Any way to search files and create reading lists?

[MarjaE]

Hi,

No, I can't use Calibre for this.

I've found a few tools which help me search files by their content. Especially Houdahspot, but also Finder Search, File Search Machine, etc.

But I want to be able to take the resulting list, remove files which turn out not to be relevant, mark ones which turn out to be relevant, and create a usable reading list which I can go through over the next few days. Without having to remove the files from their original folders, and preferably without creating duplicate files in a project folder.

I am using a Mac.

[Karellen]

What are you actually searching for? Are you searching tags or authors or filenames or something else? Give an example.

[MarjaE]

Right now, I'm searching for files which contain the phrases "wrought iron" "steel" and either "figure of merit" or "Garzke".

This turns up 29 files, including some duplicates where I've processed a pdf and kept both versions, or excerpted part to read later, or imported a file into Calibre and kept the original.

[salamanderjuice]

Did you have a look at the Power Search plugin for Calibre that allows for searching the contents of files? That might do what you want.

You could also do this somewhat easily with shell commands and scripting in the terminal although how you do it depends on exactly what kinds of files you are searching and what kinds of content and how you have it organized. For example on my Ubuntu machine I can get a list of all my epubs in my Calibre library containing the word "goose" using

 find . -name "*.epub" -exec ../mysearch.sh "goose" {} \;

The find command finds all the .epub files within the current folder (that's the "." after find) and then on each of those file names it runs the mysearch script that I put one folder up with arguments "goose" (what I'm searching for) and {} which represents the current file path from find and then \; tells it there's no more arguments.

mysearch.sh is just a simple wrapper around the utility "zipgrep" which searches inside zip files which epubs are. This is needed because otherwise it won't report the epub file name but the name of the file in the epub which isn't helpful so the wrapper returns the file name if there has been a match.

mysearch.sh:

#!/bin/bash

zipgrep "$1" "$2"
if [ $? -eq 0 ]; then
    echo "$2"
fi

It's not quick I'll admit though. And the above may not work on Mac since they have some command differences.

[MarjaE]

No, I haven't looked at Powersearch. It might help within Calibre, but I doubt it would help with the books I haven't imported because I want to know whether it'll be worth the hassle. It takes time to pre-process the pdf files, in whole or in part, import them, type up the metadata, etc.

Do any of these build a possible-reading list which I can go through and check off?

Thanks.

P.S. Another possibility would be a search tool which can search for separate words, so, for example, a search for "plates" will not return "contemplates," and which can show results where the terms occur within siome reasonable distance of each other, and where they don't overlap, so that 1894 doesn't count as a hit for both "89" and "94."

[salamanderjuice]

Originally Posted by MarjaE

No, I haven't looked at Powersearch. It might help within Calibre, but I doubt it would help with the books I haven't imported because I want to know whether it'll be worth the hassle. It takes time to pre-process the pdf files, in whole or in part, import them, type up the metadata, etc.

Do any of these build a possible-reading list which I can go through and check off?

Thanks.

P.S. Another possibility would be a search tool which can search for separate words, so, for example, a search for "plates" will not return "contemplates," and which can show results where the terms occur within siome reasonable distance of each other, and where they don't overlap, so that 1894 doesn't count as a hit for both "89" and "94."

You can do that with regular expressions if you wanted which a lot of things including the Calibre Power Search plugin or the homemade grep method I suggested. e.g

pdfgrep -P "\bplates\b" foo.pdf

will find all the matches of plates but not contemplates or templates or whatever in the PDF foo.pdf. Lots of programs support regular expressions or regex but sometimes their are differences between them so you have to read the manual.

You could also look into a desktop wide tool like Recoll. It can export it's results as a spreadsheet (csv).

[MarjaE]

Thank you.

I currently use Houdahspot for searches, but I think it can do the same: https://blog.houdah.com/2016/01/expo...s-spreadsheet/