What Is Application Octet Stream
I don't have a specific response to this 2 years later, but I did want to chime in and mention that I've seen this too. We are working with an org that has lots of legacy customizations. I pulled down all the metadata and converted to DX format for easier browsing. But about half the ZIP static resources did not auto-unzip in the conversion because the platform considered them application/octet-stream instead of zip files. The ZIPs still work perfectly well as static resources.
But SFDX gets confused. Going forward, probably the best way to generate resource ZIPs will just be to work with them in DX which auto-handles the zip/unzipping on deployment. It also means you can now source-track the individual files of the ZIP as part of your release management.
Hi Chris, I've managed to reproduce the problem, it seems that the Flash module sends the uploaded files with hard-coded content type which is application/octet-stream. I've logged the issue into our bug tracking system and we will try to get it resolved as soon as possible. I've also updated your telerik points as a token of gratitude for the report. A possible workaround might be to manually set the ContentType based on the extension of the file.
A list with the most common content types can be found Regards, Genady Sergeev the Telerik team. Hi guys, It turned out that it is not that trivial to implement such feature in the Flash/Silverlight modules. The problem is that there is no way to force Flash/Silverlight to determine the content type based on the meta data of the file. The only way we found to do it, was to scan the selected file name, and match its extension against a list of a known mime types.
While this is a good for a workaround, we haven't yet implemented it, but the task is in progress. It should be ready for the upcoming SP2 release. However, If the issue is urgent for you, I can prepare a workaround that you can use immediately. Regards, Genady Sergeev the Telerik team. Thanks for keeping us informed, much appreciated. For my project(s), this is essential for implementing versioning.
Currently, we have told our clients that it is a known issue and just have versioning 'off' until it is fixed. My main issue is that items can be uploaded using the single uploader in the file explorer or a bulk uploader based on the asyncUpload. The single uploader works as expected. I took a stab myself at some process for reading the extension of the file and determining the file type based on some list I found on the internet. I was through the major types (office products and pdf) and then compared against the data gathered through the single uploader by Telerik. The types were very different which does not solve the disparity between the two upload methods.
Again, the users can upload now, but as our database grows it will become more difficult for us to implement versioning. For me (don't know about other users), I can wait on the service pack. I can also code up myself an extension checker if given a list of what the Telerik File Explorer uses to populate the file type field so there isn't the disparity between the two upload methods. Again, thanks for staying on top of this. Basically, the workaround you have described was what I used to start with when I reported the problem in August. I have a SQL Server table that contains a list of file extensions. Once the file is uploaded, we grab the extension and look it up in the SQL table.
This gives us the flexibility to add new extensions at any time. We also save a FK reference to the extension table in the Attachment (where we save the File) table inside of SQL. If a file extension is not found in the table, it sets the FK reference of the attachment record to the 'unknown' extension.
This way we can query and see if any 'new' file extensions have been uploaded to that we don't have in the table. Then a update can be done once we've added the extension. Here is a link to an excel file with the file extensions we have in the table.
What Is Mediatype.application_octet_stream
Mime Type Application Octet Stream
Thanks for the reply, some issues with the list: It has several mime type for common extensions. For instance, for excel spreadsheets the supplied list has:.xls application/excel.xls application/vnd.ms-excel.xls application/x-excel.xls application/x-msexcel So, there is no guarantee that if we use a method to match to a list for multi upload that we will now match what was done through the single uploader. I am going to try something now that could introduce a small margin of error possibility. Currently I check MIMe type, folder location, and file name to determine if a file is duplicate.
I will remove MIME type checking so it is done ONLY on location and file name. Will report back if this does it for me.