FYI @erfajo - your packages were deleted from package manager yesterday. We currently have in our backlog a task to figure out a mechanism by which package authors can delete their own packages. But, we are still trying to figure out how to handle this properly since other customers could depend on these packages.
If I can chime in to the topic, I agree with you that the package manager is sometimes a bit cumbersome (in particular when updating a package), but pretty decent. Bear in mind that not all of the 483 authors to date are software developers and all these issues you comment might just be affecting a smaller part of the whole community (being the vast majority Dynamo users).
Also being Dynamo an open source project, I find more proactive to build the change if you think is needed than demanding it.
I am far from an expert on the matter, but regarding your four major issues I can only recommend to read the Semantic Versioning Specification. While is not mandatory to adhere to it, I think Dynamo follows it and is very reasonable.
-
When renaming a package, seems sensible to deprecate/archive adding a note pointing to the new package and transfer it to the new available name. Just think about the user with the âoldâ version, he/she should be properly informed rather than not finding it anywhere.
-
Version numbering is up to the author, but following common rules helps the overall community.
-
For parallel versions, you can always create packages on the Package Manager targeting those versions (Orchid 1.x, Orchid 2.x?). So users will know straigth away what to download.
-
To err is human. Authors have the responsability with the users to test and make sure the package is bug free, but is something that can easily happen. I recommend this semantic versioning QA., which can also be translated to âif you messed it up, realease a fixed update and apologise but not remove itâ. Imagine if someone has version X.Y.Z with bugs. If you delete it and use the same number on a bug free version, there will exist two versions with same number but different, and that is just wrong.
It also seems that this post has gone a bit off topic. Might worth to close it and follow the discussion on another if needed or GitHub itself?
This thread is getting way off topic, the latter posts have nought to do with searching in Dynamo and a little overly critical and rude in places. This is more of a github wishlist item and should be chased on there (or most certainly a different thread named âWhy I hate PM and other thingsâ)
We all know the rules. Keep calm and keep relevant!
However, Iâm sure the Dynamo Dev team are plenty busy but well aware of any shortcomings in the way packages are deployed and managed. By the sounds of it, from reading Racelâs comment, they are planning to deal with it but have some thinking to do - this sounds like a pretty good time to make any suggestions if you have them and try make it higher up on their priorities list.
It would be more useful to link the github issue raised so people interested in this part of this thread can add to it and maybe make a new topic for this where you can rally for change migrating this rant somewhere else.
Dynamo 2.0.1 I have about 15 packages installed. Search is still very slow for me as well. It was in the past and it still is.
Additionally, It doesnât always return the nodes with the searched words immediately. I.e. When I search for âAll Elements of Categoryâ. After C it appears on the list.
Another example âParametersâ. In my search results it appears like 30 after typing the whole name.
Reducing the number of external dlls used in your packages might speed things along.
Are you saving packages to your C drive or a network drive?
Iâd think that having all packages on a local SSD would be the best. Antivirus software may also have an effect.
So still noone knows how to deal with it. Its been the same for a too long time for me. I just bought new pc, I9 32gb ram and fast ssd and its still the same. Sometimes searching freezes for couple seconds, dynamo is unusable for me. Anyone?
please see here:
I have also been experiencing slowness using the search function, for years now. My theory on this is related to the way it searches.
If I search for âAll Elementsâ it returns âName-Roomâ as the first search item. If I search for âAll Elements ofâ it returns âBy-Face-Wallâ as the first. Doesnât seem to follow a logic, although Iâm sure it does.
Is the search trying to be too clever? In other words, is it trying to do some sort of âPageRankâ search based on an ever updating database? It searches as you type, trying to offer immediate results. It also searches in the Library and description.
Would it be better for the search to search alphabetically for Node names and wait until youâre finished typing?
Just a few ideasâŚ
try this:
AllElementsOfCategory is found quicker if you type âAll El Of Caâ
This.
You basically have to learn what you need to search for to find what youâre looking for rather than just searching for what youâre looking for.
I have no idea what the âlogicâ behind the search function is, but there has to be something. It would be nice if this was updated.
programmers blind spot
We all know it happens.
i do it, skipping parts that make it logical when i explain things
Hi guys, I had the same issue and after uninstalling Autodesk Point Layout (which was apparently the latest add-in that we have installed) Dynamo started working properly again.
Apparently I was about to post this as a thread.
In my case, sometimes it works too fast (2 out of 10) and rest of the time, I have to cancel the search query because it just Freezes!!!
Currently, I have 34 packages installed .
This thread really got off-topic quickly.
We just installed Windows 10 and I am experiencing the same issue as described in the original post.
Before installing Windows 10 I did not have issues. With running Dynamo Revit 2017 / 2018.
The issue only occurs after running a graph.
Dynamo in this mode is not usable.
-
Autodesk Point Layout is not (was never) installed.
-
Packages are installed locally.
-
Althought latency ist still there and noticable; after closing the file (with dynamo still open) and reopning, the latency during search improved significantly.
-
Strangely enough, Ethernet in Task Manager went through the roof during search and returned to baseline afterwards.
-
Problem persists in âManualâ execution mode.
-
Analytics reporting has been disabled.
-
After disabling analytics in Dynamo, the problem persisted. Iâve restarted dynamo, and run a graph. The problem persists.
- After disabling analytics in Revit, the latency has improved considerably.
I also recommend creating a script to run in dynamo player to release RAM when finished using dynamoâŚexpecially when working with large data sets. For example, I use a script which adds 1 + 1.
Still working to improve / test latency and settings. Iâll let you know if analytics is part of the issue.
I find this issue after running larger graphs several times. In these cases, I often find myself restarting revit so that I can work efficiently in the Dynamo environment. But if I am feeling particularly stubborn, Iâll start typing design script into code blocks. Somehow the search mechanisms that are enabled while typing script are never over encumbered. Kinda strangeâŚ
Uninstalling Autodesk Desktop Connector also helps.
I have windows 10 running on two machines. One of them a laptop.
the more powerful machine strugles to run simple scripts, while the weaker laptop tears through large data sets.
This doesnât seem to have anything to do with DLLs or graph sizes. (My graphs are broken up into smaller individual graphs and run sequentially.)
I donât have Autodesk Desktop Connector installed.