Activity for Brian Tiffin

  • Brian Tiffin Brian Tiffin modified a comment on discussion Contributions

    COBOL embedding Java, calling Nashorn Javascript (they get to call it Javascript), calling Java. /* Sample of Nashorn from GnuCOBOL embedding of the JVM */ import javax.script.ScriptEngine; import javax.script.ScriptEngineManager; import javax.script.ScriptException; public class Jjs { public static int sample() { ScriptEngine nashorn = new ScriptEngineManager().getEngineByName("nashorn"); System.out.println("in Jjs sample"); System.out.println("Java version: " + System.getProperty("java.version"));...

  • Brian Tiffin Brian Tiffin posted a comment on discussion Contributions

    No it's next Monday - going to an island to soak it up.

  • Brian Tiffin Brian Tiffin posted a comment on discussion Contributions

    COBOL embedding Java, calling Nashorn Javascript (they get to call it Javascript), calling Java. /* Sample of Nashorn from GnuCOBOL embedding of the JVM */ import javax.script.ScriptEngine; import javax.script.ScriptEngineManager; import javax.script.ScriptException; public class Jjs { public static int sample() { ScriptEngine nashorn = new ScriptEngineManager().getEngineByName("nashorn"); System.out.println("in Jjs sample"); System.out.println("Java version: " + System.getProperty("java.version"));...

  • Brian Tiffin Brian Tiffin posted a comment on discussion GnuCOBOL

    I'm going to say no. During dev, it can be done in tightly controlled stages perhaps (knowing safe path through the minefield), but it would not be something to unleash on the unsuspecting. Brian

  • Brian Tiffin Brian Tiffin posted a comment on discussion GnuCOBOL

    The biggest TODO (I actually hope Brian comes back for this after 3.0 or better even starts after 2.2 release) is to have a "thread local current state" (with only one instance in the beginning) which includes every global variable. It's in the plans. Can talk more once I'm back from a week off. Cheers, Brian

  • Brian Tiffin Brian Tiffin posted a comment on discussion GnuCOBOL

    Going backwards through your post, Pat. Doable. There is a fork of OpenCOBOL that advertises as thread safe. So, doable. And we'd have source to use as a leg up starting point. 3 to 5 years is well within reason Not really any work arounds with the current state of things. GnuCOBOL 2 is utterly and only single threaded at the runtime library level. Even calling in as a "system call" would need to invoke separate copies of libcob if said system calls are coming from a threaded host. That could get...

  • Brian Tiffin Brian Tiffin modified a comment on discussion Help getting started

    Below is an example that may be helpful. 000010 IDENTIFICATION DIVISION. 000020 PROGRAM-ID. SAMPLE4J. 000030 000040 ENVIRONMENT DIVISION. 000050 CONFIGURATION SECTION. 000060 SPECIAL-NAMES. 000070 CLASS MYALPHA IS "A" THRU "Z", " " 000080* CRT STATUS IS SCREEN-STATUS 000090 CURSOR IS LLCC. 000100 INPUT-OUTPUT SECTION. 000110 FILE-CONTROL. 000120 SELECT OPTIONAL INDEXED-FILE 000130 ASSIGN TO OUTFILE 000140 ORGANIZATION IS INDEXED 000150 RECORD KEY IS PH-NUMBER 000160 ALTERNATE RECORD KEY IS NAME-VALUE...

  • Brian Tiffin Brian Tiffin modified a comment on discussion Contributions

    Command logging. XForms includes a browser object. One of the goodies is a command log browser. An original design consideration for XForms was wrapping console applications in graphics, so the command log was added to show a view of any background command executions. This is an FL-NORMAL-BROWSER, that means no selections are allowed from the window, it is a static view only scrolling informational window. Still handy ... *> Show output from command, blocking gives exit status call "fl_exe_command"...

  • Brian Tiffin Brian Tiffin posted a comment on discussion Contributions

    Not for long. ;-) Cheers, Brian

  • Brian Tiffin Brian Tiffin modified a comment on discussion Contributions

    Command logging. XForms includes a browser object. One of the goodies is a command log browsers. An original design consideration for XForms was wrapping console applications in graphics, so the command log was added to show a view of any background command executions. This is an FL-NORMAL-BROWSER, that means no selections are allowed from the window, it is a static view only scrolling informational window. Still handy ... *> Show output from command, blocking gives exit status call "fl_exe_command"...

  • Brian Tiffin Brian Tiffin modified a comment on discussion Contributions

    Command logging. XForms includes a browser object. One of the goodies is a command log browsers. An original design consideration for XForms was wrapping console applications in graphics, so the command log was added to show a view of any background command executions. This is an FL-NORMAL-BROWSER, that means no selections are allowed from the window, it is a static view only scrolling informational window. Still handy ... *> Show output from command, blocking gives exit status call "fl_exe_command"...

  • Brian Tiffin Brian Tiffin modified a comment on discussion Contributions

    Command logging. XForms includes a browser object. One of the goodies is a command log browsers. An original design consideration for XForms was wrapping console applications in graphics, so the command log was added to show a view of any background command executions. This is an FL-NORMAL-BROWSER, that means no selections are allowed from the window, it is a static view only scrolling informational window. Still handy ... *> Show output from command, blocking gives exit status call "fl_exe_command"...

  • Brian Tiffin Brian Tiffin posted a comment on discussion Contributions

    Command logging. XForms includes a browser object. One of the goodies is a command log browsers. An original design consideration for XForms was wrapping console applications in graphics, so the command log was added to show a view of any background command executions. This is an FL-NORMAL-BROWSER, that means no selections are allowed from the window, it is a static view only scrolling informational window. Still handy ... *> Show output from command, blocking gives exit status call "fl_exe_command"...

  • Brian Tiffin Brian Tiffin modified a comment on discussion Contributions

    These are pretty easy to change up. call "fl_add_chart" using by value FL-SPIKE-CHART 5 5 240 200 by reference "Spike chart" returning chart-object(1) spike call "fl_add_chart" using by value FL-FILL-CHART 5 5 240 200 by reference "Fill chart" returning chart-object(1) Fill call "fl_add_chart" using by value FL-SPECIALPIE-CHART 5 5 240 200 by reference "Pull out pie chart" returning chart-object(1) Specialpie Opinionated whine follows (But before digging in with Pie charts for anything beyond three...

