MobileRead Forums - View Single Post - Kindle index files: file path hash method

Doitsu · 12-23-2014, 12:22 PM

I had a look at the files in:

Code:

/mnt/us/system/Search Indexes

which appears to contain the actual search indexes and the following two files:

Code:

Index.db 
Index.lg

Index.db appears to be a B-Tree database that contains words as well as file paths and hash values.

I've got the following questions:

Is there any database program that can open index.db or is it a custom Amazon file format?
What kind of hash method does the Kindle firmware usually use to convert Unix file paths to hashes? (Index file names always start with a lowercase y followed by a 32 character hexadecimal number, which is not an MD5 hash.)

12-23-2014, 12:22 PM	#1
Doitsu Grand Sorcerer Posts: 5,739 Karma: 24031403 Join Date: Dec 2010 Device: Kindle PW2	Kindle index files: file path hash method I had a look at the files in: Code: /mnt/us/system/Search Indexes which appears to contain the actual search indexes and the following two files: Code: Index.db Index.lg Index.db appears to be a B-Tree database that contains words as well as file paths and hash values. I've got the following questions: Is there any database program that can open index.db or is it a custom Amazon file format? What kind of hash method does the Kindle firmware usually use to convert Unix file paths to hashes? (Index file names always start with a lowercase y followed by a 32 character hexadecimal number, which is not an MD5 hash.)