Reliability of metadata in file uri

[titusnachbauer]

Hi there,

I am doing an analysis of a client's Alfresco system. I posted about this before. The problem is that they lost admin access to the system, so we cannot use any API or the node browser. We only have direct access to the database. I was able to extract a lot of information from that. It checks out completely with the frontend so everybody is happy. One of the fields is the content URL, which I get from the ALF_NODE_PROPERTIES table. It looks like this:

contentUrl=store://2010/5/18/9/46/f1f58d86-715f-4d34-a00d-641476bdaeb1.bin|mimetype=application/vnd.powerpoint|size=447488|encoding=UTF-8|locale=en_US_

Two questions: 1. Are the size and MIMEtype really realiable? I am assuming the file is frozen by Alfresco, so the answer would be yes. 2. Is knowing the store and the original name of the file enough to retrieve it? In other words, if the API is unavailable, can I just download the binary from this URL and then rename it and be done. I will try that of course, but it would be nice to know of any risks involved.

[mrogers]

"The problem is that they lost admin access to the system."   Well then the solution is to fix up whatever is broken.   Can you not create a temporary adminstrator?

mimetype and size are "reliable".

You also need to take account of versions.   There will potentially be more than one version per file.

[titusnachbauer]

Thanks! The problem with the admin access is a bit complicated, this being a very large organisation. I cannot go into the details, but rest assured that I tried to fix that first of course 😉

We are lucky that they made almost no use of versioning. I will tackle that problem when we are going to move the files.