Recoll Tips

If you have read this far, you already know enough Recoll to use it productively, but why stop there? Following are some tips to help you enter "Recoll brainstorming mode" and go even further.

First, keep your files in good order. The most efficient search engine is one that can work in a well ordered directory tree, on files with good metadata and names that are meaningful, but also portable and future-proof: As the Recoll manual says, "non-ASCII UNIX file paths are an unending source of trouble and are best avoided." Basically, if Recoll works well, it means that your documents are also easy to back up or process with any other tool. In fact, Recoll can help you put your files in order: You may, for example, ask Recoll to list all the files about backpacking that are not in your backpacking folder, just so you can move them there.

Second, restrict the indexed area as much as you can, while taking advantage of multiple indexes. Ideally, documents that you need to keep for whatever reason, but that you will never need to search, should go in a separate folder, excluded from any Recoll indexing: Should you need to search these files some day, you can always generate a new index on the spot. If all the files you care about indexing are in your $HOME/Documents/work folder, set that folder as the top directory, instead of the default $HOME, in the Preferences | Indexing configuration panel. If you need to search your work files every day, but your personal files maybe once a month, create separate indexes for $HOME/Documents/work and $HOME/Documents/personal, with different update schedules for each index, and then use the work index as your default index, using or adding the other only when necessary.

Third, spend some time playing with search queries in the desktop GUI, both to save the best ones to reuse (via the File menu) and to learn how to write good ones for the command-line interface. Alas, the GUI saves searches (in the configuration folder's savedqueries subfolder) in an XML format not usable on the command line by recollq. However, you can learn how the queries are expanded and passed to Xapian by clicking on Show Query at the top of the results list. Another great way to master the Recoll query language is comparing the results obtained in the GUI with those returned by recollq.

Fourth, remember that you can use Recoll also as a semi-automatic, general-purpose analyzer of all of your documents, in different ways, and for many different purposes. As just one example, running recollq with the -b switch outputs only the names of the files matching the current query. Therefore, you can write a shell script that runs that command and, if it returns at least one line, notifies you by email that somebody added a document with that string inside. Other recollq options print out only the abstracts (-A) or metadata (-m) of the files they find. Also available are excellent options to import all the documents you want inside any database, as well as text analysis software. Even with the desktop GUI, you can click the button in the top bar that displays the search results as a table; the Results menu will let you save the same results as a spreadsheet in CSV format.

The Author

Marco Fioretti is a freelance author, trainer, and researcher based in Rome, Italy, who has been working with Free and Open Source software since 1995 and on open digital standards since 2005. Marco also is a board member of the Free Knowledge Institute (http://freeknowledge.eu).