TSE 4.50 RC 18

[S.E. Mitchell]

Hopefully the last one before the official 4.50 release (famous
last words :( )

https://semware.com/files/tse-pro-install/tse-setup-4.50-rc18.zip
https://semware.com/files/tse-pro-install/tse-linux-4.50-rc18.tgz
https://semware.com/files/tse-pro-install/sanity1.s

sanity1 is a "make sure TSE kinda sorta works" macro.
If you get a chance, please try running sanity1 on your machine.
Suggestions/additions for sanity1.s are greatly appreciated!

---------------------------------------
16 Feb 2024 v4.50 release candidate 18
---------------------------------------
New FF* and PB* commands:
FFisDir()
FFisLink()
PBisDir()
PBisLink()

_SYMLINK_ - New constant that can be applied to the returns of
FFAttribute() and PBAttribute()

RenameDiskFile() did not handle ~\ - fixed.

"Capture OS Output", the capture macro: updated to go to the
capture buffer window, if open. Otherwise, open a horizontal
window to display the captured information in.

Compile macro: also look in loaddir\mac for supplemental files

"File Manager" f macro - Update rename function to use
_USE_HOME_PATH_ in displaying filenames.

e "-a -s c:\name.ext" can cause TSE to crash. Fixed. Thanks to
Carlo Hogeveen for the report.

Macro runtime error <unknown opcode nn>
Caused by an expression such as:

string-constant + str(numeric-constant) + string-variable
This has been fixed. Thanks to Eckhard Hillmann for the report.

Much fiddling with the time as used in picklists and the various
FF*() and PB*() commands. Thanks to Carlo Hogeveen for his help
with this.

Fix FindFirstFile() segmentation fault. Thanks to Carlo Hogeveen
for the report.

"Full Configuration", iconfig: allow setting the keyboard macro
path on Linux. Thanks to H. Pikaar for the suggestion.

Now works on Win98 :)

[Carlo Hogeveen]

> Hopefully the last one before the official 4.50 release (famous last words :( )

That is adorably optimistic given the number of changes, for which I thank you, of course.

> _SYMLINK_ - New constant that can be applied to the returns of FFAttribute() and PBAttribute()

Summary: _SYMLINK_ is misleading and should be _LINK_ instead.

Details:

I object, your honor!
While I confess to nitpicking, that is an unfortunate name, especially given the lack of accompanying TSE documentation.
It suggests that the link attribute bit is specifically set for a symbolic link.
Not TRUE! :-)
Windows NTFS has "reparse points".
There are 44 predefined ones and an indefinite amount of programmable ones.
ALL of them set the link bit.
One of the predefined ones (MOUNT_POINT) is just for junctions, which is an older, limited way to link a directory to another directory on the same drive.
A second predefined one (SYMLINK) is used for symbolic links, which can link directories and link files, across drives too.
The other 42 + ? reparse points "link" to executable code that does other things.
DirList41, the next future version of "DirList" already lists these other types too.
https://ecarlo.nl/tse/InDevelopment.html#DirList41
TSE's _SYMLINK_ suggests that we can find a symbolic link if the link bit is set, which both in theory and in practice is not true in most cases.

In conclusion, _SYMLINK_ is misleading and should be _LINK_ instead.

Carlo

[eli...@partner.net.il]

Hi.
Thanks for the new version.
BUT:
Missed syntax highlighting and help for added new commands.
Please, correct.
Thaks
Eliyahu Trigoub

--

---
You received this message because you are subscribed to the Google Groups "SemWare TSE Pro text editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email to semware+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/semware/CA%2BgEvkZKb5HFFmBsL7R-%2BZSVkLMTxevTVxLn54_P45AqNZjpMg%40mail.gmail.com.

[Miguel Farah]

So I got off my lazy ass and installed and ran this.



