LDCad 1.7 Alpha 2a (win+linux)

[Roland Melkert]

Hello all,

I've solved all the reported Alpha 2 issues and made a new version.

Besides bug fixes I've also added 2 new minor features to the part bin:
- Hiding of zero count items during count down mode is now optional.
- Negative searching using the '!' character, eg: "1x1 !round" to get all 1x1 parts except the round ones.




So be sure to make backups of your work before changing them with this Alpha version.

The new version can be downloaded here:
http://www.melkert.net/LDCad/nextVer

[HWQ]

Is it possible to add a new feature in the parts bin: show parts that are used but not in the PBG list

[David Manley]

- Negative searching using the '!' character, eg: "1x1 !round" to get all 1x1 parts except the round ones.

<snip/>

Marvellous, thank you.

David

[Milan Vančura]

Besides bug fixes I've also added 2 new minor features to the part bin:
- Hiding of zero count items during count down mode is now optional.
- Negative searching using the '!' character, eg: "1x1 !round" to get all 1x1 parts except the round ones.

Thank you, Roland. Both the part countdown and 'not' operator in searches are great help for me. Part countdown is superb for alternative models making and similar tasks, the 'not' operator helps with searching, a lot.

My quick check: some parts miss their understud(s) in shadow library. For example the cheese slope 1x1 (54200) although I have the latest shadow.cf - am I doing anything wrong, please? Or are they really missing and it would be helpful if I send my findings?

[N. W. Perry]

My quick check: some parts miss their understud(s) in shadow library. For example the cheese slope 1x1 (54200) although I have the latest shadow.cf - am I doing anything wrong, please? Or are they really missing and it would be helpful if I send my findings?

I've noticed the same, actually. I'll try and pay more attention, if it's not just me.

[Roland Melkert]

I've noticed the same, actually. I'll try and pay more attention, if it's not just me.

54200 should have bottom connectivity. It gets this from s\54200s01.dat

Are you using the 22-03 library, given the mentioned subpart is new to the 22-03 library.

This because I do move things around based on new subparts etc from time to time.

[Roland Melkert]

Is it possible to add a new feature in the parts bin: show parts that are used but not in the PBG list

I was actually thinking about that awhile back and already put it on my Beta 1 'maybe' list.

[N. W. Perry]

54200 should have bottom connectivity. It gets this from s\54200s01.dat

Are you using the 22-03 library, given the mentioned subpart is new to the 22-03 library.

This because I do move things around based on new subparts etc from time to time.

That one comes up OK for me. The most recent one I noticed was 30150; it has a generic snap on the bottom but not its underside studs.

[Roland Melkert]

That one comes up OK for me. The most recent one I noticed was 30150; it has a generic snap on the bottom but not its underside studs.

That one misses:

Code:

0 !LDCAD SNAP_CYL [gender=F] [caps=one] [secs=R 6 4] [pos=0 8 0] [grid=C 3 C 4 20 20]

Not sure how that got left out

[N. W. Perry]

I got a startup crash after (manually) syncing my config files. Turns out I forgot to update the LDraw paths to the Alpha 2a directory. My fault and easy to fix, but perhaps there could be a dummy check and ask for user input when an LDraw path loaded from the config file is not found?

Speaking of config files, the new .hkc file has an entry under partBinWinHotKeys that is just an = sign. Perhaps this was meant for the new showZeroCount option?

(Also spotted a minor typo in the same section: flipCountDiown -> flipCountDown.)

[Roland Melkert]

I got a startup crash after (manually) syncing my config files. Turns out I forgot to update the LDraw paths to the Alpha 2a directory. My fault and easy to fix, but perhaps there could be a dummy check and ask for user input when an LDraw path loaded from the config file is not found?

Speaking of config files, the new .hkc file has an entry under partBinWinHotKeys that is just an = sign. Perhaps this was meant for the new showZeroCount option?

(Also spotted a minor typo in the same section: flipCountDiown -> flipCountDown.)

Thanks,

Non existing library paths should not cause a crash.

I'm not able to replicate the problem by adding a '2' to all my own paths.

Do you still have the log file(s)?

The '=' entry seems to be from the new hide zero count option, I forgot to define its ini property name

[N. W. Perry]

Thanks,

Non existing library paths should not cause a crash.

I'm not able to replicate the problem by adding a '2' to all my own paths.

Do you still have the log file(s)?

