Error on view() about trelliscope HOT 6 CLOSED

I'll get this fixed right away! I just heard another report about this yesterday - it must be Windows-specific. We do have Windows users, so must not be a problem for all Windows users, which might make it more difficult to track down. Unfortunately things like travis-ci don't help us catch cross-platform things like this in testing. I'll play around with it in my Windows VM. In the mean time, if you could post your full R console from startup to running through the commands, that would be great.

from trelliscope.

I know that MSWindows use to not like forward slashes in directory locations. The warning had it looking for an image in:

'C:\Users\Brian\Documents\housingjunk\vdb/displays/common/list_sold_vs_time/thumb_small.png

The directory exists, but no image is there. Perhaps the killing error is related to that as well. (and perhaps newer Windows OS doesn’t have that problem: I’m on Win7)

Here is some R output (everything before looked OK):

makeDisplay(byCounty,

•         name = "list_sold_vs_time",
  

•         desc = "List and sold price over time",
  

•         panelFn = timePanel, cogFn = priceCog,
  

•         width = 400, height = 400,
  

•         lims = list(x = "same"))
  

• Validating 'panelFn'...
• Testing cognostics function on a subset ... ok
• Precomputed limits not supplied. Computing axis limits...

Testing 'prepanelFn' on a subset...

Using 'trellis' panelFn to determine limits... dx and dy will not be computed.

At least one of the variables is not numeric. Casting as numeric for quantile calculation purposes.

• Updating displayList...
• Storing display object...

Error in data.frame(name = nms[i], tmp, stringsAsFactors = FALSE) :

arguments imply differing number of rows: 1, 0

In addition: Warning message:

package ‘base64enc’ was built under R version 3.0.3

view()

attempting to launch shiny vdb viewer...

press Ctrl+C or Esc to stop viewer

Loading required package: shiny

Warning: package ‘shiny’ was built under R version 3.0.3

Listening on http://127.0.0.1:8100

Warning: package ‘jsonlite’ was built under R version 3.0.3

Attaching package: ‘jsonlite’

The following object is masked from ‘package:shiny’:

validate

vdbPrefix is C:\Users\Brian\Documents\housingjunk\vdb

Warning in file(con, "rb") :

cannot open file 'C:\Users\Brian\Documents\housingjunk\vdb/displays/common/list_sold_vs_time/thumb_small.png': No such file or directory

Error in file(con, "rb") : cannot open the connection

From: hafen [mailto:[email protected]]
Sent: Wednesday, November 05, 2014 10:39 AM
To: tesseradata/trelliscope
Cc: bmasch
Subject: Re: [trelliscope] Error on view() (#62)

I'll get this fixed right away! I just heard another report about this yesterday - it must be Windows-specific. We do have Windows users, so must not be a problem for all Windows users, which might make it more difficult to track down. Unfortunately things like travis-ci don't help us catch cross-platform things like this in testing. I'll play around with it in my Windows VM. In the mean time, if you could post your full R console from startup to running through the commands, that would be great.

—
Reply to this email directly or view it on GitHub #62 (comment) .

from trelliscope.

I don’t know if you changed anything, but I closed and restarted RStudio, reran the code, and got it working.

The first time through, I had to install java and a few other things. Perhaps things were not completely in sync. You might have others with the same problem try a restart.

From: hafen [mailto:[email protected]]
Sent: Wednesday, November 05, 2014 10:39 AM
To: tesseradata/trelliscope
Cc: bmasch
Subject: Re: [trelliscope] Error on view() (#62)

I'll get this fixed right away! I just heard another report about this yesterday - it must be Windows-specific. We do have Windows users, so must not be a problem for all Windows users, which might make it more difficult to track down. Unfortunately things like travis-ci don't help us catch cross-platform things like this in testing. I'll play around with it in my Windows VM. In the mean time, if you could post your full R console from startup to running through the commands, that would be great.

—
Reply to this email directly or view it on GitHub #62 (comment) .

from trelliscope.

Interesting. When you re-ran the code, did that include the install_github(...) parts? If so, I updated Trelliscope this morning, so maybe one of the updates there fixed the problem and your code re-installed with the newer version.

I think the windows path thing was a red herring - I checked the code and I use file.path(...) everywhere so path should not be an issue.

Looking at your output, what happened is that there was an error in creating the display. It shows up after it says "* Storing display object...". So the command to create the display did not succeed therefore viewing it did not work. I'm going to make that error message more prominent and if there is an error, have it remove the display completely. Also when you type view, if it doesn't see any displays, it should let you know that - I'll do that as well.

from trelliscope.

No, I didn’t re-install from GH, so of course it didn’t reflect any update you made. Subconciously, seeing something in a browser makes me think it is coming from a remote server.

From: hafen [mailto:[email protected]]
Sent: Wednesday, November 05, 2014 2:35 PM
To: tesseradata/trelliscope
Cc: bmasch
Subject: Re: [trelliscope] Error on view() (#62)

Interesting. When you re-ran the code, did that include the install_github(...) parts? If so, I updated Trelliscope this morning, so maybe one of the updates there fixed the problem and your code re-installed with the newer version.

I think the windows path thing was a red herring - I checked the code and I use file.path(...) everywhere so path should not be an issue.

Looking at your output, what happened is that there was an error in creating the display. It shows up after it says "* Storing display object...". So the command to create the display did not succeed therefore viewing it did not work. I'm going to make that error message more prominent and if there is an error, have it remove the display completely. Also when you type view, if it doesn't see any displays, it should let you know that - I'll do that as well.

—
Reply to this email directly or view it on GitHub #62 (comment) .

from trelliscope.

This should be fixed in latest update.

from trelliscope.