks for @INC in f:/perllib/lib, and you want to install the library in h:/opt/gnu, do
set PERLLIB_PREFIX=f:/perllib/lib;h:/opt/gnu
This will cause Perl with the prebuilt @INC of
f:/perllib/lib/5.00553/os2 f:/perllib/lib/5.00553 f:/perllib/lib/site_perl/5.00553/os2 f:/perllib/lib/site_perl/5.00553 .
to use the following @INC:
h:/opt/gnu/5.00553/os2 h:/opt/gnu/5.00553 h:/opt/gnu/site_perl/5.00553/os2 h:/opt/gnu/site_perl/5.00553 .
Should not be set with newer Perls, since this may hide some real problems.
In addition to possible bugs and an inability to follow changes to the translation policy with off/on switches of TERMIO translation, this introduces a serious incompatible change: before sysread() on text-mode filehandles would go through the translation layer, now it would not.
It may be possible to code a simple workaround which would
When resolving a request for a global DLL, the table of already-loaded specific DLLs is (effectively) ignored; moreover, specific DLLs are always loaded from the prescribed path.
There is/was a minor twist which makes this scheme fragile: what to do with DLLs loaded from
Unless "LIBPATHSTRICT" is set to "T" (and the kernel is after 2000/09/01), such DLLs are considered to be global. When loading a global DLL it is first looked in the table of already-loaded global DLLs. Because of this the fact that one executable loaded a DLL from "BEGINLIBPATH" and "ENDLIBPATH", or . from "LIBPATH" may affect which DLL is loaded when another executable requests a DLL with the same name. This is the reason for version-specific mangling of the DLL name for perl DLL.
Since the Perl extension DLLs are always loaded with the full path, there is no need to mangle their names in a version-specific ways: their directory already reflects the corresponding version of perl, and @INC takes into account binary compatibility with older version. Starting from 5.6.2 the name mangling scheme is fixed to be the same as for Perl 5.005_53 (same as in a popular binary release). Thus new Perls will be able to resolve the names of old extension DLLs if @INC allows finding their directories.
However, this still does not guarantee that these DLL may be loaded. The reason is the mangling of the name of the Perl DLL. And since the extension DLLs link with the Perl DLL, extension DLLs for older versions would load an older Perl DLL, and would most probably segfault (since the data in this DLL is not properly initialized).
There is a partial workaround (which can be made complete with newer OS/2 kernels): create a forwarder DLL with the same name as the DLL of the older version of Perl, which forwards the entry points to the newer Perl's DLL. Make this DLL accessible on (say) the "BEGINLIBPATH" of the new Perl executable. When the new executable accesses old Perl's extension DLLs, they would request the old Perl's DLL by name, get the forwarder instead, so effectively will link with the currently running (new) Perl DLL.
This may break in two ways:
With support for "LIBPATHSTRICT" this may be circumvented - unless one of DLLs is started from . from "LIBPATH" (I do not know whether "LIBPATHSTRICT" affects this case).
REMARK. Unless newer kernels allow . in "BEGINLIBPATH" (older do not), this mess cannot be completely cleaned. (It turns out that as of the beginning of 2002, . is not allowed, but .\. is - and it has the same effect.)
REMARK. "LIBPATHSTRICT", "BEGINLIBPATH" and "ENDLIBPATH" are not environment variables, although cmd.exe emulates them on "SET ..." lines. From Perl they may be accessed by Cwd::extLibpath and Cwd::extLibpath_set.
LIBRARY 'perlE0AC' INITINSTANCE TERMINSTANCE DESCRIPTION '@#perl5-porters@perl.org:5.006001#@ Perl module for 5.00553 -> Perl 5.6.1 forwarder' CODE LOADONCALL DATA LOADONCALL NONSHARED MULTIPLE EXPORTS
modifying the versions/names as needed. Run
perl -wnle "next if 0../EXPORTS/; print qq( \"$1\") if /\"(\w+)\"/" perl5.def >lst
in the Perl build directory (to make the DLL smaller replace perl5.def with the definition file for the older version of Perl if present).
cat perl5shim.def-leader lst >perl5shim.def gcc -Zomf -Zdll -o perlE0AC.dll perl5shim.def -s -llibperl
(ignore multiple "warning L4085").
This was needed to compile "Perl/Tk" for XFree86-OS/2 out-of-the-box, and link with DLLs for other useful libraries, which typically are compiled with "-Zmt -Zcrtdll".
Thus means that you need to get some copy of a sh.exe as well (I use one from pdksh). The path F:/bin above is set up automatically during the build to a correct value on the builder machine, but is overridable at runtime,
Reasons: a consensus on "perl5-porters" was that perl should use one non-overridable shell per platform. The obvious choices for OS/2 are cmd.exe and sh.exe. Having perl build itself would be impossible with cmd.exe as a shell, thus I picked up "sh.exe". This assures almost 100% compatibility with the scripts coming from *nix. As an added benefit this works as well under DOS if you use DOS-enabled port of pdksh (see ``Prerequisites'').
Disadvantages: currently sh.exe of pdksh calls external programs via fork()/exec(), and there is no functioning exec() on OS/2. exec() is emulated by EMX by an asynchronous call while the caller waits for child completion (to pretend that the "pid" did not change). This means that 1 extra copy of sh.exe is made active via fork()/exec(), which may lead to some resources taken from the system (even if we do not count extra work needed for fork()ing).
Note that this a lesser issue now when we do not spawn sh.exe unless needed (metachars found).
One can always start cmd.exe explicitly via
system 'cmd', '/c', 'mycmd', 'arg1', 'arg2', ...
If you need to use cmd.exe, and do not want to hand-edit thousands of your scripts, the long-term solution proposed on p5-p is to have a directive
use OS2::Cmd;
which will override system(), exec(), ``, and "open(,'...|')". With current perl you may override only system(), readpipe() - the explicit version of ``, and maybe exec(). The code will substitute the one-argument call to system() by "CORE::system('cmd.exe', '/c', shift)".
If you have some working code for "OS2::Cmd", please send it to me, I will include it into distribution. I have no need for such a module, so cannot test it.
For the details of the current situation with calling external programs, see ``2 (and DOS) programs under Perl'' in Starting OS. Set us mention a couple of features:
Combination of perl's malloc() and rigid DLL name resolution creates a special problem with library functions which expect their return value to be free()d by system's free(). To facilitate extensions which need to call such functions, system memory-allocation functions are still available with the prefix "emx_" added. (Currently only DLL perl has this, it should propagate to perl_.exe shortly.)
Most notable problems:
Note that these problems should not discourage experimenting, since they have a low probability of affecting small programs.
Закладки на сайте Проследить за страницей |
Created 1996-2024 by Maxim Chirkov Добавить, Поддержать, Вебмастеру |