| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| readme.txt | 2015-03-12 | 2.3 kB | |
| PWGen-2.5.4-Src.zip | 2015-03-12 | 369.6 kB | |
| PWGen-2.5.4-Setup.exe | 2015-03-12 | 1.4 MB | |
| PWGen-2.5.4.zip | 2015-03-12 | 1.4 MB | |
| Totals: 4 Items | 3.1 MB | 0 | |
NOTES ON THE DOWNLOADABLE FILES:
--------------------------------
PWGen-xxx.zip : contains PWGen binary (PWGen.exe) & manual *without*
installer/setup; simply extract the contents of the .zip
file into a folder of your choice and execute PWGen.exe;
useful for portable applications (USB stick etc.)
PWGen-xxx-Setup.exe : contains setup for installing PWGen on your hard disk
PWGen-xxx-Src.zip : contains source code (C/C++, Borland C++ Builder)
RELEASE NOTES:
--------------
CHANGES & IMPROVEMENTS:
- generation of unique password lists (i.e., lists where each entry must occur
only once; activated by the option "Exclude duplicate entries" in the
"Advanced Password Options") has been optimized for speed (at the expense
of a roughly doubled memory requirement) and is much faster now even for very
large password lists (>1 million entries)
- when generating password lists to files with the "Exclude duplicates" option
activated, checking for duplicates is performed in-memory now instead of
searching the entire file contents; as a consequence, the amount of passwords
that can be generated in this case is limited by the memory available to a
32-bit application like PWGen (... but it's also much faster now)
(thanks to Hussein for reporting & valuable suggestions!); note that when the
list is appended to an existing file, the file contents will NOT be searched
anymore for already existing password entries, i.e., a new unique password
list will be appended to the file
FIXES:
- generation of unique password lists didn't work properly for passphrases
(composed of words from a word list)
- fixed a bug in the internal word list ("diceware8k" module), which reduced its
effective size to 8191 instead of 8192 words (list contained an empty string
due to the faulty encoding of a string in the C source code file); apart from
(slightly!) reducing the entropy (12.9998 vs. 13), this empty word reduced the
number of words in passphrases when randomly chosen and sometimes caused
errors when converting passwords to UTF-8 Unicode encoding
- "out of memory" errors were not displayed correctly, a cryptic message was
shown without the actual information