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.


Messages - Rich Walsh

Pages: [1] 2 3 ... 23
1
Applications / Re: XWorkplace Xview - Icon Size
« on: March 21, 2024, 05:52:16 pm »
Do you know if there is a way to increase the icon size on the "Xview" view in XWP?

If you want full-sized icons (40px), you can select View->Options and uncheck "Small icons".

Quote
It is something on xWorkplace or in DynamicIcons?

Dynamic Icons was designed to support one window per folder - it does not understand Xview's file-manager design. Now that I have a better idea of how Dynamic Icons works internally, I can see several ways to add support for Xview. The question I have is whether it is really appropriate to do so.

XWP is OSS licensed under the GPL. D.I. is proprietary software owned by ecoSoft that can only be distributed as part of AOS. Should I add support to XWP for a feature that only benefits some users - even if I can do it in a way that doesn't violate the GPL?

It would be more appropriate to add this support to D.I. but I don't own or control it. And quite honestly, I'm unwilling to work on profit-generating software without being paid for my efforts. As it happens, I just finished working on a major update to D.I. that AN did pay me for. But, are they willing to put out more money for yet another round of enhancements? I don't know.

2
Applications / Re: XWP 1.0.17: any availability date ?
« on: March 20, 2024, 05:11:07 pm »
Out of interest but also need (I had raised 2 tickets): is there a rough availability date for XWP 1.0.17 ?

Normally, I don't release a new version of XWP until there are enough new features and bug fixes to make it worth the user's time and effort to install it. That typically occurs every 12-18 months (the last one was released 4 1/2 months ago). As it happens, I think I've achieved this "critical mass" sooner than usual, so I expect to do a new one in the next several months.

Quote
I was curious and looked into the XWorkplace/XWPHelpers SVN repos but could not see any checkins from Rich. I therefore suspect that currently, he only has done the changes to his local copy.

I usually keep the Netlabs repo at the same level as the most recent release. Anyone who builds it himself and then runs into problems can compare his results to the released binaries to determine if he introduced the issue or if it was already there but hadn't been noticed before. Further, I don't think it's helpful for other builders to have to wade through my mistakes and delays (i.e. a commit, 2 fixes for that commit, another commit a month later to finish it, etc, etc.). Before I commit anything, I consolidate all of my patches for a given feature or bug-fix into a single coherent patch that can be rolled-back easily if the builder wishes.

3
Programming / Re: FAT32 vs FAT16
« on: March 20, 2024, 03:33:24 pm »
indeed I'm missing information but it doesn't exactly answer the question as to why lvmgui reports a partitions entry as corrupt if it's FAT16 and not when it's FAT32?

I didn't "exactly" answer your question because there are no valid answers to invalid questions.

You seem to be convinced that this is a FAT16/FAT32 problem. If it is, then changing the '06' to '0C' should fix it. Have you tried that? I'd be really, really amazed if it did because LVM doesn't care what filesystem is in use. It cares about whether the partition's start/end/size look valid and whether the corresponding DLAT info is present and valid. And from what you say, it isn't.

BTW... neither of the questions you posted provide any clue as to what the _real_ problem is or how your system came to be in its current state. Without context, it's awfully hard to provide a useful answer. For example, you say that you're "missing information" without identifying whether both images are missing this info or just one is (and if just one, which one). Stuff like this matters - and without it, it's just another case of "Garbage In, Garbage Out".

4
Programming / Re: FAT32 vs FAT16
« on: March 19, 2024, 06:42:47 pm »
There's more to an OS/2 partition than just its partition table entry: there's also its entry in LVM's DLAT (Drive Letter Assignment Table). For primary partitions, this is stored in the last sector of the disk's first track. Using C/H/S notation 0/0/63 on most disks (i.e. LBA 0062).

This isn't something that's easily edited because the DLAT header contains a CRC for that sector - changing anything renders the CRC invalid. Instead, you may want to try copying this sector from one image to the other.


5
Programming / Re: Partition alignment
« on: March 18, 2024, 05:41:06 pm »
Looking at the Partition Alignment section at https://www.arcanoae.com/faqwd/how-can-i-tell-if-my-current-disk-drive-is-compatible-with-arcaos/ it says:

 > ArcaOS expects partitions to be CHS-aligned

Does anyone know what it means by CHS-aligned?

