Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
    GautierMichelin
    @gautiermichelin
    tty soon, I do have some pull request ideas and really would have to improve the french translation.
    Kehan Harman
    @kehh
    That sounds great. Some good looking projects you've got under your belt anyway.
    GautierMichelin
    @gautiermichelin
    thanks, musee.sacem.fr was a tough work, but really interesting and challenging. I've loved digging inside som belgium private archives too. Difficulty is here dealing with museum regulations and national requirements (with some 70s way of exporting data).
    Best to all of you, bye
    Karl Becker
    @karbecker
    Hi all, I was wondering if its possible to batch-export all media of a CA installation automnatically either by mapping of using caUtils? I’ve looked into caUtils but couldn’t find anything that exports media. This would be helpful if one would have to sync to installations not only in terms of metadata but also to keep the images in sync. thanks for your help 😊
    CollectiveAccess
    @collectiveaccess
    @karbecker No this is not possible currently.
    Make a JIRA for it an describe how you think it should work, and we'll consider implementing it
    @gautiermichelin It has been possible to scrape tilepics since they were first spec'ed back ~2000
    I remember seeing some Delphi(!) code to do that.
    If you want watermarking for tilepics we can add that
    michi-gato
    @michi-gato
    Hello- Anyone able to recommend cloud-hosting best practices? AWS, Google Cloud, DigitalOcean, etc.? Pawtucket is on our local servers and we're looking for affordable, stable options that we should look into. Thanks!
    Sean A.
    @ColonelSanderson
    It's generally much for muchness, especially when it comes to AWS/Google cloud/Microsoft/IBM.
    Both in terms of cost and stability.
    We use AWS EC2, if that helps. Downtime is practically non-existent and it's relatively easy to automate deployment.
    michi-gato
    @michi-gato
    Good to note. Thanks @stralopple
    CollectiveAccess
    @collectiveaccess
    @michi-gato We use GoogleCloud and are happy. It's not cheap though.
    GautierMichelin
    @gautiermichelin
    Hi @collectiveaccess and thanks, already done that on musee.sacem.fr, you can look at https://musee.sacem.fr/index.php/Detail/objects/7290, click on the object. Look at the icons on the left, they do move for an offset (1, 2, 3) the watermark on the tilepic havoiding it to fall too badly.
    I did myself have a peek on how to reverse tilepics in the past, problem was there to have to step-by-step process, with URL copy-paste to download the best resolution version. That was shaggy for some museums here. I can commit this tilepic version if you want. It uses gmagick to add the watermark depending on the tile number (modding 3).
    GautierMichelin
    @gautiermichelin
    Hum, just a peek, I do have a request for an undelete plugin here (some Belgian users are really enclines to play the board game BANG! with their records). Idea would be for admin to check deleted records with their idno, and click an undelete button.
    I do think that could be of some use within the community, just a question, do you prefer a pull request for this really really simple plugin in the dev branch, or do you think it's enough to have a distinct repository ?
    GautierMichelin
    @gautiermichelin
    @stralopple I do use OVH infrastructure (French company), 3 levels of dedicated servers, first usable for CA comes at 16.99$ but you'll have 2TB for data storage. Those are dedicated physical machines so have your backups, hard disks have a nasty habit to fail every 3 or 4 years.
    you can check here for CAN hosting for example : https://www.kimsufi.com/us/en/servers.xml
    CollectiveAccess
    @collectiveaccess
    @gautiermichelin It should be done as a pull request
    It's probably not a plugin. It seems like core functionality.
    michi-gato
    @michi-gato
    @collectiveaccess Thanks!
    GautierMichelin
    @gautiermichelin
    @collectiveaccess I imagine it inside Manage > Administration, ok for you ?
    does anyone have a clue on how to lower the indexing weight of representations pdf text content ? I've pushed up the BOOST limits inside search_indexing but that does not to be the right way.
    CollectiveAccess
    @collectiveaccess
    @gautiermichelin You'd need to set a boost level for media_content
    naeemmuhammad
    @naeemmuhammad
    Hi, Is it possible to put put pdf exports into background process? If yes, how can we configure it? We have to download a large number of records as a pdf but execution is timed-out, perhaps executing them in the background can solve this problem.