The sanity check macro first failed on... test_gethomepath()

The reason? My HOME variable uses two other environment variables:

W:\>set HO
HOME=%HOMEDRIVE%%HOMEPATH%
HOMEDRIVE=C:
HOMEPATH=\Users\migue

These are not expanded upon reading %HOME%, so the comparison becomes
"%HOMEDRIVE%%HOMEPATH%\" against "C:\Users\migue\" and (unduly) fails.

Note that not few applications recommend setting the HOME variable like
that, to avoid anomalies derived from possible different values between
HOME and HOMEDRIVE+HOMEPATH. The macro should be able to detect and
expand "inner" environment variables.

I manually altered HOME to "C:\Users\migue" and the test passed.




Then... G32.EXE froze on the Capture() test. I commented it out, and got
got the "All test passed" message.


I went back to test_capture(). By the judicious use of Warn() lines,
I determined the procedure got hung at the line:

if not Capture(internal)

... after a few seconds. Every time, I had to kill the window and start
over. Weirder yet, E32.EXE did not exhibit this problem and passed all
tests without issue, several times. The result of "not Capture(internal)"
was FALSE every time, so the procedure kept chugging along and finished
normally at the endif.


I then decided to test what happened if the editor was run from CMD.EXE
instead of TCC.EXE; both G32.EXE and E32.EXE passed the test without
issue. Same thing when running both from the file explorer instead of a
command line. Finally, I ran both from TCMD itself (as opposed to a
stand-alone TCC window); E32.EXE worked and G32.EXE froze, same as
before.

So, whatever Capture(internal) is doing, Take Command and G32 are not
speaking the same thing.



Windows 10, 64-bit machine here. Nothing unusual installed on it (Take

Command most definitely DOES NOT count as "unusual").

--

---
You received this message because you are subscribed to the Google Groups "SemWare TSE Pro text editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email to semware+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/semware/CA%2BgEvkZKb5HFFmBsL7R-%2BZSVkLMTxevTVxLn54_P45AqNZjpMg%40mail.gmail.com.

--

Miguel Farah
mig...@farah.cl

[Miguel Farah]

I've kept at it. Replacing line 973 with:

        internal = "/c dir"

solved the hanging, but the capture fails and the sanity check exits with the "Capture 02: dir failed" message (this time both G32 and E32). Now I really want to see the innards of Capture(), to understand how it's calling the command-line interpreter.

--

Miguel Farah
mig...@farah.cl

[Carlo Hogeveen]

> e "-a -s c:\name.ext" can cause TSE to crash. Fixed.

Yes, the command itself typically no longer crashes.
It is unlikely, but it still could, because default TSE loads the first
found file, and if that is one of the problematic ones then it would look
like the command aborts.

And yes, what follows is maybe still only a problem for users of Windows
Sandbox, like me.
This is partly because probably Windows Sandbox creates circularly linked
files which are listable but not loadable.
The fixed "-a -s" command opens all the specified files with not-too-long
paths, and loads the first one.

The problem is, that among those opened files are also ones that cannot be
loaded by TSE.
When File Next or a search command switches to them, all sorts of errors
happen, from a gentle "access denied" warning, to an "I'm sorry, Dave."
memory abort, to an "access denied" abort that creates a new "++noname++<a
counter>" buffer.
When I tried to open one such a file directly with the File Open prompt, I
not only got the same warnings and aborts, it sometimes made TSE close, and
I even just had Windows spontaneously restart (without a blue screen of
death).


Possible solutions:

1.
Exclude files that are not accessible by TSE.
There are two precedents:
- Files with too long paths are already excluded.
- TSE refuses to open files like "C:\swapfile.sys". (luckily)
In DirList41 I test a file's Windows TSE accessibility with an extra
FileExists(); in Linux TSE this does not work.
A side benefit would be, that this would also exclude files with non-TSE
characters in their name.
Which is a good thing, because currently TSE pointlessly opens a new empty
file with an adjusted name instead.

2.
Do not follow directory links.
I mention it because in my current encounters all my aborting files are
circularly linked non-accessible files.
I do not favor this solution because there are other reasons a file is not
accessible, so it solves a symptom, not the real problem.

3.
Do not solve it.
Currently I am the only one with the problem, and I have a me-specific
work-around.

Carlo

[Carlo Hogeveen]

The following new installation thing happened on only one Linux machine, not the others.
Just to be sure I deleted my TSE directory, restored it (TSE works normally), and tried the upgrade again, with the same result.

During my TSE installation procedure:
https://ecarlo.nl/tse/TseDownloads.html#linux_tse_download
-> Installation tips -> Next time installers

After I copy the new tmp/tse directory over the tse directory and start TSE I get two warnings:
"Macro compiled with wrong sc: /home/carlo/tse/tsecfg.dat"
"Macro compiled with wrong sc: /home/carlo/tse/tseui.dat"

After opening and recompiling my win_carlo.ui file the first warning disappears.
After opening my saved settings file, compiling and executing it, and saving TSE's settings, the second warning disappears too.

I checked the internal version numbers of tsecfg.dat and tseui.dat with my TSEfilVrs tool
https://ecarlo.nl/tse/ToolsAndExtensions.html#TSEfilVrs
and their internal version numbers were consistent with the other TSE files, both before and after the upgrade.

Carlo

[S.E. Mitchell]

On Tue, Feb 20, 2024 at 4:31 PM Miguel Farah <mig...@farah.cl> wrote:
>
> I've kept at it. Replacing line 973 with:
>
> internal = "/c dir"
>
> solved the hanging, but the capture fails and the sanity check exits with the "Capture 02: dir failed" message (this time both G32 and E32). Now I really want to see the innards of Capture(), to understand how it's calling the command-line interpreter.

capture.c is attached.
Only difference between the actual code is in read_file() - in the
actual code, the output is loaded into an editor buffer.
Let me know if you find anything!
--

> To view this discussion on the web visit https://groups.google.com/d/msgid/semware/CAJWaEctTmE4pivsD5uCg8WNb0yVCyTU6RyovBKwSOQ3k%2BcUDvw%40mail.gmail.com.

[Carlo Hogeveen]

@ Miguel,
Not a solution, just some input for the problem.
> internal = "/c dir"
I would have expected:
internal = GetEnvStr("ComSpec") + " /c dir"
or bluntly, maybe with a hardcoded path added:
internal = "cmd.exe /c dir"

@Sammy,
Apparently just like me, Miguel did not (fully) set his computer to his native language.
If a non-English user did, then Sanity1 fails on line 974:
internal_f = "Directory of"

Carlo

[S.E. Mitchell]

>@Sammy,
>Apparently just like me, Miguel did not (fully) set his computer to his native language.
>If a non-English user did, then Sanity1 fails on line 974:
> internal_f = "Directory of"

Can you show me a sample non-english directory listing?

--
Sammy

> --
>
> ---
> You received this message because you are subscribed to the Google Groups "SemWare TSE Pro text editor" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to semware+u...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/semware/000201da65f3%2408847e10%24198d7a30%24%40ecarlo.nl.

[Miguel Farah]

I don't think that's my case. The output I get from Alt-F9:

is:

--

---
You received this message because you are subscribed to the Google Groups "SemWare TSE Pro text editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email to semware+u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/semware/000201da65f3%2408847e10%24198d7a30%24%40ecarlo.nl.

--

Miguel Farah
mig...@farah.cl

[Carlo Hogeveen]

Sammy,

> Can you show me a sample non-english directory listing?

Below is a listing of the same directory in 4 languages.

For your current Sanity1 macro consider the following points.
I guestimate that most of us foreigners use code page 850, while you
probably use code page 437.
The listed languages only represent limited differences, because they all
use a 1-byte character set.
The below list was created on Windows 10: Other Windows versions might have
different "dir" output.
And, as we are finding out again, users might use other command line
processors.

For other macros that process "dir" output consider the following
additional point.
In the below list I only changed the language itself, not its date and time
formats, and not its decimal and digit-grouping characters.

