Hi Stefan,?
Thanks for the hints. I used Rexx a lot (since VM/SP3 on a 4341
to the abandon of OS/2) [Then I switched to Java in order to avoid
Windows even if my shop was using it) and since my retirement, I
have played with lua (quite nice) so my rexx skills are quite
rusted.
My discovery of Hercules and the ability (mailny due to René
Ferland so kind help) of having a vm-esa working machine will
certainly help me in re-mastering language.
Thanks again. have a nice day
Jean-Pierre
Le 03/04/2025 à 14:52, Stefan A.
Haubenthal via groups.io a écrit?:
Hi Jean-Pierre,
Just two hints:
#!/usr/bin/env rexx
makes your script more portable
Do l = 1 while lines(csvFile) > 0
is a short form of your condition
--
VM/370 CE V1R1.2, Hercules
3.13 on macOS
--
Jπr
|
Hi Jean-Pierre,
Just two hints:
#!/usr/bin/env rexx
makes your script more portable
Do l = 1 while lines(csvFile) > 0
is a short form of your condition
--
VM/370 CE V1R1.2, Hercules 3.13 on macOS
|
Alas, it was far from perfect. I've enhanced the results (finding
out all the titles I could and doing my best for finding was
hidden before 8 letters ones and then sort on the documents
numbers and remove the duplicate entry. And as I'm not fully
satisfied with the results, I thought that each of us would find a
need for some improvement.?
Formerly, I used the lua language (easy to handle bunches of line
and providing a good json interface) I thought that it would be
more suitable to use a language we all share : REXX.? My lua
program generated a csv file (easily modifed, handled with
libreoddice-calc like programs [who said Excel ?]? or a plain text
editor where the url, the document number and the title on every
document are on one line separated by a tab (x09) character.
The csv2html file is just a way to transform the contents of the
csv file into a table displayed by any web browser. It is not a
state of the art program just a tool cvs2html ibmdocs.csv cat.html
will produce a cat.html file identical to the catalog.html which
is in the zip file. Warning The output file is not erased by the
program. you may end up with a double size catalog, the new
version of it being at the end of the file. (it is also a way to
have a table with the MVS entries, then one with VM, JES, JES2,
JES3 ...
Just keep the imbdocs.csv file in a safe place and do whatever
you wish .
Enjoy !
Tanks again for the deep knowledge you make available for
everybody
Jean-Pierre Cabanié
?
Le 30/03/2025 à 05:21, Fish Fish via
groups.io a écrit?:
PERFECT!
THANK YOU, René!
--
Jπr
|
The following items have been added to the Files area of the [email protected] group.
By: Jean-Pierre Cabanié <jean-pierre@...>
Description:
IBM Boulder documentation catalog (HTML and CSV) and REXX translator
|
Dear h390-vm ?followers,
?
Till Winkler is currently conducting a survey on the usage of REXX.
?
If you are using REXX, please participate and forward this link:
?
The results will be presented on the REXX symposium in Vienna in May.
?
Thank you & best regards
?
Mike
?
|
I'm really sorry, but a typo in the program missing.lua made some
entries undetected and hence the html output René Ferland uploaded
for me a few days ago was not 100% correct.? I'm currently working
on a new set of tools that will (I hope) not only be bug-free but
produce a better looking html output (a table instead of a list).
Please accept my most sincere apologies for this poor
contribution.
Jean-Pierre Cabanié (a bit ashamed)
Le 30/03/2025 à 21:57, Andre via
groups.io a écrit?:
On Sun, Mar 30, 2025 at 10:35 PM, René Ferland wrote:
?
Waterloo SCRIPT is available on VM/370 CE, type "help
script". The manual is in SCRIPT LISTING U.
?
Cheers,
?
Rene FERLAND, Montreal
?
P.S. -- I remember using SCRIPT on MUSIC/SP for course
material when I taught Fortran and SAS back in 1982.
Wow!! This is so cool! I will definitely try it out!
?
Thanks, Rene!