  • Brian Tiffin Brian Tiffin posted a comment on discussion Contributions

    I'll pipe in too. Yep, current conversations with Jens, and exploring XForms over the last few days, I think it'll be worth a kick at trying to build this into codegen as a SCREEN SECTION builder. Personal opinion slipping through It won't work with native Windows though, XForms is X11, so it might be a little incentive to get developers moved off the spyware OS and into some freedoms. Cygwin support for XForms 1.2.4 was announced in June this year, so it will work with Cygwin/X under Windows which...

  • Brian Tiffin Brian Tiffin posted a comment on discussion Contributions

    Code for the fl_clock_hide_seconds was accepted in XForms, 1.2.5 is scheduled for release soon. Woohoo, less peripheral vision distractions and still an easy to wield clock. For 1.2.4 and release XForms, GnuCOBOL code is still best using the on exception continue version of the CALL to that function and put up with the seconds hand ticking by. http://git.savannah.nongnu.org/cgit/xforms.git/log/?h=clock Cheers, Brian

  • Brian Tiffin Brian Tiffin modified a comment on discussion Help getting started

    As a for instance with the difference between using the default normal and STATIC CALL. /* Author: Brian Tiffin Dedicated to the public domain Date started: August 2017 Modified: 2017-08-15/17:24-0400 btiffin */ /* alib.c tectonics: gcc -fPIC -shared -o alib.so alib.c */ #include <stdio.h> void thing(float num) { printf("%f\n", num); return; } alib.c prompt$ gcc -fPIC -shared -o alib.so alib.c building alib.so *>-<* *> Author: Brian Tiffin *> Dedicated to the public domain *> *> Date started: July...

  • Brian Tiffin Brian Tiffin modified a comment on discussion Help getting started

    As a for instance with the difference between using the default normal and STATIC CALL. /* Author: Brian Tiffin Dedicated to the public domain Date started: August 2017 Modified: 2017-08-15/17:24-0400 btiffin */ /* alib.c tectonics: gcc -fPIC -shared -o alib.so alib.c */ #include <stdio.h> void thing(float num) { printf("%f\n", num); return; } alib.c prompt$ gcc -fPIC -shared -o alib.so alib.c building alib.so *>-<* *> Author: Brian Tiffin *> Dedicated to the public domain *> *> Date started: July...

  • Brian Tiffin Brian Tiffin posted a comment on discussion Help getting started

    As a for instance with the difference between using the default normal and STATIC CALL. /* Author: Brian Tiffin Dedicated to the public domain Date started: August 2017 Modified: 2017-08-15/17:24-0400 btiffin */ /* alib.c tectonics: gcc -fPIC -shared -o alib.so alib.c */ #include <stdio.h> void thing(float num) { printf("%f\n", num); return; } alib.c prompt$ gcc -fPIC -shared -o alib.so alib.c building alib.so *>-<* *> Author: Brian Tiffin *> Dedicated to the public domain *> *> Date started: July...

  • Brian Tiffin Brian Tiffin posted a comment on discussion Help getting started

    Forgot to login... Ubuntu? You may need to add prompt$ export COB_LDFLAGS='-Wl,--no-as-needed' Before any and all GnuCOBOL cobc compiles that use external libraries via -l. There was a change to Ubuntu link cache rules that adds as-needed and the linker will drop named -llibrary names that aren't touched by the C compiler. CALL uses runtime lookup by name. Simon's suggestion of using CALL STATIC might work, but on Fedora that often breaks if the C function prototypes aren't known (and we have no...

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    There is a CVE that effects revision control (or any other system) that supports remote access via ssh:// protocols; specially crafted path portions of the URL can trigger ssh to evaluate commands under the permissions of the listening server. Git, SVN, Hg, Fossil, CVS, ..., have all been patched, so if you use Fossil, it might be a good time to clone out a new copy and build from source. Fossil release 2.4 won't be too far off, but for now upstream trunk is the safest bet. Initial fix for Fossil...

  • Brian Tiffin Brian Tiffin modified a comment on discussion The Lounge

    These are pretty easy to change up. call "fl_add_chart" using by value FL-SPIKE-CHART 5 5 240 200 by reference "Spike chart" returning chart-object(1) spike call "fl_add_chart" using by value FL-FILL-CHART 5 5 240 200 by reference "Fill chart" returning chart-object(1) Fill call "fl_add_chart" using by value FL-SPECIALPIE-CHART 5 5 240 200 by reference "Pull out pie chart" returning chart-object(1) Specialpie Opinionated whine follows (But before digging in with Pie charts for anything beyond three...

  • Brian Tiffin Brian Tiffin modified a comment on discussion The Lounge

    These are pretty easy to change up. call "fl_add_chart" using by value FL-SPIKE-CHART 5 5 240 200 by reference "Spike chart" returning chart-object(1) spike call "fl_add_chart" using by value FL-FILL-CHART 5 5 240 200 by reference "Fill chart" returning chart-object(1) Fill call "fl_add_chart" using by value FL-SPECIALPIE-CHART 5 5 240 200 by reference "Pull out pie chart" returning chart-object(1) Specialpie Opiniated whine follows (But before digging in with Pie charts for anything beyond three...

  • Brian Tiffin Brian Tiffin posted a comment on discussion Help getting started

    Good points, Steve. If the usage scenarios don't include random user SQL input, then binding can be a fair bear of a tail chase to pull off in COBOL, and it might be better left alone. But... if there is potential for injection, then it gets a little more critical to support binding, no? I'll admit to very little expertise in the domain of whitehat/blackhat SQL think. Cheers, Brian

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    These are pretty easy to change up. call "fl_add_chart" using by value FL-SPIKE-CHART 5 5 240 200 by reference "Spike chart" returning chart-object(1) spike call "fl_add_chart" using by value FL-FILL-CHART 5 5 240 200 by reference "Fill chart" returning chart-object(1) Fill call "fl_add_chart" using by value FL-SPECIALPIE-CHART 5 5 240 200 by reference "Pull out pie chart" returning chart-object(1) Specialpie Opiniated whine follows (But before digging in with Pie charts for anything beyond three...

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    Charts! *>-<* *> Author: Brian Tiffin *> Dedicated to the public domain *> *> Date started: August 2017 *> Modified: 2017-08-14/20:49-0400 btiffin *>+<* *> *> xforms-chart.cob, demonstrate a bar chart *> Tectonics: cobc -xj xforms-chart.cob -lforms *> >>SOURCE FORMAT IS FREE identification division. program-id. sample. environment division. configuration section. repository. function all intrinsic. data division. working-storage section. COPY xforms. 01 argc usage binary-long. 01 argv usage pointer....