For a comprehensive discussion of hard drive geometry (including some helpful illustrations) see https://en.wikipedia.org/wiki/Cylinder-head-sector. After reading that, the following will make sense: a CHS-aligned partition can start and end at any cylinder but it must start at the very beginning of the first cylinder (i.e. Head 0 Sector 1) and end at the very end of the last cylinder (typically, Head 255 Sector 63).

F.Y.I.  the AOS FAQ you cited is out-of-date: AOS 5.1 supports both NVME drives and the GPT partitioning scheme. It is also incorrect since OS/2 _does_ support partitions that are not cylinder-aligned (though no OS/2 tool will ever create a non-aligned partition).

6
Programming / Re: About somFree
« on: March 15, 2024, 07:06:29 am »
I was wondering it as an experiment to have a WPS class that has Qt elements on it, instead of PM.

No experiments needed, the results are already known: if the thing is well-crafted, it will look nice. But at what cost?

Here are the DLLs that will drawn into the WPS if this widget or whatever links to nothing but 'qt5core.dll':
libcn0 stdcpp6 gcc1 libcx0 libc066 z1 icuuc icuin icudt pcre260

Each one of these links to other stuff - and more than likely, this thing will link in still more stuff - none of which exists in the WPS process currently. Do you really want to burden your entire system from the time it boots up with the shared memory demands of some marginal new widget? I hope not.

And why? What is in such desperate need of whatever Qt has to offer? If you were proposing replacing the Desktop with Qt, I'd be a lot more sympathetic than I am to the idea of tacking on some Qt-whatever "just because you can".

7
Applications / Re: XWP v1.0.16
« on: March 14, 2024, 11:46:16 pm »
Is the "IFX Options" GUI also yours?
Can it please be also possible to respect the XWP font size settings?

This dialog uses 10.Helv and I'd like to continue using that typeface. I'll see what I can do about the sizing...

The next version will use 10/12/14/18pt Helv. You won't see any change unless your 'Text Size' setting is X-Large or greater (there isn't a 9pt or 11pt Helv, so Small/Medium/Large all use the current 10.Helv).

8
Programming / Re: About somFree
« on: March 14, 2024, 07:18:21 pm »
Is it possible to create a SOM Class in OS/2 that runs a Qt GUI interface? Or some kind of "bridge" is missing? Of course it would not be as fast as PM, but just wondering.

IMHO, it would be a collossal mistake to weigh the WPS down with all of the baggage that Qt requires (libc, libcx, libcn, etc, etc). Do you really want to risk having your Desktop fail to start because of some ill-considered change to any one of a dozen DLLs? In any case, it isn't SOM that determines what the GUI looks like - it's the developer.

9
Applications / Re: XWP v1.0.16
« on: March 14, 2024, 07:01:23 pm »
Is the "IFX Options" GUI also yours?
Can it please be also possible to respect the XWP font size settings?

This dialog uses 10.Helv and I'd like to continue using that typeface. I'll see what I can do about the sizing...

10
Programming / Re: Bigicons
« on: March 10, 2024, 03:31:17 am »
my guess it is a workplace shell limitation at this time.  Rich had said WPShell was a mess

No, I said Dynamic Icons was a mess. Multiple shortcomings prevent Details view support and limit Tree view to the same sized icons as Icon view. Here's why:

1) DI's "dirty little secret" is that it creates a separate, standalone PNG image for every single object you see during a session. When you open a folder with 100 JPGs, it generates 100 png images each requiring 16k. The same is true for every folder, every datafile, every everything. Over the course of a session you could end up with thousands of images.

2) DI caches these images to avoid the compute-intensive job of recreating them. However, each object can only cache one icon. If you display an object in Icon view at 64 pixels, then open the same folder in Details view using 32 pixel icons, DI will have to destroy all of the 64px images and recreate them at 32px. In a worst-case scenario where both views are open, DI might have to destroy and recreate every icon for every refresh!

