date:20230706

OK thanks. It's my mistake when editing my settings.

On Friday, July 7, 2023 at 10:43:30 AM UTC+10 tbp1...@gmail.com wrote:

> I asked about the duplicate somewhere else (can't find it just now).  
> Edward said the extra node wasn't harmful and would be a reminder for him 
> which he would remove later.
>
> On Thursday, July 6, 2023 at 8:37:03 PM UTC-4 iamap...@gmail.com wrote:
>
>> I noticed myLeoSettings.leo file has node structure:
>>> @settings
>>> UNLs
>>> >@string unl-status-kind = gnx
>>>
>>> What is the reason for using clones in leoSetting.leo file?
>>>
>>
>> I guess this is just Edward's cff workflow :D It's just not cleaned up
>>
>>
>> -- 
>> --
>> Sincerely,
>>
>> HaveF
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/8dfa6fc4-fb1f-4187-ae76-32f4ed2e1febn%40googlegroups.com.

Re: Why gnx-based unls are important

I asked about the duplicate somewhere else (can't find it just now).  
Edward said the extra node wasn't harmful and would be a reminder for him 
which he would remove later.

On Thursday, July 6, 2023 at 8:37:03 PM UTC-4 iamap...@gmail.com wrote:

> I noticed myLeoSettings.leo file has node structure:
>> @settings
>> UNLs
>> >@string unl-status-kind = gnx
>>
>> What is the reason for using clones in leoSetting.leo file?
>>
>
> I guess this is just Edward's cff workflow :D It's just not cleaned up
>
>
> -- 
> --
> Sincerely,
>
> HaveF
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/96f8c152-e90f-4ddf-96c7-60678451dfben%40googlegroups.com.

Re: Heh. I changed a gnx in the big PR by mistake



On Thursday, July 6, 2023 at 7:39:06 PM UTC-5 Edward K. Ream wrote:

PR #3430 


Sent too early by mistake. I meant to say that PR 3430 needs more testing.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/20d0b8f1-0688-4c87-b243-704c21cf71f3n%40googlegroups.com.

Re: Why gnx-based unls are important

On Thu, Jul 6, 2023 at 7:37 PM HaveF HaveF  wrote:

>
> I noticed myLeoSettings.leo file has node structure:
>> @settings
>> UNLs
>> >@string unl-status-kind = gnx
>>
>> What is the reason for using clones in leoSetting.leo file?
>>
>
> I guess this is just Edward's cff workflow :D It's just not cleaned up
>

Right. I make clones of what I'm doing and move the clones to the end of
the outline.

One of the checklist items for any release is to remove top-level clones
from various outlines.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS0xc1-QuVQ4-iV-%3DCoagUKz9%3DpYywfG0p5gP6p1oEE%2BEQ%40mail.gmail.com.

Re: Heh. I changed a gnx in the big PR by mistake

On Thu, Jul 6, 2023 at 7:02 PM Thomas Passin  wrote:

> : )
>

Congratulations on another excellent collaboration.

We have, yet again, accomplished what neither would have done alone.

PR #3430 

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS3%3Dndc_QsOysDh%2BU8%2BXo11UE8MnN5c5Sqng7hR0wejq4A%40mail.gmail.com.

Re: Why gnx-based unls are important

> I noticed myLeoSettings.leo file has node structure:
> @settings
> UNLs
> >@string unl-status-kind = gnx
>
> What is the reason for using clones in leoSetting.leo file?
>

I guess this is just Edward's cff workflow :D It's just not cleaned up


-- 
--
Sincerely,

HaveF

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CA%2BUQrQxctF%3DLxu%2BBpK0pUO9oY-t1TpEz2RFPfhGM-Wpf%3D-YzUw%40mail.gmail.com.

Re: Why gnx-based unls are important

I noticed myLeoSettings.leo file has node structure:
@settings
UNLs
>@string unl-status-kind = gnx

Any Leo user knows that "This @settings tree contains all active settings. 
Settings outside this tree have no effect."
I moved the UNLs node under @settings and all fixed. :)

Why did I make that mistake? I copied the UNLs node from leoSettings.leo 
file. The first UNLs node I saw was a clone which is outside the @settings 
tree.
What is the reason for using clones in leoSetting.leo file?

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/d0e06f9b-a293-4ffa-b1d3-0dd07c39c1cen%40googlegroups.com.

Re: Heh. I changed a gnx in the big PR by mistake

: )

On Thursday, July 6, 2023 at 1:44:36 PM UTC-4 Edward K. Ream wrote:

> On Thu, Jul 6, 2023 at 11:58 AM Thomas Passin wrote:
>
> Say you (Edward, I imagine) move a subtree that contains Terry's nodes to 
>> the Attic.  If you forget and just do a cut-and-paste, suddenly they become 
>> your nodes.  Or maybe just the head of the subtree becomes your node,  I 
>> don't even know.  
>>
>> Furthermore, I have been developing plugins and enhancing Leo internals 
>> for some time and I have *never* once thought that by creating a node I was 
>> claiming to be its creator.  Or that by moving a node by cut-and-paste I 
>> was changing its creator.  Why would I, and how would I have known about 
>> that subtlety?  After all, node ids are ordinarily invisible to the creator 
>> and user.
>>
>
> This has been a useful discussion. I never realized the side effects of 
> cut-node followed by copy-node.
>
> Let's just make the normal behavior agree with normal expectations.
>>
>
> I think #3429  will 
> do that in many cases.
>
> Edward
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/5d322ef4-54f6-4ea2-9691-38404b97n%40googlegroups.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario

On Thu, Jul 6, 2023 at 2:09 PM jkn  wrote:

> Thomas' list sounds pretty sensible to me at first blush. Nevertheless, I
> do wonder whether keeping things in a discussion/proposal phase for a while
> would be a good idea, rather than (for instance) jumping onto #3429.
>

Why wait :)

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS0EnEGN8ZH8dYwkCTRwNKj8UtO%3D-vVaYukdNGz2PHEm5Q%40mail.gmail.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario

Thomas' list sounds pretty sensible to me at first blush. Nevertheless, I 
do wonder whether keeping things in a discussion/proposal phase for a while 
would be a good idea, rather than (for instance) jumping onto #3429.