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    Yes. :-) Read the current xforms.cpy attachment and give me a start date for roughly when you did the original, Pat. Much thanks. Do you mind an LGPL 3+ license with FSF owner? With CONSTANT there is no storage space taken up and, so far, none of these have been needed to pass by anything other than BY VALUE. Stance might change if there is ever an XForms UDF that needs to pass the numbers BY REFERENCE which CONSTANT (rightly) doesn't support. Cheers, Brian

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    Clocks. *>-<* *> Author: Brian Tiffin *> Dedicated to the public domain *> *> Date started: August 2017 *> Modified: 2017-08-14/19:10-0400 btiffin *>+<* *> *> xforms-clock.cob, demonstrate some clock objects *> Tectonics: cobc -xj xforms-clocks.cob -lforms *> >>SOURCE FORMAT IS FREE identification division. program-id. sample. environment division. configuration section. repository. function all intrinsic. data division. working-storage section. COPY xforms. 01 argc usage binary-long. 01 argv usage...

  • Brian Tiffin Brian Tiffin posted a comment on discussion Help getting started

    This looks awesome, Robert. Just to add to the complexity a little bit, are there plans for supporting numbered parameter binds? And I think we can reasonably assume character data, bind_text... as SQLite doesn't really seem to care if numbers are strings or vice versa. To help manage the UDF side, I think just supporting ?NNN numbered versions versus the $VVV named parameters would also suffice. Looking forward to more bragging about GnuCOBOL powers, regardless. This will be a nice add. Cheers,...

  • Brian Tiffin Brian Tiffin posted a comment on discussion GnuCOBOL

    Just a heads up, the package in Fedora is promp$ sudo dnf install help2man Lowercase H. Or, to avoid that prompt$ touch bin/cobcrun.1 cobc/cobc.1 And the need for help2man will go away. Cheers, Brian

  • Brian Tiffin Brian Tiffin modified a comment on discussion The Lounge

    And then, because of https://sourceforge.net/p/open-cobol/discussion/lounge/thread/71be191e/ Doesn't look like it will need much tweaking, Dave. Compiled FOCAL with -fPIC and removed a few static keywords to be able to get at symbols, then a not smart hack experiment (which will be fixed now that it tests as working) *>-<* *> Author: Brian Tiffin *> Dedicated to the public domain *> but don't use this, as it is a dirty hack, wait for a real cut *> FOCAL interpreter is Copyright 1978-2011 Dave Pitts,...

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    And then, because of https://sourceforge.net/p/open-cobol/discussion/lounge/thread/71be191e/ Doesn't look like it will need much tweaking Dave; Compiled FOCAL with -fPIC and removed a few static keywords to be able to get at symbols, then a not smart hack experiment (which will be fixed now that it tests as working) *>-<* *> Author: Brian Tiffin *> Dedicated to the public domain *> but don't use this, as it is a dirty hack, wait for a real cut *> FOCAL interpreter is Copyright 1978-2011 Dave Pitts,...

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    I just noticed Dave Pitts name in that article, as having written a cross-assembler for a 704 emulator. Is that you, Dave? If so, way cool. Our -t listing code comes with some pedigree along with being the nice add that it is. And, having just dropped the name, the Focal interpeter is yours too, isn't it. https://sourceforge.net/p/open-cobol/discussion/lounge/thread/75c89205/ I got some FAQ entries to update for some braggin's. :-) Cheers, Brian

  • Brian Tiffin Brian Tiffin modified a comment on discussion The Lounge

    Ok one more, The XForms kit comes with a libflimage for image support, and it includes a Postscript dump routine for the XYPlot objects. prompt$ cobc -xj -debug -w xforms.cob -lforms -lflimage "Yes" pressed, end form Quit Quit plot prompt$ ls xforms* xforms xforms.cob xforms-xyplot.ps PS file output of the XYPlot via ... call "fl_do_forms" returning form-button display "Quit plot" call "fl_object_ps_dump" using by value xyplot by reference z"xforms-xyplot.ps" call "fl_finish" ... fl_object_ps_dump...

  • Brian Tiffin Brian Tiffin modified a comment on discussion The Lounge

    Charts and graphs The data sample # XForms sample plot data 01.0, 0.3 02.0 0.2 03.0 0.6 04.0 0.7 05.0 0.1 06.0 0.3 07.0 0.8 08.0 0.4 09.0 0.0 10.0 1.0 11.0 0.4 12.0 2.3 plot-data.txt; data can be space, tab, comma separated (mix freely) and # marks comment lines. This data could just as easily be in a table of float (x,y) pairs. The new code *>-<* *> Author: Brian Tiffin *> Dedicated to the public domain *> *> Date started: August 2017 *> Modified: 2017-08-11/23:35-0400 btiffin *>+<* *> *> xforms.cob,...

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    Ok one more, The XForms kits comes with a libflimage for image support, and it includes a Postscript dump routine for the XYPlot objects. prompt$ cobc -xj -debug -w xforms.cob -lforms -lflimage "Yes" pressed, end form Quit Quit plot prompt$ ls xforms* xforms xforms.cob xforms-xyplot.ps PS file output of the XYPlot via ... call "fl_do_forms" returning form-button display "Quit plot" call "fl_object_ps_dump" using by value xyplot by reference z"xforms-xyplot.ps" call "fl_finish" ... fl_object_ps_dump...

  • Brian Tiffin Brian Tiffin modified a comment on discussion The Lounge

    Charts and graphs The data sample # XForms sample plot data 01.0, 0.3 02.0 0.2 03.0 0.6 04.0 0.7 05.0 0.1 06.0 0.3 07.0 0.8 08.0 0.4 09.0 0.0 10.0 1.0 11.0 0.4 12.0 2.3 plot-data.txt; data can be space, tab, comma separated (mix freely) and # marks comment lines. This data could just as easily be in a table of float (x,y) pairs. The new code *>-<* *> Author: Brian Tiffin *> Dedicated to the public domain *> *> Date started: August 2017 *> Modified: 2017-08-11/23:35-0400 btiffin *>+<* *> *> xforms.cob,...

