[Bug 1915971] [NEW] pandas needs pyarrow, which doesn't exist

2021-02-17 Thread Ross Boylan
Public bug reported:

python3-pandas needs certain packages which are not listed among its 
dependencies, even as suggests, and are not available in Ubuntu.  I ran into 
this while using feather.  Here's a debug session; earlier import pandas as pd 
we executed.
(Pdb) c
> /usr/local/fahydev/Kornak/bifs/examples/investigate09.py(59)()
-> subsample = pd.read_feather(SUBFILE).astype({"id": int})
(Pdb) n
ImportError: Missing optional dependency 'pyarrow'.  Use pip or conda to 
install pyarrow.

The removal of support for feather and the absence of a suitable package
is noted in the changelog for 0.25.2+dfsg-1.

apache provides packaging and instructions upstream, but it would be
really nice if this just worked.

Note that python3-arrow, an existing Ubuntu package, has absolutely
nothing to do with the pyarrow library.

Desired Outcome:
pyarrow packaged like other python libraries.
The relevant package listed as a dependency of python3-pandas (I'd be fine with 
recommends though suggests would do).

In case the info I uploaded didn't get through, this is on Ubuntu 20.04
with python3-pandas 0.25.3+dfsg-7

ProblemType: Bug
DistroRelease: Ubuntu 20.04
Package: python3-pandas 0.25.3+dfsg-7
ProcVersionSignature: Ubuntu 5.4.0-1039.41-azure 5.4.78
Uname: Linux 5.4.0-1039-azure x86_64
ApportVersion: 2.20.11-0ubuntu27.16
Architecture: amd64
CasperMD5CheckResult: skip
Date: Wed Feb 17 16:01:02 2021
InstallationDate: Installed on 2020-10-02 (138 days ago)
InstallationMedia: Ubuntu-Server 18.04.5 LTS "Bionic Beaver" - Release amd64 
(20200810)
PackageArchitecture: all
ProcEnviron:
 TERM=screen
 PATH=(custom, no user)
 XDG_RUNTIME_DIR=
 LANG=en_US.UTF-8
 SHELL=/bin/bash
SourcePackage: pandas
UpgradeStatus: Upgraded to focal on 2020-10-02 (138 days ago)

** Affects: pandas (Ubuntu)
 Importance: Undecided
 Status: New


** Tags: amd64 apport-bug focal

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1915971

Title:
  pandas needs pyarrow, which doesn't exist

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/pandas/+bug/1915971/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

[Bug 521533]

2018-11-28 Thread Ross-boylan
I don't want my previous response to be taken as a vote for the status
quo.  The behavior I would expect is:

1. If I don't hit save my work disappears.(*)  The current application
does not have a save function (as distinct from save as), and I'm pretty
sure that if I fill in a form, exit, and then open the form my old
values will still be there.  Worse, if someone else using the same
account opens the form, they will see my info.

2. If I do save (not just save as) my work will be saved with the file.

In this case I might not expect, but would be pleased if 
3. there were an option to save the form data to a separate file and restore it 
from a separate file.  I'd guess such a facility is consistent with the 1600 
page XFA spec, though I can't say I know where :)

Because the current behavior violates these expectations, it is a
security risk.   Someone's personal information may be exposed in ways
unanticipated, and operations that usually assure security, like not
saving a file or deleting it, will not work.  And operations that are
expected to reveal info, like copying/mailing a pdf or operating on it
with a different program, may instead conceal/disappear the information.

(*) Some usability experts argue that "work disappears if I don't save"
is not the expectation of the lay user, and that our current model of
"you must save to keep your work" is aggravating and unintuitive to
them.  That may well be correct.  But unless the surrounding programs
all start behaving this way, this behavior is undesirable.  An
application that may be dealing with sensitive private information is
not the place to pioneer new interface models.

-- 
You received this bug notification because you are a member of Kubuntu
Bugs, which is subscribed to kdegraphics in Ubuntu.
https://bugs.launchpad.net/bugs/521533

Title:
  Okular stores form data in a different directory (possible leak of
  private data)

To manage notifications about this bug go to:
https://bugs.launchpad.net/kdegraphics/+bug/521533/+subscriptions

-- 
kubuntu-bugs mailing list
kubuntu-b...@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/kubuntu-bugs


[Bug 521533]

2018-11-28 Thread Ross-boylan
(In reply to lutz.wrage from comment #50)
> Is there any use case that justifies storing form data in
> ~/.kde/share/apps/okular/docdata/ given that the same data can be stored in
> the pdf itself? 
> It seems to me that there isn't.

While I consider the current behavior undesirable, it does have its
advantages.  In fact, it's the reason I decided to use okular for my
taxes.  Here's the use case:

1. Generate some forms automatically (e.g., opentaxsolver computes my taxes and 
fills in government  forms).
2. Resulting forms require some manual tweaks (e.g., check boxes, fill in 
additional fields).
3. Discover forms need to be regenerated to correct a mistake.  Modify inputs 
and return to 1.

In this scenario, the work in 2 is lost if the results have been stored
in the pdf, but is retained if the values are stored elsewhere, as
okular currently does.  Even if the pdf in 2 is saved under a different
name, so that the results are not literally lost, one must manually
identify the changed information and reenter it.

There is another scenario in which the recreation of the information is
less desirable.  If some of the manually entered information in 2
depends on the values from 1, e.g., you manually enter line 55 as a copy
of line 32 but the automatically generated line 32 changes, then the
previous manual data may be invalid.

-- 
You received this bug notification because you are a member of Kubuntu
Bugs, which is subscribed to kdegraphics in Ubuntu.
https://bugs.launchpad.net/bugs/521533

Title:
  Okular stores form data in a different directory (possible leak of
  private data)

To manage notifications about this bug go to:
https://bugs.launchpad.net/kdegraphics/+bug/521533/+subscriptions

-- 
kubuntu-bugs mailing list
kubuntu-b...@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/kubuntu-bugs


[Bug 1655584] Re: systemd-udevd busyloops when nvidia kernel module fails to load

2017-01-31 Thread Ross Boylan
I was able to stop the loop by blacklisting the nvidia modules and restarting 
the system.
I added /etc/modprobe.d/nvidia-kill.conf, deliberately chose to appear after 
the existing nvivida conf files in the same directory (not sure if that's 
essential):

blacklist nvidia_367
blacklist nvidia_367_uvm
blacklist nvidia_367_modeset
blacklist nvidia_367_drm

alias nvidia off
alias nvidia-uvm off
alias nvidia-modeset off
alias nvidia-drm off

This was all guesswork; I don't know how much is essential.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1655584

Title:
  systemd-udevd busyloops when nvidia kernel module fails to load

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/1655584/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs


[Bug 1655584] Re: systemd-udevd busyloops when nvidia kernel module fails to load

2017-01-31 Thread Ross Boylan
I tried 
systemctl stop nvidia-persistenced
nvidia-persistenced --no-persistence-mode
nvidia-smi --persistence-mode=Disabled
but none of them helped.  nvidia-smi failed because it couldn't communicate 
with the nvidia driver.

The system was created using a raw disk in a virtual machine
(VirtualBox).  I then booted off the disk so it was running for real,
and accepted the offered nvidia drivers.  Finally, I shutdown and
restarted in the VM.  That is the context for the loop.  The VM has no
nvidia hardware, and so obviously there is no reason for the driver to
load.

Ross

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1655584

Title:
  systemd-udevd busyloops when nvidia kernel module fails to load

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/1655584/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs