Change log for duff
===================

Maintained since version 0.3.1.

2012-01-28  Camilla Berglund  <elmindreda@elmindreda.org>

 * dufffile.c (compare_files): Removed superflous physical_flag test.
 * duff.h dufffile.c (init_file_list): Renamed from file_list_init.
 * duff.h dufffile.c (alloc_file): Renamed from file_list_alloc.
 * duff.h dufffile.c (file_list_empty): Renamed from empty_file_list.
 * duff.h dufffile.c (free_entry_list): Renamed from file_list_free.
 * duffdriver.c dufffile.c duffutil.c: Put non-static functions on top.
 * duffdriver.c (process_clusters): Initialized cluster index.
 * duffdriver.c (report_cluster): Made use of get_field_terminator.
 * duffutil.c (read_path): Made use of get_field_terminator.
 * duffutil.c (get_field_terminator): Added function.
 * duffutil.c (read_path): Removed limit on path length, updated signature.

2012-01-27  Camilla Berglund  <elmindreda@elmindreda.org>

 * dufffile.c (compare_entry_contents): Made warnings take -q into account.
 * dufffile.c (get_file_sample): Tagged file INVALID on read error.
 * dufffile.c (compare_file_digests): Replaced for-loop with memcmp.
 * dufffile.c (get_file_sample, get_file_digest): Replaced pointer checks with
 status checks.
 * duff.h (Status): Added SAMPLED and HASHED values.
 * duffdriver.c (process_clusters): Added test for whether base item has entered
 INVALID state.
 * dufffile.c (get_file_sample, get_file_digest): Removed duplicate check for
 INVALID state.

2012-01-25  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.h (Status): Removed unused DUPLICATE status.
 * duffdriver.c (process_file): Removed read permission check on files (credits
 to Kamal Mostafa).

2012-01-24  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.h (HASH_BITS): Increased to 10.
 * duffdriver.c (print_terminator): Added function.
 * duff.h duffdriver.c dufffile.c (file_list_*): Renamed from entry_list_*.
 * duff.h duffdriver.c dufffile.c (free_file): Renamed from free_entry.
 * duff.h duffdriver.c dufffile.c (init_file): Renamed from fill_entry.
 * duff.h duffdriver.c dufffile.c duffutil.c (FileList): Renamed from EntryList.
 * duff.h duffdriver.c dufffile.c duffutil.c (File): Renamed from Entry.
 * dufffile.c: Renamed from duffentry.c.
 * duffdriver.c (has_recorded_directory): Renamed from has_recursed_directory.
 * duffdriver.c (DirList): Added array of Dir structs.
 * duffdriver.c (Dir): Renamed from Directory and removed list link.
 * duff.h duffdriver.c: Moved Directory to duffdriver.
 * duffdriver.c (process_directory): Renamed from recurse_directory.
 * duff.h (EntryList): Renamed from List.
 * duffentry.c (get_entry_sample, get_entry_digest, compare_entry_contents):
 Improved error reporting.
 * duff.h duffentry.c: Added BUFFER_SIZE macro for size of read buffers.
 * duff.1: Improved grammar for -z flag.
 * duff.1: Clarified language for -e flag.
 * duff.1: Fixed statement claiming digests are calculated in thorough mode.

2012-01-23  Camilla Berglund  <elmindreda@elmindreda.org>

 * duffdriver.c (process_args, process_clusters): Moved freeing of entries to
 directly after each bucket has been processed.
 * duffdriver.c (process_clusters): Renamed from report_clusters.
 * duffdriver.c: Moved from single to BUCKET_COUNT entry lists, indexed by size.
 * duffdriver.c (process_args): Added function.
 * duffdriver.c (process_path, report_clusters): Made private to duffdriver.
 * duff.c duffdriver.c: Moved argument processing to duffdriver.
 * duffentry.c (fill_entry): Renamed function from make_entry.
 * duffdriver.c (free_entry_list): Removed function.
 * duffdriver.c: Replaced doubly linked lists with List.
 * duffutil.c (entry_list_init, entry_list_alloc, entry_list_empty,
 entry_list_free): Added entry list functions.
 * duffutil.c (link_entry, unlink_entry): Removed functions.
 * duff.h (List): Added linear entry list struct and typedef.
 * duff.h (Entry): Removed list links.
 * duff.c (main): Saved header digest use before processing entries.
 * duff.c (header_uses_digest): Added global flag.
 * duffentry.c (report_cluster): Ensured that first cluster entry has digest.
 * duffentry.c (compare_entries): Added more early-out cases.
 * duffentry.c (get_entry_digest): Stopped reading empty files.
 * duffentry.c (generate_entry_digest): Added function.
 * duffentry.c (get_entry_digest): Hash sample if it contains entire file.
 * duffutil.c (error, warning): Prefixed program name to output.
 * duffutil.c (set_digest_function): Added function.
 * duffutil.c (digest_init): Added function.
 * duffutil.c (digest_update): Added function.
 * duffutil.c (digest_finish): Added function.
 * duff.c duffutil.c duffentry.c: Moved all calls to sha to duffutil.
 * duff.c duffutil.c duffentry.c: Made digest_function private to duffutil.
 * duff.1 duff.c: Changed default sampling limit to zero.