J^n


On Thursday, July 6, 2023 at 6:40:33 PM UTC+1 Edward K. Ream wrote:

> On Thu, Jul 6, 2023 at 11:09 AM Thomas Passin  wrote:
>
> I propose that whether a node gets cut or copied, that [snip].
>>
>
> Many thanks for this great idea. 
>
> See #3429 . I'll do 
> this asap. 
>
> Edward
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/ac3811e4-cfb9-423c-ad72-ae35dfd12a4cn%40googlegroups.com.

Re: Heh. I changed a gnx in the big PR by mistake

On Thu, Jul 6, 2023 at 11:58 AM Thomas Passin wrote:

Say you (Edward, I imagine) move a subtree that contains Terry's nodes to
> the Attic.  If you forget and just do a cut-and-paste, suddenly they become
> your nodes.  Or maybe just the head of the subtree becomes your node,  I
> don't even know.
>
> Furthermore, I have been developing plugins and enhancing Leo internals
> for some time and I have *never* once thought that by creating a node I was
> claiming to be its creator.  Or that by moving a node by cut-and-paste I
> was changing its creator.  Why would I, and how would I have known about
> that subtlety?  After all, node ids are ordinarily invisible to the creator
> and user.
>

This has been a useful discussion. I never realized the side effects of
cut-node followed by copy-node.

Let's just make the normal behavior agree with normal expectations.
>

I think #3429  will
do that in many cases.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS2YshKi8aQiN2eNdzROdNMmsoWZ-szn8%2B%2Bdd%2BPUhKXM%2Bw%40mail.gmail.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario

On Thu, Jul 6, 2023 at 11:09 AM Thomas Passin  wrote:

I propose that whether a node gets cut or copied, that [snip].
>

Many thanks for this great idea.

See #3429 . I'll do
this asap.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS0o989CMx9Ouk2nAz0NT_CFkCwxx--YULjXLDeYm-eAuQ%40mail.gmail.com.

Re: Heh. I changed a gnx in the big PR by mistake

On Thursday, July 6, 2023 at 12:31:55 PM UTC-4 Edward K. Ream wrote:

On Thursday, July 6, 2023 at 11:11:51 AM UTC-5 Thomas wrote:

Reminding people (including devs) to "be careful" isn't enough.

It's not about being careful. It's about not needlessly cutting/pasting
nodes.

Devs must not cut/paste nodes if they intend to issue a PR that changes
those nodes.

Devs should never falsely claim to be the creator of a node. Got it?

No I don't. Say you (Edward, I imagine) move a subtree that contains
Terry's nodes to the Attic. If you forget and just do a cut-and-paste,
suddenly they become your nodes. Or maybe just the head of the subtree
becomes your node, I don't even know.

Furthermore, I have been developing plugins and enhancing Leo internals for
some time and I have *never* once thought that by creating a node I was
claiming to be its creator. Or that by moving a node by cut-and-paste I
was changing its creator. Why would I, and how would I have known about
that subtlety? After all, node ids are ordinarily invisible to the creator
and user.

Let's just make the normal behavior agree with normal expectations.

--
You received this message because you are subscribed to the Google Groups
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/leo-editor/2084-60f0-496c-9328-4a21f992efabn%40googlegroups.com.

Re: Heh. I changed a gnx in the big PR by mistake

On Thursday, July 6, 2023 at 11:04:18 AM UTC-5 Edward K. Ream wrote:

The diffs for PR #3215 
 show that the 
gnx changed for p.get_GNX, renamed.

- #@+node:tbrown.20111010104549.26758: *4* p.get_UNL
+ #@+node:ekr.20230628174317.1: *5* p.get_full_gnx_UNL 


Correcting this mistake is fraught because both the name and the contents 
of the original node changed. 

In other words, so much has changed that I might actually *be* the author 
of all affected nodes. See the details in the last to-do item of the tweaks 
PR. 

*Summary*

I'll restore Terry's gnx only if it makes sense to do so.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/5b782196-c775-455d-8317-476e343a4182n%40googlegroups.com.

Re: Heh. I changed a gnx in the big PR by mistake

On Thursday, July 6, 2023 at 11:11:51 AM UTC-5 Thomas wrote:

Reminding people (including devs) to "be careful" isn't enough.


It's not about being careful. It's about not needlessly cutting/pasting 
nodes.

Devs must not cut/paste nodes if they intend to issue a PR that changes 
those nodes.

Devs should never falsely claim to be the creator of a node. Got it?

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/fe4dc785-b26d-44b8-929d-5b61c7e921f6n%40googlegroups.com.

Re: Heh. I changed a gnx in the big PR by mistake

See, it even happens to Edward.  Let's make things so that it won't happen 
to anyone.  Reminding people (including devs) to "be careful" isn't enough.

On Thursday, July 6, 2023 at 12:04:18 PM UTC-4 Edward K. Ream wrote:

> I'm glad we've had the conversation about cut-node/paste-node. It reminded 
> me of something that has been bothering me subconsciously.
>
> The diffs for PR #3215 
>  show that the 
> gnx changed for p.get_GNX, renamed.
>
> - #@+node:tbrown.20111010104549.26758: *4* p.get_UNL
> + #@+node:ekr.20230628174317.1: *5* p.get_full_gnx_UNL 
>
> How did I discover this, you ask? How did I know to look for this diff?
>
> The answer is simple: Weeks ago, Leo's git-diff-pr command showed that I 
> had deleted p.get_UNL and then added p.get_UNL. This kind of diff is the 
> classic symptom of cut-node/paste-node.
>
> The reason I did the cut/paste of this node is interesting, but not 
> interesting enough to tell here :-)
>
> The tweak PR, #3424 , 
> will fix this misattribution. Terry deserves all the credit(??) for this 
> method.
>
> *Summary*
>
> Leo's devs (and *only *devs) should beware when Leo's git-diff or 
> git-diff-pr commands show nodes that have been deleted and added. This kind 
> of diff is the classic symptom of cutting/pasting nodes.
>
> The tweaks PR will restore the correct attribution for Terry's original 
> g.get_UNL method.
>
> Edward
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/d9761661-7886-4da9-b0f6-23a1948f54b4n%40googlegroups.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario

On Thursday, July 6, 2023 at 11:36:41 AM UTC-4 Edward K. Ream wrote:

On Thu, Jul 6, 2023 at 9:28 AM jkn wrote:

> So, "once you have learned about gnx's, Cut- and paste- nodes is lost to
you"?