Carlo


English (American :-)

D:\LinkTest>dir
Volume in drive D is Data
Volume Serial Number is AC67-D8CB

Directory of D:\LinkTest

2024-02-23 10:03 <DIR> .
2024-02-23 10:03 <DIR> ..
2024-02-21 13:27 <JUNCTION> Application Data
[d:\LinkTest\ProgramData]
2024-02-23 09:57 <SYMLINKD> BaseLayer [ProgramData\BaseLayer]
2024-02-21 13:55 <DIR> BITS
2024-02-23 09:54 <SYMLINK> bitsctr.h [BITS\bitsctr.h]
2024-02-21 13:49 <DIR> ProgramData
2024-02-21 13:22 <DIR> Users
2018-10-22 20:36 756,030 winnt.h
2 File(s) 756,030 bytes
7 Dir(s) 729,255,055,360 bytes free

D:\LinkTest>



Español (Spanish)

D:\LinkTest>dir
El volumen de la unidad D es Data
El número de serie del volumen es: AC67-D8CB

Directorio de D:\LinkTest

2024-02-23 10:03 <DIR> .
2024-02-23 10:03 <DIR> ..
2024-02-21 13:27 <JUNCTION> Application Data
[d:\LinkTest\ProgramData]
2024-02-23 09:57 <SYMLINKD> BaseLayer [ProgramData\BaseLayer]
2024-02-21 13:55 <DIR> BITS
2024-02-23 09:54 <SYMLINK> bitsctr.h [BITS\bitsctr.h]
2024-02-21 13:49 <DIR> ProgramData
2024-02-21 13:22 <DIR> Users
2018-10-22 20:36 756,030 winnt.h
2 archivos 756,030 bytes
7 dirs 729,255,051,264 bytes libres

D:\LinkTest>



Français (French)

D:\LinkTest>dir
Le volume dans le lecteur D s’appelle Data
Le numéro de série du volume est AC67-D8CB

Répertoire de D:\LinkTest

2024-02-23 10:03 <DIR> .
2024-02-23 10:03 <DIR> ..
2024-02-21 13:27 <JUNCTION> Application Data
[d:\LinkTest\ProgramData]
2024-02-23 09:57 <SYMLINKD> BaseLayer [ProgramData\BaseLayer]
2024-02-21 13:55 <DIR> BITS
2024-02-23 09:54 <SYMLINK> bitsctr.h [BITS\bitsctr.h]
2024-02-21 13:49 <DIR> ProgramData
2024-02-21 13:22 <DIR> Users
2018-10-22 20:36 756,030 winnt.h
2 fichier(s) 756,030 octets
7 Rép(s) 729,255,051,264 octets libres

D:\LinkTest>



Deutsch (German)

D:\LinkTest>dir
Datenträger in Laufwerk D: ist Data
Volumeseriennummer: AC67-D8CB

Verzeichnis von D:\LinkTest

2024-02-23 10:03 <DIR> .
2024-02-23 10:03 <DIR> ..
2024-02-21 13:27 <JUNCTION> Application Data
[d:\LinkTest\ProgramData]
2024-02-23 09:57 <SYMLINKD> BaseLayer [ProgramData\BaseLayer]
2024-02-21 13:55 <DIR> BITS
2024-02-23 09:54 <SYMLINK> bitsctr.h [BITS\bitsctr.h]
2024-02-21 13:49 <DIR> ProgramData
2024-02-21 13:22 <DIR> Users
2018-10-22 20:36 756,030 winnt.h
2 Datei(en), 756,030 Bytes
7 Verzeichnis(se), 729,255,051,264 Bytes frei

D:\LinkTest>

[S.E. Mitchell]

I'm convinced :)
dir is not a good test!
--
Sammy

> --
>
> ---
> You received this message because you are subscribed to the Google Groups "SemWare TSE Pro text editor" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to semware+u...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/semware/000001da6643%24a7ac2650%24f70472f0%24%40ecarlo.nl.

