purr-data issueshttps://git.purrdata.net/jwilkes/purr-data/-/issues2016-02-11T18:09:38Zhttps://git.purrdata.net/jwilkes/purr-data/-/issues/18gui_start_array can create malformed GUI msg2016-02-11T18:09:38ZJonathan Wilkesgui_start_array can create malformed GUI msgcurrently you can't send a message to the GUI with a single array parameter:
```c
gui_start_vmess("some_message", "");
gui_start_array();
gui_s("hello world");
gui_end_array();
gui_end_vmess();
```
This will output:
```c
"s...currently you can't send a message to the GUI with a single array parameter:
```c
gui_start_vmess("some_message", "");
gui_start_array();
gui_s("hello world");
gui_end_array();
gui_end_vmess();
```
This will output:
```c
"some_message ,[\"hello world\"]"
```
which will cause a parser errorJonathan WilkesJonathan Wilkeshttps://git.purrdata.net/jwilkes/purr-data/-/issues/22canvas selection rectangle2016-02-24T05:30:44ZJonathan Wilkescanvas selection rectangle1. create [pd]
2. click an empty part of the canvas to instantiate
3. now that [pd] has focus, click the parent canvas window to give it focus
Bug! The canvas selection rectangle is anchored and resizes on drag, as if you had the mo...1. create [pd]
2. click an empty part of the canvas to instantiate
3. now that [pd] has focus, click the parent canvas window to give it focus
Bug! The canvas selection rectangle is anchored and resizes on drag, as if you had the mouse button depressed.Jonathan WilkesJonathan Wilkeshttps://git.purrdata.net/jwilkes/purr-data/-/issues/20perf mode quit dialog looks strange2016-03-05T02:41:59ZJonathan Wilkesperf mode quit dialog looks strangeThe confirmation dialog in perf mode (`[perf 1(--[send pd]`) has a big long chrome app hash at the top and looks different than the canvas close dialog. Both the dialogs should look the same.The confirmation dialog in perf mode (`[perf 1(--[send pd]`) has a big long chrome app hash at the top and looks different than the canvas close dialog. Both the dialogs should look the same.Jonathan WilkesJonathan Wilkeshttps://git.purrdata.net/jwilkes/purr-data/-/issues/19Four Declarations Of Independence in an Object Box messes up the GUI viewport2016-03-07T00:33:45ZJonathan WilkesFour Declarations Of Independence in an Object Box messes up the GUI viewportIf you put an enormous amount of data into an object or message box, you can no longer click on an empty part of the patch to instantiate the object. `<ctrl-Enter>` still works, though.If you put an enormous amount of data into an object or message box, you can no longer click on an empty part of the patch to instantiate the object. `<ctrl-Enter>` still works, though.Jonathan WilkesJonathan Wilkeshttps://git.purrdata.net/jwilkes/purr-data/-/issues/23[draw image] disappears if image path doesn't exist2016-03-24T17:58:10ZJonathan Wilkes[draw image] disappears if image path doesn't exist[draw image] will disappear after you create it. It happens either with no path argument, or when the path argument doesn't exist[draw image] will disappear after you create it. It happens either with no path argument, or when the path argument doesn't existJonathan WilkesJonathan Wilkeshttps://git.purrdata.net/jwilkes/purr-data/-/issues/21properly handle reading/writing utf-8 filenames on Windows2016-06-04T04:48:45ZJonathan Wilkesproperly handle reading/writing utf-8 filenames on WindowsThe wrappers to properly convert utf-8 filenames for reading/writing on Windows have not yet been ported.
See Pd Vanilla git: https://sourceforge.net/p/pure-data/pure-data/ci/master/tree/#
commit: 78b81aa3cb903d923da9eec8ad104935a5...The wrappers to properly convert utf-8 filenames for reading/writing on Windows have not yet been ported.
See Pd Vanilla git: https://sourceforge.net/p/pure-data/pure-data/ci/master/tree/#
commit: 78b81aa3cb903d923da9eec8ad104935a5ae9ce4
plus some other commits after that oneJonathan WilkesJonathan Wilkeshttps://git.purrdata.net/jwilkes/purr-data/-/issues/15menu item shortcuts don't consistently propagate to the DOM2016-11-03T17:31:39ZJonathan Wilkesmenu item shortcuts don't consistently propagate to the DOMSome menuitem shortcuts don't pass their events on to the dom, while others do. I think it happens when a double-modifier has keys in common with a single-modifier.