2012-01-22  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.c duff.h duffentry.c: Changed sampling to read (up to) the first
 SAMPLE_SIZE bytes (credits to Kamal Mostafa).
 * duff.c: Clarified help for -H and -L flags.
 * duff.c: Added list of supported digest functions to help for -d flag.

2012-01-20  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.1: Added missing -type f to example (credits to Stephan Hegel).

2012-01-19  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.h: Tagged functions error and warning with GCC attributes.
 * duff.1: Clarified default sampling limit.
 * duffdriver.c: Renamed cluster index variable to index.

2012-01-18  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.1: Clarified language regarding -t and %d.
 * duff.1: Made the output of examples human-readable.

2012-01-16  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.1: Added new cluster header to description of -f flag.
 * duff.c (main): Added error if thorough mode output uses digest (credits to
 Kamal Mostafa and Mika Kuoppala).
 * duff.c (main): Added non-digest default cluster header for thorough mode.
 * duffutil.c duff.h: Added function cluster_header_uses_digest.

2011-04-11  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.c (main): Made cluster header translatable.
 * duffutil.c (error): Changed exit status to EXIT_FAILURE.
 * duffdriver.c (process_path): Split file processing into process_file function.
 * duffdriver.c (process_file): Reverted to old behaviour of skipping block devices.

2011-04-10  gettextize  <bug-gnu-gettext@gnu.org>

 * Makefile.am (SUBDIRS): Add po.
   (ACLOCAL_AMFLAGS): New variable.
   (EXTRA_DIST): Add config.rpath, m4/ChangeLog.
 * configure.ac (AC_OUTPUT): Add po/Makefile.in.

2011-04-10  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.c duff.h duffdriver.c duffentry.c: Changed gettext macro to _.
 * gettext.h: Added file to lib directory.
 * duff.c (usage): Improved grammar in help text.
 * duff.c duff.h duffdriver.c duffentry.c: Enabled use of gettext.
 * duff.c duff.h duffdriver.c duffentry.c duffutil.c
   duffstring.c: Moved to use autoconf large file support detection.

2010-08-15  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.c (usage): Annotated strings for i18n.
 * duff.c (main): Made digest function name comparison case-insensitive.

2009-09-21  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.h: Added enum typedefs.

2009-09-20  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.c (version, usage, bugs): Changed to output to stdout.

2009-09-19  Camilla Berglund  <elmindreda@elmindreda.org>

 * bootstrap.sh: Added instruction to run configure. 
 * duffentry.c (get_entry_digest): Removed broken optimization.
 * duff.1: Added note on file modification.
 * duff.h: Added struct typedefs.

2009-08-09  Camilla Berglund  <elmindreda@elmindreda.org>

 * duffdriver.c (recurse_directory): Added error handling for asprintf.
 * duffentry.c (get_entry_samples): Added error handling for fread.
 * duffutil.c (error, warning): Added error handling for asprintf.
 * duff.c (usage): Clarified wording of help text.

2009-01-16  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.c (read_path, kill_trailing_slashes): Moved functions into duffutil.c.
 * duff.h: Added declarations for read_path and kill_trailing_slashes.
 * duff.c (version): Annotated strings for i18n.

2009-01-05  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.c (main): Moved option default values into source file.
 * duff.c (main) (usage): Renamed -c flag to -d.

2009-01-04  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.c duffdriver.c duffentry.c duffstring.c duffutil.c: Annotated strings for i18n.
 * duff.c (main): Added call to setlocale(3).

2009-01-03  Camilla Berglund  <elmindreda@elmindreda.org>

 * sha256.c: Added file.
 * sha256.h: Added file.
 * sha384.c: Added file.
 * sha384.h: Added file.
 * sha512.c: Added file.
 * sha512.h: Added file.
 * duffutil.c (print_message_digest): Added function.
 * duffutil.c (print_cluster_header): Added call to print_message_digest.
 * duffentry.c (get_entry_digest): Added support for digest function selection.
 * duffentry.c (get_entry_digest_sha1) (get_entry_digest_sha256)
   (get_entry_digest_sha384) (get_entry_digest_sha512): Added function.
 * duff.c (main) (usage): Added -c option for selecting digest function.
 * duff.h: Added digest function enumeration.

