*Version: Pd-0.47.1 Windows 7* *Desc: * *Save-as should change current context to new file* A habit I have developed is to Ctrl-S frequently, to save current patch under development (in case of a crash).
This is my habit not just for Pd, but for almost any editing tool.
Sometimes, I am working on a patch or abstraction, where I want to do a CHECKPOINT/FORK.
I mean Checkpoint as, "I am using this abstraction elsewhere. So I do not want to change the behavior there. So I will SAVE-AS to create a new context."
However Pd editor GUI behaves differently here than almost any other tool.
Pd will Save-As to the new filename, but keep the present context the EXISTING file.
So now I am patching, patching, changing things.... HABIT: Control-S.
*BAM.*
I have just *clobbered my Checkpoint,* my last good, working copy of the patch!
This is very annoying, and I think I have mistakenly clobbered abstractions I wanted to save as stable/working, and inadvertently altered behavior of patches I wanted to leave alone.
So let's say I had OrigFile.pd and want to fork that to OrigFileNEW.pd
So I Save-As to OrigFileNEW.pd *But the exiting context stays as OrigFile.pd*
When I am done hacking, and close down for the night, I end up with *OrigFile.pd, which is actually the altered abstraction*.
EVERY INSTANCE of this in other patches has the new behavior. and *OrigFileNEW.pd *is actually the Checkpoint of OrigFile.pd!
Doesn't that seem weird?
I guess this is OK, as long as people understand what is going on and anticipate it.
I would love to see this fixed, or else the existing behavior explained/justified with a BIG WARNING in the docs so that people can be prepared.
*With Current behavior, my Checkpoint procedure should be:*
1) Save-AS (new filename). 2) CLOSE existing context. (which destroys settings) 3) OPEN new context 4) continue patching.
or else 1) Save-AS (Patch.CHECKPOINT.pd). 2) continue patching 3) DONE, close everything down. 4) Rename *OrigFile.pd * to *OrigFileDEV.pd *, and *OrigFileNEW.pd * to *OrigFile.pd *
Thanks, BH
On 10/28/2016 06:50 AM, William Huston wrote:
I would love to see this fixed, or else the existing behavior explained/justified with a BIG WARNING in the docs so that people can be prepared.
which docs? while there are docs, i doubt that any BIG FAT WARNING will have the desired effect.
i think the best (and simplest) solution would be to rename the "Save As..." menu to "Save a Copy..." for abstraction-instances (as opposed to top-level-patches).
switching/destroying contexts seems not such a good idea (for an instance-oriented programming language such as Pd).
gfamrd IOhannes
On 28-10-16 22:17, IOhannes m zmölnig wrote:
On 10/28/2016 06:50 AM, William Huston wrote:
I would love to see this fixed, or else the existing behavior explained/justified with a BIG WARNING in the docs so that people can be prepared.
which docs? while there are docs, i doubt that any BIG FAT WARNING will have the desired effect.
i think the best (and simplest) solution would be to rename the "Save As..." menu to "Save a Copy..." for abstraction-instances (as opposed to top-level-patches).
The GNU program Notepad++ has just such an option, next to Save As...
switching/destroying contexts seems not such a good idea (for an instance-oriented programming language such as Pd).
In the past I vaguely remember this type of behaviour, but just now with Linux and 0.47.1, it behaves as expected for Save As... I should try more platforms... What I saw looked more like a menu bar which was not updated.
@William: What is the platform and Pd-version?
gfamrd IOhannes
Greetings,
Fred Jan
Pd-dev mailing list Pd-dev@lists.iem.at https://lists.puredata.info/listinfo/pd-dev