README revision 1.1.1.1.4.2
1 1.1.1.1.4.2 martin List of bugs fixed. 2 1.1.1.1.4.2 martin 3 1.1.1.1.4.2 martin 1. ofs-rebuild: OFS value used to rebuild the record was incorrect. 4 1.1.1.1.4.2 martin Fixed August 19, 2014. Revised fix August 2018. 5 1.1.1.1.4.2 martin 6 1.1.1.1.4.2 martin 2. system-status: Instead of a floating-point division by 256, use 7 1.1.1.1.4.2 martin the wait(2) macros to create a reasonable exit status. Fixed March 12, 2016. 8 1.1.1.1.4.2 martin 9 1.1.1.1.4.2 martin 3. space: Use provided xisblank() function instead of ispace() for 10 1.1.1.1.4.2 martin matching [[:blank:]]. 11 1.1.1.1.4.2 martin 12 1.1.1.1.4.2 martin 4. a-format: Add POSIX standard %a and %A to supported formats. Check 13 1.1.1.1.4.2 martin at runtime that this format is available. 14 1.1.1.1.4.2 martin 15 1.1.1.1.4.2 martin 5. decr-NF: Decrementing NF did not change $0. This is a decades-old 16 1.1.1.1.4.2 martin bug. There are interactions with the old and new value of OFS as well. 17 1.1.1.1.4.2 martin Most of the fix came from the NetBSD awk. 18 1.1.1.1.4.2 martin 19 1.1.1.1.4.2 martin 6. string-conv: String conversions of scalars were sticky. Once a 20 1.1.1.1.4.2 martin conversion to string happened, even with OFMT, that value was used until 21 1.1.1.1.4.2 martin a new numeric value was assigned, even if OFMT differed from CONVFMT, 22 1.1.1.1.4.2 martin and also if CONVFMT changed. 23 1.1.1.1.4.2 martin 24 1.1.1.1.4.2 martin 7. unary-plus: Unary plus on a string constant returned the string. 25 1.1.1.1.4.2 martin Instead, it should convert the value to numeric and give that value. 26 1.1.1.1.4.2 martin 27 1.1.1.1.4.2 martin 8. concat-assign-same: Concatenation previously evaluated both sides of the 28 1.1.1.1.4.2 martin expression before doing its work, which, since assign() evaluates to the cell 29 1.1.1.1.4.2 martin being assigned to, meant that expressions like "print (a = 1) (a = 2)" would 30 1.1.1.1.4.2 martin print "22" rather than "12". 31 1.1.1.1.4.2 martin 32 1.1.1.1.4.2 martin 9. missing-precision: When using the format string "%*s", the precision 33 1.1.1.1.4.2 martin argument was used without checking if it was present first. 34 1.1.1.1.4.2 martin 35 1.1.1.1.4.2 martin 10. missing-precision: When using the format string "%*s", the precision 36 1.1.1.1.4.2 martin argument was used without checking if it was present first. 37 1.1.1.1.4.2 martin 38 1.1.1.1.4.2 martin 11. fmt-overflow: The buffer used for OFMT/CONVFMT conversions was written 39 1.1.1.1.4.2 martin to with sprintf(), which meant that some conversions could write past the 40 1.1.1.1.4.2 martin end. 41 1.1.1.1.4.2 martin 42 1.1.1.1.4.2 martin 12. numeric-subsep, numeric-fs, numeric-output-seps, numerics-rs: If SUBSEP, 43 1.1.1.1.4.2 martin FS, RS, OFS, or ORS were set to a numeric value, then their string values 44 1.1.1.1.4.2 martin wouldn't always be generated before being needed. 45 1.1.1.1.4.2 martin 46 1.1.1.1.4.2 martin 13. subsep-overflow: The length of SUBSEP needs to be rechecked after 47 1.1.1.1.4.2 martin calling execute(), in case SUBSEP itself has been changed. 48 1.1.1.1.4.2 martin 49 1.1.1.1.4.2 martin 14. split-fs-from-array: If the third argument to split() comes from the 50 1.1.1.1.4.2 martin array passed as the second argument, then split() would previously read 51 1.1.1.1.4.2 martin from the freed memory and possibly produce incorrect results (depending 52 1.1.1.1.4.2 martin on the system's malloc()/free() behaviour.) 53 1.1.1.1.4.2 martin 54 1.1.1.1.4.2 martin 15. getline-numeric: The `getline xx < file' syntax did not check if 55 1.1.1.1.4.2 martin values were numeric, in discordance from POSIX. Test case adapted from 56 1.1.1.1.4.2 martin one posted by Ben Bacarisse <ben.usenet (a] bsb.me.uk> in comp.lang.awk, 57 1.1.1.1.4.2 martin January 2019. 58
Indexes created Sat Oct 03 00:25:10 UTC 2026