How to get book paths from command linee file

[groob]

Hello,
I'm trying to analyze my library space usage, I've identified some books take a very large amount of space on my disk.
I want to analyze this further and automate some tasks.

I went on a first selection of my books and

- Create catalog to export / CSV/XML Options

But in there there is no option to export each book file path.

I've looked into calibredb command and also there could find a way to obtain this information.

Have I missed something? I don't need to export the books, I just need to find the folder paths containing those books as I need to further analyze the disk usage.

[theducks]

Did you grab {id} in your CSV?

Calibre has a pretty consistent path name
Library_folder\{author}\{title} (id)
eg

xxx MainLib\Martha Carr\Goth Drow Unleashed Boxed Set Two_ (21560)

[theducks]

BTW You can show / sort on the {size} column in the GUI.
Illustrated PDF's are my big hogs, followed by lots of embedded fonts in EPUB

[groob]

Originally Posted by theducks

BTW You can show / sort on the {size} column in the GUI.
Illustrated PDF's are my big hogs, followed by lots of embedded fonts in EPUB

Thanks! I didn't notice the Size column!
Do you if this is just for the main version of the book? I might have several formats for the same book: audiobook as a zip file, pdf, epub and so on.

I've made a quick test, it's not completely reliable, it's directional (I mean I can see the books with bigger file sizes first).
Example, the GUI tells me a book is 522MB, then by opening the folder the total size of all formats is 558MB. Even the single biggest file (547MB) is already bigger than what shown on the GUI. 🤷*♂️

[DNSB]

If you have, for example, a PDF, azw3, ePub and txt file, the size shown is the largest file.

If you have a 250MB file and and 247MB file, the file size will be shown as 250MB.

For the file size, when I look in calibre, there is a file shown as 367MB, when I open the folder the largest file shows as 375,772KB. Dividing 375,772 by 1024 gives me 366.96MB so round up to 367MB while Windows file properties will round down to 366MB.

In your example, the 547 is likely the first three characters of the file size in bytes. 547,000,000 bytes gives me 521.65MB (divided twice by 1024) while 547,999,999 gives me 522.61MB.

[capink]

This can be done by combination of Action Chains plugin and calibre template language from within the GUI. Attached below is a chain that calculates total size for selected books in human readable format. You can import the attached chain by: Action Chains > Add/modify chains > right click on the chains dialog > import > browse to the zip file.

Note: Also see what DNSB said in the post above about the difference between MB and MiB.

[chaley]

This calibre python script might help. It produces a tab-separated CSV file for each book id provided on the command line. The CSV file contains 4 columns:

• book id
• extension
• size in bytes
• full path to file

To use it, save this script somewhere convenient:

Spoiler:

import json, os
from calibre.db.legacy import LibraryDatabase

'''
argument 1: path to the library

argument 2: list of book ids, e.g., 208,209. Quote the argument if there
            are spaces between the ids. calibredb search can be used to
            produce a list of ids. For example,
            calibredb search "size:>100000"
            finds all books with size > 100,000, printing the list of ids

output: a tab-separated csv list written to standard output
'''

# open the library database
db = LibraryDatabase(sys.argv[1]).new_api

# convert the book ids to a list of integers
book_ids = [int(bid.strip()) for bid in sys.argv[2].split(',')]

# get the library path from calibre. This removes symlinks and the like
library_path = db.backend.library_path

# print the header for the CSV output
print('\t'.join(('book_id', 'extension','size', 'path')))

# loop over the books generating the output for each format
for book_id in book_ids:
    # get the list of formats for the book
    formats = db.formats(book_id)

    # loop over the formats, generating the csv line
    for ext in formats:
        # get the metadata for the format: the extension, size, path, and modtime
        fmt_data = db.format_metadata(book_id, ext)
        # write the csv line for the format
        print('\t'.join((str(book_id), ext, str(fmt_data['size']), fmt_data['path'])))

Execute the script with

calibre-debug -e script_file_path library_path id1,id2,id3 > output.csv

Example using one of my test libraries:

1. I saved the script as formats_sizes.py
2. I used calibredb to get a list of ids, in this case books with 163 in the title
   

   calibredb search "title:163"

   that produced the output
   

   1337,1343,1361,1434

   You might want to use something like
   

   calibredb search "size:>100000"

3. I ran the script using
   

   calibre-debug -e formats_sizes.py C:\CBH_Data\calibre.git\Library.test_small 1337,1343,1361,1434 > aaaaa.csv

The output is

Spoiler:

book_id	extension	size	path
1337	EPUB	790778	C:\CBH_Data\calibre.git\Library.test_small\Eric Flintt\16332 (1337)\16332 - Eric Flintt.epub
1337	ORIGINAL_EPUB	786992	C:\CBH_Data\calibre.git\Library.test_small\Eric Flintt\16332 (1337)\16332 - Eric Flintt.original_epub
1343	EPUB	736765	C:\CBH_Data\calibre.git\Library.test_small\Eric Flint\1632 (1343)\1632 - Eric Flint.epub
1343	MOBI	1029315	C:\CBH_Data\calibre.git\Library.test_small\Eric Flint\1632 (1343)\1632 - Eric Flint.mobi
1343	PRC	1093048	C:\CBH_Data\calibre.git\Library.test_small\Eric Flint\1632 (1343)\1632 - Eric Flint.prc
1361	EPUB	888796	C:\CBH_Data\calibre.git\Library.test_small\Weber, David\1633 (1361)\1633 - Weber, David.epub
1434	EPUB	990171	C:\CBH_Data\calibre.git\Library.test_small\A. B. C. Personn\1632 - Flint Eric (1434)\1632 - Flint Eric - A. B. C. Personn.epub

Libre office calc shows this file as


Edit: it would be easy for the script to use a search expression instead of a list of ids. Let me know if you want an example script.

[kovidgoyal]

calibredb list --fields=formats