Jump to content


Most Liked Content


#6531 P4VS hangs Visual Studio 2012 for several minutes frequently

Posted solidstore on 08 November 2012 - 11:42 AM

I would love to use the Visual Studio plugin for perforce however in our solution we have found that it hangs the entire IDE for atleast 5 minutes every so often and its extremely frustrating. The problem disappears when the Perforce plugin is disabled. I'm not sure what causes it, it seems to happen randomly but quite frequently.

Does this happen to anyone else? Or is there a workaround?


#24588 P4Python undefined symbol SSL_library_init

Posted PompeyPaul on 12 March 2019 - 06:50 PM

I solved this under Fedora Core 29 as I describe here: https://forums.perfo...fedora-core-29/


#24292 P4Python undefined symbol SSL_library_init

Posted P4Jen on 07 January 2019 - 04:49 PM

Hi,

Thanks for bringing this to our attention.

I have tested here with Ubuntu 18.04 and have reproduced the behaviour you describe.

I have also had a look through the latest P4Python release notes, and do not see any mention of
Ubuntu 18.04 not being a supported Linux flavour.

I will check to see if we have an existing report of this issue.

If we do I will add details of your post to it.

If not I will log one and add details of the post to it.

Thanks again,
Jen!



#24650 P4Python undefined symbol SSL_library_init

Posted TTimoT on 03 April 2019 - 06:47 PM

There are several fixes and workarounds floating out there by now, but it would be really useful to get an official fix from Perforce to support newer libssl releases.


#24463 P4Python undefined symbol SSL_library_init

Posted bdurbin on 21 February 2019 - 06:31 PM

I also get this ImportError when I try to use p4python. Is there any work around or plan for an update to fix this?


#24284 P4Python undefined symbol SSL_library_init

Posted p4user101 on 03 January 2019 - 11:21 PM

I've recently installed P4Python on Ubuntu 18.04, which uses libssl-1.1. This version of libssl has deprecated SSL_library_init (https://www.openssl....brary_init.html).

$ aptitude show libssl-dev
some code
Package: libssl-dev
Version: 1.1.0g-2ubuntu4.3
State: installed
Automatically installed: no
Multi-Arch: same
Priority: optional
Section: libdevel
Maintainer: Ubuntu Developers <ubuntu-devel-discuss@lists.ubuntu.com>
Architecture: amd64
Uncompressed Size: 6870 k
Depends: libssl1.1 (= 1.1.0g-2ubuntu4.3)
Recommends: libssl-doc
Conflicts: libssl1.0-dev
Description: Secure Sockets Layer toolkit - development files
This package is part of the OpenSSL project's implementation of the SSL and TLS cryptographic protocols for secure communication over the Internet.

It contains development libraries, header files, and manpages for libssl and libcrypto.
Homepage: [url="https://www.openssl.org/"]https://www.openssl.org/[/url]

$ pip3 show p4python
Name: p4python
Version: 2018.2.1743033
Summary: P4Python - Python interface to Perforce API
Home-page: http://www.perforce.com
Author: Perforce Software Inc
Author-email: sknop@perforce.com
License: LICENSE.txt
Location: /home/.../.local/lib/python3.6/site-packages
Requires:
Required-by:
 
$ python3 -c '
>try:
>	from P4 import P4
>except Exception as e:
>	print(e)'
/home/.../.local/lib/python3.6/site-packages/P4API.cpython-36m-x86_64-linux-gnu.so: undefined symbol: SSL_library_init



#17239 Allow user to set whitespace preferences globally

Posted matthauck on 22 April 2015 - 06:40 PM

Currently you can only ignore whitespace on a per-file setting even within a given review. It would be great to be able to set it per review, and have that default to a users' preference.


#16573 Almost impossible to answer "Registration Question" needed to create...

Posted dragon on 20 February 2015 - 05:36 AM

I was trying to answer the "Registration Question" in order to create a new account on these forums and found it extremely frustrating.  The question asked is "Name a Perforce command line command (all lower case)".  So I tried:
p4
p4merge
p4 merge
p4 help add
p4 set
p4v
perforce
P4Merge (even though it says all lower case)
P4

And probably a few others.  Each one gave me the error "Your answer to the challenge question was not valid. Please try again."  Except I wasn't sure if that cryptic error was referring to the Perforce command question or the additional security question that asks you to type numbers contained in an image.  I intentionally got the numbers wrong and got a different error, so it was definitely complaining about the command.