[Carlo Hogeveen]

Ugh,
In the previous mail the line containing "Application Data" was not split by the "dir" command but, but by an Outlook email setting. Below it should be corrected.

Carlo



English (American):

[S.E. Mitchell]

I'm switching to "echo hello world", which works under both Windows and Linux.
--
Sammy

> --
>
> ---
> You received this message because you are subscribed to the Google Groups "SemWare TSE Pro text editor" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to semware+u...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/semware/000a01da6648%2451c22f50%24f5468df0%24%40ecarlo.nl.

[Carlo Hogeveen]

1.
I do not have a "HOME" environment variable, so the test_gethomepath() test fails.
Replacing GetEnvStr("HOME") with GenEnvStr('HOMEDRIVE') + GetEnvStr('HOMEPATH') would return the expected path on my system.

2.
The test_ffisdir() test would try to remove the directory "\Windows" instead of the temporary directory it created.
Luckily I did not run it, although it probably lacks permission to do so anyway.
On Linux it would try to remove the directory "/bin".

Carlo

[Carlo Hogeveen]

3.
The test_pbisdir() test also tries to remove the directory "\Windows" or "/bin" instead of the temporary directory it created.
How rude!

4.
The test_editfilebug() test fails if the LoadWildFromInside setting is ON,
because then EditFile('*.*', _DONT_PROMPT_) succeeds
and EditFile("", _DONT_PROMPT_) and EditFile(" ", _DONT_PROMPT_) fail.
I tried three other editors, and none of them loads all files when opening "" or " ", so in my view the test is wrong, not EditFile().
I suggest saving the setting, turning it off, doing the test, and restoring the setting.

5.
Minor error in test_FileStuff().
The other test procedures create their own test directory if necessary.
test_FileStuff() writes to the current directory and deleted foo.bar.
The test will fail if the current directory is not writable,
which might occur as a user's security precaution or just happen.

6.
Logic error:
test_GetTempPath() is executed after a lot of other tests have already used it.
It should be the first test.
I would like to see the procedure test if what GetTempPath() returns is a directory and is writable.

7.
Minor error in test_gotomark().
By just adding a GotoMark("a") at the start of the procedure, the procedure would not delete bookmark "a".

8.
Programming error in test_sc_define_bug().
The 3rd lFind() should use options "^+", otherwise it finds the same error line as the 2nd lFind().

Additionally test_sc_define_bug(), test_sc_menu_bug() and test_sc_str_bug() do not clean up their temporary .s files, and potentially not their .mac files either.

9.
Nitpicking:
In the test_filemanager_sort() procedure I would expect 4 QuotePath()s.

Carlo

[S.E. Mitchell]

I've uploaded:
https://semware.com/files/tse-pro-install/g32-2024-02-23.zip
https://semware.com/files/tse-pro-install/sanity1.s

No new features in that g32, just updated the one capture works internally.
Also an updated sanity1, with numerous bug fixes.
--

[Miguel B. Farah]

S.E. Mitchell [24/02/2024 08:54] dijo/said:

>
>I've uploaded:
>https://semware.com/files/tse-pro-install/g32-2024-02-23.zip
>https://semware.com/files/tse-pro-install/sanity1.s
>
>No new features in that g32, just updated the one capture works internally.
>Also an updated sanity1, with numerous bug fixes.

Does this one count as "RC 18 bis"? ;-)


--
Miguel Farah
mig...@farah.cl

[S.E. Mitchell]

You got me - what does that mean? e.g., "bis"
--
Sammy

> --
>
> ---
> You received this message because you are subscribed to the Google Groups "SemWare TSE Pro text editor" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to semware+u...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/semware/0100018ddb070885-3823729f-77c3-43b4-beba-dc1c97b38f44-000000%40email.amazonses.com.

[Miguel Farah]

Downloaded, tested.

The home path test works fine now (more on this later).

The sanity macro continues to hang when G32 is started from TCC, but runs properly when started from CMD or Windows File Explorer.

