Shrine clear cached images on persistence success - shrine

Background
I am using file system storage, with the Shrine::Attachment module in a model setting (my_model), with activerecord (Rails). I am also using it in a direct upload scenario, therefore i need the response from the file upload (save to cache).
my_model.rb
class MyModel < ApplicationRecord
include ImageUploader::Attachment(:image) # adds an `image` virtual attribute
omitted relations & code...
end
my_controller.rb
def create
#my_model = MyModel.new(my_model_params)
# currently creating derivatives & persisting all in one go
#my_model.image_derivatives! if #my_model.image
if #my_model.save
render json: { success: "MyModel created successfully!" }
else
#errors = #my_model.errors.messages
render 'errors', status: :unprocessable_entity
end
Goal
Ideally i want to clear only the cached file(s) I currently have hold of in my create controller as soon as they have been persisted (the derivatives and original file) to permanent storage.
What the best way is to do this for scenario A: synchronous & scenario B: asynchronous?
What i have considered/tried
After reading through the docs i have noticed 3 possible ways of clearing cached images:
1. Run a rake task to clear cached images.
I really don't like this as i believe the cached files should be cleaned once the file has been persisted and not left as an admin task (cron job) that cant be tested with an image persistence spec
# FileSystem storage
file_system = Shrine.storages[:cache]
file_system.clear! { |path| path.mtime < Time.now - 7*24*60*60 } # delete files older than 1 week
2. Run Shrine.storages[:cache] in an after block
Is this only for background jobs?
attacher.atomic_persist do |reloaded_attacher|
# run code after attachment change check but before persistence
end
3. Move the cache file to permanent storage
I dont think I can use this as my direct upload occurs in two distinct parts: 1, immediately upload the attached file to a cached store then 2, save it to the newly created record.
plugin :upload_options, cache: { move: true }, store: { move: true }
Are there better ways of clearing promoted images from cache for my needs?

Synchronous solution for single image upload case:
def create
#my_model = MyModel.new(my_model_params)
image_attacher = #my_model.image_attacher
image_attacher.create_derivatives # Create different sized images
image_cache_id = image_attacher.file.id # save image cache file id as it will be lost in the next step
image_attacher.record.save(validate: true) # Promote original file to permanent storage
Shrine.storages[:cache].delete(image_cache_id) # Only clear cached image that was used to create derivatives (if other images are being processed and are cached we dont want to blow them away)
end

Related

How To Upload A Large File (>6MB) To SalesForce Through A Lightning Component Using Apex Aura Methods

I am aiming to take a file a user attaches through an Lightning Component and create a document object containing the data.
So far I have overcome the request size limits by chunking the data being uploaded into 1MB chunks. When the Apex Aura method receives these chunks of data it will either create a new document (if it is the first chunk), or will retrieve the existing document and add the new chunk to the end.
Data is received Base64 encoded, and then decoded server-side.
As the document data is stored as a Blob, the original file contents will be read as a String, and then appended with the chunk received. The new contents are then converted back into a Blob to be stored within the ContentVersion object.
The problem I'm having is that strings in Apex have a maximum length of 6,000,000 or so. Whenever the file size exceeds 6MB, this limit is hit during the concatenation, and will cause the file upload to halt.
I have attempted to avoid this limit by converting the Blob to a String only when necessary for the concatenation (as suggested here https://developer.salesforce.com/forums/?id=906F00000008w9hIAA) but this hasn't worked. I'm guessing it was patched because it's still technically allocating a string larger then the limit.
Code's really simple when appending so far:
ContentVersion originalDocument = [SELECT Id, VersionData FROM ContentVersion WHERE Id =: <existing_file_id> LIMIT 1];
Blob originalData = originalDocument.VersionData;
Blob appendedData = EncodingUtil.base64Decode(<base_64_data_input>);
Blob newData = Blob.valueOf(originalData.toString() + appendedData.toString());
originalDocument.VersionData = newData;
You will have hard time with it.
You could try offloading the concatenation to asynchronous process (#future/Queueable/Schedulable/Batchable), they'll have 12MB RAM instead of 6. Could buy you some time.
You could try cheating by embedding an iframe (Visualforce or lightning:container tag? Or maybe a "canvas app") that would grab your file and do some manual JavaScript magic calling normal REST API for document upload: https://developer.salesforce.com/docs/atlas.en-us.api_rest.meta/api_rest/dome_sobject_insert_update_blob.htm (last code snippet is about multiple documents). Maybe jsforce?
Can you upload it somewhere else (SharePoint? Heroku?) and have that system call into SF to push them (no Apex = no heap size limit). Or even look "Files Connect" up.
Can you send an email with attachments? Crude but if you write custom Email-to-Case handler class you'll have 36 MB of RAM.
You wrote "we needed multiple files to be uploaded and the multi-file-upload component provided doesn't support all extensions". That may be caused by these:
In Experience Builder sites, the file size limits and types allowed follow the settings determined by site file moderation.
lightning-file-upload doesn't support uploading multiple files at once on Android devices.
if the Don't allow HTML uploads as attachments or document records security setting is enabled for your organization, the file uploader cannot be used to upload files with the following file extensions: .htm, .html, .htt, .htx, .mhtm, .mhtml, .shtm, .shtml, .acgi, .svg.

