PageSucker versions

v3.2 [Apr 23, 2010]
- Fixed a couple of bugs which affected the correct reloading of pattern matching settings when DOS style patterns were being used.
- Fixed a bug which could cause a corrupt prefs file to be written when running on MacOS X 10.2 (Jaguar).
- In previous versions PageSucker removed multiple consecutive occurrences of slashes at the end of URLs. This could create problems for certain Web servers, and thus multiple trailing slashes will now no longer be removed.
- Worked around a problem with Apple's Java 1.3.1 update 1 on MacOS X, which made it impossible or very difficult to enter proxy authentication information.
- Worked around another problem with Apple's Java 1.3.1 update 1 on MacOS X, which could cause the control window's menus to get deactivated when the "end of download" dialog was dismissed by hitting the "Return" key while the log window was active.
- Enhanced the log window such as to only show the last 200 lines of log output. Previously the number of lines shown was unlimited, which could lead to memory shortages upon long downloads.
- Made sure that the log window shows newly added text on Windows machines. Previously, the log window would remained scrolled all the way up on Windows.
- A relative URL in a redirected page will now be correctly interpreted relative to the redirected URL. Previously it would be interpreted relative to the original URL, which could lead to incomplete downloads, error messages or even infinite loops.
- Fixed a couple of bugs which could cause mysterious "File Not Found" messages to appear when the "Complement Existing File" mode was used and PageSucker encountered name clashes on the site being downloaded. Due to this bug, multiple threads could attempt to write to a given local file at the same time, which would cause a "FileNotFoundException" message to be displayed.
- Previously, PageSucker's JavaScript interpretation would handle a single dot as a potential URL, which would corrupt a downloaded page when a single dot was used with some other meaning inside that page. This has been changed, so that single dots inside blocks of JavaScript are now always left alone.
- Corrected a bug which could cause problems when downloading a JavaScript include file the URL of which does not have an extension. Such a file would incorrectly be treated like an HTML file.
- Fixed a bug which would cause certain %xx encoded characters (such as ampersands) in a URL to be decoded, which could lead to the impossibility to correctly download certain pages.
- In previous releases, initial whitespace inside a quoted string would be moved out of the string, to show up before the quote character. Now, initial and trailing whitespace inside quoted strings will simply be removed.
- Added support for the most common cases of cascading stylesheets (CSS).
- Added two options to delete unused empty files, and/or incompletely downloaded files when a download process is interrupted by the user.
- Enabled the "Recognize unterminated quoted strings" option in the Miscellaneous settings window by default, as this option makes PageSucker more tolerant toward an HTML error which is very common in today's Web pages.
- The "Parse HTML pages not in hierarchy" option in the HTML Files settings window is now also available in the demo version of PageSucker. In previous versions, this option was reserved for registered users, but this restriction turned out to be too severe considering the structure of many modern Web sites.