#16 Additional metadata for files & index unparsable files

[Anonymous]

I'm apologize for not being particularly familiar with underlying technologies and their particular limitations, but I have been playing with DocFetcher and think it would be very useful to be able to associate additional metadata with files without altering their content.

For instance, I have a bunch of PDF files I'd like to index (some protected). I might like to augment content with my own custom keywords without altering a file (you might not even have the option of altering protected files). This would permit combining full text search features with my one's own way of indexing and finding particular things of interest, with keywords like, e.g. ZZNB, ZZFISHING, ZZHOBBY, ZZARCHITECTURE, ZZWORK, ZZHOME, ZZOPENSOURCE, etc. or to simply add additional common synonym keywords to documents without altering them.

Having an option that would allow parsing and indexing filenames would be one way to work around this, but not a nice one.

A general solution would be to search for an extra file with known name in each directory being indexed (e.g. .index, .metadata, index.me, etc) that if found gets parsed or optionally parsed. One could associate extra keywords with files in the directory, whether the file was of known type or had parse errors. File could be a simple XML or custom format. This would allow DocFetcher to handled unparsable PDFs or files of unrecognized type in a limited way. You would want to have the option to allow files with parse errors to be added to the index to support this.

[Nam-Quang Tran]

What you're asking for is basically a file manager with built-in indexing and tagging. I'm afraid this goes way beyond the scope of this project (meaning that it would require rewriting some of the main parts of this program).

Anything we could do in this direction would be nothing but a suboptimal fix for the core of the problem: the flawed files-and-folders metaphor.