Can Fiddler .SAZ capture files be combined?

Fiddler has an autosave feature which unfortunately clears the captured sessions each time it saves to an .SAZ. Rather than have a folder of Fiddler save sessions (.SAZ's), I'd prefer to have one master .SAZ, albeit saved at regular intervals.
Since there doesn't appear to be an option in Fiddler to do this, is there a way to combine or merge .SAZ files?
There are two possibilities:
Use the Fiddler UI: When you execute the menu command "Load Archive..." you can append the data from the loaded SAZ file to the current session list. Therefore it is possible to load multiple SAZ files and them save them into a new SAZ file.
Use Fiddler Core: Using Fiddler Core you can develop an own .net program that merges multiple SAZ files into one new SAZ file. The methods of loading and saving sessions is pretty simple:
using Fiddler;
Session[] oLoaded1 = Utilities.ReadSessionArchive("FileToLoad1.saz", false);
Session[] oLoaded2 = Utilities.ReadSessionArchive("FileToLoad2.saz", false);
Session[] oAllSessions = ... //<merge the two arrays>
Utilities.WriteSessionArchive("Combined.saz", oAllSessions, null, false);
Sources:
http://fiddler.wikidot.com/fiddlercore-demo
Utilities.ReadSessionArchive
Utilities.WriteSessionArchive

What are the best practices for pre-fetching in backbone.js?

What's the best way to pre-fetch the next page in backbone.js?
Is there a build-in mechanism to do that, or do I have to take care of it myself by making Ajax calls and storing the results.
Also, is there a way to preload the entire page like in JQuery mobile( http://jquerymobile.com/demos/1.2.0/docs/pages/page-cache.html)
There is no built in support for a such a thing. It's dependent on your use case but you could do a number of things.
1) Use setTime() to wait a short time before fetching the data you might be needing shortly. (Probably not a good solution)
2) Set up an event handler to fetch the data on a specific event, or something similar:
$('#my-button').on('hover',function() {
//fetch data
});
To fetch the data you can use the fetch() function on a backbone model or collection, which will return a jqXHR (or you can use a straight $.ajax() call). You can then wait and see if it failed or passed:
var fetch = myModel.fetch();
fetch.done(function(data) {
// Do something with data, maybe store it in a variable than you can later use
})
.fail(function(jqXHR) {
// Handle the failed ajax call
// Use the jqXHR to get the response text and/or response status to do something useful
});
No build-in support, but actually easy to add. Please referer to the concept of View Manager, it is able to handle both "view-keeping" tasks and transitions.
In short, the concept is: view manager is component, which is reposible for switching from one application view to another. It will dispose existing view, so it prevents zombies and memory leaks. As well it could handle transitions between view switch.
Here my way how I handle the loading of pages into an "endless scrolling" list.
Make you backend paginate aware
First of all you require a DB backend which is capable of handlling page load requests:
As an example refer to my git modelloader project which provides a small Coffee based framework integrated into a Node.js/Moongoose environment.
Model Loader on GIT will contain additional information and samples.
Here the most important points:
You backend should support the following Pagination features
Each request will return partial response only limiting it to for example 20 records (size of a page).
By default the last JSON record entry returned by a request, will contain additional technical and meta information about the request, necessary for allowing consumers to implement a paging.
{
_maxRec: "3",
_limit: "20",
_offset: "0"
}
_maxRec will list the total amount of records in the collection
_limit will list the maximum number of requests which are given back
_offset will tell you which set of records was passed back, i.e. an _offset of 200 would mean that result list skipped the first 199 records and presents the records from 200-220
The backend should support the following pagination control parameters:
http(s)://<yourDomainName>/<versionId</<collection>?offset=<number>
Use offset to skip a number of records, as for example with a limit of 20, you would send a first request with offset=0 then offset=20, then offset=40 etc. until you reached _maxRec.
In order to reduce the db activities you should provide a possiblity to reduce the _maxRec calculation for subsequent requests:
http(s)://<yourDomainName>/<versionId</<collection>?maxRec=<number>
By passing in a maxRec parameter (normally the one gotten by an earlier paging requerst), the request handler will by pass the database count objects statement, which results in one db activity less (performance optimization). The passed in number will passed back via _maxRec entry. Normally a consumer will fetch in the first request the _maxRec number and pass it back for the subsequent request, resulting in a faster data access request.
Fire of Backbone Model requests if necessary
So now you have to implement on the Backbone side the firing of page loading requests when necessary.
In the example below we assume to have a Backbone.View which has a list loaded into a jquery.tinyscrollbar based HTML element. The list contains the first 20 records loaded via the URL when built up initially:
http(s)://<yourDomainName>/<versionId</<collection>?offset=0
The View would listen in this case to the following scrolling events
events:
'mousewheel': 'checkScroll'
'mousemove': 'checkScroll'
Goal is as soon the user has scrolled down to the bottom of the scrollable list (e.g. he reaches a point which is 30px above the scrollable list end point) a request will be fired to load the next 20 entries. The following code sample desrcribes the necessary step:
checkScroll: () =>
# We calculate the actual scroll point within the list
ttop = $(".thumb").css("top")
ttop = ttop.substring(0,ttop.length-2)
theight = $(".thumb").css("height")
theight = theight.substring(0,theight.length-2)
triggerPoint = 30 # 30px from the bottom
fh = parseFloat(ttop)+parseFloat(theight)+triggerPoint
# The if will turn true if the end of the scrollable list
# is below 30 pixel, as well as no other loading request is
# actually ongoing
if fh > #scrollbarheight && ! #isLoading && #endLessScrolling
# Make sure that during fetch phase no other request intercepts
#isLoading = true
log.debug "BaseListView.checkscroll "+ ttop + "/"+ theight + "/"+ fh
# So let's increase the offset by the limit
# BTW these two variables (limit, offset) will be retrieved
# and updated by the render method when it's getting back
# the response of the request (not shown here)
skipRec = #offset+#limit
# Now set the model URL and trigger the fetch
#model.url = #baseURL+"?offset="+skipRec+"&maxRec="+#maxRec
#model.fetch
update: true
remove: false
merge: false
add: true
success: (collection, response, options) =>
# Set isLoading to false, as well as
# the URL to the original one
#isLoading = false
#model.url = #baseURL
error: (collection, xhr, options) =>
#isLoading = false
#model.url = #baseURL
The render method of the view will get the response back and will update the scrollable list, which will grow in size and allows the user again to start scrolling down along the new loaded entries.
This will load nicely all the data in a paged manner.

