On Thu, 19 May 2022 13:10:27 -0500, Lynn McGuire
<lynnmcguire5@gmail.com> wrote:

>On 5/19/2022 11:36 AM, Paul S Person wrote:
>> On Wed, 18 May 2022 17:17:31 -0500, Lynn McGuire
>> <lynnmcguire5@gmail.com> wrote:
>>
>>> On 5/18/2022 11:48 AM, Paul S Person wrote:
>>>> On Tue, 17 May 2022 14:51:51 -0500, Lynn McGuire
>>>> <lynnmcguire5@gmail.com> wrote:
>>>>
>>>>> On 5/16/2022 11:00 PM, Gary R. Schmidt wrote:
>>>>>> On 17/05/2022 06:14, Lynn McGuire wrote:
>>>>>>> On 5/16/2022 12:35 AM, Gary R. Schmidt wrote:
>>>>>>>> On 15/05/2022 07:07, John W Kennedy wrote:
>>>>>>>>> On 5/13/22 11:28 PM, Lynn McGuire wrote:
>>>>>>>>>> On 5/13/2022 5:26 PM, John W Kennedy wrote:
>>>>>>>>>>> On 5/13/22 4:11 PM, Lynn McGuire wrote:
>>>>>>>>>>>> xkcd: Crêpe
>>>>>>>>>>>>     https://www.xkcd.com/2619/
>>>>>>>>>>>>
>>>>>>>>>>>> I love crêpes !  My Dad, myself, my son, and three of my nephews
>>>>>>>>>>>> stood at a fancy crêpe stand and ate about 20 to 30 of them in
>>>>>>>>>>>> Paris by the Eiffel Tower back in 2009.  We had a lot more fun
>>>>>>>>>>>> than standing in that hours long line to ride the Eiffel Tower
>>>>>>>>>>>> elevator.
>>>>>>>>>>>>
>>>>>>>>>>>> And funky non-ASCII characters are tough on us programmers.
>>>>>>>>>>>>
>>>>>>>>>>>> Explained at:
>>>>>>>>>>>>     https://www.explainxkcd.com/wiki/index.php/2619:_Cr%C3%AApe
>>>>>>>>>>>>
>>>>>>>>>>>> Lynn
>>>>>>>>>>>
>>>>>>>>>>> Well, this mess isn’t even a Unicode character. And, on my Mac, I
>>>>>>>>>>> can’t even fake it by putting a combining circumflex over a
>>>>>>>>>>> combining inverted breve over the e, though it ought to work in
>>>>>>>>>>> theory. (I tried both straight output to the command-line window
>>>>>>>>>>> and a trivial SwiftUI app.)
>>>>>>>>>>
>>>>>>>>>> So you are saying that the ê is not a Unicode character ?  Or are
>>>>>>>>>> you talking about the little hat over the e ?
>>>>>>>>>>
>>>>>>>>>> Lynn
>>>>>>>>>
>>>>>>>>> It’s not a proper Ê. It either E with an inverted breve above it (?)
>>>>>>>>> and a circumflex ^ above the inverted breve, or else it’s an E with
>>>>>>>>> a tiny A (or a Starfleet A badge or even a simple 2-D line drawing
>>>>>>>>> of Duke, the Java mascot) above it.
>>>>>>>>>
>>>>>>>> Do you often find your hair being parted as the joke whooshes by???
>>>>>>>>
>>>>>>>>      Cheers,
>>>>>>>>          Gary    B-)
>>>>>>>
>>>>>>> Sorry, I did not intend a joke here.  As a computer programmer, I find
>>>>>>> Unicode both fascinating and horrifying.
>>>>>>>
>>>>>> UNICODE??  It's a joke, Joyce.  (If only it was!!)
>>>>>>
>>>>>> And I've been fighting with UNICODE (and incompetence/lack of
>>>>>> understanding) of UNICODE for decades.
>>>>>>
>>>>>> The only thing less (or wrongly) understood than UNICODE in computing is
>>>>>> time and date.
>>>>>>
>>>>>>     Cheers,
>>>>>>         Gary    B-)
>>>>>
>>>>> We are in the middle of converting our software distribution of 1.3
>>>>> million lines of C++ and F77 from ASCII to Unicode (UTF-8). The
>>>>> conversion is not going well and we have yet to start on the Fortran code.
>>>>
>>>> Stupid question time:
>>>>
>>>> you mean that you are revising the code to read, store, and write
>>>> Unicode, right?
>>>>
>>>> not that you are rewriting the code itself in Unicode
>>>>
>>>> I /said/ it was a stupid question.
>>>
>>> No questions are stupid.
>>>
>>> We are updating our Win32 diagrammatic user interface C++ code (450
>>> KLOC) to be able read, store, and display UTF-8. We are converting from
>>> the ASCII Win32 API to the Wide Win32 API, which, we already supported
>>> in our OLE2 code and in some of our file handling code. Microsoft has
>>> been screaming at me and threatening me for years to do this.
>>> https://www.winsim.com/media/refinery.png
>>>
>>> Converting our mostly Fortran 77 (850 KLOC) calculation engine is
>>> turning into a freaking disaster. The first item is converting from the
>>> old 1995 Watcom Fortran compiler to the Intel Fortran compiler. I have
>>> already failed twice by crashing various aspects of the Ifort compiler,
>>> starting about 15 years ago. Intel fixed them for me after great
>>> periods of time but the integration with Microsoft Visual Studio is not
>>> very good, especially for mixed Fortran and C++. I am moving from my
>>> two pound hammer to my twenty pound hammer to make it happen.
>>
>> Too bad you couldn't continue with Watcom, but if you can't, you
>> can't. Perhaps different FORTRAN standards are involved.
>>
>>> I hope that we do not end up with very much code in Unicode. That could
>>> suck.
>>
>> IIRC, C pretty much defines it's standard character set, and, lo and
>> behold, it is 7-bit ASCII. So Unicode code would probably violate the
>> standard.
>>
>> Unless, of course, the standard I am referring to has been updated.
>> And not just by allowing 8-bit ASCII to accomodate the Europeans.
>>
>> I suspect FORTRAN would be, if anything, even more restrictive.
>> Although surely lower case is allowed.
>
>C / C++ has wide characters, 16 bit.