 * duff.c duffdriver.c duffentry.c duffstring.c duffutil.c: Made off_t 64-bit on glibc.

2006-10-07  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.c (main): Always kill trailing slashes (and in a nicer way).
 * duff.c (kill_trailing_slashes): Added function.
 * duff.c (main) (read_path): Don't kill newlines for -0 option.

2006-09-29  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.c (main) (usage): Added -0 option (credits to Clemens Lucas Fries).
 * duff.c (read_path): Added function.
 * duffdriver.c (report_clusters): Added -0 option (credits to Clemens Lucas Fries).
 * duffutil.c (print_cluster_header): Removed newline termination.

2006-01-11  Camilla Berglund  <elmindreda@elmindreda.org>

 * duffdriver.c (stat_file): Added parenthesis to remove warning.

2006-01-09  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.h (copy_entry): Removed unused function.
 * duffentry.c (copy_entry): Removed unused function.

 * duffdriver.c (process_path): Only check access for non-empty files.
 * duffentry.c (get_entry_samples)
   (get_entry_checksum): Only open and read from non-empty files.

 * join-duplicates.sh: Removed logging, added some error checking and mktemp.

2006-01-08  Camilla Berglund  <elmindreda@elmindreda.org>

 * join-duplicates.sh: Added file (credits to Ross Newell).

2006-01-07  Camilla Berglund  <elmindreda@elmindreda.org>

 * duffentry.c (get_entry_samples): Fixed seek bug (credits to Alexander Bostrom).

 * duffdriver.c (stat_file): Removed collection of symlinks to files.

 * duff.h: Added enum for symlink dereferencing modes.
 * duffdriver.c (stat_file): Implemented symlink dereferencing modes.
 * duffdriver.c (process_path, recurse_directory)
   (stat_file): Added recursion depth tracking.
 * duff.c (main, usage): Added -H option for following only symlinks listed as arguments.

 * duff.c (main): Reordered options for better clarity.

 * duff.c (main): Improved warnings for -l option.

2006-01-04  Camilla Berglund  <elmindreda@elmindreda.org>

 * duffentry.c (compare_entries): Removed code for -z option.
 * duffdriver.c (stat_file): Do not collect empty files when using -z option.

 * duffdriver.c (report_clusters): Fixed unlink bug.
 * duffdriver.c (report_clusters): All entries are now freed by the cluster search phase.

2006-01-03  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.h: Added inode and device fields to struct Entry (credits to Ross Newell).
 * duffdriver.c (process_path): Implemented physical mode (credits to Ross Newell).
 * duff.c (main, usage): Added -p option for physical mode.

 * duff.h: Added backlink to struct Entry.
 * duffentry.c (unlink_entry, link_entry): Added functions.
 * duffentry.c (make_entry): Added storing of device and inode fields.
 * duffdriver.c: Moved to using doubly-linked entries.

2005-12-14  Camilla Berglund  <elmindreda@elmindreda.org>

 * duffdriver.c (stat_file): Eliminated duplicate stat call for -L option.

 * duffdriver.c (stat_file): Added inital sketch for symlink-aware -e option.

2005-12-01  Camilla Berglund  <elmindreda@elmindreda.org>

 * duffutil.c (print_cluster_header): Fixed bug when printing off_t values on Mac OS X.

 * duffdriver.c: Added includes for building on Solaris.
 * duff.c: Added includes for building on Solaris.

2005-11-28  Camilla Berglund  <elmindreda@elmindreda.org>

 * duffdriver.c (has_recursed_directory)
   (record_directory): Added primitive directory loop detection.
 * duffutil.c: Added hash function for device/inode pair.
 * duff.h: Added struct Directory for recording of recursed directories.

 * duffdriver.c: Added several additional TODO items.

2005-11-27  Camilla Berglund  <elmindreda@elmindreda.org>

 * duffdriver.c (report_clusters): Added freeing of entries.

 * duffentry.c (compare_entries): Added code for -z option.

 * duff.c (main): Added file name reading from stdin.

2005-11-26  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.c (main, usage): Added -z option for ignoring empty files.

 * duff.c: Moved primary driver logic into duffdriver.c.
 * duffdriver.c: Created new file for primary driver logic.

2005-11-24  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.c: Added missing 't' to the getopt string (credits to Richard Levitte).

 * duffentry.c (compare_entries): Fixed bug in ordering of phases.

2005-11-23  Camilla Berglund  <elmindreda@elmindreda.org>

 * duff.c (usage): Fixed typo in -t option (credits to Patrik Jarnefelt).