Example: `<ctrl-shift-v>` creates a vsl, but the key gets sent to Pd....Some menuitem shortcuts don't pass their events on to the dom, while others do. I think it happens when a double-modifier has keys in common with a single-modifier.
Example: `<ctrl-shift-v>` creates a vsl, but the key gets sent to Pd. However, `<ctrl-shift-h>` creates an hsl, and does _not_ send a key to Pd.
The result is that for one you get a "floating" iemgui that follows the mouse, and the other gets anchored. (Can't remember which happens with vsl and which with hsl, but I'm not in front of a running instance of Pd atm.)
Since `<ctrl-v>` is bound to "paste" and `<ctrl-h>` isn't bound to anything, that may somehow be causing an inconsistency.Jonathan WilkesJonathan Wilkeshttps://git.purrdata.net/jwilkes/purr-data/-/issues/24pwd is wrong on Windows2016-11-03T17:36:20ZJonathan Wilkespwd is wrong on WindowsOn windows,the working directory is set to HOMEPATH. This breaks the user experience.
nw.js doesn't currently have a way to retrieve the pwd on Windows. As a workaround we could retrieve it with a windows system call in C, and send th...On windows,the working directory is set to HOMEPATH. This breaks the user experience.
nw.js doesn't currently have a way to retrieve the pwd on Windows. As a workaround we could retrieve it with a windows system call in C, and send that as an argument when starting the GUI.Jonathan WilkesJonathan Wilkeshttps://git.purrdata.net/jwilkes/purr-data/-/issues/289invalid read with hidden iemgui on GOP canvas2017-03-23T02:40:46ZJonathan Wilkesinvalid read with hidden iemgui on GOP canvas1. run Pd with `valgrind pd-l2ork -nrt`
2. create a new patch
3. create a `[pd]` ~~and make it graph-on-parent~~
4. open `[pd]`
5. create an iemgui (I tested with `[bng]` and `[nbx]`
6. choose "Properties" and click the "init" check...1. run Pd with `valgrind pd-l2ork -nrt`
2. create a new patch
3. create a `[pd]` ~~and make it graph-on-parent~~
4. open `[pd]`
5. create an iemgui (I tested with `[bng]` and `[nbx]`
6. choose "Properties" and click the "init" checkbox
7. ~~position the new iemgui *outside* the red graph-on-parent rectangle~~
8. close the subpatch
9. select the `[pd]` object on the parent canvas
10. click `<Delete>` or `<Backspace>` (I tested with `<Delete>`)
Bug!
A chain of invalid reads gets triggered for the iemgui.
Doesn't happen if the iemgui is visible on the graph-on-parent. Also doesn't happen if the iemgui doesn't have "init" checked.
Edit: also can happen if the subpatch isn't a GOP subpatch.https://git.purrdata.net/jwilkes/purr-data/-/issues/286change t_int to int in gui_do_vmess2017-03-23T02:40:46ZJonathan Wilkeschange t_int to int in gui_do_vmessThere's a stray `t_int` in line 843 of gui_do_vmess in s_inter.c. That should be changed to `int`.There's a stray `t_int` in line 843 of gui_do_vmess in s_inter.c. That should be changed to `int`.https://git.purrdata.net/jwilkes/purr-data/-/issues/284feature request: display version of purr data in Purr Data -> About Purr Data2017-03-23T02:40:46ZEsa Ruohofeature request: display version of purr data in Purr Data -> About Purr DataCurrently, if I boot up Purr Data on OS X and go to Purr Data -> About Purr Data, it says:
Version 54.0.2840.71 (2840.71)
Copyright 2016.
Yet, @jwilkes posted about some sort of Purr Data 2.1.1 or something on the pd-mailinglist?
...Currently, if I boot up Purr Data on OS X and go to Purr Data -> About Purr Data, it says:
Version 54.0.2840.71 (2840.71)
Copyright 2016.
