Bug 1909950 - ghostscript v9.53.3 opvp is still disable
Summary: ghostscript v9.53.3 opvp is still disable
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: None
Product: Fedora
Classification: Fedora
Component: ghostscript
Version: 32
Hardware: x86_64
OS: Linux
unspecified
medium
Target Milestone: ---
Assignee: Anna Khaitovich
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2020-12-22 07:13 UTC by tess
Modified: 2021-01-14 11:56 UTC (History)
5 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2021-01-14 11:56:00 UTC
Type: Bug
Embargoed:


Attachments (Terms of Use)

Description tess 2020-12-22 07:13:00 UTC
Description of problem:
On Fedora32, I updated the ghostscript to the newest version v9.53.3-3.
The changelog says:
------
* Tue Nov 24 2020 Michael J Gruber <mjg> - 9.53.3-3
- Restore opvp device (#1899885)
------
But I tested the ghostscript, the opvp is still disable.


Version-Release number of selected component (if applicable):


How reproducible:


Steps to Reproduce:
1. Update ghostscript to newest version 9.53.3-3
2. Check the opvp status by inputting command "devicenames =="


Actual results:
opvp is not included in the device names.


Expected results:
opvp should be included in the device names.


Additional info:
(1) related git commit:
https://git.ghostscript.com/?p=ghostpdl.git;a=commitdiff;h=c6ce09aa5c9e

(2) I downloaded the newest source, and patch the ghostscript-9.53.3-restore-opvp-device.patch manually, then run ./autogen.sh & make. After building I checked the opvp status via "devicenames ==", the opvp is included.

(3) I am not sure, does the configure script in the ghostscript-9.53.3-3.fc32.src.rpm also need updated?

Comment 1 Michael J Gruber 2020-12-22 10:50:31 UTC
Over in the original bug 1899885 the OP reported "works for me" for the test build from the same source.

Are you sure opvp is not there, or is it just not listed because it is deprecated?

Comment 2 tess 2020-12-22 11:05:52 UTC
I builded the code with the patch, it works for me. But the newest rpm package in the fedora32 source doesn’t work for me. Also I tested the canon driver. Please confirm the configure script file is correct or not.

Comment 3 tess 2020-12-22 11:13:37 UTC
With the ghostscript-9.53.3-restore-opvp-device.patch, the configure script file should be updated.
But the configure script file in the gostscript-9.53.3-3.fc32.src.rpm and gostscript-9.53.3-1.fc32.src.rpm are same. Please confirm again.

Comment 4 Michael J Gruber 2020-12-22 11:52:20 UTC
This is somehow backwards.

How did you build the code? From a git checkout, from a release tarball or from the source rpm? How did you test the canon driver?

As I said, over in bug 1899885  the original reporter confirmed that the bug was solved for him, and it would have been easier to have the discussion there rather than opening a new bug.

Fedora development is done in the open, there is nothing we could or should "confirm" - and you already looked at the source, apparantly.

Upcoming scratch builds with a patched configure after autogen.sh rerun:

F34:
https://koji.fedoraproject.org/koji/taskinfo?taskID=58026171

F33:
https://koji.fedoraproject.org/koji/taskinfo?taskID=58026277

F32:
https://koji.fedoraproject.org/koji/taskinfo?taskID=58026425

Comment 5 tess 2020-12-22 12:13:23 UTC
Thanks for the rerun autogen.sh. I will test it tomorrow.
I builded the code from source rpm. And it’s easy to test the canon driver: print and get the spooler file. 
And I don’t know the bug 1899885 before I report this one :)

Comment 6 tess 2020-12-23 01:09:49 UTC
[Test Distribution]: Fedora 32 & Fedora 33

[Test Steps]: 
1. Download ghostscript-9.53.3-4.fc32.x86_64.rpm & libgs-9.53.3-4.fc32.x86_64.rpm from above links.
2. Install libgs-9.53.3-4.fc32.x86_64.rpm & ghostscript-9.53.3-4.fc32.x86_64.rpm by using command "rpm -i --force"

[Test Result]: OK
After inputting the command "devicenames ==", the opvp is included in the device names. And the Canon driver prints OK.

