geiser/elisp, branch 0.1.4 emacs and scheme talk to each other https://jao.io/cgit/geiser/atom?h=0.1.4 2011-11-26T06:11:26Z Racket: capturing and displaying standard error during evaluation 2011-11-26T06:11:26Z Jose Antonio Ortega Ruiz jao@gnu.org 2011-11-26T06:11:26Z urn:sha1:8a61de0ab4d29d777122a15734c68e9744a69d30 This bugs was exposed by using rackunit, where all the output of, say, check-eq? was lost for good (it was being sent to the stderr black hole). Hat tip Grant Retkke. Bug fix: avoiding sending unbalanced sexps in C-c C-r and friends 2011-10-01T21:03:41Z Jose Antonio Ortega Ruiz jao@gnu.org 2011-10-01T21:03:41Z urn:sha1:d08f649e8bd6ddad39a4e9d4766f88dbb5d24048 We were not checking that the region sent to the scheme process was balanced, resulting in said process waiting for ever on `read' (or its moral equivalent in our current implementation). We now just refuse to evaluate an improper region in the first place. Compatibility with filladapt. 2011-09-15T01:15:18Z Jose Antonio Ortega Ruiz jao@gnu.org 2011-09-15T01:15:18Z urn:sha1:fa250dbf7620a050f61c4ccc5496507ade9bff82 Seems like the add-on package filladapt.el is broken in that its version of fill-adapt uses a non-optional first argument. Aquamacs users were filling the pain. Fixed by passing nil in our call to fill-paragraph. Hat tip Jonathan Oddie. bug fix: make C-c C-z behave with run-geiser 2011-09-08T21:15:36Z Jose Antonio Ortega Ruiz jao@gnu.org 2011-09-08T21:15:36Z urn:sha1:ebad1fd4e47d87d2780ef29ed209f3cd0fed928c The nice go-back-to-previous-scheme-buffer behaviour of C-c C-z wasn't working when the jump from a scheme file to the REPL was initiated via run-geiser. Thanks, Marijn. REPL: restore TAB indenting behaviour around whitespace 2011-08-20T20:47:53Z Jose Antonio Ortega Ruiz jao@gnu.org 2011-08-20T20:47:53Z urn:sha1:6752fa5cd559e49449ccb03e53c8d19fda7afb4b REPL: new customizable faces for input and prompt 2011-08-20T18:47:53Z Jose Antonio Ortega Ruiz jao@gnu.org 2011-08-20T18:47:53Z urn:sha1:0f441d62d9cd89e6e1486af1b4283abe77d30822 Namely, geiser-font-lock-repl-prompt and geiser-font-lock-repl-input. elisp: customizable patience amount on racket startup 2011-08-20T01:23:30Z Jose Antonio Ortega Ruiz jao@gnu.org 2011-08-20T01:23:30Z urn:sha1:7614e02ca555a166c29b110f83e7033339e0abd1 In this little notebook i'm using, racket takes its time to start. In fact, it can take more the previously slotted 10 seconds. Hence the new geiser-repl-startup-time variable. elisp: add-hook doesn't need a defvared hook 2011-08-02T15:07:27Z Jose Antonio Ortega Ruiz jao@gnu.org 2011-08-02T15:07:27Z urn:sha1:ad7434be210344a6fb90384008ad9653daf4590a Thanks, Leo. Irrelevant compilation warning fix 2011-06-22T22:40:37Z jao jao@gnu.org 2011-06-22T22:40:37Z urn:sha1:b0d6c15ad6801fb46093c85aa59e772373ca81a9 Using called-interactively-p instead of interactive-p, if you have to know. The latter is deprecated as of Emacs 23.2, which the lowest version supported by Geiser. Avoid (read) breakage (fixes #33090) 2011-06-22T22:32:27Z jao jao@gnu.org 2011-06-22T22:32:27Z urn:sha1:4396a8cf2d27c03589a60e14c1eb95a3ba21fff8 Autodoc was firing while the REPL was waiting for input of a (read) call, causing all kinds of misbehaviour. We now inhibit autodoc on sending a form for evaluation and re-inhibit it once a prompt is read back again.