Yes, I think it should be these two:

  Archive.zip (Size: 100.2 KB / Downloads: 1)

[Milan Vančura]

54200 should have bottom connectivity. It gets this from s\54200s01.dat

Are you using the 22-03 library, given the mentioned subpart is new to the 22-03 library.

This because I do move things around based on new subparts etc from time to time.

You are right, I needed to make sure both shadowfiles and ldraw library are updated. Now, 54200 works.
But the other parts I had seen are still with no understuds. According to my notes, it's a problem of these (at least): 78256, 44861, 73825, 28802, 39262 (tricky: the hole is there but understud not), 86996, 3023pb0, 98302, 3183b, 3183c, 26599, 61280p0* (all of them), 32064c (the middle minibar shadow is missing there as well), 73230, 20310,44865, 2454a, 79756, 27266, 57910, 62712, 92013, 66961 (snap on handles missing as well there), 67810, 39893, 54661, 30389c, 18455

Of course, I'm sorry the list is not complete, probably. This is just what I met last weeks.

[Roland Melkert]

Of course, I'm sorry the list is not complete, probably. This is just what I met last weeks.

Seems the first few are from libraries I haven't processed yet.

So they just don't have info yet.

I'm way (way way way) behind shadow wise, hoping to catch up to at least 2022 for the beta version.

[Milan Vančura]

Seems the first few are from libraries I haven't processed yet.

So they just don't have info yet.

I'm way (way way way) behind shadow wise, hoping to catch up to at least 2022 for the beta version.

OK, so a simple question: how can I help? Do you have a TODO list? You know, we are all short of time but if we were two on this work it's much better than being alone - and others may join our effort as well. I'm not a professional LDCADer (I love this term ) but I already made some shadow files in past (if you remember) and I may help, for example, starting with those easier to do saving your time to make the more complicated ones. Or to work on LDCAD Beta

[Roland Melkert]

OK, so a simple question: how can I help? Do you have a TODO list? You know, we are all short of time but if we were two on this work it's much better than being alone - and others may join our effort as well. I'm not a professional LDCADer (I love this term ) but I already made some shadow files in past (if you remember) and I may help, for example, starting with those easier to do saving your time to make the more complicated ones. Or to work on LDCAD Beta

Alpha 2 has shadow info up to the 20-01 library, so 20-02 is next on my list.

I usually open a search bin, search for "20-02" (while the tags option is enabled) also using F11 and a clear transparent color.

Next I go trough all items one by one to ether add information or change/improve existing info.

I also have some tools in the DEV version of LDCad to weave out 'moved to' parts and sync descriptions etc.

[Roland Melkert]

Yes, I think it should be these two:

both logs seem to be loading parts from:

Z:\Applications\ldraw\complete.zip
and
Z:\Applications\ldraw\ldrawunf.zip

The log stops after it just stated loading 6 parts so there might be a multi threading bug at play.

Those bugs are very hard to replicate though, has it happened before/again?

Good thing is both logs end at the same place (while loading the same 6 parts).

Do you know what part bin group was open at the time?

[Orion Pobursky]

I created an official LDraw.org account on Github with the intent of creating repositiories for some of LDraw.org's custom software. I haven't done that (yet) but I could create a shadow library repostory so that there's a place for people to contribute.

[Roland Melkert]

I created an official LDraw.org account on Github with the intent of creating repositiories for some of LDraw.org's custom software. I haven't done that (yet) but I could create a shadow library repostory so that there's a place for people to contribute.

Seems like a great plan.

You could use the (unpacked) shadow of Alpha 2a for the initial commit.

I could also look into how to get just the shadow seed info node from my svn repo into git if you think the history of changes might be interesting.

[Orion Pobursky]

Awesome. I try to get something set up soon.

[N. W. Perry]

both logs seem to be loading parts from:

Z:\Applications\ldraw\complete.zip
and
Z:\Applications\ldraw\ldrawunf.zip

The log stops after it just stated loading 6 parts so there might be a multi threading bug at play.

Those bugs are very hard to replicate though, has it happened before/again?

Good thing is both logs end at the same place (while loading the same 6 parts).

Do you know what part bin group was open at the time?

I think it was one of my downloaded inventories in the sets group (set 10190).

I did get the problem to go away, simply by renaming Alpha 2 to Alpha 2a in the config file (the rest of the pathname is identical). This would have affected the shadow lib paths associated with both complete.zip and ldrawunf.zip, viz.:

Code:

<LDrawPaths>
unOff->Z:\Applications\ldraw\User\->none
unOff->Z:\Applications\ldraw\ldrawunf.zip->C:\users\nwperry\Desktop\LDCad-1-7-Alpha-2a-Win\shadow\unoffLib\
off->Z:\Applications\ldraw\complete.zip->Z:\Users\nwperry\Desktop\LDCad-1-7-Alpha-2a-Win\shadow\offLib\offLibShadow.csl
donors->Z:\Users\nwperry\Desktop\LDCad-1-7-Alpha-2a-Win\donors\->none

The Alpha 2 directory had been moved off the desktop by this time, so until I fixed it it would have been searching an invalid location.

[Milan Vančura]

Seems like a great plan.

You could use the (unpacked) shadow of Alpha 2a for the initial commit.

I could also look into how to get just the shadow seed info node from my svn repo into git if you think the history of changes might be interesting.

I used the latest shadow file released as the initial commit and started my personal git repo. Of course, I'm ready to replay my history to the common git repo when it's ready.

So far, I checked 155 parts of the latest ldraw update set (as Roland suggested) - see the status file 'status_2203'. I started at the bottom of (numerical) file name list and go up. Filenames 'ticked' are OK according to my knowledge: either they were OK or I fixed them. Filenames marked with a cross sign are not OK and I do not have enough knowledge to fix them. Sometimes, there are notes, too. I.e. if file is fixed by another (included) one.

EDIT: Additional set: 2203_m2 - apply on top of 2203_m1. In sum, 230 parts checked.

EDIT: I passed through all the 2203 update, see below, so zip files attached to this message, are obsoleted.

[Orion Pobursky]

I used the latest shadow file released as the initial commit and started my personal git repo. Of course, I'm ready to replay my history to the common git repo when it's ready.

So far, I checked 155 parts of the latest ldraw update set (as Roland suggested) - see the status file 'status_2203'. I started at the bottom of (numerical) file name list and go up. Filenames 'ticked' are OK according to my knowledge: either they were OK or I fixed them. Filenames marked with a cross sign are not OK and I do not have enough knowledge to fix them. Sometimes, there are notes, too. I.e. if file is fixed by another (included) one.

The repo exists. I created it a week ago. I was waiting for some SteerCo discussion before officially. announcing.

[Milan Vančura]

The repo exists. I created it a week ago. I was waiting for some SteerCo discussion before officially. announcing.

Great. So, please, just let me note if/when we can start using it (not to be faster than SteerCo )
No problem, I can use my local copy until then.

[Milan Vančura]

Hi Roland. I think I'm done with 2203, please let me know your opinion. I zipped all changes together, for your convenience. In the zip archive, check status_2203 file for details.

In general: I fixed all basic snap info: studs, antistuds, pinholes, pins, duplo studs and duplo antistuds. Plus I added snap infos for dinosaur parts so one can build these monsters with body, legs, neck, tails of several kinds... Another part-specific addition is for train wheels: I was surprised one has the axle hole of diameter 4 and another of diameter 3.

I did NOT add any other part-specific snap infos, i.e. for electric part: like special snap bounding boxes for battery lids etc. I'm not sure it's worth the work on it so I let it on more interested users

Also, I did not work on parts/s/* and parts/p/* subfiles on their own. Of course, I always tried to go as deep in included files as possible when editing real parts so some improvements of parts/s/* are there but only what I met during my work.

When I was unsure or saw another problem, I noted that in the status file.

[Roland Melkert]

Hi Roland. I think I'm done with 2203, please let me know your opinion. I zipped all changes together, for your convenience. In the zip archive, check status_2203 file for details.

In general: I fixed all basic snap info: studs, antistuds, pinholes, pins, duplo studs and duplo antistuds. Plus I added snap infos for dinosaur parts so one can build these monsters with body, legs, neck, tails of several kinds... Another part-specific addition is for train wheels: I was surprised one has the axle hole of diameter 4 and another of diameter 3.

I did NOT add any other part-specific snap infos, i.e. for electric part: like special snap bounding boxes for battery lids etc. I'm not sure it's worth the work on it so I let it on more interested users

Also, I did not work on parts/s/* and parts/p/* subfiles on their own. Of course, I always tried to go as deep in included files as possible when editing real parts so some improvements of parts/s/* are there but only what I met during my work.