  • Brian Tiffin Brian Tiffin modified a comment on discussion The Lounge

    Charts and graphs The data sample # XForms sample plot data 01.0, 0.3 02.0 0.2 03.0 0.6 04.0 0.7 05.0 0.1 06.0 0.3 07.0 0.8 08.0 0.4 09.0 0.0 10.0 1.0 11.0 0.4 12.0 2.3 plot-data.txt; data can be space, tab, comma separated (mix freely) and # marks comment lines. This data could just as easily be in a table of float (x,y) pairs. The new code *>-<* *> Author: Brian Tiffin *> Dedicated to the public domain *> *> Date started: August 2017 *> Modified: 2017-08-11/23:35-0400 btiffin *>+<* *> *> xforms.cob,...

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    Charts and graphs The data sample # XForms sample plot data 01.0, 0.3 02.0 0.2 03.0 0.6 04.0 0.7 05.0 0.1 06.0 0.3 07.0 0.8 08.0 0.4 09.0 0.0 10.0 1.0 11.0 0.4 12.0 2.3 plot-data.txt The new code *>-<* *> Author: Brian Tiffin *> Dedicated to the public domain *> *> Date started: August 2017 *> Modified: 2017-08-11/23:35-0400 btiffin *>+<* *> *> xforms.cob, initial trial of XForms *> Tectonics: cobc -xj xforms.cob -lforms *> >>SOURCE FORMAT IS FREE identification division. program-id. sample. environment...

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    Here's a quick sample, all CALL. Requires xforms and xforms-devel packages. xforms-docs also has a set of C based examples. Simple, two forms; first just returns from do_forms after any click, but stays in a loop. second with callbacks, only the Quit button returns (it has no callback) so do_forms keep running. *>-<* *> Author: Brian Tiffin *> Dedicated to the public domain *> *> Date started: August 2017 *> Modified: 2017-08-11/15:43-0400 btiffin *>+<* *> *> xforms.cob, initial trial of XForms *>...

  • Brian Tiffin Brian Tiffin posted a comment on discussion Artists Wanted

    Can't help but be impressed by the work that gets posted to this Discussion group. The GnuCOBOL project, https://sourceforge.net/projects/open-cobol/ could use that level of professional flair. We currently use an image of a work horse, Sire the Old English Black, https://commons.wikimedia.org/wiki/File:Old_English_Black.jpg but it would be nice to get that integrated with the name and with a little touch of modern style as befits a logo. And a banner would be a sweet touch for across the top of...

  • Brian Tiffin Brian Tiffin posted a comment on discussion Contributions

    Cool; when the Intrinsic is more complete, supporting that syntax will just be a few tweaks and exposing some slightly different input output arguments. A fair portion of that API is already covered in the Intrinsic but the system routine is farther ahead when it comes to array handling. Thanks for the link reminder. Cheers, Brian

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    http://basic-converter.proboards.com/thread/959/bacon-3-6-released Get it at http://www.basic-converter.org/#downloads Highlights are UTF-8 support throughout (optional, with an OPTION UTF8 statement) and C++ binding possibilities. Rave on follows; you may already know this from previous ravings on BaCon... If you happen to need some UTF-8 data manipulation, and can't wait for the hinted at GnuCOBOL support, then perhaps a BaCon module can come to the rescue. bacon -f program.bac Creates GnuCOBOL...

  • Brian Tiffin Brian Tiffin modified a comment on discussion The Lounge

    And just because, not wasting enough time is wasteful Had to change cofunge.cob to mock up an env pointer for the Befunge sys info command (and to add the -ux option to start up the "TURT" engine X11 interface). I also modified the quine program to end with an &q to accept an integer (which is returned to GNuCOBOL) to allow for the screen capture before image rundown. So the "@" on the fifth line from the bottom is "&q" in the new cut. The Funge code draws an image of itself using TURT, but as explained...

  • Brian Tiffin Brian Tiffin modified a comment on discussion The Lounge

    Just modified rcfunge ; Called from GnuCOBOL ; 0a"degnuf">:#,_$v v ; Factorial of number at (2,1) ; < >9 :>>#;1-:48*+22p*22g48*-#;1#+-#1:#<_$v v q# < >"ESAB"4#^($$:3Na,q cobol.b98, display a message, compute factorial 9, show in base 3, return value prompt$ cobc -x cofunge.cob bin/*.o -lX11 prompt$ ./cofunge funged 200102210000 GnuCOBOL: +0000362880 9 factorial; result limited to int cell size, which rcfunge has defined as 32bits. Just had to change rcfunge/src/main.c to separate out a bemain function...

  • Brian Tiffin Brian Tiffin modified a comment on discussion The Lounge

    And just because, not wasting enough time is wasteful Had to change cofunge.cob to mock up and env pointer for Befunge sys info command (and to add the -ux option to start up the "TURT" engine X11 interface). I also modified the quine program to end with an &q to accept an integer (which is returned to GNuCOBOL) to allow for the screen capture before image rundown. So the "@" on the fifth line from the bottom is "&q" in the new cut. The Funge code draws an image of itself using TURT, but as explained...

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    And just because, not wasting enough time is wasteful "TRUT"4(001-0p002-0p9a+y03-0p8a+y1+04-0pf2+f*1+::**1-N0C>0PQA01-0g02-0gg 84*1+- v v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v>v> :-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-:-: #1#1#1#1#2#1#1#1#1#2#2#1#1#1#1#1#1#1#1#1#1#1#2#2#2#1#1#1#1#2#3#3#2#2#1#1#1#1#1#5#4#1#2#1#1#3#1#9#6#3#3# >^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>^>v...

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    Just modified rcfunge ; Called from GnuCOBOL ; 0a"degnuf">:#,_$v v ; Factorial of number at (2,2) ; < >9 :>>#;1-:48*+22p*22g48*-#;1#+-#1:#<_$v v q# < >"ESAB"4#^($$:3Na,q cobol.b98, display a message, compute factorial 9, show in base 3, return value prompt$ cobc -x cofunge.cob bin/*.o -lX11 prompt$ ./cofunge funged 200102210000 GnuCOBOL: +0000362880 9 factorial; result limited to int cell size, which rcfunge has defined as 32bits. Just had to change rcfunge/src/main.c to separate out a bemain function...

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    Then I'd give Flint a try. Can't answer your orginal question really, but the Chiselapp team seems well regarded by the Fossil inner circles. Richard has pulled a few of their ease of use changes into upsteam. They seem to be on the frontline when it comes to fighting off spiders and bots, so I can only assume Flint is about as safe as it gets for public facing sites, (but I can't really speak to how much time is spent/wasted in the background to make it look easy - could be none, could be lots,...

