NEWS revision 126432
1Version 2.4.2 2 3 - Added more check in configure to default the grep-${version}/src/regex.c 4 instead of the one in GNU Lib C. 5 6Version 2.4.1 7 8 - If the final byte of an input file is not a newline, grep now silently 9 supplies one. 10 11 - The new option --binary-files=TYPE makes grep assume that a binary input 12 file is of type TYPE. 13 --binary-files='binary' (the default) outputs a 1-line summary of matches. 14 --binary-files='without-match' assumes binary files do not match. 15 --binary-files='text' treats binary files as text 16 (equivalent to the -a or --text option). 17 18 - New option -I; equivalent to --binary-files='without-match'. 19 20Version 2.4: 21 22 - egrep is now equivalent to `grep -E' as required by POSIX, 23 removing a longstanding source of confusion and incompatibility. 24 `grep' is now more forgiving about stray `{'s, for backward 25 compatibility with traditional egrep. 26 27 - The lower bound of an interval is not optional. 28 You must use an explicit zero, e.g. `x{0,10}' instead of `x{,10}'. 29 (The old documentation incorrectly claimed that it was optional.) 30 31 - The --revert-match option has been renamed to --invert-match. 32 33 - The --fixed-regexp option has been renamed to --fixed-string. 34 35 - New option -H or --with-filename. 36 37 - New option --mmap. By default, GNU grep now uses read instead of mmap. 38 This is faster on some hosts, and is safer on all. 39 40 - The new option -z or --null-data causes `grep' to treat a zero byte 41 (the ASCII NUL character) as a line terminator in input data, and 42 to treat newlines as ordinary data. 43 44 - The new option -Z or --null causes `grep' to output a zero byte 45 instead of the normal separator after a file name. 46 47 - These two options can be used with commands like `find -print0', 48 `perl -0', `sort -z', and `xargs -0' to process arbitrary file names, 49 even those that contain newlines. 50 51 - The environment variable GREP_OPTIONS specifies default options; 52 e.g. GREP_OPTIONS='--directories=skip' reestablishes grep 2.1's 53 behavior of silently skipping directories. 54 55 - You can specify a matcher multiple times without error, e.g. 56 `grep -E -E' or `fgrep -F'. It is still an error to specify 57 conflicting matchers. 58 59 - -u and -U are now allowed on non-DOS hosts, and have no effect. 60 61 - Modifications of the tests scripts to go around the "Broken Pipe" 62 errors from bash. See Bash FAQ. 63 64 - New option -r or --recursive or --directories=recurse. 65 (This option was also in grep 2.3, but wasn't announced here.) 66 67 - --without-included-regex disable, was causing bogus reports .i.e 68 doing more harm then good. 69 70Version 2.3: 71 72 - When searching a binary file FOO, grep now just reports 73 `Binary file FOO matches' instead of outputting binary data. 74 This is typically more useful than the old behavior, 75 and it is also more consistent with other utilities like `diff'. 76 A file is considered to be binary if it contains a NUL (i.e. zero) byte. 77 78 The new -a or --text option causes `grep' to assume that all 79 input is text. (This option has the same meaning as with `diff'.) 80 Use it if you want binary data in your output. 81 82 - `grep' now searches directories just like ordinary files; it no longer 83 silently skips directories. This is the traditional behavior of 84 Unix text utilities (in particular, of traditional `grep'). 85 Hence `grep PATTERN DIRECTORY' should report 86 `grep: DIRECTORY: Is a directory' on hosts where the operating system 87 does not permit programs to read directories directly, and 88 `grep: DIRECTORY: Binary file matches' (or nothing) otherwise. 89 90 The new -d ACTION or --directories=ACTION option affects directory handling. 91 `-d skip' causes `grep' to silently skip directories, as in grep 2.1; 92 `-d read' (the default) causes `grep' to read directories if possible, 93 as in earlier versions of grep. 94 95 - The MS-DOS and Microsoft Windows ports now behave identically to the 96 GNU and Unix ports with respect to binary files and directories. 97 98Version 2.2: 99 100Bug fix release. 101 102 - Status error number fix. 103 - Skipping directories removed. 104 - Many typos fix. 105 - -f /dev/null fix(not to consider as an empty pattern). 106 - Checks for wctype/wchar. 107 - -E was using the wrong matcher fix. 108 - bug in regex char class fix 109 - Fixes for DJGPP 110 111Version 2.1: 112 113This is a bug fix release(see Changelog) i.e. no new features. 114 115 - More compliance to GNU standard. 116 - Long options. 117 - Internationalisation. 118 - Use automake/autoconf. 119 - Directory hierarchy change. 120 - Sigvec with -e on Linux corrected. 121 - Sigvec with -f on Linux corrected. 122 - Sigvec with the mmap() corrected. 123 - Bug in kwset corrected. 124 - -q, -L and -l stop on first match. 125 - New and improve regex.[ch] from Ulrich Drepper. 126 - New and improve dfa.[ch] from Arnold Robbins. 127 - Prototypes for over zealous C compiler. 128 - Not scanning a file, if it's a directory 129 (cause problems on Sun). 130 - Ported to MS-DOS/MS-Windows with DJGPP tools. 131 132See Changelog for the full story and proper credits. 133 134Version 2.0: 135 136The most important user visible change is that egrep and fgrep have 137disappeared as separate programs into the single grep program mandated 138by POSIX 1003.2. New options -G, -E, and -F have been added, 139selecting grep, egrep, and fgrep behavior respectively. For 140compatibility with historical practice, hard links named egrep and 141fgrep are also provided. See the manual page for details. 142 143In addition, the regular expression facilities described in Posix 144draft 11.2 are now supported, except for internationalization features 145related to locale-dependent collating sequence information. 146 147There is a new option, -L, which is like -l except it lists 148files which don't contain matches. The reason this option was 149added is because '-l -v' doesn't do what you expect. 150 151Performance has been improved; the amount of improvement is platform 152dependent, but (for example) grep 2.0 typically runs at least 30% faster 153than grep 1.6 on a DECstation using the MIPS compiler. Where possible, 154grep now uses mmap() for file input; on a Sun 4 running SunOS 4.1 this 155may cut system time by as much as half, for a total reduction in running 156time by nearly 50%. On machines that don't use mmap(), the buffering 157code has been rewritten to choose more favorable alignments and buffer 158sizes for read(). 159 160Portability has been substantially cleaned up, and an automatic 161configure script is now provided. 162 163The internals have changed in ways too numerous to mention. 164People brave enough to reuse the DFA matcher in other programs 165will now have their bravery amply "rewarded", for the interface 166to that file has been completely changed. Some changes were 167necessary to track the evolution of the regex package, and since 168I was changing it anyway I decided to do a general cleanup. 169