Happily, there's a solution in the pipeline for #1. After our earlier discussion, I rewrote large parts of DI to eliminate duplicate icons as much as possible. Some users may see an 80-90% reduction in DI's memory usage when AN releases it. However, it will require a thoroughly revised 'bigicons.txt' file to achieve these savings. @David, could you send me a PM with your email address so we can discuss this and exchange files?

Sadly, the shortcomings that prevent Details view support and limit Tree view will remain for the foreseeable future...

11
Programming / Re: DosDevIOCtl disabling keyboard in VIO mode
« on: March 07, 2024, 07:19:01 pm »
Quote
I'm surprised that the program you wrote about originally ever ran if it had this kind of layout.
The original program (almost same source code) was built with Watcom tools. What you see above is a new suite of tools.

If you expect these programs to run on OS/2 - even if only to prove that (some) OS/2 programs run on your OS - then fix your tools to create bona-fide OS/2 executables. It's not optional.

12
Programming / Re: DosDevIOCtl disabling keyboard in VIO mode
« on: March 07, 2024, 05:56:27 pm »
pdld has been updated to fix a couple of errors, but result is still the same (OS/2 says it can't run it).
I have suggested that the MZ header is too small for OS/2 to cope with, but he thought it was one of the other reasons.

How did you link this thing? Its layout doesn't look anything like an OS/2 LX executable.
Using 'exehdr' (from the OS/2 Toolkit) I find this:

Code: [Select]
no. virtual  virtual  map      map      flags
     address   size    index    size
0001 00400000 00005fcd 00000001 00000006 EXECUTABLE, READABLE, 32-bit
0002 00406000 000006d8 00000007 00000001 READABLE, WRITEABLE, 32-bit
0003 00407000 000067b0 00000008 00000007 READABLE, WRITEABLE, 32-bit

The virtual addresses (2nd column) for a small program like this should be:
0001 00010000
0002 00020000
0003 00030000

The "00400000" looks more like a 16-bit segmented address (i.e. 0040:0000).

Then too, we have:

Code: [Select]
Initial CS:EIP:                 object 1 offset 00000000
Initial SS:ESP:                 object 4 offset 00008000

Your stack is in a non-existent segment. Your entrypoint _could_ be at offset 0x00 but seldom is in most programs. I'm surprised that the program you wrote about originally ever ran if it had this kind of layout.

13
Programming / Re: DosDevIOCtl disabling keyboard in VIO mode
« on: March 04, 2024, 06:46:11 am »
Quote
What are you trying to do?
I am trying to switch off echo and line buffering so that I can write a fullscreen text application, like microemacs 3.6. As a pure 32-bit app (no thunking within my executable). As such - OS/2 should not do keyboard character echoing. The application will deal with that.

I've never had the slightest interest in writing a true text-mode app, so I have no hand-on experience. However, perusing the 'Control Program Reference', I found KbdSetStatus(pkbdinfo, hkbd). The description of the KBDINFO structure's "fsMask" member lists these flags:

Code: [Select]
fsMask (USHORT)
State mask.

  Bits 15-9 Reserved.
  Bit 8 Shift return is on.
  Bit 7 Length of the turn-around character (meaningful only if bit 6 is on).
  Bit 6 Turn-around character is modified.
  Bit 5 Interim character flags are modified.
  Bit 4 Shift state is modified.
  Bit 3 ASCII mode is on.
  Bit 2 Binary mode is on.
  Bit 1 Echo off.
  Bit 0 Echo on.

Isn't this what you're looking for? (I _assume_ this is a 32-bit entry point to a 16-bit function.)

14
Applications / Re: XWP v1.0.16
« on: March 04, 2024, 05:30:04 am »
Query os2.ini->PM_SystemFonts->WindowText to get the font name & size, then use it.

Which specific lswitch282.ini key are we talking about here?

I have no idea. I was simply identifying how you get the current general-purpose font from os2.ini. How lswitcher uses it is up to Gregg...

15
Applications / Re: XWP v1.0.16
« on: March 03, 2024, 07:26:08 pm »
What do you think it is needed for lSwitcher to also respect the XWP font sizes too?

Query os2.ini->PM_SystemFonts->WindowText to get the font name & size, then use it.

Pages: [1] 2 3 ... 23