  • Brian Tiffin Brian Tiffin modified a comment on discussion The Lounge

    No experience, Russell, but from a quick glance this looks like a very thin PHP layer around multiple Fossils with some session management. Can't really go wrong as the code looks easily digestable. For personal use, you might want to just use raw Fossil and something like #!/bin/bash fossil serve --repolist --localauth /home/user/fossils/ & The --repolist option will scan the given dir for (explicitly) .fossil files and throw up a small page to let you pick your repo. I use the above script and...

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    No experience, Russell, but from a quick glance this looks like a very thin PHP layer around multiple Fossils with some session management. Can't really go wrong as the code looks easily digestable. For personal use, you might want to just use raw Fossil and something like 1 2#!/bin/bash fossil serve --repolist --localauth /home/user/fossils/ & The --repolist option will scan the given dir for (explicitly) .fossil files and throw up a small page to let you pick your repo. I use the above script and...

  • Brian Tiffin Brian Tiffin posted a comment on discussion GnuCOBOL

    OCCURS DEPENDING ON The warning was added because COBOL usually wants 1 relative numbering, but an ODO field can be of indeterminent range. Zero lower bound being one common option versus the other common choice of one. I'd almost prefer the warning was dropped in favour of a documented assumption of zero; but then again, it does lead to questionable runtime test when bounds checking is requested. Maybe zero is not a valid number of entries. Kind of a 50/50, hence a warning. Cheers, Brian

  • Brian Tiffin Brian Tiffin posted a comment on discussion Help getting started

    Nice one, Eugenio. Have good, Brian

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    Should be working, but instead 3.a,0"1 peels"=2.a,0"1 peels"=v v < >>#;>:#,_@;"rock"a0?0a"srossics">:#,_@ 0 a " r e p a p " v : # , | @ Rock Scissors Paper, on 3 rsp.b98 I was 3-3-1 on the first 7 goes, so yayy, didn't lose to the computer; didn't win, but didn't lose. prompt$ cfunge rsp.b98 3 2 paper cfunge is a nice implementation; does Perl too. "LREP"4#v($$057*1-"n\rekcah lreP degnufeb rehtona tsuJ"f2*4+" tnirp"E>:#,_"LREP"4)a,@ >0a"elbaliava ton LREP">:#,_$@ perlit.b98 prompt$ cfunge perlit.bf...

  • Brian Tiffin Brian Tiffin modified a comment on discussion GnuCOBOL

    Federico; Try void gtk_window_set_interactive_debugging (gboolean enable); C as call "gtk_window_set_interactive_debugging" using by value 1 returning omitted end-call any time after gtk_init tres marvelous for exploring. GTK 3+ version 3.14 or better. I use COBOL external items a lot in GTK trials to get at fields inside event handlers. Probably going to change that around to properly nested sub programs or other form of sane refactoring. A sample GUI was using the same data names as the library,...

  • Brian Tiffin Brian Tiffin modified a comment on discussion GnuCOBOL

    Federico; Try void gtk_window_set_interactive_debugging (gboolean enable); C as call "gtk_window_set_interactive_debugging" using by value 1 returning omitted end-call any time after gtk_init tres marvelous for exploring. GTK 3+ version 3.14 or better. I use COBOL external items a lot in GTK trials to get at fields inside event handlers. Probably going to change that around to properly nested sub programs or other form of sane refactoring. A sample GUI was using the same data names as the library,...

  • Brian Tiffin Brian Tiffin posted a comment on discussion GnuCOBOL

    Federico; Try void gtk_window_set_interactive_debugging (gboolean enable); C as call "gtk_window_set_interactive_debugging" using by value 1 returning omitted end-call any time after gtk_init tres marvelous for exploring. GTK 3+ version 3.14 or better. I use COBOL external items a lot in GTK trials to get at fields inside event handlers. Probably going to change that around to properly nested sub programs or other form of sane refactoring. A sample GUI was using the same data names as the library,...

  • Brian Tiffin Brian Tiffin modified a comment on discussion GnuCOBOL

    I'd have to run some code to verify, and I'm not up on what the post processed code looks like, but move valore to "TEXT" of edit2. seems wrong. edit2 doesn't seem to have a field called "TEXT", not sure what C is being emitted, but it's likely borked. One thing to try is maybe move "TEXT" to valore of edit2 but that gets into a mismatched external definition, which is going to lead to other problems. The way GnuCOBOL handles EXTERNAL data items is by managing a small table at runtime. References...

  • Brian Tiffin Brian Tiffin posted a comment on discussion GnuCOBOL

    I'd have to run some code to verify, but move valore to "TEXT" of edit2. seems wrong. edit2 doesn't seem to have a field called "TEXT", not sure what C is being emitted, but it's likely borked. One thing to try is maybe move "TEXT" to valore of edit2 but that gets into a mismatched external definition, which is going to lead to other problems. The way GnuCOBOL handles EXTERNAL data items is by managing a small table at runtime. References use a table lookup to get an address. If not found, an entry...

  • Brian Tiffin Brian Tiffin posted a comment on discussion Contributions

    Can we discuss a little bit, Pat? First, add the license one liner to the source, otherwise it'll get lost in the mists of time. The listing might not always be near a forum post description. Next, passing on Bill's wisdoms, use FILE STATUS IS. See some of Steve's samples, worldcities for a grand tutorial on professional grade COBOL. https://sourceforge.net/p/open-cobol/contrib/HEAD/tree/trunk/samples/worldcities/ What can happen with AT END, is that it may never fire. Error status codes can defeat...

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    Seems this has affected perceptions from the top. During a discussion about the recent exploit potentials regarding stack clashing and sane values for rlimit And yes, a large part of this may be that I no longer feel like I can trust "init" to do the sane thing. You all presumably know why. Linus https://www.theregister.co.uk/2017/07/17/linux_4_13_rc1/ I wonder if enough people will pick up on the lack of faith to turn the tide. Jeers, Brian

  • Brian Tiffin Brian Tiffin posted a comment on discussion GnuCOBOL

    Here be dragons. These files are not for the fient of heart and there is a level of expertise that is off the chart. I still get confused and often times resort to flailing, becoming one of the inifinite monkeys relying on chance to come up with a "blarg, fine, good enough" solution before retiring from the battle field. :-) Cheers, Brian

  • Brian Tiffin Brian Tiffin posted a comment on discussion GnuCOBOL

    Well, there are a couple of ways to fix that Pat, but it's time for a bit of a redo on the whole Vim COBOL thing. Vim 8 ships with a new highlighter, updated in 2015 by Tim Pope, a Vim guru of some renown. I haven't gotten round to updating it for GnuCOBOL yet, but it starts out on some better footing than the one listed above. In the short term, for that file, change syn match cobolNumber "\<-\=\d*\.\=\d\+\>" contains=cobolMarker,cobolComment old to syn match cobolMarker "^.\{6\}" syn match cobolNumber...