As a data type, sure.

As something you can write code in ... well, I suppose it would depend
on the compiler (which would make it not portable).

>Fortran has allowed lower case since Fortran 77. Maybe even some of the
>Fortran 66 compilers.
>
>Lynn
--
"I begin to envy Petronius."
"I have envied him long since."

On Thu, 19 May 2022 13:08:57 -0500, Lynn McGuire
<lynnmcguire5@gmail.com> wrote:

>On 5/19/2022 11:36 AM, Paul S Person wrote:
>> On Wed, 18 May 2022 17:17:31 -0500, Lynn McGuire
>> <lynnmcguire5@gmail.com> wrote:
>>
>>> On 5/18/2022 11:48 AM, Paul S Person wrote:
>>>> On Tue, 17 May 2022 14:51:51 -0500, Lynn McGuire
>>>> <lynnmcguire5@gmail.com> wrote:
>>>>
>>>>> On 5/16/2022 11:00 PM, Gary R. Schmidt wrote:
>>>>>> On 17/05/2022 06:14, Lynn McGuire wrote:
>>>>>>> On 5/16/2022 12:35 AM, Gary R. Schmidt wrote:
>>>>>>>> On 15/05/2022 07:07, John W Kennedy wrote:
>>>>>>>>> On 5/13/22 11:28 PM, Lynn McGuire wrote:
>>>>>>>>>> On 5/13/2022 5:26 PM, John W Kennedy wrote:
>>>>>>>>>>> On 5/13/22 4:11 PM, Lynn McGuire wrote:
>>>>>>>>>>>> xkcd: Crêpe
>>>>>>>>>>>>     https://www.xkcd.com/2619/
>>>>>>>>>>>>
>>>>>>>>>>>> I love crêpes !  My Dad, myself, my son, and three of my nephews
>>>>>>>>>>>> stood at a fancy crêpe stand and ate about 20 to 30 of them in
>>>>>>>>>>>> Paris by the Eiffel Tower back in 2009.  We had a lot more fun
>>>>>>>>>>>> than standing in that hours long line to ride the Eiffel Tower
>>>>>>>>>>>> elevator.
>>>>>>>>>>>>
>>>>>>>>>>>> And funky non-ASCII characters are tough on us programmers.
>>>>>>>>>>>>
>>>>>>>>>>>> Explained at:
>>>>>>>>>>>>     https://www.explainxkcd.com/wiki/index.php/2619:_Cr%C3%AApe
>>>>>>>>>>>>
>>>>>>>>>>>> Lynn
>>>>>>>>>>>
>>>>>>>>>>> Well, this mess isn’t even a Unicode character. And, on my Mac, I
>>>>>>>>>>> can’t even fake it by putting a combining circumflex over a
>>>>>>>>>>> combining inverted breve over the e, though it ought to work in
>>>>>>>>>>> theory. (I tried both straight output to the command-line window
>>>>>>>>>>> and a trivial SwiftUI app.)
>>>>>>>>>>
>>>>>>>>>> So you are saying that the ê is not a Unicode character ?  Or are
>>>>>>>>>> you talking about the little hat over the e ?
>>>>>>>>>>
>>>>>>>>>> Lynn
>>>>>>>>>
>>>>>>>>> It’s not a proper Ê. It either E with an inverted breve above it (?)
>>>>>>>>> and a circumflex ^ above the inverted breve, or else it’s an E with
>>>>>>>>> a tiny A (or a Starfleet A badge or even a simple 2-D line drawing
>>>>>>>>> of Duke, the Java mascot) above it.
>>>>>>>>>
>>>>>>>> Do you often find your hair being parted as the joke whooshes by???
>>>>>>>>
>>>>>>>>      Cheers,
>>>>>>>>          Gary    B-)
>>>>>>>
>>>>>>> Sorry, I did not intend a joke here.  As a computer programmer, I find
>>>>>>> Unicode both fascinating and horrifying.
>>>>>>>
>>>>>> UNICODE??  It's a joke, Joyce.  (If only it was!!)
>>>>>>
>>>>>> And I've been fighting with UNICODE (and incompetence/lack of
>>>>>> understanding) of UNICODE for decades.
>>>>>>
>>>>>> The only thing less (or wrongly) understood than UNICODE in computing is
>>>>>> time and date.
>>>>>>
>>>>>>     Cheers,
>>>>>>         Gary    B-)
>>>>>
>>>>> We are in the middle of converting our software distribution of 1.3
>>>>> million lines of C++ and F77 from ASCII to Unicode (UTF-8). The
>>>>> conversion is not going well and we have yet to start on the Fortran code.
>>>>
>>>> Stupid question time:
>>>>
>>>> you mean that you are revising the code to read, store, and write
>>>> Unicode, right?
>>>>
>>>> not that you are rewriting the code itself in Unicode
>>>>
>>>> I /said/ it was a stupid question.
>>>
>>> No questions are stupid.
>>>
>>> We are updating our Win32 diagrammatic user interface C++ code (450
>>> KLOC) to be able read, store, and display UTF-8. We are converting from
>>> the ASCII Win32 API to the Wide Win32 API, which, we already supported
>>> in our OLE2 code and in some of our file handling code. Microsoft has
>>> been screaming at me and threatening me for years to do this.
>>> https://www.winsim.com/media/refinery.png
>>>
>>> Converting our mostly Fortran 77 (850 KLOC) calculation engine is
>>> turning into a freaking disaster. The first item is converting from the
>>> old 1995 Watcom Fortran compiler to the Intel Fortran compiler. I have
>>> already failed twice by crashing various aspects of the Ifort compiler,
>>> starting about 15 years ago. Intel fixed them for me after great
>>> periods of time but the integration with Microsoft Visual Studio is not
>>> very good, especially for mixed Fortran and C++. I am moving from my
>>> two pound hammer to my twenty pound hammer to make it happen.
>>
>> Too bad you couldn't continue with Watcom, but if you can't, you
>> can't. Perhaps different FORTRAN standards are involved.
>>
>>> I hope that we do not end up with very much code in Unicode. That could
>>> suck.
>>
>> IIRC, C pretty much defines it's standard character set, and, lo and
>> behold, it is 7-bit ASCII. So Unicode code would probably violate the
>> standard.
>>
>> Unless, of course, the standard I am referring to has been updated.
>> And not just by allowing 8-bit ASCII to accomodate the Europeans.
>>
>> I suspect FORTRAN would be, if anything, even more restrictive.
>> Although surely lower case is allowed.
>
>We have to move to 64 bit (x64 / Win64). Watcom does not have 64 bit
>compilers and linkers.

Sad but true.

Unless Jiri's 2.0 has made the jump.
--
"I begin to envy Petronius."
"I have envied him long since."

