| 1 | * ABOUT BUGS |
|---|
| 2 | |
|---|
| 3 | Before reporting a bug, please check the list of known bugs |
|---|
| 4 | and the list of oft-reported non-bugs (below). |
|---|
| 5 | |
|---|
| 6 | Bugs and comments may be sent to bug-gnu-utils@gnu.org; please |
|---|
| 7 | include the word ``sed'' in the Subject: header. |
|---|
| 8 | |
|---|
| 9 | Please do not send a bug report like this: |
|---|
| 10 | |
|---|
| 11 | [while building frobme-1.3.4] |
|---|
| 12 | $ configure |
|---|
| 13 | sed: file sedscr line 1: Unknown option to 's' |
|---|
| 14 | $ sed --version |
|---|
| 15 | GNU sed version 3.01 |
|---|
| 16 | |
|---|
| 17 | If GNU sed doesn't configure your favorite package, take a few extra |
|---|
| 18 | minutes to identify the specific problem and make a stand-alone test |
|---|
| 19 | case. |
|---|
| 20 | |
|---|
| 21 | A stand-alone test case includes all the data necessary to perform the |
|---|
| 22 | test, and the specific invocation of sed that causes the problem. The |
|---|
| 23 | smaller a stand-alone test case is, the better. A test case should |
|---|
| 24 | not involve something as far removed from sed as ``try to configure |
|---|
| 25 | frobme-1.3.2''. Yes, that is in principle enough information to look |
|---|
| 26 | for the bug, but that is not a very practical prospect. |
|---|
| 27 | |
|---|
| 28 | |
|---|
| 29 | |
|---|
| 30 | * KNOWN BUGS |
|---|
| 31 | |
|---|
| 32 | Regular expression evaluation performance sucks. Fixing this is |
|---|
| 33 | a high priority for the next release, but I did not want to open |
|---|
| 34 | this can of worms for the 3.01 release because there were more |
|---|
| 35 | important functional bugs which had been wanting fixing. And |
|---|
| 36 | another issue is that the regular expression routines provided |
|---|
| 37 | do not _fully_ support POSIX.2 BREs. |
|---|
| 38 | |
|---|
| 39 | Improvement of the documentation and the testsuite are also in the |
|---|
| 40 | plans for 3.02. Specific suggestions are welcome, but I am |
|---|
| 41 | already aware of the general complaint of "this ain't good |
|---|
| 42 | enough", so you don't need to tell me that. |
|---|
| 43 | |
|---|
| 44 | |
|---|
| 45 | * NON-BUGS |
|---|
| 46 | |
|---|
| 47 | `sed -n' and `s/regex/replace/p' |
|---|
| 48 | |
|---|
| 49 | Some versions of sed ignore the `p' (print) option of an `s' command |
|---|
| 50 | unless the `-n' command switch has been specified. Other versions |
|---|
| 51 | always honor the `p' option. GNU sed is the latter sort. (Both |
|---|
| 52 | approaches are allowed by POSIX.2.) |
|---|
| 53 | |
|---|
| 54 | |
|---|
| 55 | regexp syntax clashes |
|---|
| 56 | |
|---|
| 57 | GNU sed uses the Posix basic regular expression syntax. According to |
|---|
| 58 | the standard, the meaning of some escape sequences is undefined in |
|---|
| 59 | this syntax; notable in the case of GNU sed are `\|', `\+', `\?', |
|---|
| 60 | `\`', `\'', `\<', `\>', `\b', `\B', `\w', and `\W'. |
|---|
| 61 | |
|---|
| 62 | As in all GNU programs that use Posix basic regular expressions, sed |
|---|
| 63 | interprets these escape sequences as meta-characters. So, `x\+' |
|---|
| 64 | matches one or more occurrences of `x'. `abc\|def' matches either |
|---|
| 65 | `abc' or `def'. |
|---|
| 66 | |
|---|
| 67 | This syntax may cause problems when running scripts written for other |
|---|
| 68 | seds. Some sed programs have been written with the assumption that |
|---|
| 69 | `\|' and `\+' match the literal characters `|' and `+'. Such scripts |
|---|
| 70 | must be modified by removing the spurious backslashes if they are to |
|---|
| 71 | be used with GNU sed. |
|---|