Yet, @jwilkes posted about some sort of Purr Data 2.1.1 or something on the pd-mailinglist?
I find this somewhat confusing. Should this 2.1.1 not read somewhere in the "About Purr Data"?https://git.purrdata.net/jwilkes/purr-data/-/issues/270wrong date/rev number when building after `git pull`2017-03-23T02:40:46ZJonathan Wilkeswrong date/rev number when building after `git pull`If I `git pull` new code on Tuesday after having successfully build Purr Data on Monday, the binary I get on Windows still has the date/rev from Monday's repository state.If I `git pull` new code on Tuesday after having successfully build Purr Data on Monday, the binary I get on Windows still has the date/rev from Monday's repository state.https://git.purrdata.net/jwilkes/purr-data/-/issues/268Cord inspector regression2017-03-23T02:40:46ZAlbert GräfCord inspector regressionI'm not sure whether that's a recent regression, but I vaguely recall this working properly at *some* point. Anyway, the tooltips of the cord inspector are too short to show the complete data (in case of an audio connection at least). Th...I'm not sure whether that's a recent regression, but I vaguely recall this working properly at *some* point. Anyway, the tooltips of the cord inspector are too short to show the complete data (in case of an audio connection at least). This is how it looks on Linux (Mac is pretty much the same); note that there's only very little of the third number visible there:
![Screenshot_2017-02-21_14-47-57](/uploads/a0ee34ec6282ae4a44fa084e5d7783cb/Screenshot_2017-02-21_14-47-57.png)https://git.purrdata.net/jwilkes/purr-data/-/issues/266give patch local paths precedence in loader search2017-03-23T02:40:46ZJonathan Wilkesgive patch local paths precedence in loader searchFirst off-- does anyone know if we ported the relatively recent changes to Vanilla's loader search order? One feature I remember is that it searches for binaries then abstraction for *each* path, rather than searching for binaries in *a...First off-- does anyone know if we ported the relatively recent changes to Vanilla's loader search order? One feature I remember is that it searches for binaries then abstraction for *each* path, rather than searching for binaries in *all* paths then searching for abstractions in all paths.
Anyhow, what I want is to give precedence to the patch local (or I guess "canvas local") paths over the system ones. Here's the behavior I want:
* If I put a single "list2symbol.pd" in the same directory as an empty patch and then open the empty patch in a freshly running Pd instance, I want to type [list2symbol] and be *guaranteed* to create my abstraction.
* Similarly-- suppose I put a single "list2symbol.pd" in a subdirectory "foo" that resides in the directory of my empty patch. If I open the empty patch in a freshly running Pd instance, I want to type [declare -path foo] then type [list2symbol] and be *guaranteed* that my abstraction loads.
This should make it possible to get minimally predictable behavior from two models of patching. The first is the Pd-extended model of just depending on the externals to be available in the distro. The second is putting everything needed into the patch directory or subdirs.https://git.purrdata.net/jwilkes/purr-data/-/issues/263loading more libraries by default2017-03-23T02:40:46ZAlexandre Porresloading more libraries by defaultI see that you inherited libraries from extended and also which ones to load by default, and how maybe loading all of them is not worthy...
i use extended 0.42-5 and can see that about 30 aren't loaded by default, but I wish to ask that...I see that you inherited libraries from extended and also which ones to load by default, and how maybe loading all of them is not worthy...
i use extended 0.42-5 and can see that about 30 aren't loaded by default, but I wish to ask that at least 3 of them could be added
- bsaylor
- tof
- windowing
thankshttps://git.purrdata.net/jwilkes/purr-data/-/issues/259loading tof by default2017-03-23T02:40:46ZAlexandre Porresloading tof by defaulthowdy, not sure how you feel about this, but i know that in extended not all libraries were loaded by defaulft, tof being one of them... I dont see much point in not loading particular libraries and it's kind of a pain to do it yourself
...howdy, not sure how you feel about this, but i know that in extended not all libraries were loaded by defaulft, tof being one of them... I dont see much point in not loading particular libraries and it's kind of a pain to do it yourself
do you have plans to load all of them by default?
i consider tof important enough for that anyway
cheershttps://git.purrdata.net/jwilkes/purr-data/-/issues/256make a dropdown object to stand in for [tof/pmenu]2017-03-23T02:40:46ZJonathan Wilkesmake a dropdown object to stand in for [tof/pmenu]Since we're not shipping `[tof/pmenu]`, we need some kind of GUI object that acts as a dropdown menu.