On 5/19/22 6:28 PM, Dimensional Traveler wrote:
> On 5/19/2022 2:57 PM, John W Kennedy wrote:
>> On 5/19/22 12:16 AM, Dimensional Traveler wrote:
>>> On 5/18/2022 6:01 PM, John W Kennedy wrote:
>>>> On 5/17/22 5:57 PM, J. Clarke wrote:
>>>>> On Tue, 17 May 2022 12:59:28 -0400, John W Kennedy
>>>>> <john.w.kennedy@gmail.com> wrote:
>>>>>
>>>>>> On 5/16/22 8:38 PM, J. Clarke wrote:
>>>>>>> On Mon, 16 May 2022 15:12:54 -0500, Lynn McGuire
>>>>>>> <lynnmcguire5@gmail.com> wrote:
>>>>>>>
>>>>>>>> On 5/16/2022 8:08 AM, J. Clarke wrote:
>>>>>>>>> On Sun, 15 May 2022 14:26:48 -0400, John W Kennedy
>>>>>>>>> <john.w.kennedy@gmail.com> wrote:
>>>>>>>>>
>>>>>>>>>> On 5/14/22 9:54 PM, Dorothy J Heydt wrote:
>>>>>>>>>>> In article <t5ncnu$jer$1@dont-email.me>,
>>>>>>>>>>> Lynn McGuire  <lynnmcguire5@gmail.com> wrote:
>>>>>>>>>>>>
>>>>>>>>>>>> My Dad used APL in the 1960s trying to control a refinery
>>>>>>>>>>>> cat-cracker
>>>>>>>>>>>> using an IBM 360.
>>>>>>>>>>>
>>>>>>>>>>> Ok, I am certain that the refinery was not cracking felines.
>>>>>>>>>>> The ASPCA would have complained.  What does "cat-" stand for in
>>>>>>>>>>> this instance?
>>>>>>>>>>>
>>>>>>>>>>>> He kinda got it working and wrote his PhD Chem
>>>>>>>>>>>> Engineering thesis on it.  He got his PhD from Princeton in
>>>>>>>>>>>> three years
>>>>>>>>>>>> because of that.
>>>>>>>>>>>
>>>>>>>>>>> Good for him.
>>>>>>>>>>
>>>>>>>>>> Actually, APL was becoming very popular among accountants and
>>>>>>>>>> the like
>>>>>>>>>> in the prehistoric days before VisiCalc (awaiting a cry from
>>>>>>>>>> the Peanut
>>>>>>>>>> Gallery, “What’s Visicalc?”).
>>>>>>>>>
>>>>>>>>> Right now a debate above my pay grade at my current employer is
>>>>>>>>> whether to keep APL or move everything that is currently on APL to
>>>>>>>>> another platform.
>>>>>>>>
>>>>>>>> I was not even aware that there was a current APL compiler.
>>>>>>>
>>>>>>> While there have been APL compilers in the past, they are
>>>>>>> aberrations.
>>>>>>> APL has historically been an interpreted language and most APL code
>>>>>>> expects that environment.
>>>>>>
>>>>>> Especially since the ? operator (small circle over inverted-T),
>>>>>> “Execute” was added, which interprets a character string as an APL
>>>>>> expression in the current context.
>>>>>>> Current commercial implementations are APL+Win, APL2 (I think--it
>>>>>>> was
>>>>>>> an IBM product and has been divested--I'm not sure that it's
>>>>>>> currently
>>>>>>> being sold) and Dyalog.  There are also the open-source Gnu APL and
>>>>>>> NARS2000.  Those are the ones that come to me off the top of my
>>>>>>> head.
>>>>>>> Then there's the whole j universe.
>>>>>>
>>>>>> It was only in 2021 that IBM sold APL2 to Log-On Software, which now
>>>>>> sells it as Log-On APL2.
>>>>>
>>>>> Is Log-On actually selling it at this point?  I knew they were
>>>>> supporting it, but I seem to recall they had some difficulty that was
>>>>> preventing actual sales.
>>>>
>>>> I can’t say; I haven’t had access to a mainframe since 1997. But the
>>>> website says nothing of a problem
>>>>
>>> Which website?  If its the company's website why would you expect it
>>> tell people about a problem?
>>>
>> You can’t ignore a fundamental problem in a signature product and hope
>> to stay in business, especially when the product has been around for
>> decades.
>>
> Microsoft.  :D

I loath Microsoft, but even they have not, I believe, withdrawn a
product for months while continuing to advertise it on their homepage
without a caveat.

--
John W. Kennedy
Algernon Burbage, Lord Roderick, Father Martin, Bishop Baldwin,
King Pellinore, Captain Bailey, Merlin -- A Kingdom for a Stage!

On 5/19/22 11:05 PM, Joy Beeson wrote:
> On Mon, 16 May 2022 15:14:51 -0500, Lynn McGuire
> <lynnmcguire5@gmail.com> wrote:
>
>> As a computer programmer, I find
>> Unicode both fascinating and horrifying.
>
> No "universal code" that adamantly refuses to distinguish between
> umlaut and diaresis is worth thinking about. The glyphs of the two
> diacritics don't even resemble each other as much as "1" and "l" do.
>
I am unaware of any font that does that. I hope you are not thinking of
the distinction between ö and ő in Hungarian; they are both umlauts, but
the second one means an umlaut with an acute accent (which in Hungarian
denotes a long vowel [as in Latin, not as in elementary-school English]).

In general, if you can show the Unicode folks any distinction that is
made in existing typographic practice, they’ll pick it up, as they did
when someone pointed out that yogh (Ȝȝ) was not the same letter as ezh
(Ʒʒ), as they had initially thought.

--
John W. Kennedy
Algernon Burbage, Lord Roderick, Father Martin, Bishop Baldwin,
King Pellinore, Captain Bailey, Merlin -- A Kingdom for a Stage!