Finally I tried "set" on the off chance it wanted an argument to the p4 command, not including the p4.  And miraculously, it accepted that.  But someone needs to make that question and possible answers a heck of a lot more clear because I was very close to giving up on registering.


#15255 Feature Request: Syntax highlighting in diff view

Posted dominik on 16 September 2014 - 11:47 AM

I would like to have the color coded syntax highlighting that I have in the file view also in the diff view. It would be tremendous helpful when doing the review and writing comments.

Dominik


#18087 Setup Swarm projects via Active Directory groups?

Posted Andrew DeFaria on 11 July 2015 - 01:11 AM

Is it possible to utilize Active Directory groups to setup Swarm projects instead of having to input each and every project member into each project?


#16883 Feature Requests

Posted jerome on 18 March 2015 - 01:15 PM

Hi, we are also evaluating swarm and gathering feature requests from our developers.

Here's a list so far, I hope that it will help
  • Preventing approval of code review when a task is still opened, also re-open the review when a task is re-opened or created.
  • Reporting feature. This is critical for us, we are in a regulated environment and we are required to regularly produce a code review report of all submitted code (for a given job, for example) with all associated tasks.
    • Retrieve comments and tasks details through the REST API. Related to the above request, we started to work on a reporting tool and the REST API does not provide any hook to comments / tasks. Querying directly the perforce server appears to be kind of wrong for that.
  • Allow the developer to set a deadline for his review. When the deadline date approaches and the code review is still opened, a notification is sent.
  • Official support of Windows for swarm server. P4D is compatible with windows, we expect to have all the P4 ecosystem compatible as well.
  • Detach a changelist from a review. We have developers preferring post-commit reviews and they use the add-commit button to associate task-fixing changelists to the original review. When the wrong commit is select and added to a review, they are stuck.
  • Aggregate notifications. When a reviewer makes three comments in a row on the same file, three emails are sent out. That would be better to aggregate them by using a two-step comment creation (batch save) or some kind of periodic notification.
  • Delete and edit of comments, already mentioned in this thread
  • Syntax highlighting in diffs. The syntax highlighting is available when browsing files but disappears in diffs, can that be enabled ? That could be an option
  • Flag a review as important and be able to filter important reviews
  • The add commit button adds the submitted changelist as a new dot in the review timeline. This is quite confusing because it is different from a file update (re-shelving). That would be easier to spot if it was, for example another icon.
  • Clarify the voting feature. Developers approve or reject reviews, why would they need to up-vote or down-vote as well? This adds complexity and people are reluctant to have multiple required reviewers because they have trouble understanding it. Maybe having a configuration option to either go all up-vote (majority and required up-vote = approve) or all approve button would make it simpler.
  • A real TODO list so the user knows what is expected from him. Already mentioned in this thread. We can partially get that from the review tab by playing on filters but that ought to be more convenient and that does not take into account tasks.
  • Review comments not aligned with the code when a file is updated. Can the review comment or task be shown on the left pane of the diff to reduce this occurrence?
  • Threaded comments
And cosmetics or less important in our eyes:
  • Possibility to select multiple lines of code for a comment
  • Group review. Being able to assign a group of people to a review
  • Analytics reports. For example have knowledge of how many tasks are still open by project, review load by user or changelists that are not reviewed.
  • Ping a reviewer, similar to the deadline feature request.
  • We would love code navigation in swarm....
  • [Added 05/22] Attach files of documentation to a review, there is a documented way to do that but that doesn't seem to be doable from the web client
  • [Added 05/22] Diff between two non consecutive shelves, we can achieve that with /reviews/{id}/v{x},{y} but that it is not intuitive at all