Not at all. Only Leo's *devs *need to take care.

To repeat: a Leo dev is someone who changes Leo's codebase.

Anyone who uses a UNL to point to a node can get bit by this, not only Leo
codebase devs.

Consider this scenario: A user clones node N at position P1 and moves the
clone to position P2. This node and its clone have gnx1. Then he deletes
P1. The node at (what used to be) P2 now has gnx1.

OTOH, if the user cuts node N at Position P1 and then pastes it into
position P2, then the node at P2 will have a different gnx, say gnx2. gnx2
!= gnx1.

Any normal person would think that these operations will end up with the
same result. I don't see any reason why there should be non-intuitive
behavior in this scenario. Of course it only matters if those gnxs get
used for something. The most likely way they would get used by a normal
user - or even a non-Leo-codebase script developer - would be for new-style
unls.

I have this problem - that I have to remember to paste as a clone - with my
zettel-kasten system and its adaptation to family history/geneological
data. That's because I use gnxs to crosslink everything. At least I knew
this (use clone rather than plain paste) would be a problem almost from the
beginning but it's taken discipline on my part to avoid wrecking the data
structures. I almost created my own ID system but gnxs were easier to use
and persist, so I went with them.

I propose that whether a node gets cut or copied, that:

1. If it gets pasted into the same outline, then it should keep the
original gnx *unless* that gnx already exists in the outline. Then the
pasted copy should get a new gnx;
2. If a node is copied and pasted into a different outline, then again it
should keep its gnx unless a node with that gnx already exists;
3. Paste-as-clone should always keep the original gnx.

This would make the behavior as intuitive as possible.

--
You received this message because you are subscribed to the Google Groups
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/leo-editor/6a5e9ca8-8069-403d-8a87-a0a8ae373840n%40googlegroups.com.

Heh. I changed a gnx in the big PR by mistake

I'm glad we've had the conversation about cut-node/paste-node. It reminded 
me of something that has been bothering me subconsciously.

The diffs for PR #3215 
 show that the 
gnx changed for p.get_GNX, renamed.

- #@+node:tbrown.20111010104549.26758: *4* p.get_UNL
+ #@+node:ekr.20230628174317.1: *5* p.get_full_gnx_UNL 

How did I discover this, you ask? How did I know to look for this diff?

The answer is simple: Weeks ago, Leo's git-diff-pr command showed that I 
had deleted p.get_UNL and then added p.get_UNL. This kind of diff is the 
classic symptom of cut-node/paste-node.

The reason I did the cut/paste of this node is interesting, but not 
interesting enough to tell here :-)

The tweak PR, #3424 , 
will fix this misattribution. Terry deserves all the credit(??) for this 
method.

*Summary*

Leo's devs (and *only *devs) should beware when Leo's git-diff or 
git-diff-pr commands show nodes that have been deleted and added. This kind 
of diff is the classic symptom of cutting/pasting nodes.

The tweaks PR will restore the correct attribution for Terry's original 
g.get_UNL method.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/a85acf39-2e61-4714-95c0-4a4ad064638bn%40googlegroups.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario

On Thu, Jul 6, 2023 at 9:28 AM jkn wrote:

> So, "once you have learned about gnx's, Cut- and paste- nodes is lost to
you"?

Not at all. Only Leo's *devs *need to take care.

To repeat: a Leo dev is someone who changes Leo's codebase.

And I don't mind if Leo devs change gnxs in their *own *plugins.

> I too think things are moving too fast.

Yes. I get that :-) But everyone can calm down:

- Nothing is likely to break.
- I'll fix anything that does break.
- There are no new rules (unless you are a Leo dev).

I don't mind answering these questions. It's part of the process of
changing Leo.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS31CRVpyS64yxt3ARHT8hGHaYF0-3QY52%3DRHWYevtqA-A%40mail.gmail.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario

On Thu, Jul 6, 2023 at 8:49 AM Jacob Peck  wrote:

> Your recent prescriptivism has been a bit frustrating as someone who has
> been using Leo for over a decade.
>

You can do anything you like provided you don't issue a PR.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS1s450a8YH1zyqeS9_inbk3nZo%3DSO18%3DyrPEcBoe9tvhA%40mail.gmail.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario

So, "once you have learned about gnx's, Cut- and paste- nodes is lost to 
you"?

Sorry, like Jake says, I am trying to be constructive here. But I too think 
things are moving too fast.

Jon N


On Thursday, July 6, 2023 at 2:49:17 PM UTC+1 Edward K. Ream wrote:

> On Thu, Jul 6, 2023 at 8:02 AM Thomas Passin  wrote:
>
>> The gnx change  may be expected, but as a user if I cut a node and paste 
>> it somewhere else, in my mind it's the *same* node and an "unbreakable" UNL 
>> should take me to that same node afterwards.  In my mind there should be no 
>> difference between moving a node using move commands (like CTRL-D) and 
>> moving it by cut-paste.  
>>
>> I would say that a user who has not thought about, or maybe not even 
>> learned about, gnx's would not expect a unl to break when he did a 
>> cut-paste on a node.
>>
>
> The average Leonista may cut and paste nodes without harm. Leo's devs *must 
> never *do so.
>
> Clear?
>
> Edward
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/aeaa7753-767e-4a50-aa2b-de6b385a00f5n%40googlegroups.com.

Re: Why gnx-based unls are important

My computer is a Dell Windows-10 laptop.  The status bar display seems to 
be working as intended.

On Thursday, July 6, 2023 at 9:57:16 AM UTC-4 lewis wrote:

> On laptop build 947ea935f5 works correctly.
> However on PC 947ea935f5 does not show gnx-style format.
>
> Here is log from PC updating to build 947ea935f5 
> $ git pull
> remote: Enumerating objects: 85, done.
> remote: Counting objects: 100% (85/85), done.
> remote: Compressing objects: 100% (43/43), done.
> remote: Total 85 (delta 58), reused 69 (delta 42), pack-reused 0
> Unpacking objects: 100% (85/85), 47.31 KiB | 13.00 KiB/s, done.
> From https://github.com/leo-editor/leo-editor
>d845b1024..947ea935f  devel -> origin/devel
>  * [new branch]  ekr-3423-node-history -> 
> origin/ekr-3423-node-history
>2534f31b7..9c1261cd8  ekr-dmypy -> origin/ekr-dmypy
>f587fb5f7..f3ae2e5cb  ekr-tweak-unls-> origin/ekr-tweak-unls
>d845b1024..947ea935f  felix-server5 -> origin/felix-server5
> Updating d845b1024..947ea935f
> Fast-forward
>  leo/plugins/freewin.py | 19 ++-
>  1 file changed, 10 insertions(+), 9 deletions(-)
>
> I repeated git pull:
> $ git pull
> Already up to date.
>
> On Thursday, July 6, 2023 at 11:42:02 PM UTC+10 Edward K. Ream wrote:
>
>> On Thu, Jul 6, 2023 at 8:26 AM lewis  wrote:
>>
>>> I updated to build 947ea935f5 and the status bar now shows gnx format.
>>> Previously build c4434257ff was installed.
>>>
>>
>> Good. Keep on testing. Detours don't matter :-)
>>
>> Edward
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/b6d69a36-d232-4ed1-85d4-099a638dea27n%40googlegroups.com.

Re: Why gnx-based unls are important

947ea935f5 works for me. MacOS


-- 
--
Sincerely,

HaveF

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CA%2BUQrQytxw%3D6fGnTRxtGKnFxh5qMqJHdzBgShNxuBPAPiT-B5g%40mail.gmail.com.

Re: Why gnx-based unls are important

On laptop build 947ea935f5 works correctly.
However on PC 947ea935f5 does not show gnx-style format.

Here is log from PC updating to build 947ea935f5 
$ git pull
remote: Enumerating objects: 85, done.
remote: Counting objects: 100% (85/85), done.
remote: Compressing objects: 100% (43/43), done.
remote: Total 85 (delta 58), reused 69 (delta 42), pack-reused 0
Unpacking objects: 100% (85/85), 47.31 KiB | 13.00 KiB/s, done.
>From https://github.com/leo-editor/leo-editor
   d845b1024..947ea935f  devel -> origin/devel
 * [new branch]  ekr-3423-node-history -> 
origin/ekr-3423-node-history
   2534f31b7..9c1261cd8  ekr-dmypy -> origin/ekr-dmypy
   f587fb5f7..f3ae2e5cb  ekr-tweak-unls-> origin/ekr-tweak-unls
   d845b1024..947ea935f  felix-server5 -> origin/felix-server5
Updating d845b1024..947ea935f
Fast-forward
 leo/plugins/freewin.py | 19 ++-
 1 file changed, 10 insertions(+), 9 deletions(-)

I repeated git pull:
$ git pull
Already up to date.

On Thursday, July 6, 2023 at 11:42:02 PM UTC+10 Edward K. Ream wrote:

> On Thu, Jul 6, 2023 at 8:26 AM lewis  wrote:
>
>> I updated to build 947ea935f5 and the status bar now shows gnx format.
>> Previously build c4434257ff was installed.
>>
>
> Good. Keep on testing. Detours don't matter :-)
>
> Edward
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/542f07a1-48de-4357-934b-5a386e677bb0n%40googlegroups.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario

2023-07-06 Thread Jacob Peck

I move nodes like this all the time.  Additionally, I rarely ever use
clones, despite being one of Leo's 'headline features'.

Your recent prescriptivism has been a bit frustrating as someone who has
been using Leo for over a decade.  The joy of Leo is that the user can do
things in so many ways.  I sincerely think that imposing these 'hidden'*
restrictions on user activity is an objectively bad move.

To be clear, I love Leo, and I genuinely appreciate all the design and work
that has gone into it.  I just think perhaps we're moving fast and breaking
things just to do so, lately.  It's frustrating and difficult to keep up
with.  And I say this as someone who has developed a handful of plugins in
the repo, was involved in the bzr->git migration, and contributed a fair
amount of bugfixes over the years.

(*I say 'hidden' because how is a user unfamiliar with Leo's internals
supposed to grok that cut-paste to move a node, an *extremely common*
method of moving data in other applications, will have unintended, internal
consequences?  This is just *bad* UX, full stop.  The user shouldn't have
to think about the internal data structure of the thing they're working on.)

Just my $0.02.  Please take this as constructive -- I believe we all want
Leo to be the best it can be.

Jake

On Thu, Jul 6, 2023 at 8:45 AM Edward K. Ream  wrote:

> On Wed, Jul 5, 2023 at 11:21 PM Thomas Passin  wrote:
>
> Trouble is, when you cut a node and paste it, its gnx changes.  To prevent
>> that you would have to remember to paste the node as a clone rather than do
>> a simple paste.  This is asking for trouble. (And it's a weakness of my
>> zettel-kasten system, which is gnx-based).
>>
>
> Then don't move nodes by cut and paste!
>
> I *never* move nodes this way. Surely you can find an alternative that
> works for you.
>
> Edward
>
> --
> You received this message because you are subscribed to the Google Groups
> "leo-editor" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to leo-editor+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/leo-editor/CAMF8tS0R7M602uaJ57SjOTnrHZYuq6w5V36vy%2BXtD6POgcdukw%40mail.gmail.com
> 
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAJ1i%2BSb_jkmkAtUTAeaLj%2BDwjvxOrKFVcWbDEWuEzqX05xgXqQ%40mail.gmail.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario

On Thu, Jul 6, 2023 at 8:02 AM Thomas Passin  wrote:

> The gnx change  may be expected, but as a user if I cut a node and paste
> it somewhere else, in my mind it's the *same* node and an "unbreakable" UNL
> should take me to that same node afterwards.  In my mind there should be no
> difference between moving a node using move commands (like CTRL-D) and
> moving it by cut-paste.
>
> I would say that a user who has not thought about, or maybe not even
> learned about, gnx's would not expect a unl to break when he did a
> cut-paste on a node.
>

The average Leonista may cut and paste nodes without harm. Leo's devs *must
never *do so.

Clear?