On 5/20/2022 10:56 AM, John W Kennedy wrote:
> On 5/19/22 6:28 PM, Dimensional Traveler wrote:
>> On 5/19/2022 2:57 PM, John W Kennedy wrote:
>>> On 5/19/22 12:16 AM, Dimensional Traveler wrote:
>>>> On 5/18/2022 6:01 PM, John W Kennedy wrote:
>>>>> On 5/17/22 5:57 PM, J. Clarke wrote:
>>>>>> On Tue, 17 May 2022 12:59:28 -0400, John W Kennedy
>>>>>> <john.w.kennedy@gmail.com> wrote:
>>>>>>
>>>>>>> On 5/16/22 8:38 PM, J. Clarke wrote:
>>>>>>>> On Mon, 16 May 2022 15:12:54 -0500, Lynn McGuire
>>>>>>>> <lynnmcguire5@gmail.com> wrote:
>>>>>>>>
>>>>>>>>> On 5/16/2022 8:08 AM, J. Clarke wrote:
>>>>>>>>>> On Sun, 15 May 2022 14:26:48 -0400, John W Kennedy
>>>>>>>>>> <john.w.kennedy@gmail.com> wrote:
>>>>>>>>>>
>>>>>>>>>>> On 5/14/22 9:54 PM, Dorothy J Heydt wrote:
>>>>>>>>>>>> In article <t5ncnu$jer$1@dont-email.me>,
>>>>>>>>>>>> Lynn McGuire  <lynnmcguire5@gmail.com> wrote:
>>>>>>>>>>>>>
>>>>>>>>>>>>> My Dad used APL in the 1960s trying to control a refinery
>>>>>>>>>>>>> cat-cracker
>>>>>>>>>>>>> using an IBM 360.
>>>>>>>>>>>>
>>>>>>>>>>>> Ok, I am certain that the refinery was not cracking felines.
>>>>>>>>>>>> The ASPCA would have complained.  What does "cat-" stand for in
>>>>>>>>>>>> this instance?
>>>>>>>>>>>>
>>>>>>>>>>>>> He kinda got it working and wrote his PhD Chem
>>>>>>>>>>>>> Engineering thesis on it.  He got his PhD from Princeton in
>>>>>>>>>>>>> three years
>>>>>>>>>>>>> because of that.
>>>>>>>>>>>>
>>>>>>>>>>>> Good for him.
>>>>>>>>>>>
>>>>>>>>>>> Actually, APL was becoming very popular among accountants and
>>>>>>>>>>> the like
>>>>>>>>>>> in the prehistoric days before VisiCalc (awaiting a cry from
>>>>>>>>>>> the Peanut
>>>>>>>>>>> Gallery, “What’s Visicalc?”).
>>>>>>>>>>
>>>>>>>>>> Right now a debate above my pay grade at my current employer is
>>>>>>>>>> whether to keep APL or move everything that is currently on
>>>>>>>>>> APL to
>>>>>>>>>> another platform.
>>>>>>>>>
>>>>>>>>> I was not even aware that there was a current APL compiler.
>>>>>>>>
>>>>>>>> While there have been APL compilers in the past, they are
>>>>>>>> aberrations.
>>>>>>>> APL has historically been an interpreted language and most APL code
>>>>>>>> expects that environment.
>>>>>>>
>>>>>>> Especially since the ? operator (small circle over inverted-T),
>>>>>>> “Execute” was added, which interprets a character string as an APL
>>>>>>> expression in the current context.
>>>>>>>> Current commercial implementations are APL+Win, APL2 (I
>>>>>>>> think--it was
>>>>>>>> an IBM product and has been divested--I'm not sure that it's
>>>>>>>> currently
>>>>>>>> being sold) and Dyalog.  There are also the open-source Gnu APL and
>>>>>>>> NARS2000.  Those are the ones that come to me off the top of my
>>>>>>>> head.
>>>>>>>> Then there's the whole j universe.
>>>>>>>
>>>>>>> It was only in 2021 that IBM sold APL2 to Log-On Software, which now
>>>>>>> sells it as Log-On APL2.
>>>>>>
>>>>>> Is Log-On actually selling it at this point?  I knew they were
>>>>>> supporting it, but I seem to recall they had some difficulty that was
>>>>>> preventing actual sales.
>>>>>
>>>>> I can’t say; I haven’t had access to a mainframe since 1997. But
>>>>> the website says nothing of a problem
>>>>>
>>>> Which website?  If its the company's website why would you expect it
>>>> tell people about a problem?
>>>>
>>> You can’t ignore a fundamental problem in a signature product and
>>> hope to stay in business, especially when the product has been around
>>> for decades.
>>>
>> Microsoft.  :D
>
> I loath Microsoft, but even they have not, I believe, withdrawn a
> product for months while continuing to advertise it on their homepage
> without a caveat.
>
That's because they never admit to a defective product and never recall
them. :D

--
I've done good in this world. Now I'm tired and just want to be a cranky
dirty old man.

