Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Topics - Alfredo Fernández Díaz

Pages: [1]
Programming / Any standard on exit codes?
« on: April 30, 2024, 12:54:24 am »
Hi everyone,
I am writing a few tools that can report a number of different conditions, e.g. "target file does not exist," "file can't be read," "wrong file type," "parameter out of range," etc. While well-chosen error messages are the best you can give users, when tools are to be chained, it is better to pass along different numeric return codes. Now the question is, in order to cooperate most properly, every tool must know at least the error codes of the previous one, so I wondered whether we have any kind of standard or guidelines regarding this, instead of having a plethora of necessarily arbitrary values.
Thank you in advance.

If you have appropriate keyboard mappings set up, you can change object titles using only a standard keyboard, which I like to do, because I don't like mice much and sometimes I just don't have one around.

In my case, I usually have F6 mapped to edit object titles, so I just select an object, hit F6, change its title (maybe splitting long ones into lines with Enter), and press the NumPad Enter to finish and save when I'm done. Naturally, the key to my question is this last NumPad Enter key in particular:

On a small laptop, I have a 80+ keys keyboard, with no num pad, so no NumPad Enter. Is there any alternate system key mapping (editable or otherwise) that lets me end editing titles, and save the new one (Esc will finish the operation too, but you get no cake)?

I know I can Alt+Enter to open the object properties dialog, navigate to the Icon tab, enter a new title, and close, but that is kind of cumbersome (and most likely the reason why editing object titles can be mapped to a single key in the first place).

Thank you in advance,


I seem to recall there is a message file where you can read the localized equivalents of "y" and "n" (for example that would be "j" and "n" --ja/nein-- for German) used in NLV OS/2 systems for stuff like simple confirmations, and such.

Any hints/help/reminders?

Thank you in advance,


I need help from anyone who knows how to use a Russian / Ukrainian keyboard layout (or any other dual-layer one I guess) on OS/2, to type Cyrillic characters on a text mode session.

Just in case you wonder, I have a Russian system, and other OS/2 systems with 866 as the secondary system codepage. If I execute "CHCP 866" as necessary at a command prompt, then execute the attached rexx script, text-mode Cyrillic characters display just fine (see attached screenshot Cyrillic.png).

If I then execute "keyb ru" as necessary, I get the expected messages stating the necessary conversion codepage tables loaded just fine. The problem seems to be that the Russian keyboard starts in English / Latin mode and I can't switch to use the Russian layer of the text mode keyboard layout.

I.e., if I start say "AE" from the same command prompt, typing will get me English / Latin characters in the system editor, but if I press Left Alt + Left Shift I can type Cyrillic characters no problem (see attached capture MoreCyrillic.png), and pressing Right Alt + Right Shift gets me back to Latin mode. Closing AE gets me back to the same prompt window, but sure enough, my Russian keyboard layout stays in Latin mode at the command prompt.

So the problem is, how do I switch between the Latin and Cyrillic layers of the text mode Russian keyboard layout? Does anyone know where the layer switch hotkeys are documented?

Thank you in advance,

Hi everyone...

Google sent me this a few weeks ago:
On May 30th, you may lose access to applications that use a less secure login technology.

To protect your account, Google will not allow to use applications or third-party devices that ask you to login to your Google account using only your user name and password. Instead, you will have to login via "Sign in with Google" or other more secure technologies, like OAuth 2.0.

What you have to do:

E-mail software, like Outlook 2016 or previous versions, accesses your Gmail account through less secure means. Switch to Office 365, Outlook 2019 (or a newer version), or another email software in which you can login through "Sign in with Google".

Somewhere else, I read how to setup gmail accounts to use OAuth 2.0 under Mail Preferences in ThunderBird, and that it just worked, so I forgot about the issue until I actually tried to do the same recently on SeaMonkey on OS/2...

To the best of my knowledge, the latest SeaMonkey we have available (2.42.9esr 20200322125440) does not support OAuth 2.0. (or I can't find it in the menus under email and news preferences), which in turn I take to mean no more GMail from OS/2. I am not big on using GMail so no problem letting it go if necessary, but this is happening at a most inconvenient time for me, and obeying Murphy's Law the 'cutoff' on the 30th will of course happen while I am on the go.

Am I missing something? Is there some way to use OAuth 2.0 to login to gmail accounts from OS/2 SM / TB?

Thank you in advance,

Hi everyone,

I have this old application that needs some files updated/replaced and a few more added after install. No biggie, except when you forget the fixes, and are reminded of them a few days later, the first time the program is started. So I thought, why not slipstream it?

You know, as in, unpack the installation media, replace / add files, repack it all, and have an installation that doesn't need upgrades and/or fixes afterwards.

This being directly based on IBM installers stuff, installation media was easy enough to unpack using, well, IBM's UNPACK or UNPACK2 (both work). However, I have no idea how to create compatible files that will be handled by the UNPACK routines later on. A quick search at Hobbes was not obviously helpful, so any ideas are welcome.

"PACK", anyone?

Thank you in advance,

Utilities / Manual for eFTE/2 / eFTE / FTE (Folding Text Editors)
« on: November 29, 2017, 07:41:08 pm »
A long time user of FTE, tester of derivatives, and writer of extended configurations, I agree with every other user I've heard say that the worst thing about this fabulous editor is the lack of a good manual -- if there had only been one when I started using it...

A few days ago I finished writing a manual for the current eFTE/2 (99.5% of which still applies to the original FTE), thinking of what I would have liked to know when I started using it, and I thought it would be a good idea to find out if polishing it a bit for others out there is worth the effort...

Do you know FTE? Do you still use it? Would this make you try the editor? Do you think I should improve this manual? How? Am I better off devoting my time to other stuff?

From the Netlabs page where it belongs:
It's still not a real manual, though, but a first step in having one.

If I were to set priorities, I would say the first thing to do is review the internal commands section for accuracy and adequacy of the descriptions. After that, maybe see if more useful information can be added to the task reference (Using eFTE2 / Standard text editing). Currently it is little more than a collection of links from menu items to internal commands.

Of course, the whole tutorial / customizing section can be reviewed for usefulness, and it might be useful to add screenshots here and there (all over the manual, actually) to further illustrate some points / examples.
but those are my own thoughts. What are yours?

Pages: [1]