#15853 Feature Requests

Posted adrianstone on 17 November 2014 - 11:03 PM

We've been evaluating Swarm for about a month now and it is a great product but we have a few requests for how it could be improved:
  • In the reviews screen, there are many filters available, but it lacks the one filter that we'd find most useful.  There is no way to see only reviews that you are part of that you haven't voted up.  We often have lots of reviews in progress, and what people want to see is all the reviews they are part of that they haven't yet completed.  Even without the filter, there is no way to scan the list and see without clicking on each review whether it is still awaiting a vote from you.
  • It would be nice to have filters on the comments view.  We'd like filters for comments with unverified tasks, comments by the reviewer, and comments by the reviewee.
  • It would be nice to edit comments.  It drives people mad when they mistype a comment and can't correct it.
  • For post-commit reviews, it would be nice if adding committed changelists to the review expanded the list of files in the review rather than replacing it.  If you check in 3 files and start a post-commit review, then fix one of those files in response to comments and check that in, the review should show the modifications to the one file but not lose the changes to the other two files.

Thanks!
Adrian


#11690 Checkpoint replay to PostgreSQL database for Insight

Posted Punith on 24 November 2013 - 08:57 AM

Hi,


We have a p4d checkpoint of size 40Gig, while replaying it to PostgresSQL database, its taking days to complete. With 30% of progress it consumed 500Gig of space.

Do you guys recommend any configuration for PostgresSQL to improve the checkpoint replay time?
Also wondering when will Insight start supporting database other than PostgresSQL?


Thanks!
Punith


#11602 Issues With Logging In

Posted spectre1989 on 22 November 2013 - 05:37 PM

One of my colleagues is having some issues with logging in. He commits stuff to his sandbox, but when he comes to copy up, he gets error messages saying that he isn't logged in. He's not prompted with a log in box or anything, the only way we've found to authenticate him is to fire up p4admin have him log in to the remote depot from there.

On a similar note, I have a similar problem - I'm prompted with a log in box when I copy up. It doesn't matter how many times I put my password in, it won't accept it. I have to fire up p4admin (for which I'm not prompted to log in for some reason), I then have to choose to change my password, where I actually keep it the same, but the process of doing it seems to authenticate me.

Help with either of these problems would be awesome :D

Cheers
Joe


#11566 GitHub to Perforce migraton - Help requested

Posted ckenkare on 21 November 2013 - 07:17 PM

Folks,

As part of merging projects, We have a requirement to import files from GITHub to Perforce (details below).

Also, please note that P4D is currently integrated with bugzilla via p4dti.
1
P4D: Server
P4D/LINUX26X86_64/2008.2/179173 (2008/12/05)- Broadview Netwokrs 24 Users
-
2
P4V: Visual Client
P4/NTX86/2008.2/179173 (2008/12/05)
-
3
P4DTI
p4dti-2.4.2-1
-
4
Bugzilla
bugzilla-3.0.8
I have never done this kind of migration before.  Kindly suggest me how it can easily achieved. Also, some suggested "GIT Fusion" - But I'm not sure how it's going to work with defect tracking component (bugzilla/p4dti with versions mentioned above).  Thanks in advance..


#11510 Unexpected "all revision already integrated" from p4 integrate using...

Posted Oskar Senft on 20 November 2013 - 04:29 PM

Hi everyone

I'm using a branchspec that maps two different source folders to the same target folder. What happened is that the files in the two folders have been moved manually into the same folder and then just added to P4 without maintaining integration history.

Here's the branchspec:
//source/MDMTests/PCK24_FileLoader/... //target/SystemTests/PCK24_FileLoader/...
"//source/MDMTests/Product Maintenance Tests/..." //target/SystemTests/...

There are changes in the //source/MDMTests/PCK24_FileLoader/... location that have not yet been integrated to the target. However, there are now changes in "//source/MDMTests/Product Maintenance Tests/..." that need integrating.

If I run
p4 integrate -b branchspec
I'm getting
All revision(s) already integrated

However, if I remove the second line from the branchspec, it does integrate the files as expected.

As far as I can tell, since the second mapping from the branchspec yields no files that need integration, P4 seems to discard the first mapping from the branchspec.

Is this correct behavior? Did anyone come across that?

Thanks
Oskar.


#10706 Interested in installing Perforce products from RPM or deb packages? Quick su...