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS1My0d3FXZOq5vBCbcG%3DrXpP6H-CATrNE_1tF1DwfJDPg%40mail.gmail.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario

On Thu, Jul 6, 2023 at 8:28 AM jkn  wrote:

 By  'devs', do you mean, erm ... 'users' ?
>

I meant anyone who commits code to Leo's code base.

The harmful effects of copy-node/paste-node would also apply to *other*
devs, that is, anyone who commits code to any *other* code base.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS1uMG-4Ja2dOFrCNKNTSimLNEj-szuNidT0zDL3Tnh_kA%40mail.gmail.com.

Re: Why gnx-based unls are important

On Thu, Jul 6, 2023 at 8:26 AM lewis  wrote:

> I updated to build 947ea935f5 and the status bar now shows gnx format.
> Previously build c4434257ff was installed.
>

Good. Keep on testing. Detours don't matter :-)

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS0VAdaazP4RXubx%2BpOu7v0a-sPYPJtWwEF35Zv%3D1M17cw%40mail.gmail.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario

On Thu, Jul 6, 2023 at 8:23 AM Thomas Passin  wrote:

> Hmm, we have a real disagreement here.  I think that cut-paste within an
> outline should be equivalent to a move.  And anyway, some fundamental
> property like node identity should not have to depend on whether someone
> happens to remember a subtle point - if they even had learned about it -
> when their mind is thinking of something else.
>

I'm glad the issue has come to the fore.

This is *not* a matter of opinion. No option can mitigate the evil effects
of copy/paste.

I don't care if you change gnxs in VR3, but you *must not *gnxs anywhere
else.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS34cPTA9xFx%2BB_1sAXTdTAc%2BTxfQLb8197Q%2B5F05aNSuA%40mail.gmail.com.

Re: Why gnx-based unls are important

On Thu, Jul 6, 2023 at 8:19 AM Thomas Passin  wrote:

> I have *@string unl-status-kind = legacy *set in myLeoSettings.leo, and
> the status bar does show the path-based UNL.  I added a setting *@string
> unl-status-kind = gnx* to my workbook, restarted Leo, and for that
> outline, I see gnx-style addresses in the status bar.
>

Thanks for testing!

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS3ZEeJZZ3c7n_Kak63gdCO%3DzzCzvpNyCWcC2XpQ-8B3Jw%40mail.gmail.com.

Re: Why gnx-based unls are important

On Thu, Jul 6, 2023 at 8:06 AM lewis  wrote:

> No, the local file does not have @string unl-status-kind = legacy
> Only myLeoSettings.leo has @string unl-status-kind = gnx
>
> Strange. Both my desktop and laptop have this behaviour. I'll keep
> investigating.
>

What branch are you in?

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS2JwfZHdSk5N4U5rf7sq4x%2B%3DFbjMciGsrZUZX5YRce19A%40mail.gmail.com.

Re: Why gnx-based unls are important

On Thu, Jul 6, 2023 at 8:05 AM Thomas Passin wrote:

What do you mean by "This"?
>
>
> What @iamap wrote:
> "Btw, I do believe gnx-based unls is important, but the current display
> format lack some readability.
> Could it possible to add some readable path string to the end of the unls?
> just for human?
>

Enough. Let's use what we have. There is no need for complex feature
proposals that solve non-existent problems.

The workaround: put a comment above the url/unl. An example from my
leoPy.leo:

=== PR #3215: mypy links
unl:gnx://leoPy.leo#ekr.20230318040137.1

Clicking on the unl:gnx: link takes me to the node whose headline is "===
PR #3215: mypy links".

Yesterday I discussed using such links as a lightweight alternative to
clones.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS1Ypm2d%3D30jpdYr2SKF%3DmFvjsJmbxZeEJEuPknx60BpDg%40mail.gmail.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario


Hi Edward
  By  'devs', do you mean, erm ... 'users' ?

I agree with Thomas, I think there is a real disagreement here.

Has it always been true that "Leo devs should *never* move nodes by 
copy/paste", or is this a recent ... limitation, or ... evolution of your 
thinking, or what?

Thanks, Jon N

On Thursday, July 6, 2023 at 2:21:11 PM UTC+1 Edward K. Ream wrote:

> On Thursday, July 6, 2023 at 8:11:56 AM UTC-5 jkn wrote:
>
> What are Outline | Cut-Node and Outline | Paste-Node for then?
>
>
> An earlier reply 
>  (in 
> this thread) explains that "copy outline" creates an independent copy.
> Useful for cherry-picking nodes from other branches.
>
> A later reply 
>  (in 
> this thread) explains why *Leo devs should never move nodes by copy/paste*
> .
>
> Edward
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/8af538a5-6e77-4953-8cde-93c7824045d9n%40googlegroups.com.

Re: Why gnx-based unls are important

I updated to build 947ea935f5 and the status bar now shows gnx format.
Previously build c4434257ff was installed.

On Thursday, July 6, 2023 at 11:19:06 PM UTC+10 tbp1...@gmail.com wrote:

> I have *@string unl-status-kind = legacy *set in myLeoSettings.leo, and 
> the status bar does show the path-based UNL.  I added a setting *@string 
> unl-status-kind = gnx* to my workbook, restarted Leo, and for that 
> outline, I see gnx-style addresses in the status bar.
>
> Leo 6.7.4-devel, devel branch, build 947ea935f5
> 2023-07-06 08:01:34 -0500
> Python 3.11.4, PyQt version 6.4.3
> Windows 10 AMD64 (build 10.0.19045) SP0
>
> On Thursday, July 6, 2023 at 9:06:47 AM UTC-4 lewis wrote:
>
>> No, the local file does not have @string unl-status-kind = legacy
>> Only myLeoSettings.leo has @string unl-status-kind = gnx 
>>
>> Strange. Both my desktop and laptop have this behaviour. I'll keep 
>> investigating.
>>
>> On Thursday, July 6, 2023 at 10:53:33 PM UTC+10 Edward K. Ream wrote:
>>
>>> On Thu, Jul 6, 2023 at 4:40 AM lewis  wrote:
>>>
 In myLeoSettings.leo I have:
 @string unl-status-kind = gnx

 However the status bar still shows legacy format, it does not show 
 links in the gnx format:
 unl:gnx//G:/My Drive/editor/workbook.leo#lewis.20170803232020.1

 Do I need to add another setting?

>>>
>>> Works for me in devel.
>>>
>>> Does your local .leo file define @string unl-status-kind = legacy ?
>>>
>>> Edward
>>>
>>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/531c927b-2538-4c3e-a326-411ff2b1c982n%40googlegroups.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario

Hmm, we have a real disagreement here.  I think that cut-paste within an 
outline should be equivalent to a move.  And anyway, some fundamental 
property like node identity should not have to depend on whether someone 
happens to remember a subtle point - if they even had learned about it - 
when their mind is thinking of something else.

On Thursday, July 6, 2023 at 9:14:25 AM UTC-4 Edward K. Ream wrote:

> On Thursday, July 6, 2023 at 7:45:21 AM UTC-5 Edward K. Ream wrote:
>
> Then don't move nodes by cut and paste! I *never* move nodes this way.
>
>
> *Devs should never move a node by cut/paste*!!
>
> Doing so destroys the node's gnx. In effect, the dev is claiming that *they 
> *created the node, not the original creator!
>
> Changing the identify of the node will also thwart Leo's upcoming 
> git-history feature. See #3423 
> .
>
> Btw, new unls make clear who created the node. Or who pasted it last, 
> hehe. 
>
> Edward
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/6abf5a9f-3df8-46c6-88ce-34e428513ec4n%40googlegroups.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario

On Thursday, July 6, 2023 at 8:11:56 AM UTC-5 jkn wrote:

What are Outline | Cut-Node and Outline | Paste-Node for then?


An earlier reply 
 (in 
this thread) explains that "copy outline" creates an independent copy.
Useful for cherry-picking nodes from other branches.

A later reply 
 (in 
this thread) explains why *Leo devs should never move nodes by copy/paste*.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/8fbd3783-bb8b-4ec6-8079-4d661754df62n%40googlegroups.com.

Re: Why gnx-based unls are important

I have *@string unl-status-kind = legacy *set in myLeoSettings.leo, and the 
status bar does show the path-based UNL.  I added a setting *@string 
unl-status-kind = gnx* to my workbook, restarted Leo, and for that outline, 
I see gnx-style addresses in the status bar.

Leo 6.7.4-devel, devel branch, build 947ea935f5
2023-07-06 08:01:34 -0500
Python 3.11.4, PyQt version 6.4.3
Windows 10 AMD64 (build 10.0.19045) SP0

On Thursday, July 6, 2023 at 9:06:47 AM UTC-4 lewis wrote:

> No, the local file does not have @string unl-status-kind = legacy
> Only myLeoSettings.leo has @string unl-status-kind = gnx 
>
> Strange. Both my desktop and laptop have this behaviour. I'll keep 
> investigating.
>
> On Thursday, July 6, 2023 at 10:53:33 PM UTC+10 Edward K. Ream wrote:
>
>> On Thu, Jul 6, 2023 at 4:40 AM lewis  wrote:
>>
>>> In myLeoSettings.leo I have:
>>> @string unl-status-kind = gnx
>>>
>>> However the status bar still shows legacy format, it does not show links 
>>> in the gnx format:
>>> unl:gnx//G:/My Drive/editor/workbook.leo#lewis.20170803232020.1
>>>
>>> Do I need to add another setting?
>>>
>>
>> Works for me in devel.
>>
>> Does your local .leo file define @string unl-status-kind = legacy ?
>>
>> Edward
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/ee1bd2d6-0c34-4634-9ee5-1d6de2b7db35n%40googlegroups.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario

On Thursday, July 6, 2023 at 7:45:21 AM UTC-5 Edward K. Ream wrote:

Then don't move nodes by cut and paste! I *never* move nodes this way.


*Devs should never move a node by cut/paste*!!

Doing so destroys the node's gnx. In effect, the dev is claiming that *they 
*created the node, not the original creator!

Changing the identify of the node will also thwart Leo's upcoming 
git-history feature. See #3423 
.

Btw, new unls make clear who created the node. Or who pasted it last, hehe. 

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/bba14ee9-387e-470b-a429-df706fee26a5n%40googlegroups.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario



On Thursday, July 6, 2023 at 1:45:21 PM UTC+1 Edward K. Ream wrote:

On Wed, Jul 5, 2023 at 11:21 PM Thomas Passin  wrote:

Trouble is, when you cut a node and paste it, its gnx changes.  To prevent 
that you would have to remember to paste the node as a clone rather than do 
a simple paste.  This is asking for trouble. (And it's a weakness of my 
zettel-kasten system, which is gnx-based).


Then don't move nodes by cut and paste!


really?!

What are Outline | Cut-Node and Outline | Paste-Node for then?


I *never* move nodes this way. Surely you can find an alternative that 
works for you.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/fa5b96e2-14cf-428a-96cd-25b44cc2fde4n%40googlegroups.com.

Re: Why gnx-based unls are important

 No, the local file does not have @string unl-status-kind = legacy
Only myLeoSettings.leo has @string unl-status-kind = gnx 

Strange. Both my desktop and laptop have this behaviour. I'll keep 
investigating.

On Thursday, July 6, 2023 at 10:53:33 PM UTC+10 Edward K. Ream wrote:

> On Thu, Jul 6, 2023 at 4:40 AM lewis  wrote:
>
>> In myLeoSettings.leo I have:
>> @string unl-status-kind = gnx
>>
>> However the status bar still shows legacy format, it does not show links 
>> in the gnx format:
>> unl:gnx//G:/My Drive/editor/workbook.leo#lewis.20170803232020.1
>>
>> Do I need to add another setting?
>>
>
> Works for me in devel.
>
> Does your local .leo file define @string unl-status-kind = legacy ?
>
> Edward
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/06d7a812-e732-4d57-85b3-202e11ca24bdn%40googlegroups.com.

Re: Why gnx-based unls are important


On Thursday, July 6, 2023 at 8:56:59 AM UTC-4 Edward K. Ream wrote:

On Wed, Jul 5, 2023 at 10:53 PM Thomas Passin  wrote:

This is why I am sticking with the legacy format for the status bar.


What do you mean by "This"?