  • Brian Tiffin Brian Tiffin modified a comment on discussion GnuCOBOL

    Hi Everyone and a Special "Hi" to Brian Brian, could you help me modify my cobol.vim file ? I just posted a small program in contributions. It adds sequence numbers. In vim, the sequence numbers are coloured the same way as variables in WORKING-STORAGE. I find it jarring to have 001002 01 foo pic x(20) when 001002 and 01 are the same colour. I have tried to hack my vim file but i can't figure out what line is responsible for the colour of the sequence numbers. Could you help me find it so i can change...

  • Brian Tiffin Brian Tiffin posted a comment on discussion Help getting started

    The SF "newsletters" are spam in disguise. That's an opt-out during registration (I think, been a long time). I do know, from jusdt looking that it is opt-in opt-out at anytime under Account Settings, Subscriptions, Newsletter. Subscriptions allow very fine control over any mail sent to the registration address, including none. I don't want people to fear having SourceForge accounts, it is not a problem. The anon posting feature is really meant for first timers, as a kind of "try before you buy"...

  • Brian Tiffin Brian Tiffin posted a comment on discussion Help getting started

    I'm going to defend this Anonymous poster, Pat. Me thinks you have crossed signals. Take a look at https://sourceforge.net/p/open-cobol/discussion/help/thread/51d48303/?page=3 and a few other places. These posts are all good natured, attempting to get clean VBISAM builds up and running. I might be wrong, but those anonymous posts have been far more helpful than neutral or troll-ey, with an eye to making the installs more robust. Seems to be a different person than "the others". ;-) Cheers, Brian

  • Brian Tiffin Brian Tiffin posted a comment on discussion Help getting started

    One of the reasons why we do not register is that we don't want your spam. What spam? You may have a mistaken impression of how SourceForge works, Anon. To the best of my knowledge, SourceForge has never leaked any registration emails. Email visible by having an account is of the form, username@users.sourceforge.net and you have to explicitly allow out bound messages being directed to your registration email. In site messaging is all that it is allowed by default, and even that can be turned off...

  • Brian Tiffin Brian Tiffin modified a comment on discussion Help getting started

    Depending on how the next few days go, Anonymous posting may be coming to an end for Help getting started. Something (I'm not going to say someone, as that elevates the persons responsible above the whale mud) is posting spam. A trickle for now, fairly easy to manage, but if it continues, the only real alternative will be to close the valve. This might not happen, but if it does, it will happen without notice or fore warning, so I'm posting a little heads up. Brian

  • Brian Tiffin Brian Tiffin posted a comment on discussion Help getting started

    Depending on how the next few days go, Anonymous posting may be coming to an end for Help getting started. Something (I'm not going to say someone, as that's elevates the persons responsible above the whale mud) is posting spam. A trickle for now, fairly easy to manage, but if it continues, the only real alternative will be to close the valve. This might not happen, but if it does, it will happen without notice or fore warning, so I'm posting a little heads up. Brian

  • Brian Tiffin Brian Tiffin posted a comment on discussion Open Discussion

    Should fork() be restricted? I have embedded Regina FUNCTION REXX() in GnuCOBOL as restricted by default, and a FUNCTION REXX-UNRESTRICTED() for explicit allow all scripting. Leaning to recommend REXX for user land scripting, but if a user script can execute fork(), things break down quickly in the hosting application. The change to unx_fork in unxfuncs.c is minor, with a TSD->restricted check, but for safer COBOL this would have to be in standard Regina installs, and patching (meaning from source...

  • Brian Tiffin Brian Tiffin posted a comment on discussion Help getting started

    Apologies to everyone that got mail. Finger slip, approved some spam instead of deleting it.

  • Brian Tiffin Brian Tiffin posted a comment on discussion Contributions

    Just clownin', Hilsen, but I grew up on OpenVMS/VAX, and well, competed with VAXen, some thought them much better. them be fightin' words! :-) Rock, scissors, paper, at dawn. Too bad the economies of scale put paid to the very well engineered middle ground of mini computers; so much less like toys than what most of us use as tools now. Cheers, Brian

  • Brian Tiffin Brian Tiffin modified a comment on discussion Contributions

    Simon Sobisch [2017-07-20 13:14]: What is ND COBOL? An old-school verison along the new one would be nice. Cobol (in this case COBOL-74) for the Norsk Data minicomputers https://en.wikipedia.org/wiki/Norsk_Data. Great machines, ran their own Sintran OS, dominated the Norwegian market, competed with VAXen, some thought them much better. The Cobol (74) had its own screen handling system, not compatible with SCREEN SECTION. They had a COBOL-85 with SCREEN SECTION, which I never used. Hilsen Harald small...

  • Brian Tiffin Brian Tiffin posted a comment on discussion GnuCOBOL

    For the peace of minded: prompt$ md5sum gnucobol-2.2-rc.tar.xz 9a0594866c358498d93f7349562ddd72 gnucobol-2.2-rc.tar.xz prompt$ sha1sum gnucobol-2.2-rc.tar.xz f79d0cbcdfd6714bb96f569fc4289e6962bf366d gnucobol-2.2-rc.tar.xz prompt$ sha3sum gnucobol-2.2-rc.tar.xz 3e2fe6094f3be2a3339bdffcc8d2d21f499d3743088f07f5d38b3e0f gnucobol-2.2-rc.tar.xz The XZ form. prompt$ md5sum gnucobol-2.2-rc_win.zip ffc8c08b9d43744d88f2bdbc16244d5e gnucobol-2.2-rc_win.zip prompt$ sha1sum gnucobol-2.2-rc_win.zip 5d02ea0c3963ac38f565a55e5bf13d92227f4517...

  • Brian Tiffin Brian Tiffin posted a comment on discussion GnuCOBOL

    .xz and _win.zip done. Dog was too jumpy to let me finish, so she took me for the night's run instead. ;-) Cheers, Brian

  • Brian Tiffin Brian Tiffin posted a comment on discussion GnuCOBOL

    gnucobol-2.2-rc.tar.gz has been uploaded to alpha.gnu.org prompt$ md5sum gnucobol-2.2-rc.tar.gz 95daec84e72034d4c1bcf216e0ebc34a gnucobol-2.2-rc.tar.gz prompt$ sha1sum gnucobol-2.2-rc.tar.gz 3724fc7cfcc3789deaa5af783ec0aca6aa1eb2ce gnucobol-2.2-rc.tar.gz prompt$ sha3sum gnucobol-2.2-rc.tar.gz 51b6527cf6fd5c18bc70a7c23bd43650271ed6ee764c69c76848f8e7 gnucobol-2.2-rc.tar.gz Signed with gnucobol-2.2-rc.tar.gz.sig GNU FTP is an anonymous ftp server prompt$ ftp anonymous ftp> cd gnu ftp> cd gnucobol ftp>...

  • Brian Tiffin Brian Tiffin modified a comment on discussion The Lounge

    LE looks like something we could add for system services. We should pick and choose some key ones from http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/CEEA3170/1.1?SHELF=CEE2BK70&DT=20060629134445 Drifting your topic a little bit now, Mickey, from GnuCOBOL in LE on frame, to LE emulation in GnuCOBOL ... anywhere. Part of the LE docs point to the z/OS XL C/C++ library. These we get with POSIX libc in probably mostly compatible form. For the curious; man pages exist (along with lots of other...

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    LE looks like something we could add for system services. We should pick and choose some key ones from http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/CEEA3170/1.1?SHELF=CEE2BK70&DT=20060629134445 Drifting your topic a little bit now, Mickey, from GnuCOBOL in LE to LE in GnuCOBOL. Part of the LE docs point to the z/OS XL C/C++ library. These we get with POSIX libc in probably mostly compatible form. For the curious; man pages exist (along with lots of other IBM generated documentation...

  • Brian Tiffin Brian Tiffin modified a comment on discussion The Lounge

    A m/f (mainframe) could do 20M record taxation processing within 1 day and more importantly spit out the individual reports out along with the assessments etc within 2 - 4 hours depending on the printer model. How? The DASD (HD) I/O is quick and by greater than a factor of 10 over one or more HD's used in a PC or server. DASD controllers handle a limited number of DASD's and there is a few of them to control a fixed number of them and speed of data going between the DASD's and the system so that...

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    Core not really equal CPU, 40,960 64bit chips, 256 cores per. What one earth could take that much processing power? Weather forecasting. Nuclear explosion simulators. Quantum physics explorations. Encryption cracking. Rendering 3D animations for big ticket movies. Anything that has to juggle huge numbers of information bits simultaneously. The higher the number of bits that can be thrown in the hopper at once, the more accurate the interaction predictions can become. Some mixed integer linear programming...

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    So, going from a z14 M05 (upto 170 core), to a TaiHuLight (10.6 million core) it would cost a little more to get a batch job fell over alert? ;-) (And yeah, z14 core not equal Sunway core, but it would be a neat billing record to see, once). Giving up some history ... a billing system we interfaced with at redacted was corrected, someone noticed a business feature was being under charged, a support tech thought the table change was minor, so in it went. The next end of month run back dated the correction,...

  • Brian Tiffin Brian Tiffin modified a comment on discussion The Lounge

    Two physicists pass each other in a hall "What's new?" The other looks over, utterly disappointed, "C over Lambda. Duh." The automated turnstile tallied 196 horses leaving the stable. When the rancher rounded them up, she had 200.

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    Two physicists pass each other in a hall "What's new?" The other looks over, utterly disappointed, "C over Lambda. Duh." The automated turnstile tallied 196 horses leaving the stable. But when the rancher rounded them up, she had 200.

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    Again, I'm still pretty envious while joking about with this, but Or, "We'll squeeze until we feel some stress, then we'll negotiate ... upwards. And don't fret too much, the machine pays itself". "But, but, bulk discounts?..."

  • Brian Tiffin Brian Tiffin modified a comment on discussion The Lounge

    New big iron. One of the bigger selling points seems to be end to end encryption. https://www.ibm.com/us-en/marketplace/z14 But this one got me: New software pricing tied to business value So, umm, the more you might make the less you pay, like a buying in bulk discount? ;-) Edit: Read a little more https://www-03.ibm.com/systems/z/resources/swprice/container.html Container Pricing will simplify pricing and billing on the IBM Z platform, by superseding a number of existing price offerings and by...

  • Brian Tiffin Brian Tiffin modified a comment on discussion The Lounge

    New big iron. One of the bigger selling points seems to be end to end encryption. https://www.ibm.com/us-en/marketplace/z14 But this one got me: New software pricing tied to business value So, umm, the more you might make the less you pay, like a buying in bulk discount? ;-) Edit: Read a little more https://www-03.ibm.com/systems/z/resources/swprice/container.html Container Pricing will simplify pricing and billing on the IBM Z platform, by superseding a number of existing price offerings and by...

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    New big iron. One of the bigger selling points seems to be end to end encryption. https://www.ibm.com/us-en/marketplace/z14 But this one got me: New software pricing tied to business value So, umm, the more you might make the less you pay, like a buying in bulk discount? ;-) Cheers, Brian

  • Brian Tiffin Brian Tiffin modified a comment on discussion The Lounge

    Red, the very big little programming stack just announced release 0.6.3 with macOS GUI support. Read all the sweet news at http://www.red-lang.org/2017/07/063-macos-gui-backend.html Big. Little. Go there. If you are not on a Mac, go there. If you are on a boat, get connected, go there. If you are stranded on a deserted island, get rescued, go there. Have good, go there, Brian

  • Brian Tiffin Brian Tiffin posted a comment on discussion The Lounge

    Red, the very big little programming stack just annouced release 0.6.3 with macOS GUI support. Read all the sweet news at http://www.red-lang.org/2017/07/063-macos-gui-backend.html Big. Little. Go there. If you are not on a Mac, go there. If you are on a boat, get connected, go there. If you are stranded on a deserted island, get rescued, go there. Have good, go there, Brian

  • Brian Tiffin Brian Tiffin posted a comment on discussion GnuCOBOL

    Adding for completeness: As Simon has pointed out; Note: if you had a checkout of branches/gnu-cobol-2.0 before the fastest way to get running again is svn switch ^/trunk. That's prompt$ svn switch ^/trunk If you are contributing, a smart move might be to pull a completely new tree svn checkout svn://svn.code.sf.net/p/open-cobol/code gnucobol-complete If you are only interested in the leading development, that is svn checkout svn://svn.code.sf.net/p/open-cobol/code/trunk gnucobol-dev If you are only...