Posted Don Marti on 15 August 2013 - 09:53 PM

We're in the process of creating RPM and deb packages for installing Perforce products. IMHO this is going to be a time saver, especially when installing something with dependencies such as Git Fusion or Swarm.

If you're interested in installing Perforce products using your system package manager (doing something like "yum install perforce-git-fusion") could you please take a minute to answer a brief survey on how you work with packages? All questions optional, all feedback appreciated.

https://www.surveymonkey.com/s/6WFDQ9R


#25703 config: startup pull batch logic

Posted Matt Janulewicz on 05 December 2019 - 02:19 AM

I think that was pretty much my experience when testing this, but you posit an interesting question I didn't think of. When I tested I had five 'big file' queues and five 'small file' queues, but I had the batch set to 10 on all of them.

What I observed is that if you had 100 large files at the start of the queue, and 1 small file at the end, the small file would not get into the pull threads until all 100 other files were at least in the queue. It still felt/smelled like the pull threads were only looking at the top of the total queue and there was no kind of smart management/distribution going on.

Of course I have all kinds of ideas on how to improve this. :) One thorn in my side is that if rdb.lbr gets big (not sure how big, seems to vary) you'll see big stalls in the queues, and getting more files into the queue takes a long time. I don't think rdb.lbr is efficient for really big transactions (for instance, every so often I build a server pedantically with 'p4 verify -qzt', but I have to break it down into small chunks.) In our configuration, five batches of a million files transfer way, way faster than one batch of 5 million files. And it doesn't hurt to remove rdb.lbr and restart the server in-between chunks.

Anyway, it would be neat if rdb.lbr was a first-class db file. Or maybe if it were managed through p4 keys. Or maybe if you had max/min/batch set, that it somehow marks the file for the target pull thread when it goes into the queue, not when it's being popped out of it.


#24707 Perforce change-submit trigger to run script on client

Posted Matt Janulewicz on 17 April 2019 - 01:16 AM

Howdy!

Not to be a buzzkill (I'm going to be a buzzkill), but if I'm reading this correctly, the first three bullet points are covered by Perforce anyway, assuming folks are using it the way it is intended.

* files in a changelist must have some sort of action associated with them to get there in the first place
* Perforce can't/won't submit a file that's missing locally, it's got to have some content to submit
* a new file would have to be opened for add to be in the changelist

The last one is the only one that is sort of a thorn in my side. That can be controlled at the client level (submit unchanged files toggle.) I've given up on trying to enforce 'no unchanged revisions' because ultimately that's been a losing battle for me over the years. I've resigned myself to accepting that there will be lots of references to revisions that don't actually have deltas. It might be worth investigating, someday, a form-out trigger that sets "SubmitOptions: leaveunchanged" by default on new clients so people that are not paying enough attention will be less likely to submit empty revisions.

You implied that the script would want to run 'reconcile offline work', which reads to me that people are not using Perforce in the way it's intended to be used. If I were going down this road (again, I've been there) I would start by asking myself _why_ I'd want to do this kind of thing. Are people editing files locally without checking them out first? Are they manually setting 'allwrite' in their clients? (which is not the default.) Is there a reason people are working around the tool instead of with it?

It might be one of my very rare type-A afflictions, but I'm always weary of architecting things that hide the functionality and nature of a tool from end users. It just allows them to continue bad habits. And in a cosmic sense, I don't won't to be responsible for sending employees off to new opportunities where they might get side-eye glances when they say they're proficient in Perforce usage when in fact they are not.

Sorry if that started to sound like a rant, I didn't mean it. :)  Over the years I've just been down this rabbit hole a few times and it has never ended well. Too many weird things you have to do to possibly get Perforce to do things on the client side, where most of the time user training (sometimes with a very large stick) will fix the problem in the long term.


#24658 Installing p4python on Fedora Core 29

Posted jaess on 04 April 2019 - 07:51 PM

Unfortunately your solution no longer works in Fedora 29.  Also, using compat-openssl isn't a great solution since it replaces the built in openssl packages and may break lots of things.

It would be a lot better if perforce would release a version that works with the current release of OpenSSL.