This time around, I *also* tested starting G32 from a TCC window *within* TCMD (hangs) and from TCMD's own File explorer (runs fine).

We're gonna have to ask Rex about this???

ALSO! Once a source code auditor, always a source code auditor. This code in sanity1.s (lines 2547..2562):

>    loop
>        p1 = Pos("%", s)
>        if p1 == 0
>            break
>        endif
>
>        p2 = Pos("%", substr(s, p1 + 1))
>        if p2 == 0
>            break
>        endif
>
>        p2 = p2 + p1
>        s2 = substr(s, p1 + 1, (p2 - p1) - 1)
>        s3 = GetEnvStr(s2)
>        s = StrReplace("%" + s2 + "%", s, s3)
>    endloop

will replace variables specified %LIKE THIS% without a problem, but will run into trouble with variables where the second percent sign is %OMITTED, a rather frequent "vice" (this used to be the norm in DOS days until a certain soMeone choSe to change the rules in obscurely incompatible ways.

Worse yet, if %SOMEONE should put %TWO variables omitting the trailing percent sign, the results will be rather bad.

To view this discussion on the web visit https://groups.google.com/d/msgid/semware/CA%2BgEvkYJ-NJWNygOHj87-OANcBe%3DWwhQT%3DAVTCanWpzNcooA_A%40mail.gmail.com.

--

Miguel Farah
mig...@farah.cl

[Miguel B. Farah]

S.E. Mitchell [24/02/2024 09:26] dijo/said:

>
>You got me - what does that mean? e.g., "bis"

"Again". Math-minded people would go with "RC18'" (with the apostrophe
as a stand-in for the prime symbol).


--
Miguel Farah
mig...@farah.cl

[S.E. Mitchell]

Note: I plan to substitute your macro for environment variables, just
haven't gotten around to it yet. Hopefully later today!
--
Sammy

> To view this discussion on the web visit https://groups.google.com/d/msgid/semware/CAJWaEcs8b%2Bmxv_tJc%2BjceMB-zO5JkWMxjCTNiAPjWyY9h04KRg%40mail.gmail.com.

[Carlo Hogeveen]

Two insignificant bugs remain:
- sanity1 deletes my "a" bookmark.
- sanity1 deletes my current directory's foo.bar file.

One significant bug remains:
If LoadWildFromInside is ON,
then sanity1 fails with error "editfile() failed on blank path",
and all the current directory's files are opened in TSE.

Carlo

[S.E. Mitchell]

re: bugs in sanity1

I have uploaded a new sanity1: v 24 Feb 2024 9:47 am
https://semware.com/files/tse-pro-install/sanity1.s
Fixes all known bugs except sc32 not working when temp path has a
space in the name.
--
Sammy

> --
>
> ---
> You received this message because you are subscribed to the Google Groups "SemWare TSE Pro text editor" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to semware+u...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/semware/001901da6726%24a533bb20%24ef9b3160%24%40ecarlo.nl.

[S.E. Mitchell]

Hmm. This is very strange.
As you know, you should only get that message with a version mismatch.
But your versions were the same.
Not really sure what to do or why it happened at this point.
--
Sammy

> --
>
> ---
> You received this message because you are subscribed to the Google Groups "SemWare TSE Pro text editor" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to semware+u...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/semware/000f01da64ab%245c4a9d70%2414dfd850%24%40ecarlo.nl.

[Carlo Hogeveen]

About this release note:

> Much fiddling with the time as used in picklists and the various
> FF*() and PB*() commands. Thanks to Carlo Hogeveen for his help
> with this.

I just tested the result of the fiddling, and came to this conclusion.

Windows TSE v4.50rc18 started rounding file times down to the closest even
second, instead of rounding them up to the closest even second. This made it
compatible with Linux TSE: They now show the same time for the same file.

Carlo

[H P]

Couldn't this be because the parameter "StartupAnsiColor" ( Set(StartupAnsiColor, 1 for example) ) was not present in the tsecfg.dat file.