What @iamap wrote:
"Btw, I do believe gnx-based unls is important, but the current display 
format lack some readability. 
Could it possible to add some readable path string to the end of the unls? 
just for human?

unl:gnx:///Users/mac/.leo/workbook.leo#mac.20230706111222.1

could be better:

unl:gnx:///Users/mac/.leo/workbook.leo#mac.20230706111222.1
*?2023-07-06/leo-note"* 

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/3acb9bd0-e260-49c8-8c46-9791b08a20c0n%40googlegroups.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario

The gnx change  may be expected, but as a user if I cut a node and paste it 
somewhere else, in my mind it's the *same* node and an "unbreakable" UNL 
should take me to that same node afterwards.  In my mind there should be no 
difference between moving a node using move commands (like CTRL-D) and 
moving it by cut-paste.  

I would say that a user who has not thought about, or maybe not even 
learned about, gnx's would not expect a unl to break when he did a 
cut-paste on a node.  

On Thursday, July 6, 2023 at 8:49:11 AM UTC-4 Edward K. Ream wrote:

> On Thu, Jul 6, 2023 at 7:42 AM Thomas Passin  wrote:
>
>> Here's what I tried in my workbook:
>>
>> 1. Create a node for testing.
>> 2. Copy its gnx to clipboard using a tiny script. *
>> 3. Paste that gnx into the node's body for future reference.
>> 4. Cut the node using the Outline/cut-node item.
>> 5. Paste the node somewhere else in the same outline.
>> 6. Copy the node's gnx to the clipboard.
>> 7. Paste the gnx into the node's body.
>> 8. Observe the 2nd gnx is not the same as the first.
>>
>
> As expected.
>
> Edward
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/8172c852-5cfd-465d-ab59-e331d5493baen%40googlegroups.com.

Re: Why gnx-based unls are important

On Thu, Jul 6, 2023 at 7:54 AM lewis  wrote:

> Sorry, I realise my previous post was unclear.
> The status bar still shows legacy format
> unl://G:/My Drive/editor/workbook.leo#Home
>

I repeat. Everything works for me.  Examples (with settings as shown in
myLeoSettings.leo):

@string unl-status-kind = gnx
unl:gnx://ekr.leo#ekr.20200316054710.1

vs:

@string unl-status-kind = legacy
unl://ekr.leo#--- recent settings & scripts--> Recursive import script

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS2Ue%2B-R0Qr-8jR91A1PHGiJR1EopMaJeDFWmBjXwqrRPA%40mail.gmail.com.

Re: Why gnx-based unls are important

On Wed, Jul 5, 2023 at 10:53 PM Thomas Passin  wrote:

This is why I am sticking with the legacy format for the status bar.
>

What do you mean by "This"?

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS1tkrdKBMLBr-Sqx3S6hDRC8MtAmCksrndCxL4YUk79Qg%40mail.gmail.com.

Re: Why gnx-based unls are important

Sorry, I realise my previous post was unclear.
The status bar still shows legacy format
unl://G:/My Drive/editor/workbook.leo#Home

On Thursday, July 6, 2023 at 7:40:29 PM UTC+10 lewis wrote:

In myLeoSettings.leo I have:
@string unl-status-kind = gnx

However the status bar still shows legacy format, it does not show links in 
the gnx format:

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/f7f34da1-b6b8-4605-bcba-6f105094f5dcn%40googlegroups.com.

Re: Why gnx-based unls are important

On Thu, Jul 6, 2023 at 4:40 AM lewis  wrote:

> In myLeoSettings.leo I have:
> @string unl-status-kind = gnx
>
> However the status bar still shows legacy format, it does not show links
> in the gnx format:
> unl:gnx//G:/My Drive/editor/workbook.leo#lewis.20170803232020.1
>
> Do I need to add another setting?
>

Works for me in devel.

Does your local .leo file define @string unl-status-kind = legacy ?

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS0DkVfTJ4WeRUT29yrK_yXH3Bhf819JteYFEVWfJhf2PA%40mail.gmail.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario

On Thu, Jul 6, 2023 at 7:42 AM Thomas Passin  wrote:

> Here's what I tried in my workbook:
>
> 1. Create a node for testing.
> 2. Copy its gnx to clipboard using a tiny script. *
> 3. Paste that gnx into the node's body for future reference.
> 4. Cut the node using the Outline/cut-node item.
> 5. Paste the node somewhere else in the same outline.
> 6. Copy the node's gnx to the clipboard.
> 7. Paste the gnx into the node's body.
> 8. Observe the 2nd gnx is not the same as the first.
>

As expected.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS1jPummcH_nSqLBCx%3DMr%3DGJ-8WCNXE364r6SEgdQKc6Gg%40mail.gmail.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario

On Wed, Jul 5, 2023 at 11:21 PM Thomas Passin  wrote:

Trouble is, when you cut a node and paste it, its gnx changes.  To prevent
> that you would have to remember to paste the node as a clone rather than do
> a simple paste.  This is asking for trouble. (And it's a weakness of my
> zettel-kasten system, which is gnx-based).
>

Then don't move nodes by cut and paste!

I *never* move nodes this way. Surely you can find an alternative that
works for you.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS0R7M602uaJ57SjOTnrHZYuq6w5V36vy%2BXtD6POgcdukw%40mail.gmail.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario

Here's what I tried in my workbook:

1. Create a node for testing.
2. Copy its gnx to clipboard using a tiny script. *
3. Paste that gnx into the node's body for future reference.
4. Cut the node using the Outline/cut-node item.
5. Paste the node somewhere else in the same outline.
6. Copy the node's gnx to the clipboard.
7. Paste the gnx into the node's body.
8. Observe the 2nd gnx is not the same as the first.

* If you are having the statusbar show the gnx form of the unl, you can of 
course copy that from the clipboard instead of using a little script.

A screenshot of the test node is attached.