  • Brian Tiffin Brian Tiffin posted a comment on discussion Contributions

    We have talked about how "call by name" routines are added. I know this is not ready for general consumption but could it be made to be? You are familiar with every scripting language under the sun. Is there low hanging fruit for people that don't really want to learn another language? Not really sure I understand the first part, Pat. Call by name, added in what sense. Assume the rest of this note is talking to everyone, Pat. What do you think is the easiest way for people who are in fact at entry...

  • Brian Tiffin Brian Tiffin posted a comment on discussion GnuCOBOL

    Just adding; WOOHOO!

  • Brian Tiffin Brian Tiffin posted a comment on discussion Contributions

    Well, it all starts as personal preference, and every one has their own, then it gets duked out via a committee of contributors to pick a style guide and then everyone strives to follow suit for the benefit of all (with no one completely happy as the common is usually a compromise). Or, as happens, the most prolific or early trail blazer gets to set a pace and then everyone else is expected to maintain the style. Comments blocks are fairly low hanging fruit and we probably won't get trapped with...

  • Brian Tiffin Brian Tiffin modified a comment on discussion GnuCOBOL

    Well met The times they are a-changin'. Trunk has finally been updated in SVN. Trunk is now the focus of lead development in the source code tree. Simon has made some sweeping changes to the repository. After many years of work, GnuCOBOL has a focal point. Trunk is going to be active source code, changes here will be leading edge. That means less stable than stable from time to time. Stable will be branches/gnucobol-2.x The old venerable trunk is now a tag, as tags/gnu-cobol-1.1 The action will be...

  • Brian Tiffin Brian Tiffin posted a comment on discussion GnuCOBOL

    Well met The times they are a-changin'. Trunk has finally been updated in SVN. Trunk is now the focus of lead development in the source code tree. Simon has made some sweeping changes to the repository. After many years of work, GnuCOBOL has a focal point. Trunk is going to be active source code, changes here will be leading edge. That means less stable than stable from time to time. Stable will be branches/gnucobol-2.x The old venerable trunk is now a tag, as tags/gnu-cobol-1.1 The action will be...

  • Brian Tiffin Brian Tiffin modified a comment on discussion Contributions

    Code sample for FUNCTION CONTENTS-OF(pointer, [length]) Reference modification allowed. Length request is optional and leads to a NUL byte scan if not given or 0. Zero length space returned for null pointers and pointers to empty strings. *>-<* *> Author: Brian Tiffin *> Dedicated to the public domain *> *> Date started: July 2017 *> Modified: 2017-07-15/11:26-0400 btiffin *>+<* *> *> contents.cob, demonstrate FUNCTION CONTENTS-OF *> Tectonics: cobc -xj contents.cob *> >>SOURCE FORMAT IS FREE identification...

  • Brian Tiffin Brian Tiffin modified a comment on discussion GnuCOBOL

    Sweet. One thing I'd like to clear up, as it's going to burn people that don't know. C calling COBOL, page 553, 9.9.5. That no longer works. IDENTIFICATION DIVISION. PROGRAM-ID. subcob. DATA DIVISION. LINKAGE SECTION. 01 Arg1 PIC X(7). 01 Arg2 PIC X(7). 01 Arg3 USAGE BINARY-LONG. PROCEDURE DIVISION USING BY VALUE Arg1, BY REFERENCE Arg2, BY REFERENCE Arg3. 000-Main. DISPLAY ’Starting cobsub.cbl’ DISPLAY ’Arg1=’ Arg1 DISPLAY ’Arg2=’ Arg2 DISPLAY ’Arg3=’ Arg3 MOVE ’X’ TO Arg1 (1:1) MOVE ’Y’ TO Arg2...

  • Brian Tiffin Brian Tiffin posted a comment on discussion GnuCOBOL

    Hmm, more work to do... I can't get that example to work as is anymore. And it's not because of what I was mistakenly all snarky about. That C sample does the proper thing when calling cob_init. My problems are more localized to gcc not scanning the proper usr/local subdirs. Foot mouth meet again. Jeers, Brian

  • Brian Tiffin Brian Tiffin modified a comment on discussion GnuCOBOL

    Sweet. One thing I'd like to clear up, as it's going to burn people that don't know. C calling COBOL, page 553, 9.9.5. That no longer works. IDENTIFICATION DIVISION. PROGRAM-ID. subcob. DATA DIVISION. LINKAGE SECTION. 01 Arg1 PIC X(7). 01 Arg2 PIC X(7). 01 Arg3 USAGE BINARY-LONG. PROCEDURE DIVISION USING BY VALUE Arg1, BY REFERENCE Arg2, BY REFERENCE Arg3. 000-Main. DISPLAY ’Starting cobsub.cbl’ DISPLAY ’Arg1=’ Arg1 DISPLAY ’Arg2=’ Arg2 DISPLAY ’Arg3=’ Arg3 MOVE ’X’ TO Arg1 (1:1) MOVE ’Y’ TO Arg2...

  • Brian Tiffin Brian Tiffin posted a comment on discussion GnuCOBOL

    Sweet. One thing I'd like to clear up, as it's going to burn people that don't know. C calling COBOL, page 553, 9.9.5. That no longer works. IDENTIFICATION DIVISION. PROGRAM-ID. subcob. DATA DIVISION. LINKAGE SECTION. 01 Arg1 PIC X(7). 01 Arg2 PIC X(7). 01 Arg3 USAGE BINARY-LONG. PROCEDURE DIVISION USING BY VALUE Arg1, BY REFERENCE Arg2, BY REFERENCE Arg3. 000-Main. DISPLAY ’Starting cobsub.cbl’ DISPLAY ’Arg1=’ Arg1 DISPLAY ’Arg2=’ Arg2 DISPLAY ’Arg3=’ Arg3 MOVE ’X’ TO Arg1 (1:1) MOVE ’Y’ TO Arg2...

1 >

Get latest updates about Open Source Projects, Conferences and News.

Sign up for the SourceForge newsletter:

JavaScript is required for this form.





No, thanks