Re: [Bug-apl] Is this output incorrect?

Hi Jürgen: Well please help me properly understand the no_readline function in Input.cc. Because it's called instead of the readline library because of the --rawCIN. I've made comments of my interpretation of the code but am probably wrong. So I'd appreciate any contribution to my very rusty C++.

Re: [Bug-apl] Messed up indentation with )DUMP

Hello! Juergen Sauermann writes: > 1. the del editor (?) removes leading spaces from lines entered: > […] > 2. The ?-editor indents lines that are not comments or labels by one > space. This was not so earlier but was requested recently: > […] Ah, now I see. When using the Emacs mode, it seems

Re: [Bug-apl] New release?

Unless things settle down for a long while, I'm probably going to hold back the package manager release until GNU APL 1.4 drops. I'd like to be able to say that the package manager works with a binary release rather than requiring "latest SVN". Also, I'll second Blake's sentiment: GNU APL is great

[Bug-apl] ⎕TC presents as Unix, not APL

See IBM Reference, page 335. ⎕tc[2] and ⎕tc[3] differ from the IBM-specified behavior, presenting as Unix format effectors rather than as APL format effectors. In particular, APL specifies that ⎕tc[2] should advance the print position to the start of the next line and ⎕tc[3] should advance the pr

Re: [Bug-apl] quad-ET doesn't follow SI

Thank you. Confirmed working as expected. On Sat, 2014-05-31 at 14:32 +0200, Juergen Sauermann wrote: > Hi David, > > thanks, fixed in SVN 304. > This changed some data structures, so make install is needed. > > /// Jürgen > > > On 05/29/2014 05:45 AM, David B. Lamkins wrote: > > quad-ET shoul

Re: [Bug-apl] quad-EC fails with function that doesn't return a value

Thank you. Confirmed working as expected. On Fri, 2014-05-30 at 18:46 +0200, Juergen Sauermann wrote: > Hi David, > > thanks, fixed in SVN 303. > > /// Jürgen > > > On 05/29/2014 05:06 AM, David B. Lamkins wrote: > >⎕fx 'bar' '1' > > bar > >⎕ec 'bar' > > 1 > > unexpected result

[Bug-apl] ⎕EC return values are incorrect for user-defined errors

See the IBM Reference, page 280, paragraph 3. When ⎕EC executes an expression that signals a user-defined error, the second and third items of the result should be the same as the expression's ⎕ET (which would be 0 1 in this case) and ⎕EM. GNU APL returns 0 0 and 'User defined error'. Note also t

Re: [Bug-apl] Boehm garbage collector

Hi, I have implemented the linked list of deleted values as discussed below. SVN 305. If you want to benchmark it change line 509 of Value.hh to enable (#if 1) or disable (#if 0) the new memory management. /// Jürgen On 05/31/2014 03:25 PM, Juergen Sauermann wrote: Hi, in GNU APL the on

Re: [Bug-apl] New release?

Hi Blake, in principle yes. Normally I wait until the number of emails on bug-apl goes down a little (which wasn't the case lately). Creating a release is more work than just committing a bug-fix to SVN. I had also planned a bit more for the next release - multi-core and better documentation

[Bug-apl] New release?

Greetings, Just a thought - IMO, GNU APL has come a long way since its last release. It is quite a bit more usable than the last release. Perhaps a new release is in order? If not a release now, perhaps there should be a short list of items before a new release. Again, just raising the issue.

Re: [Bug-apl] Is this output incorrect?

Hi Peter, looks OK to me. (Note: as of recently *apl -s* does the same as your command line options). --rawCIN reads directly from the file (stdin in this case) without outputting any prompts. The line numbers in the ?-editor count as prompts and are therefore suppressed as well. However w

Re: [Bug-apl] Messed up indentation with )DUMP

Hi Daniel, I cant confirm this but would need additional information. A few hints, though: 1. the del editor (?) removes leading spaces from lines entered: * ?foo [1] 1 [2] 2 [3] 3 [4] [?] ? [0] foo [1] 1 [2] 2 [3] 3 ? [4] ?* This is a bit strange but for IBM APL2 comp

[Bug-apl] Is this output incorrect?

Please note the command line options. My question is: Should the following be (a) indented by 6 spaces and (b) Nabla::edit ought to provide line numbering? respect…. Peter

Re: [Bug-apl] Limit what )LIB shows

Hi Blake, correct. And I believ it doesn't. )load xxx.apl loading )DUMP file /home/eedjsa/projects/juergen/apl-1.3/src/workspaces/xxx.apl... WAS CLEAR WS )load xxx.xml SAVED 2014-5-28 18:3:45 (GMT+2) /// Jürgen On 05/31/2014 04:17 PM, Blake McBride wrote: Dear Juergen, Okay.

Re: [Bug-apl] Limit what )LIB shows

Dear Juergen, Okay. I see. But this does lead to one more question. Presumably, if a .apl and .xml file exists for the same workspace _and_ the user specifies the full file name (with the .apl or .xml) then you wouldn't have to issue an error about the two versions of the same ws since the user

Re: [Bug-apl] Limit what )LIB shows

Hi Blake, In MS windows, file extensions are everything (so important that they have spent extra efforts to hide them for the dumb user). In Linux they are more like a convention but not a must. GNU APL searches first for a file with the name provided by the user and only looks for .xml or .ap

Re: [Bug-apl] Revisiting the APL 2 empty-array display behavior

Hi Jay, as the GNU APL info manual says: /GNU APL supports direct functions (aka. lambdas), but only in a rather// //limited form./ No plan to change this. /// Jürgen On 05/29/2014 04:44 PM, Jay Foad wrote: On 29 May 2014 15:28, Juergen Sauermann > wrote

Re: [Bug-apl] Revisiting the APL 2 empty-array display behavior

Hi, I believe the dyadic ⍕ with character left argument (called *Format by Example* in IBM APL2) is more powerful than ⎕FMT (which is a bit too FORTRAN-like for my taste) and is already implemented in GNU APL. And then we have good old printf() in FILE_IO. /// Jürgen On 05/29/2014 04:33 PM

Re: [Bug-apl] Boehm garbage collector

Hi, in GNU APL the only performance relevant data structures are APL values. There are two kinds of values: small values and large values. small and large refer to the number of ravel elements, i.e. *?,VALUE*. The borderline between small and large is ./configurable (SHORT_VALUE_LENGTH_WANTED,

Re: [Bug-apl] quad-ET doesn't follow SI

Hi David, thanks, fixed in SVN 304. This changed some data structures, so make install is needed. /// Jürgen On 05/29/2014 05:45 AM, David B. Lamkins wrote: quad-ET should always reflect the error that caused the top suspension on the SI. When the SI is empty, quad-ET should yield 0 0.