On Thursday, July 6, 2023 at 8:31:07 AM UTC-4 Edward K. Ream wrote:

> On Thu, Jul 6, 2023 at 6:40 AM Thomas Passin  wrote:
>
> Maybe a paste of a node should always maintain the gnx unless it would 
>> create a clone.
>>
>
> Hmm. Under what circumstance would a paste-node *not* create a clone?
>
> I sometimes use paste-node to create an independent copy of a node (and 
> its descendants). It's a way of cherry-picking a node from one branch to 
> another:
>
> - Copy and paste a node *N* in branch A to a copy of N, say *N-Copy*.
> - Close Leo, change to branch B.  *Never leave Leo open when changing 
> branches*.
> - Open Leo. Now we're in branch B, but *N-Copy* contains the 
> cherry-picked node from branch A.
> - Copy the body text of nodes from *N-Copy* to N. This changes N's text 
> while preserving N's gnx.
>
> Edward
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/3ae02cba-ac44-4384-a466-5c97d74754c6n%40googlegroups.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario

On Thu, Jul 6, 2023 at 6:40 AM Thomas Passin  wrote:

Maybe a paste of a node should always maintain the gnx unless it would
> create a clone.
>

Hmm. Under what circumstance would a paste-node *not* create a clone?

I sometimes use paste-node to create an independent copy of a node (and its
descendants). It's a way of cherry-picking a node from one branch to
another:

- Copy and paste a node *N* in branch A to a copy of N, say *N-Copy*.
- Close Leo, change to branch B.  *Never leave Leo open when changing
branches*.
- Open Leo. Now we're in branch B, but *N-Copy* contains the cherry-picked
node from branch A.
- Copy the body text of nodes from *N-Copy* to N. This changes N's text
while preserving N's gnx.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS2wHc3FwdjwPa-U12jn9yu4xtqv2s9iNp7kNbBgLtCYkQ%40mail.gmail.com.

Re: Proposal: abandon support for .json files and jupyter notebooks.

On Thu, Jul 6, 2023 at 7:20 AM HaveF HaveF  wrote:

> I'm using another function, *g.execGitCommand*, as the basis for #3423
>> .
>>
>
> Thanks, Edward, I'll check it later
>

You're welcome.  Happy coding :-)

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS3efP3ozsu%2BegDXu7qEovbAdsPBrf%2BvBbR_0C4gh6%3DmLQ%40mail.gmail.com.

Re: Why gnx-based unls are important

On Wed, Jul 5, 2023 at 10:27 PM HaveF HaveF  wrote:

Everything will work provided you define *@data unl-path-prefixes* in your
> *myLeoSettings.leo* file *on each platform*. These data should define an
> absolute path (platform-specific) for various .leo files.
>
>

> Does that means I can set  @data unl-path-prefixes  in my current leo file
> @settings node to specify a specific x.leo file path?
>

Yes. I have been meaning to point out this fact.

This morning I'll be writing a new Engineering Notebook post discussing the
fine points of PR #3215 ,
aka the *big *PR.

This post will be pre-writing for comments in PR #3424
, aka the *tweaks *PR.

The big PR already has too many comments. Comments in the tweaks PR will
form the basis for our discussions.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS06_UntuJEJdT68MCkqEsodAkV2HuGFOXEb4i22-CJhng%40mail.gmail.com.

Re: Proposal: abandon support for .json files and jupyter notebooks.

>
> I'm using another function, *g.execGitCommand*, as the basis for #3423
> .
>

Thanks, Edward, I'll check it later

-- 
--
Sincerely,

HaveF

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CA%2BUQrQz_s6MeUTeO%2B70Hm8y1aZWOTuR1DMrQXrvjn4ssDmScpg%40mail.gmail.com.

Re: Proposal: abandon support for .json files and jupyter notebooks.

>
> I might be more receptive to this idea if we can come up with a way for a
> user to change in and out of the stay-on-top property and the opacity.  It
> won't work as one-size-fits-all, in my experience.
>

A checkbox and a spinbox along with the current "Rendered < -- > Plain"
button for every separate FW window?

-- 
--
Sincerely,

HaveF

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CA%2BUQrQxJ_K5tMx-_ah%2BKy7e-8gjBxbGpOC%3DFbdrugBAA90L0Fg%40mail.gmail.com.

Re: Proposal: abandon support for .json files and jupyter notebooks.

On Wed, Jul 5, 2023 at 8:32 PM HaveF HaveF  wrote:

I found the g.getGitIssues function yesterday, it seems very useful.
> I'm trying to add support to handle private repo issues.
>

I'm using another function, *g.execGitCommand*, as the basis for #3423
.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/CAMF8tS2ZndnaK7MHk46rYEbR%2BYouJ-gEARqrvy9bqo%3DqQLskLw%40mail.gmail.com.

Re: Proposal: abandon support for .json files and jupyter notebooks.

I might be more receptive to this idea if we can come up with a way for a 
user to change in and out of the stay-on-top property and the opacity.  It 
won't work as one-size-fits-all, in my experience.

On Thursday, July 6, 2023 at 1:20:46 AM UTC-4 iamap...@gmail.com wrote:

> Btw, I believe making some FW windows always stay on the top and 
> set opacity of windows are also useful.
>
>
> https://stackoverflow.com/questions/19097323/setwindowflagsqtwindowstaysontophint-hides-qt-window
> https://doc.qt.io/qt-6/qwidget.html#windowOpacity-prop
>
> -- 
> --
> Sincerely,
>
> HaveF
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/8b005763-4b2a-4a52-bb20-46b42c3d2676n%40googlegroups.com.

Re: "Unbreakable UNLs" Can Break Under A Common Scenario

Maybe a paste of a node should always maintain the gnx unless it would 
create a clone.

On Thursday, July 6, 2023 at 12:21:12 AM UTC-4 Thomas Passin wrote:

> I think it's not uncommon to cut a node and paste it somewhere else in the 
> outline.  For example, if I want to move a node from near the top to near 
> the bottom of a long outline, it would be impractical to move it down node 
> by node or to drag it.  I simply cut and paste it.
>
> Trouble is, when you cut a node and paste it, its gnx changes.  To prevent 
> that you would have to remember to paste the node as a clone rather than do 
> a simple paste.  This is asking for trouble. (And it's a weakness of my 
> zettel-kasten system, which is gnx-based).
>
> It's true that legacy path-based UNLs have the same problem.  But at least 
> that's fairly obvious, whereas the new gnx-based UNLs are supposed to be 
> unbreakable.
>
> I don't have a ready solution to offer at this point, but I think it's a 
> real scenario.
>

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to leo-editor+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/0eb73578-0f04-45f2-bd03-c8ec2f729d02n%40googlegroups.com.

Re: Why gnx-based unls are important