--
Jπr
|
On Sun, Mar 30, 2025 at 10:35 PM, René Ferland wrote:
?
Waterloo SCRIPT is available on VM/370 CE, type "help script". The manual is in SCRIPT LISTING U.
?
Cheers,
?
Rene FERLAND, Montreal
?
P.S. -- I remember using SCRIPT on MUSIC/SP for course material when I taught Fortran and SAS back in 1982.
Wow!! This is so cool! I will definitely try it out!
?
Thanks, Rene!
|
On Sun, Mar 30, 2025 at 12:30 PM, Andre wrote:
Is there SCRIPT/VS foreground processor for MVS 3.8 or VM/370 by any chance?
?
?
Waterloo SCRIPT is available on VM/370 CE, type "help script". The manual is in SCRIPT LISTING U.
?
Cheers,
?
Rene FERLAND, Montreal
?
P.S. -- I remember using SCRIPT on MUSIC/SP for course material when I taught Fortran and SAS back in 1982.
|
Thank you, Ross! Very interesting info.
Is there SCRIPT/VS foreground processor for MVS 3.8 or VM/370 by any chance?
?
Best wishes,
Andre
|
What program IBM used in 1970s and 1980s to prepare, format and typeset those docs??
I know that UNIX crowd usually used something like TROFF or NROFF, then TeX.
But I doubt that IBM didn't have their own software for this task.
IBM did indeed have its own software for this task.? And it made it available to their customers, for a fee, of course.? The core was the IBM "SCRIPT" product,?sold under several names,?but most-lately "Document Composition Facility" (DCF).??
SCRIPT was inspired by the ur-formatter, RUNOFF[1], which ran on MIT's Compatible Time Sharing System (CTSS), and goes back to the late 1960s.? Soon afterward, some IBM'ers built a set of SCRIPT macros and a pre-processor to support writing documents in Generalized Markup Language[2].? IBM released that work as the "GML Starter Set" for SCRIPT/VS.? Internally, it continued to develop the markup language and came up with what was eventually?released to customers as "BookMaster", a fully-fledged book composition and formatting system.? The ".BOO" files that we've been discussing were produced as one form of BookManager output (it could also do text-only, fully-typeset, etc.).? IBM also released a set of programs that read those files, called "BookManager", and even a free version called "IBM Library Reader" for OS/2 and Windows, that could read documents specially "stamped" for it.
In the 1990s, as HTML[3] took off, Gary Richtmeyer, of IBM and later AT&T, wrote "B2H", a BookMaster-to-HTML converter that did a great job of putting BookMaster books on the web (from the source form, not the .BOO files).? That code was released freely and runs today just fine on VM/SP and later (I haven't tried it on VM/370 yet), OS/2, Windows, Linux (and other unices), and MVS (and presumably z/OS).
Ross
[1] RUNOFF was also the inspiration for roff,?the original Unix formatter. [2] GML was the inspiration for XML. [3] GML was also the inspiration for HTML.
|
I have a strange question, but maybe someone here knows the answer.
What program IBM used in 1970s and 1980s to prepare, format and typeset those docs??
I know that UNIX crowd usually used something like TROFF or NROFF, then TeX.
But I doubt that IBM didn't have their own software for this task.
?
Best wishes,
Andre
|
[...] By: René Ferland <ferland.rene@...>
Description:
Yet another program (by Jean-Pierre Cabanié) that convert the JSON file of
Jon Templin to a HTML file. The HTML output is included in the ZIP. PERFECT! THANK YOU, René! -- "Fish" (David B. Trout) Software Development Laboratories mail: fish@...
|
The following items have been added to the Files area of the [email protected] group.
By: René Ferland <ferland.rene@...>
Description:
Yet another program (by Jean-Pierre Cabanié) that convert the JSON file of Jon Templin to a HTML file. The HTML output is included in the ZIP.
|
Re: IBM Documentation Hidden At IBM
On Sat, Mar 29, 2025 at 07:41 AM, Ken Dreger wrote:
Would you mind posting the code for your JSON effort here? Maybe some of us can take a look at it and figure a way to make it work different??
