diff options
Diffstat (limited to 'RELNOTES')
| -rw-r--r-- | RELNOTES | 45 |
1 files changed, 45 insertions, 0 deletions
@@ -1,3 +1,48 @@ + TXR 157 + 2016-11-14 + + + Features + + - New #; syntax for commenting out Lisp object. + - #: syntax for uninterned symbols now works as an input notation. + - Packages: + - Overhauled implementation with increased scope. + - New *package* special variable holds current package. + - New concepts in package system: foreign symbols, hidden symbols. + - New functions: package-local-symbols, package-foreign-symbols, use-sym, + unuse-sym, use-package, unuse-package, unintern. + - New defpackage macro. + - Improvements in Vim syntax coloring; don't forget to install new txr.vim + and tl.vim files. + - New convenience functions for file or process I/O in one call: file-get, + file-put, file-append, file-get-string, file-put-string, + file-append-string, file-get-lines, file-put-lines, file-append-lines, + command-get, command-put, command-get-string, command-put-string, + command-get-lines, and command-put-lines. + + Bugs + + - Fixed an infinite loop in the processing of syntactic places: regression + introduced in 156. + - Fixed bug in processing of nested syntactic places like [place arg] where + the inner place is a macro. + - Fixed regression in awk macro introduced in 156, related to the above bug: + not being able to assign to a field as in (set [f 3] "foo"). The bug was + there before, but the change suddenly removed the accidental circumstances + which protected the awk macro from reproducing the problem. + - Fixed some additional problems in circle printing, in the printing + of package objects and and interpreted functions. + - A few GC-unsafe mutations were found in the TXR kernel, by inspection, and + fixed. No actual instability was reproduced. + - Fixed a bug in the caseq, caseql and casequal macros: single-atom key + values were being exposed to evaluation rather than taken as themselves. + This behavior can still be obtained with the compatibility switch; + but it is recommended to first see whether switching to one of the new + case constructs caseq*, caseql* or casequal* is feasible. + + + TXR 156 2016-11-05 |