Also, what I did to get there was a dirty hack at best. Another reason why
I didn't commit it. A little more thought should probably go into such a
migration.
But anyway, I can use it in my spellcheck plan. ~200 errors I can work
with. Mine won't drown completely.
I must admit though, that I don't quite get your reluctance migrating to
x64.
Ditto. I mean, the next release of Mac OS will drop support for x86, so
really we wouldn't be doing anything ALL THAT drastic, but I suspect that a
lot of the Eudora users have inertia on their side, so I say "let's migrate
if we must, but ONLY if we must".
As I mentioned I suspect our smartphones are already there.
Precisely. Smartphone platforms have been 64-bit (not, mind you, Intel
64-bit) for a long time now. Not that it's really comparable; what
32/64-bit refers to is word size, and the differences between, say, x64 and
ARM architectures are much bigger than that (RISC v. CISC instruction sets,
and that's just a start).
I don't know
for certain, I'm just painting a mental picture of how pervasive it's
become.
Regards
On Monday, October 29, 2018, Soren Bro sbrothy@users.sourceforge.net
wrote:
(AFK)
Yes. And that's why I didn't. Not only does it become a 64 app, but all the
ULONGs must be converted to __int64s.
I'll see if I can recreate this general pattern in 32 bit, but I doubt it.
Regards
On Sunday, October 28, 2018, Pete Maclean petemaclean@users.sourceforge.
net
wrote:
Push what exactly? If it is something that would confine us to 64-bit
builds then I would have to say no.
--
----- BEGIN TECO SIGNATURE BLOCK -----
32UD44UE97UR99UM101UA104UT106UO107UG110UL111UY114UP115UH116UI117UC$
QA:^US$QP:^US$QD:^US$QI:^US$QA:^US$QM:^UQ$QG:^UQ$QA:^UQ$QP:^UQ$
QE:^UQ$QO:^UU$QC:^UU$QH:^UU$QI:^UU$QD:^UU$QM:^UI$QY:^UI$QD:^UI$
QT:^UI$QR:^UI$QR:^UB$QL:^UB$QY:^UB$QI:^UB$QT:^UB$GI-5CGUGS-5CGB10CGQ0JT$$
----- END TECO SIGNATURE BLOCK -----
(Don't forget: ^ in TECO means just that, and $ means press the Esc key!)
My answer to this is, let's hold out for now, or perhaps create a
parallel x64 branch as a backup plan. I'm terrified of excessively
narrowing the scope of the project, so that "compatible with all PC's"
turns into "compatible with all Windows PC's" turns into "compatible with
x64-equipped Windows PC's" turns into "compatible with x64-equipped Windows
PC's on the first full moon of February when the church bells ring".
That said, actually having a product ready to download trumps that
consideration, in my view, and besides, aren't most machines x64-based by
now? Windows XP only has a 5% usage share, in any case. Then again, if we
commit this, it'll be very hard to "uncommit" it, so I think my final
verdict will be "let's hold our horses for now".
--
----- BEGIN TECO SIGNATURE BLOCK -----
32UD44UE97UR99UM101UA104UT106UO107UG110UL111UY114UP115UH116UI117UC$
QA:^US$QP:^US$QD:^US$QI:^US$QA:^US$QM:^UQ$QG:^UQ$QA:^UQ$QP:^UQ$
QE:^UQ$QO:^UU$QC:^UU$QH:^UU$QI:^UU$QD:^UU$QM:^UI$QY:^UI$QD:^UI$
QT:^UI$QR:^UI$QR:^UB$QL:^UB$QY:^UB$QI:^UB$QT:^UB$GI-5CGUGS-5CGB10CGQ0JT$$
----- END TECO SIGNATURE BLOCK -----
(Don't forget: ^ in TECO means just that, and $ means press the Esc key!)
I like what you say. I see several reasons to focus on only a 32-bit Hermes for the time being. The primary one is that, for an initial release, in order to avoid unnecessary work and the introduction of subtle problems, we should change nothing unless there is a compelling reason to do so.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
--
----- BEGIN TECO SIGNATURE BLOCK -----
32UD44UE97UR99UM101UA104UT106UO107UG110UL111UY114UP115UH116UI117UC$
QA:^US$QP:^US$QD:^US$QI:^US$QA:^US$QM:^UQ$QG:^UQ$QA:^UQ$QP:^UQ$
QE:^UQ$QO:^UU$QC:^UU$QH:^UU$QI:^UU$QD:^UU$QM:^UI$QY:^UI$QD:^UI$
QT:^UI$QR:^UI$QR:^UB$QL:^UB$QY:^UB$QI:^UB$QT:^UB$GI-5CGUGS-5CGB10CGQ0JT$$
----- END TECO SIGNATURE BLOCK -----
(Don't forget: ^ in TECO means just that, and $ means press the Esc key!)