Met vriendelijke groet,
With kind regards,

Muy atentamente,
Mit Freundliche Gruß,

Sinceramente,


H. Pikaar
Henri...@gmail.com

Op di 5 mrt 2024 om 12:30 schreef S.E. Mitchell <sammy.m...@gmail.com>:

To view this discussion on the web visit https://groups.google.com/d/msgid/semware/CA%2BgEvkavD1%2BJyKghsXanPcgror%3DN7t9FR885yzYy9%2BDg3cVU%2BA%40mail.gmail.com.

[Carlo Hogeveen]

Good thought.
It cannot have been the cause by itself, but maybe in combination with something else.
As apparently I was the only one affected on only one machine and it was easily solved, this does not rate any further investment of anyone's time.
Thanks,

Carlo

From: sem...@googlegroups.com <sem...@googlegroups.com> On Behalf Of H P
Sent: Tuesday, 5 March, 2024 14:37
To: sem...@googlegroups.com
Subject: Re: Linux installation surprise - WAS: [TSE] TSE 4.50 RC 18

Couldn't this be because the parameter "StartupAnsiColor" ( Set(StartupAnsiColor, 1 for example) ) was not present in the tsecfg.dat file.

Met vriendelijke groet,
With kind regards,
Muy atentamente,
Mit Freundliche Gruß,
Sinceramente,


H. Pikaar

mailto:Henri...@gmail.com


Op di 5 mrt 2024 om 12:30 schreef S.E. Mitchell <mailto:sammy.m...@gmail.com>:

Hmm. This is very strange.
As you know, you should only get that message with a version mismatch.
But your versions were the same.
Not really sure what to do or why it happened at this point.
--
Sammy

On Wed, Feb 21, 2024 at 4:50 AM Carlo Hogeveen <mailto:t...@ecarlo.nl> wrote:
>
>
> The following new installation thing happened on only one Linux machine, not the others.
> Just to be sure I deleted my TSE directory, restored it (TSE works normally), and tried the upgrade again, with the same result.
>
> During my TSE installation procedure:
> https://ecarlo.nl/tse/TseDownloads.html#linux_tse_download
> -> Installation tips -> Next time installers
>
> After I copy the new tmp/tse directory over the tse directory and start TSE I get two warnings:
> "Macro compiled with wrong sc: /home/carlo/tse/tsecfg.dat"
> "Macro compiled with wrong sc: /home/carlo/tse/tseui.dat"
>
> After opening and recompiling my win_carlo.ui file the first warning disappears.
> After opening my saved settings file, compiling and executing it, and saving TSE's settings, the second warning disappears too.
>
> I checked the internal version numbers of tsecfg.dat and tseui.dat with my TSEfilVrs tool
> https://ecarlo.nl/tse/ToolsAndExtensions.html#TSEfilVrs
> and their internal version numbers were consistent with the other TSE files, both before and after the upgrade.
>
> Carlo
>
>
>
> --
>
> ---
> You received this message because you are subscribed to the Google Groups "SemWare TSE Pro text editor" group.

> To unsubscribe from this group and stop receiving emails from it, send an email to mailto:semware%2Bunsu...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/semware/000f01da64ab%245c4a9d70%2414dfd850%24%40ecarlo.nl.

--

---
You received this message because you are subscribed to the Google Groups "SemWare TSE Pro text editor" group.

To unsubscribe from this group and stop receiving emails from it, send an email to mailto:semware%2Bunsu...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/semware/CA%2BgEvkavD1%2BJyKghsXanPcgror%3DN7t9FR885yzYy9%2BDg3cVU%2BA%40mail.gmail.com.

--

---
You received this message because you are subscribed to the Google Groups "SemWare TSE Pro text editor" group.

To unsubscribe from this group and stop receiving emails from it, send an email to mailto:semware+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/semware/CAAR3jqGKydiuoGgpU1nYdpRFmmm6ZnVb36qCd09gLR2tTiuH8Q%40mail.gmail.com?utm_medium=email&utm_source=footer.