    Kehan Harman
    @kehh
    @naeemmuhammad experiment with using the wkhtmltopdf method of generating PDFs - we've found it to be many times faster than DOMPDF (and also supports modern HTML5 / CSS3). Then maybe you won't have to set the timeout for the download?
    naeemmuhammad
    @naeemmuhammad
    @kehh we are already using wkhtmltopdf, however it only reduces the time it takes to generate the pdf itself. Indeed Dompdf took a large time to generate a pdf as compared to wkhtmltopdf tool. Regardless of the time taken by pdf generation tool, the time it takes to create input file (a html file) for pdf generation tool is quite high. A search result of about 800 record for example times out with wkhtmltopdf. And it fails before it passes html file to wkhtmltopdf tool.
    CollectiveAccess
    @collectiveaccess
    @naeemmuhammad No you can't background them. It's possible to do, but we have not implemented it.
    naeemmuhammad
    @naeemmuhammad
    @collectiveaccess Ok, can we request (in Jira) that as a new feature for upcoming versions?
    CollectiveAccess
    @collectiveaccess
    @naeemmuhammad Yes of course
    Sean A.
    @ColonelSanderson
    @gautiermichelin As I recall the big PaaS providers are on par, possibly a little more expensive than a virtual server. As you mention, you have to do your own risk assessment on their infrastructure and where I work, we extensively use the other products on the platform (storage, serverless compute, db, notifications, etc)
    so it's cost efficient in that sense.
    Kehan Harman
    @kehh
    @collectiveaccess when using movement / workflow based location tracking is it possible to show the contents of a location using a display template? I'm imagining a <unit/> tag with an isCurrent?
    Current config is:
    # Location tracking options
    object_storage_location_tracking_relationship_type = storage_loc
    movement_storage_location_tracking_relationship_type = moved_to
    movement_object_tracking_relationship_type = part
    record_movement_information_when_moving_storage_location = 1
    movement_storage_location_date_element = removal_date
    
    current_location_criteria = {
        ca_movements = {
            movement = { 
                date = pickup_date,
                template = ^ca_storage_locations.hierarchy.preferred_labels%delimiter=_➜_ 
            }
        }
    }
    Julia
    @juliaweist
    @kehh Yes, the current location template accepts display syntax such as unit tags. What are you imagining with isCurrent?
    Kehan Harman
    @kehh
    @juliaweist current_location bundle works well from the objects side - I'm talking about from the storage location side finding out which objects are currently at that particular storage location.
    Julia
    @juliaweist
    @kehh have you tried the UI bundle setting showCurrentOnly (https://docs.collectiveaccess.org/wiki/Bundles#User_interface_Settings)? You can set a display template as you would normally for a UI element along with that setting.
    sheggen
    @sheggen
    Inside of the Dimensions metadata element, I created a new container, and inside the container created three text field subelements. I want to delete them all now because they aren't used. The server returns a 500 error, likely after it times out waiting for mysql. I'm not seeing anything helpful in my log files for apache, mysql, or CA. Any suggestions where else to look?
    sheggen
    @sheggen
    Here's the only log that points to the cause, in apache's error.log: PHP Fatal error: Maximum execution time of 30 seconds exceeded in /var/www/html/providence/app/lib/core/Configuration.php on line 146, referer: http://localhost/providence/index.php/administrate/setup/Elements/Delete/parent_id/45/element_id/47
    Kehan Harman
    @kehh
    @sheggen you can increase the execution time to a much higher number temporarily if you edit your php ini file (depends on the operating system where this is set. It sounds like you might have some values associated with these attributes. If that doesn't work then manually deleting the values from ca_attribute_values where the element_id matches the element_id of these elements might be able to help you work around this problem.
    sheggen
    @sheggen
    Thanks @kehh . Turns out the query to delete was trying to hit 40k records, and it was doing exactly what it said; timing out. updating the php.ini and my.cnf temporarily resolved the issue.
    Tim Duckworth
    @EdgarCayceARE_gitlab
    Hi all, I recently installed CA, when I load up the login page I get Warning: sizeof(): Parameter must be an array or an object that implements Countable in C:\xampp\htdocs\app\lib\core\BaseObject.php on line 90
    When I log in, I get numerous errors and skewed visuals, did I miss something?
    CollectiveAccess
    @collectiveaccess
    @EdgarCayceARE_gitlab You're using PHP 7.2 I gather?
    The current release of CA doesn't full support it. We're working on a new release that does
    Tim Duckworth
    @EdgarCayceARE_gitlab
    Yes,