?
The code is a .NET 8 console app written in C# and developed with Visual Studio 2022 under Windows. It makes use of the PdfPig NuGet package to parse the PDFs.
?
The input file with the list of URLs is included with the project. The longest time taken by the app is fetching each PDF off the internet. On my i5-8400 dev box, it takes about 40 minutes to run all the files.
?
Take what's useful "as-is".
?
Jon T.
|
Re: IBM Documentation Hidden At IBM
Hi Ken,
?
thank you. I wrote a little JavaScript program for the transformation. You need NodeJs to run it.
?
Here is the source code (it includes Jon's JSON data):
?
You run it (e.g. on the Windows command line) with: node generate_docu.js >ibm_documentation.html
?
As I wrote, it is quick and dirty and no example of elegant JavaScript coding. It generates the ouptput just with console.log statements.
?
Maybe it even runs on a modern IBM mainframe :-). As far as I know, NodeJS is available for z/OS.
?
Feel free to adapt it to your needs.
?
Best regards
?
Norbert
|
Re: IBM Documentation Hidden At IBM
Would you mind posting the code for your JSON effort here? Maybe some of us can take a look at it and figure a way to make it work different??
?
BTW, thank you for your fine effort.
|
Re: IBM Documentation Hidden At IBM
On Fri, Mar 28, 2025 at 10:48 PM, Fish Fish wrote:
Not to look a gift horse in the mouth (because I very much do appreciate your effort, Jon!), but I was kind of looking for something like simple HTML
?
You're welcome.
?
My thinking: JSON is so ubiquitous nowadays, you can shape it into anything you want. Or even have Chat-GPT do it for you.?
I was in a hurry, and in C#, I could serialize JSON with one or two statements.
?
Jon T.
|
Re: IBM Documentation Hidden At IBM
Hi Jon,
?
thanks. This is really great. I made a very quick and dirty conversion of your JSON to HTML.
?
You find it here:
?
Greetings
?
Norbert
|
Re: IBM Documentation Hidden At IBM
Hello guys!
?
Please find attached excel database. I made it from JSON for myself, but decided to share. Hope it helps.
https://docs.google.com/spreadsheets/d/14gY-ZKwfgf_srx0lY0aRQQZ-Wf1ltV2d/edit?usp=sharing&ouid=111879433608677265990&rtpof=true&sd=true
?
Best wishes,
Andre
|
Re: IBM Documentation Hidden At IBM
Jon Templin wrote: {
"Url": ",
"Title": "ID6TUMST",
"Author": "",
"Subject": "",
"Keywords": [],
"DocumentNumber": "GA22-1030-03",
"FirstPage": "S/390\uF6DAParallel Enterprise Server - Generation 6IBM
System Overview Level 03a, February 8, 2001 GA22-1030-03"
}, [...] Not to look a gift horse in the mouth (because I very much do appreciate your effort, Jon!), but I was kind of looking for something like simple HTML, such as: <ul> <li><a href=">GA22-1030-03 ID6TUMST</a> <li><a href=">SA22-7832-04 Principles of Operation</a> <li><a href=">SA22-7832-05 IBM z/Architecture Principles of Operation</a> ...etc... </ul> That is to say, a simple html unordered list of hyperlinks, whose name consists of your "DocumentNumber" value followed by tour "Title" value (with a space in between), which RESOLVES to your "Url" value, so that when published on a web page, allows one to simply click on the link to obtain the actual document. To find the document they want, they can just do a simple search of the web page in question (i.e. ctrl+f in most browsers). Maybe someone could take the data that Jon extracted and throw together a simple script to do that? And then publish the results somewhere? But THANK YOU for doing the hard part for us, Jon! MUCH appreciated! Enjoy your vacation! -- "Fish" (David B. Trout) Software Development Laboratories mail: fish@...
|
Jean-Pierre Cabanié
Stefan A. Haubenthal
René Ferland
Ken Dreger