I was just about to suggest making a separate branch. Rolling out the new
spellchecking might well hinge upon having a compilable version. As I said:
"it's a bit all over the place", and just swapping out the SPELL32.DLL
won't cut it. A little early to say for sure but, I might reintroduce some
order there and make it possible anyway. A bit early to say.
But yeah. Just because a project has 0 errors doesn't necessarily mean it
works as intended, or indeed at all.
--
----- BEGIN TECO SIGNATURE BLOCK -----
32UD44UE97UR99UM101UA104UT106UO107UG110UL111UY114UP115UH116UI117UC$
QA:^US$QP:^US$QD:^US$QI:^US$QA:^US$QM:^UQ$QG:^UQ$QA:^UQ$QP:^UQ$
QE:^UQ$QO:^UU$QC:^UU$QH:^UU$QI:^UU$QD:^UU$QM:^UI$QY:^UI$QD:^UI$
QT:^UI$QR:^UI$QR:^UB$QL:^UB$QY:^UB$QI:^UB$QT:^UB$GI-5CGUGS-5CGB10CGQ0JT$$
----- END TECO SIGNATURE BLOCK -----
(Don't forget: ^ in TECO means just that, and $ means press the Esc key!)
There was a 64-bit edition of Windows XP but I think very few people used it. (I was one who did.) Would it be reasonable to make only a 64-bit Hermes? Possibly but I am doubtful. For one thing, there is probably nobody who exclusively needs a 64-bit version. I am inclined to stick with 32-bits to start with. With a clean 32-bit version, converting to 64-bit is generally very easy. Which I can say having made several 32- to 64-bit conversions.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
A lot of errors seem to be removable shifting to x64. At one time playing
around here I was down to < 200.
In reality, I'm almost pretty sure even our smartphones are running 64 bit.
https://en.wikipedia.org/wiki/Comparison_of_Microsoft_Windows_versions
x64 creeps up, first in Windows 2000, then Windows XP. Who's running Eduora
on anything older than XP?
Pete, please tell me it's not that simple.
Regards.
Case in point (See attached file).
Regards.
On Sun, Oct 28, 2018 at 7:18 PM sbrothy@gmail.com wrote:
Should I push it commit you think? :P
:)
On Sun, Oct 28, 2018 at 7:28 PM sbrothy@gmail.com wrote:
At least it makes my spellchecker plan implementable.
Regards.
On Sun, Oct 28, 2018 at 7:39 PM sbrothy@gmail.com wrote:
Push what exactly? If it is something that would confine us to 64-bit builds then I would have to say no.
(AFK)
Yes. And that's why I didn't. Not only does it become a 64 app, but all the
ULONGs must be converted to __int64s.
I'll see if I can recreate this general pattern in 32 bit, but I doubt it.
Regards
On Sunday, October 28, 2018, Pete Maclean petemaclean@users.sourceforge.net
wrote:
--
Søren Bro Thygesen
(AFK)
I must admit though, that I don't quite get your reluctance migrating to
x64.
As I mentioned I suspect our smartphones are already there. I don't know
for certain, I'm just painting a mental picture of how pervasive it's
become.
Regards
On Monday, October 29, 2018, Soren Bro sbrothy@users.sourceforge.net
wrote:
--
Søren Bro Thygesen
(AFK)
Also, what I did to get there was a dirty hack at best. Another reason why
I didn't commit it. A little more thought should probably go into such a
migration.
But anyway, I can use it in my spellcheck plan. ~200 errors I can work
with. Mine won't drown completely.