When can I update the ghostcript from the fedora source?


PS: 
I come from here: https://bugs.ghostscript.com/show_bug.cgi?id=703262

Comment 7 tess 2021-01-06 01:51:53 UTC
Hi,

(In reply to Michael J Gruber from comment #4)
> This is somehow backwards.
> 
> How did you build the code? From a git checkout, from a release tarball or
> from the source rpm? How did you test the canon driver?
> 
> As I said, over in bug 1899885  the original reporter confirmed that the bug
> was solved for him, and it would have been easier to have the discussion
> there rather than opening a new bug.
> 
> Fedora development is done in the open, there is nothing we could or should
> "confirm" - and you already looked at the source, apparantly.
> 
> Upcoming scratch builds with a patched configure after autogen.sh rerun:
> 
> F34:
> https://koji.fedoraproject.org/koji/taskinfo?taskID=58026171
> 
> F33:
> https://koji.fedoraproject.org/koji/taskinfo?taskID=58026277
> 
> F32:
> https://koji.fedoraproject.org/koji/taskinfo?taskID=58026425

When will the Ghostscript be updated to the Version 9.53.4-4 in the Fedora Repository?

Comment 8 tess 2021-01-06 01:55:01 UTC
(In reply to tess from comment #7)

> When will the Ghostscript be updated to the Version 9.53.4-4 in the Fedora
> Repository?

Sorry for written mistake, 
When will the Ghostscript be updated to the Version 9.53.3-4 in the Fedora Repository?

Comment 9 Michael J Gruber 2021-01-06 09:29:01 UTC
I have created pull requests now.

The thing is that we had a positive report without this fix already, and now one negative report without it and one positive report with it. That's not much. But should be OK.

Note also that I'm not a ghostscript maintainer. I maintain a library which gs depends on, and I have commit rights for gs only to submit coordinated rebuilds of gs and other packages when the library (jbig2dec) changes. So, while technically I could do the update I don't feel entitled to.

Comment 10 tess 2021-01-06 09:51:55 UTC
Thanks, Michael! I am waiting :)

Comment 11 Anna Khaitovich 2021-01-13 15:57:34 UTC
Hi, there is an update including this fix - https://bodhi.fedoraproject.org/updates/FEDORA-2021-644648d619
Could you please confirm if its working for you?
Thanks in advance

Comment 12 tess 2021-01-14 01:39:04 UTC
It works for me.

<Test Case1>
-----------------
[Architecture]: Fedora32 +x86_64

[Steps]: 
1. Download ghostscript-9.53.3-4.fc32.x86_64.rpm & libgs-9.53.3-4.fc32.x86_64.rpm from the link https://bodhi.fedoraproject.org/updates/FEDORA-2021-644648d619.
2. Install libgs-9.53.3-4.fc32.x86_64.rpm & ghostscript-9.53.3-4.fc32.x86_64.rpm by using command "rpm -i --force"

[Result]: OK
After inputting the command "devicenames ==", the opvp is included in the device names. And the Canon driver prints OK.


<Test Case2>
-----------------
[Architecture]: Fedora33 + aarch64

[Steps]: 
1. Download ghostscript-9.53.3-4.fc32.aarch64.rpm & libgs-9.53.3-4.fc32.aarch64.rpm from the link https://bodhi.fedoraproject.org/updates/FEDORA-2021-644648d619.
2. Install libgs-9.53.3-4.fc32.aarch64.rpm & ghostscript-9.53.3-4.fc32.aarch64.rpm by using command "rpm -i --force"

[Result]: OK
After inputting the command "devicenames ==", the opvp is included in the device names. But Canon has no aarch64 driver to be tested.


Note You need to log in before you can comment on or make changes to this bug.