On 5/20/2022 11:14 AM, Paul S Person wrote:
> On Thu, 19 May 2022 13:08:57 -0500, Lynn McGuire
> <lynnmcguire5@gmail.com> wrote:
>
>> On 5/19/2022 11:36 AM, Paul S Person wrote:
>>> On Wed, 18 May 2022 17:17:31 -0500, Lynn McGuire
>>> <lynnmcguire5@gmail.com> wrote:
>>>
>>>> On 5/18/2022 11:48 AM, Paul S Person wrote:
>>>>> On Tue, 17 May 2022 14:51:51 -0500, Lynn McGuire
>>>>> <lynnmcguire5@gmail.com> wrote:
>>>>>
>>>>>> On 5/16/2022 11:00 PM, Gary R. Schmidt wrote:
>>>>>>> On 17/05/2022 06:14, Lynn McGuire wrote:
>>>>>>>> On 5/16/2022 12:35 AM, Gary R. Schmidt wrote:
>>>>>>>>> On 15/05/2022 07:07, John W Kennedy wrote:
>>>>>>>>>> On 5/13/22 11:28 PM, Lynn McGuire wrote:
>>>>>>>>>>> On 5/13/2022 5:26 PM, John W Kennedy wrote:
>>>>>>>>>>>> On 5/13/22 4:11 PM, Lynn McGuire wrote:
>>>>>>>>>>>>> xkcd: Crêpe
>>>>>>>>>>>>>     https://www.xkcd.com/2619/
>>>>>>>>>>>>>
>>>>>>>>>>>>> I love crêpes !  My Dad, myself, my son, and three of my nephews
>>>>>>>>>>>>> stood at a fancy crêpe stand and ate about 20 to 30 of them in
>>>>>>>>>>>>> Paris by the Eiffel Tower back in 2009.  We had a lot more fun
>>>>>>>>>>>>> than standing in that hours long line to ride the Eiffel Tower
>>>>>>>>>>>>> elevator.
>>>>>>>>>>>>>
>>>>>>>>>>>>> And funky non-ASCII characters are tough on us programmers.
>>>>>>>>>>>>>
>>>>>>>>>>>>> Explained at:
>>>>>>>>>>>>>     https://www.explainxkcd.com/wiki/index.php/2619:_Cr%C3%AApe
>>>>>>>>>>>>>
>>>>>>>>>>>>> Lynn
>>>>>>>>>>>>
>>>>>>>>>>>> Well, this mess isn’t even a Unicode character. And, on my Mac, I
>>>>>>>>>>>> can’t even fake it by putting a combining circumflex over a
>>>>>>>>>>>> combining inverted breve over the e, though it ought to work in
>>>>>>>>>>>> theory. (I tried both straight output to the command-line window
>>>>>>>>>>>> and a trivial SwiftUI app.)
>>>>>>>>>>>
>>>>>>>>>>> So you are saying that the ê is not a Unicode character ?  Or are
>>>>>>>>>>> you talking about the little hat over the e ?
>>>>>>>>>>>
>>>>>>>>>>> Lynn
>>>>>>>>>>
>>>>>>>>>> It’s not a proper Ê. It either E with an inverted breve above it (?)
>>>>>>>>>> and a circumflex ^ above the inverted breve, or else it’s an E with
>>>>>>>>>> a tiny A (or a Starfleet A badge or even a simple 2-D line drawing
>>>>>>>>>> of Duke, the Java mascot) above it.
>>>>>>>>>>
>>>>>>>>> Do you often find your hair being parted as the joke whooshes by???
>>>>>>>>>
>>>>>>>>>      Cheers,
>>>>>>>>>          Gary    B-)
>>>>>>>>
>>>>>>>> Sorry, I did not intend a joke here.  As a computer programmer, I find
>>>>>>>> Unicode both fascinating and horrifying.
>>>>>>>>
>>>>>>> UNICODE??  It's a joke, Joyce.  (If only it was!!)
>>>>>>>
>>>>>>> And I've been fighting with UNICODE (and incompetence/lack of
>>>>>>> understanding) of UNICODE for decades.
>>>>>>>
>>>>>>> The only thing less (or wrongly) understood than UNICODE in computing is
>>>>>>> time and date.
>>>>>>>
>>>>>>>     Cheers,
>>>>>>>         Gary    B-)
>>>>>>
>>>>>> We are in the middle of converting our software distribution of 1.3
>>>>>> million lines of C++ and F77 from ASCII to Unicode (UTF-8). The
>>>>>> conversion is not going well and we have yet to start on the Fortran code.
>>>>>
>>>>> Stupid question time:
>>>>>
>>>>> you mean that you are revising the code to read, store, and write
>>>>> Unicode, right?
>>>>>
>>>>> not that you are rewriting the code itself in Unicode
>>>>>
>>>>> I /said/ it was a stupid question.
>>>>
>>>> No questions are stupid.
>>>>
>>>> We are updating our Win32 diagrammatic user interface C++ code (450
>>>> KLOC) to be able read, store, and display UTF-8. We are converting from
>>>> the ASCII Win32 API to the Wide Win32 API, which, we already supported
>>>> in our OLE2 code and in some of our file handling code. Microsoft has
>>>> been screaming at me and threatening me for years to do this.
>>>> https://www.winsim.com/media/refinery.png
>>>>
>>>> Converting our mostly Fortran 77 (850 KLOC) calculation engine is
>>>> turning into a freaking disaster. The first item is converting from the
>>>> old 1995 Watcom Fortran compiler to the Intel Fortran compiler. I have
>>>> already failed twice by crashing various aspects of the Ifort compiler,
>>>> starting about 15 years ago. Intel fixed them for me after great
>>>> periods of time but the integration with Microsoft Visual Studio is not
>>>> very good, especially for mixed Fortran and C++. I am moving from my
>>>> two pound hammer to my twenty pound hammer to make it happen.
>>>
>>> Too bad you couldn't continue with Watcom, but if you can't, you
>>> can't. Perhaps different FORTRAN standards are involved.
>>>
>>>> I hope that we do not end up with very much code in Unicode. That could
>>>> suck.
>>>
>>> IIRC, C pretty much defines it's standard character set, and, lo and
>>> behold, it is 7-bit ASCII. So Unicode code would probably violate the
>>> standard.
>>>
>>> Unless, of course, the standard I am referring to has been updated.
>>> And not just by allowing 8-bit ASCII to accomodate the Europeans.
>>>
>>> I suspect FORTRAN would be, if anything, even more restrictive.
>>> Although surely lower case is allowed.
>>
>> We have to move to 64 bit (x64 / Win64). Watcom does not have 64 bit
>> compilers and linkers.
>
> Sad but true.
>
> Unless Jiri's 2.0 has made the jump.