When I was unsure or saw another problem, I noted that in the status file.

Thanks, I will merge/review them.

however you should add some more header information to them, e.g.

Code:

0 LDCad info for "same description as original"
0 Author: Milan Vančura [ldrawPtUserNameIfYouGotOne]
0 !LICENSE Free for non-commercial use.

The license line is something we need to work on though.

I'm thinking (thanks to Orion) about changing them all to CC NC 4.0.

But I guess contributors would be free to use CC BY 4.0 or something similar (but not <=2.0) if they want.

Given it doesn't conflict with distributing the shadow as a whole like I'm used to.

So maybe it would be easier to force a single license kind for the whole library (CC NC 4.0?)

Also if we really go the GIT route there should also be some more edit tracking (like the pt) if people start changing existing files etc.

[Orion Pobursky]

Feel free to send a pull request to the ldraw-org repo. We'll change the license via a mass edit once we've landed on what we're going to use.

[Milan Vančura]

Feel free to send a pull request to the ldraw-org repo. We'll change the license via a mass edit once we've landed on what we're going to use.

Thanks, Orion. Please, how can I access the repo?

[Milan Vančura]

Thanks, I will merge/review them.

however you should add some more header information to them, e.g.

I added the needed header lines to all files. New files are authored by me and licensed by CC-BY-4.0. Modified files keep the original authorship and "0 !HISTORY" line is added in usual manner.
Check new zip file - OK this way?

[Milan Vančura]

Thanks, I will merge/review them.

The question is how to proceed further. Now, I'm on 22-02 which is a set you probably edit as well. Is it OK for you I'm going, again, from the bottom of name-sorted list up? And where should I stop? Or do you have a list of already fixed parts by you?

EDIT: First (last ) 100 parts of update 22-02 checked, see attached zip file.

[Travis Cobbs]

Thanks, Orion. Please, how can I access the repo?

I assume it is here:

https://github.com/ldraw-org/ldraw-connection-library

[Roland Melkert]

I assume it is here:

https://github.com/ldraw-org/ldraw-connection-library

I always found the term 'pull request' confusing as it, imho, is more like a push request given you want to 'push' changes into the repo.

But I guess it's from the repo's pov, like upload/download which they very often use wrong in movies and tv shows

[Milan Vančura]

I assume it is here:

https://github.com/ldraw-org/ldraw-connection-library

Thank you, Travis.

The current state of my work is requested to be pulled. And, as Roland did not say anything against that, I continue with 22-02 parts - I'm not sure when I get another free-time-window so I try to use the current one to help with LDCAD as much as I can

[Orion Pobursky]

Thank you, Travis.

The current state of my work is requested to be pulled. And, as Roland did not say anything against that, I continue with 22-02 parts - I'm not sure when I get another free-time-window so I try to use the current one to help with LDCAD as much as I can

After some consultation with Roland, we've decided to leave this under his umbrella. You posted your work above so my repo will be shutting down.

[Roland Melkert]

And, as Roland did not say anything against that, I continue with 22-02 parts - I'm not sure when I get another free-time-window so I try to use the current one to help with LDCAD as much as I can

I haven't had the time to do a real review but it looks ok at first glance

Only thing I did notice is you added p/axl5hol8.dat, that one I don't use on purpose because it's always part of a bigger thing.

Adding it will also cause the need to revisit all parts using it

After some consultation with Roland, we've decided to leave this under his umbrella. You posted your work above so my repo will be shutting down.

I will start a repo on my @rolandMelkert github account once I decided to go CC 4.0 BY or NC license wise.

Also sure about an efficient way to review pull requests though.

[Roland Melkert]

once I decided to go CC 4.0 BY or NC license wise.

I just discovered there might be a middle ground, namely: CC-BY-SA-4.0

It adds a 'must be similar licensed' part to derivatives.

Anyone know of a reason not to use that variant?

[Roland Melkert]

As the git thing is a bit off topic, I started a new thread here:
https://forums.ldraw.org/thread-26553.html

[Milan Vančura]

I haven't had the time to do a real review but it looks ok at first glance

Only thing I did notice is you added p/axl5hol8.dat, that one I don't use on purpose because it's always part of a bigger thing.