The problem with using a native html dropdown is the following:
* it's difficult to sync with Pd's widget behavior. A user might w...Since we're not shipping `[tof/pmenu]`, we need some kind of GUI object that acts as a dropdown menu.
The problem with using a native html dropdown is the following:
* it's difficult to sync with Pd's widget behavior. A user might want to click-drag it in editmode, but the HTML5 menu will want to pop up the options on a click. One could disable the button in editmode, but then `<ctrl-click>` becomes a problem, etc.
* html5 inputs either have to be put on top of the patch's svg (in which case bring to front/send to back won't work), or included inside the svg as a foreignObject. The foreignObject part of the svg spec isn't particularly detailed and probably doesn't work consistently across browsers (or even browser versions...)
* `<select>` sizes the button widget to fit the containing text. That's obviously the right behavior, but it conflicts with Pd's absolute positioning system. Users won't be able to know how wide the widget will be on other people's systems and collisions are therefore likely to occur.
One solution would be to add a t_text behavior for a dropdown box. It could be drawn to look like an atom box (or maybe slightly different border), but a click would bring up the options. To display it, we could use a `<select>` object with the button's visibility set to 'hidden'. Then when we get a click we invoke the click method to bring up the options list. That way we leverage the browser's ability to place the options properly (handling the literal corner cases correctly), and get Pd's text sizing/positioning for the button.https://git.purrdata.net/jwilkes/purr-data/-/issues/254add pd-l2ork version method for [pdinfo]2017-03-23T02:40:46ZJonathan Wilkesadd pd-l2ork version method for [pdinfo]We should make it possible to print out the pd-l2ork version number from patch.We should make it possible to print out the pd-l2ork version number from patch.https://git.purrdata.net/jwilkes/purr-data/-/issues/251Copy/Paste - Strange behavior2017-03-23T02:40:46ZPierreCopy/Paste - Strange behaviorI don't know if it is intentional or not but I think that the copy/paste action has a strange behavior. When I paste an object, it acts like if I had selected the object and kept the mouse down so the object follows the mouse's moves unt...I don't know if it is intentional or not but I think that the copy/paste action has a strange behavior. When I paste an object, it acts like if I had selected the object and kept the mouse down so the object follows the mouse's moves until I click down. And when I click down, that activates the edition of the object (see the gif). In my opinion, this behavior is strange and I think when you paste an object, this object should be selected but the moves of the mouse shouldn't move the object.
![Copy-Paste](/uploads/9ffc0b125d89465bb4e37ff2c204fb36/Copy-Paste.gif)
I made this video with two patches but it's the same behavior with one patch.
Furthermore, if you copy/paste an object in the same patch with the original object still selected, it adds the new object after the original one and creates a connection between them (like if you create an object whith another selected). In my opinion, it's not annoying (perhaps it's even useful, I don't know yet) but I don't know if this is intentional or not so I just want to notify you of it.
macOS Sierra 10.12.3 (16D32)
Version 54.0.2840.71 (2840.71) (pd-l2ork-2.0-osx-10.11-x86_64)
ps: It seems that the Purr-Data version of the 2.0 is the same version than the previous binary.https://git.purrdata.net/jwilkes/purr-data/-/issues/292apple/sudden_motion_sensor2017-03-23T22:46:05Z60-hzapple/sudden_motion_sensor[apple/sudden_motion_sensor] is created but doesn't work.
error: [sudden_motion_sensor]: could not read device
Hitting "info" message give:
print: sensor SMCMotionSensor
Other objects from apple library works correctly.
OSX 10.9.5, Ve...[apple/sudden_motion_sensor] is created but doesn't work.
error: [sudden_motion_sensor]: could not read device
Hitting "info" message give:
print: sensor SMCMotionSensor
Other objects from apple library works correctly.
OSX 10.9.5, Version 2.1.2 (20170322-rev.6f652fe)