Jiri is a long way off from releasing 64 bit. And he has totally
fragged the debugger in his 2.0 version.

And I am having serious Watcom debugger problems on Windows 10 with
Fortran code. It will not let me set breakpoints in DLLs written in
Fortran. I have a Windows 7 PC for any serious debugging.

Lynn

On Fri, 20 May 2022 15:17:52 -0500, Lynn McGuire
<lynnmcguire5@gmail.com> wrote:

>On 5/20/2022 11:14 AM, Paul S Person wrote:
>> On Thu, 19 May 2022 13:08:57 -0500, Lynn McGuire
>> <lynnmcguire5@gmail.com> wrote:
>>
>>> On 5/19/2022 11:36 AM, Paul S Person wrote:
>>>> On Wed, 18 May 2022 17:17:31 -0500, Lynn McGuire
>>>> <lynnmcguire5@gmail.com> wrote:
>>>>
>>>>> On 5/18/2022 11:48 AM, Paul S Person wrote:
>>>>>> On Tue, 17 May 2022 14:51:51 -0500, Lynn McGuire
>>>>>> <lynnmcguire5@gmail.com> wrote:
>>>>>>
>>>>>>> On 5/16/2022 11:00 PM, Gary R. Schmidt wrote:
>>>>>>>> On 17/05/2022 06:14, Lynn McGuire wrote:
>>>>>>>>> On 5/16/2022 12:35 AM, Gary R. Schmidt wrote:
>>>>>>>>>> On 15/05/2022 07:07, John W Kennedy wrote:
>>>>>>>>>>> On 5/13/22 11:28 PM, Lynn McGuire wrote:
>>>>>>>>>>>> On 5/13/2022 5:26 PM, John W Kennedy wrote:
>>>>>>>>>>>>> On 5/13/22 4:11 PM, Lynn McGuire wrote:
>>>>>>>>>>>>>> xkcd: Crêpe
>>>>>>>>>>>>>>     https://www.xkcd.com/2619/
>>>>>>>>>>>>>>
>>>>>>>>>>>>>> I love crêpes !  My Dad, myself, my son, and three of my nephews
>>>>>>>>>>>>>> stood at a fancy crêpe stand and ate about 20 to 30 of them in
>>>>>>>>>>>>>> Paris by the Eiffel Tower back in 2009.  We had a lot more fun
>>>>>>>>>>>>>> than standing in that hours long line to ride the Eiffel Tower
>>>>>>>>>>>>>> elevator.
>>>>>>>>>>>>>>
>>>>>>>>>>>>>> And funky non-ASCII characters are tough on us programmers.
>>>>>>>>>>>>>>
>>>>>>>>>>>>>> Explained at:
>>>>>>>>>>>>>>     https://www.explainxkcd.com/wiki/index.php/2619:_Cr%C3%AApe
>>>>>>>>>>>>>>
>>>>>>>>>>>>>> Lynn
>>>>>>>>>>>>>
>>>>>>>>>>>>> Well, this mess isn’t even a Unicode character. And, on my Mac, I
>>>>>>>>>>>>> can’t even fake it by putting a combining circumflex over a
>>>>>>>>>>>>> combining inverted breve over the e, though it ought to work in
>>>>>>>>>>>>> theory. (I tried both straight output to the command-line window
>>>>>>>>>>>>> and a trivial SwiftUI app.)
>>>>>>>>>>>>
>>>>>>>>>>>> So you are saying that the ê is not a Unicode character ?  Or are
>>>>>>>>>>>> you talking about the little hat over the e ?
>>>>>>>>>>>>
>>>>>>>>>>>> Lynn
>>>>>>>>>>>
>>>>>>>>>>> It’s not a proper Ê. It either E with an inverted breve above it (?)
>>>>>>>>>>> and a circumflex ^ above the inverted breve, or else it’s an E with
>>>>>>>>>>> a tiny A (or a Starfleet A badge or even a simple 2-D line drawing
>>>>>>>>>>> of Duke, the Java mascot) above it.
>>>>>>>>>>>
>>>>>>>>>> Do you often find your hair being parted as the joke whooshes by???
>>>>>>>>>>
>>>>>>>>>>      Cheers,
>>>>>>>>>>          Gary    B-)
>>>>>>>>>
>>>>>>>>> Sorry, I did not intend a joke here.  As a computer programmer, I find
>>>>>>>>> Unicode both fascinating and horrifying.
>>>>>>>>>
>>>>>>>> UNICODE??  It's a joke, Joyce.  (If only it was!!)
>>>>>>>>
>>>>>>>> And I've been fighting with UNICODE (and incompetence/lack of
>>>>>>>> understanding) of UNICODE for decades.
>>>>>>>>
>>>>>>>> The only thing less (or wrongly) understood than UNICODE in computing is
>>>>>>>> time and date.
>>>>>>>>
>>>>>>>>     Cheers,