Regards
On Monday, October 29, 2018, Soren Bro sbrothy@users.sourceforge.net
wrote:
--
Søren Bro Thygesen
On Mon, 29 Oct 2018 at 03:54, Soren Bro sbrothy@users.sourceforge.net
wrote:
--
----- BEGIN TECO SIGNATURE BLOCK -----
32UD44UE97UR99UM101UA104UT106UO107UG110UL111UY114UP115UH116UI117UC$
QA:^US$QP:^US$QD:^US$QI:^US$QA:^US$QM:^UQ$QG:^UQ$QA:^UQ$QP:^UQ$
QE:^UQ$QO:^UU$QC:^UU$QH:^UU$QI:^UU$QD:^UU$QM:^UI$QY:^UI$QD:^UI$
QT:^UI$QR:^UI$QR:^UB$QL:^UB$QY:^UB$QI:^UB$QT:^UB$GI-5CGUGS-5CGB10CGQ0JT$$
----- END TECO SIGNATURE BLOCK -----
(Don't forget: ^ in TECO means just that, and $ means press the Esc key!)
My answer to this is, let's hold out for now, or perhaps create a
parallel x64 branch as a backup plan. I'm terrified of excessively
narrowing the scope of the project, so that "compatible with all PC's"
turns into "compatible with all Windows PC's" turns into "compatible with
x64-equipped Windows PC's" turns into "compatible with x64-equipped Windows
PC's on the first full moon of February when the church bells ring".
That said, actually having a product ready to download trumps that
consideration, in my view, and besides, aren't most machines x64-based by
now? Windows XP only has a 5% usage share, in any case. Then again, if we
commit this, it'll be very hard to "uncommit" it, so I think my final
verdict will be "let's hold our horses for now".
On Sun, 28 Oct 2018 at 14:40, Soren Bro sbrothy@users.sourceforge.net
wrote:
--
----- BEGIN TECO SIGNATURE BLOCK -----
32UD44UE97UR99UM101UA104UT106UO107UG110UL111UY114UP115UH116UI117UC$
QA:^US$QP:^US$QD:^US$QI:^US$QA:^US$QM:^UQ$QG:^UQ$QA:^UQ$QP:^UQ$
QE:^UQ$QO:^UU$QC:^UU$QH:^UU$QI:^UU$QD:^UU$QM:^UI$QY:^UI$QD:^UI$
QT:^UI$QR:^UI$QR:^UB$QL:^UB$QY:^UB$QI:^UB$QT:^UB$GI-5CGUGS-5CGB10CGQ0JT$$
----- END TECO SIGNATURE BLOCK -----
(Don't forget: ^ in TECO means just that, and $ means press the Esc key!)
I like what you say. I see several reasons to focus on only a 32-bit Hermes for the time being. The primary one is that, for an initial release, in order to avoid unnecessary work and the introduction of subtle problems, we should change nothing unless there is a compelling reason to do so.
That is impressive that you have got down to 168 errors! I started off with about 10,000! Very encouraging!!
On Sun, 28 Oct 2018 at 17:51, Pete Maclean petemaclean@users.sourceforge.net wrote:
It is, isn't it? Maybe this won't be vapourware after all.
--
----- BEGIN TECO SIGNATURE BLOCK -----
32UD44UE97UR99UM101UA104UT106UO107UG110UL111UY114UP115UH116UI117UC$
QA:^US$QP:^US$QD:^US$QI:^US$QA:^US$QM:^UQ$QG:^UQ$QA:^UQ$QP:^UQ$
QE:^UQ$QO:^UU$QC:^UU$QH:^UU$QI:^UU$QD:^UU$QM:^UI$QY:^UI$QD:^UI$
QT:^UI$QR:^UI$QR:^UB$QL:^UB$QY:^UB$QI:^UB$QT:^UB$GI-5CGUGS-5CGB10CGQ0JT$$
----- END TECO SIGNATURE BLOCK -----
(Don't forget: ^ in TECO means just that, and $ means press the Esc key!)
(AFK)
I was just about to suggest making a separate branch. Rolling out the new
spellchecking might well hinge upon having a compilable version. As I said:
"it's a bit all over the place", and just swapping out the SPELL32.DLL
won't cut it. A little early to say for sure but, I might reintroduce some
order there and make it possible anyway. A bit early to say.
But yeah. Just because a project has 0 errors doesn't necessarily mean it
works as intended, or indeed at all.
Regards
On Monday, October 29, 2018, Ted Matavka nmatavka@users.sourceforge.net
wrote:
--
Søren Bro Thygesen
There was a 64-bit edition of Windows XP but I think very few people used it. (I was one who did.) Would it be reasonable to make only a 64-bit Hermes? Possibly but I am doubtful. For one thing, there is probably nobody who exclusively needs a 64-bit version. I am inclined to stick with 32-bits to start with. With a clean 32-bit version, converting to 64-bit is generally very easy. Which I can say having made several 32- to 64-bit conversions.