Thanks, this is exactly the experience I need to learn. I found your comment in stud2a.dat so I fixed that before I pushed the result to you but I overlooked p/axl5hol8.dat

BTW, I have more work done for 22-02 but it got deleted together with all Orion's repo. I think it's not a problem because you probably set up your own repo before reviewing all my files but in case you want to get something more to read in a bed, tell me and I prepare new zip files for you

About more tools in LDCAD: there are already several features helpful in shadow editing, for example the fact LDCAD checks modifications on the filesystem automatically. Or a warning if I try to save a file with unsaved dependencies. This is both very useful.

What I'd like to get is a partBin with a list of already prepared snap objects often used: a stud, antistud, technic hole full sized, technic hole half-deep, bar, clip, basic hinges. It would be more comfortable than to look for them in corresponding parts and their subfiles, again and again.

Do you have another idea(s)?

[Milan Vančura]

After some consultation with Roland, we've decided to leave this under his umbrella. You posted your work above so my repo will be shutting down.

As I wrote in another reply, your repo contained more than it is in zip files but it is not a problem, I prepare commits in the new repository again or publish new zip files, whatever will be needed first.

But you surprised me. I'd lied if I said that not Your repo is/was the fastest and most surprisingly disappeared upstream I've ever seen!

[Orion Pobursky]

Sorry about that. I assumed you had issued a pull request from your own repo. My fault for that assumption.

[Milan Vančura]

Sorry about that. I assumed you had issued a pull request from your own repo. My fault for that assumption.

No problem, this is a power of git - I've lost nothing but the pull request message - everything stays in my repo, too. And even if github removed my public repo I still have a local copy. Only I've lost is "upstream" to create pull requests for. This can be restarted for Roland's repo in future, and I hope it will be easy.

As I wrote, it was just surprising

BTW, files in your repo were all with LF line endings - probably by mistake so it's good to switch to CRLF in the new repo anyway as we all agreed on it, it seems.

[Roland Melkert]

What I'd like to get is a partBin with a list of already prepared snap objects often used: a stud, antistud, technic hole full sized, technic hole half-deep, bar, clip, basic hinges. It would be more comfortable than to look for them in corresponding parts and their subfiles, again and again.

This is a great idea, I'll look into it.

Do you have another idea(s)?

Things like an option to force 'repo headers' and auto history lines on shadow content.

[Orion Pobursky]

I found what I think is a bug:

When using "Clean Up" to embed unofficial parts, LDCad inserts these part but doesn't preserve the decimal precision. Instead all embedded parts go to 8 decimal places. This only happens for numbers with decimals, whole numbers without a decimal are unaffected.

[Roland Melkert]

I found what I think is a bug:

When using "Clean Up" to embed unofficial parts, LDCad inserts these part but doesn't preserve the decimal precision. Instead all embedded parts go to 8 decimal places. This only happens for numbers with decimals, whole numbers without a decimal are unaffected.

It is the result of the default digit count for new/modified part lines.

That said it might be cleaner to force the new line to use the exact string of its donor using the same mechanism currently in place for existing unmodified lines.

[HWQ]

I want to add a feature to show used parts but not in the parts bin

[N. W. Perry]

I want to add a feature to show used parts but not in the parts bin

My first thought was to try making a .pbg file with a part having 0 quantity, as a workaround. I tested this, and while it makes the part show up in the part bin, it doesn't show any quantity (zero or otherwise) for the part.

So, if you add the part to a model, it does not show as quantity -1 when countdown is enabled. (Nor does it hide the part when Hide Zero Countdown is enabled.) So that workaround doesn't work.

Could the countdown behavior be changed to (optionally) show all parts that are in the model, not just those listed in the .pbg file?

And then, secondarily, should countdown behavior properly reflect zero-quantity parts that are in the .pbg file?

[Roland Melkert]

My first thought was to try making a .pbg file with a part having 0 quantity, as a workaround. I tested this, and while it makes the part show up in the part bin, it doesn't show any quantity (zero or otherwise) for the part.

So, if you add the part to a model, it does not show as quantity -1 when countdown is enabled. (Nor does it hide the part when Hide Zero Countdown is enabled.) So that workaround doesn't work.

Could the countdown behavior be changed to (optionally) show all parts that are in the model, not just those listed in the .pbg file?

And then, secondarily, should countdown behavior properly reflect zero-quantity parts that are in the .pbg file?

I will look into this, but somehow it seems 'wrong' to display stuff that does not match the rules of the pbg file.

This might need a new pbg kind (combination of basic and overview)