>>>>>>>>         Gary    B-)
>>>>>>>
>>>>>>> We are in the middle of converting our software distribution of 1.3
>>>>>>> million lines of C++ and F77 from ASCII to Unicode (UTF-8). The
>>>>>>> conversion is not going well and we have yet to start on the Fortran code.
>>>>>>
>>>>>> Stupid question time:
>>>>>>
>>>>>> you mean that you are revising the code to read, store, and write
>>>>>> Unicode, right?
>>>>>>
>>>>>> not that you are rewriting the code itself in Unicode
>>>>>>
>>>>>> I /said/ it was a stupid question.
>>>>>
>>>>> No questions are stupid.
>>>>>
>>>>> We are updating our Win32 diagrammatic user interface C++ code (450
>>>>> KLOC) to be able read, store, and display UTF-8. We are converting from
>>>>> the ASCII Win32 API to the Wide Win32 API, which, we already supported
>>>>> in our OLE2 code and in some of our file handling code. Microsoft has
>>>>> been screaming at me and threatening me for years to do this.
>>>>> https://www.winsim.com/media/refinery.png
>>>>>
>>>>> Converting our mostly Fortran 77 (850 KLOC) calculation engine is
>>>>> turning into a freaking disaster. The first item is converting from the
>>>>> old 1995 Watcom Fortran compiler to the Intel Fortran compiler. I have
>>>>> already failed twice by crashing various aspects of the Ifort compiler,
>>>>> starting about 15 years ago. Intel fixed them for me after great
>>>>> periods of time but the integration with Microsoft Visual Studio is not
>>>>> very good, especially for mixed Fortran and C++. I am moving from my
>>>>> two pound hammer to my twenty pound hammer to make it happen.
>>>>
>>>> Too bad you couldn't continue with Watcom, but if you can't, you
>>>> can't. Perhaps different FORTRAN standards are involved.
>>>>
>>>>> I hope that we do not end up with very much code in Unicode. That could
>>>>> suck.
>>>>
>>>> IIRC, C pretty much defines it's standard character set, and, lo and
>>>> behold, it is 7-bit ASCII. So Unicode code would probably violate the
>>>> standard.
>>>>
>>>> Unless, of course, the standard I am referring to has been updated.
>>>> And not just by allowing 8-bit ASCII to accomodate the Europeans.
>>>>
>>>> I suspect FORTRAN would be, if anything, even more restrictive.
>>>> Although surely lower case is allowed.
>>>
>>> We have to move to 64 bit (x64 / Win64). Watcom does not have 64 bit
>>> compilers and linkers.
>>
>> Sad but true.
>>
>> Unless Jiri's 2.0 has made the jump.
>
>Jiri is a long way off from releasing 64 bit. And he has totally
>fragged the debugger in his 2.0 version.

That's not surprising. Major upgrades take a lot of effort.

But the fragging of the debugger is disappointing.

Even the 1.9 debugger has its problems and, of course, I have no idea
what it does with Fortran or really large DLLs.

>And I am having serious Watcom debugger problems on Windows 10 with
>Fortran code. It will not let me set breakpoints in DLLs written in
>Fortran. I have a Windows 7 PC for any serious debugging.

So I recall. IIRC, the Fortran compiler shares a problem with the
DOS-specific stuff: the compiler/assembler is writtin in C, but the
people most concerned work in Fortran and assembler. This makes these
features hard to get fixed.

Oh, at some point I should have said this:

When I talk about 7-bit ASCII, I am, of course, not including text
strings, but only tokens that the compiler actually processes.