Grails database migration plugin - Java heap space

I am running grails 1.3.7 and using the grails database migration plugin version database-migration-1.0
The problem I have is I have a migration change set. That is pulling blobs out of a table and writing them to disk. When running through this migration though I am running out of heap space. I was thinking I would need to flush and clear the session to free up some space however I am having difficulty getting access to the session from within the migration. BTW The reason it's in a migration is we are moving away from storing files in oracle and putting them on disk
I have tried
SessionFactoryUtils.getSession(sessionFactory, true)
I have also tried
SecurityRequestHolder.request.getSession(false) //request in null -> not surprising
changeSet(author: "userone", id: "saveFilesToDisk-1") {
grailsChange{
change{
def fileIds = sql.rows("""SELECT id FROM erp_file""")
for (row in fileIds) {
def erpFile = ErpFile.get(row.id)
erpFile.writeToDisk()
session.flush()
session.clear()
propertyInstanceMap.get().clear()
}
ConfigurationHolder.config.erp.ErpFile.persistenceMode = previousMode
}
}
}
Any help would be greatly appreciated.
The application context will be automatically available in your migration as ctx. You can get the session like this:
def session = ctx.sessionFactory.currentSession
To access session, you can use withSession closure like this:
Book.withSession { session ->
session.clear()
}
But, this may not be the reason why your app run out of heap space. If the data volume is large, then
def fileIds = sql.rows("""SELECT id FROM erp_file""")
for (row in fileIds) {
..........
}
will consume up your space. Try to process the data with pagination. Don't load all the data at once.

make a copy of an image in blobstore

I have an image in blob store which is uploaded by users(their profile pic). I want to make a copy of the same and and re-size the copy so that it can be displayed as a thumbnail. I want to make a copy of the same instead of using the ImageService because this would be used more often compared to the profile image.
What I am doing here is this:
reader = profile_image.open() #get binary data from blob
data = reader.read()
file_name = files.blobstore.create(mime_type=profile_image.content_type)#file to write to
with files.open(file_name, 'a') as f:
f.write(data)
files.finalize(file_name)
blob_key = files.blobstore.get_blob_key(file_name)
image = images.Image(blob_key = blob_key)
image.resize(width=32, height=32)
entity.small_profile_pic = <MyImageModel>(caption=<caption given by user>,
picture=str(blob_key))
This is giving me error:
BadValueError: Image instance must have a complete key before it can be stored as a reference.
I think this is because the blob is not saved(put()) into the datastore, but how do i do it. Doed files.blobstore.get_blob_key(file_name) not do it ?
I would also like to ask: does the blobstore also cache the dynamically transformed images images served using get_serving_url() ...
I would use the get_serving_url method. In the doc is stated that:
The get_serving_url() method allows you to generate a stable, dedicated URL for serving web-suitable image thumbnails. You simply store a single copy of your original image in Blobstore, and then request a high-performance per-image URL. This special URL can serve that image resized and/or cropped automatically, and serving from this URL does not incur any CPU or dynamic serving load on your application (though bandwidth is still charged as usual). Images are served with low latency from a highly optimized, cookieless infrastructure.
Also the code you posted doesn't seem to follow the exampled posted in the docs. I would use something like this
img = images.Image(blob_key=original_image_key)
img.resize(width=32, height=32)
thumbnail = img.execute_transforms(output_encoding=images.JPEG)
file_name = files.blobstore.create(mime_type='image/jpeg')#file to write to
with files.open(file_name, 'a') as f:
f.write(thumbnail)
files.finalize(file_name)
blob_key = files.blobstore.get_blob_key(file_name)

Resources