Am 13.05.2015 um 09:41 schrieb Sven Schreiber: > Hi, > > a new thingy: > > <hansl> > function void scheck (int a, string *s) > printf "%s\n", s > end function > > string hey = "ho" > scheck(2, &hey) > </hansl> And I just discovered that not using the pointer form seems to work with a null default value: <hansl> function void scheck (string s[null]) if isnull(s) print "null" else print s endif end function scheck() </hansl> I'm mentioning this because according to the docs under the heading "Function parameters: optional refinements" it says: "For optional pointer and list arguments (see section 13.4), the special default value null." So this case isn't covered by the description.

OTOH if I try this null value with an integer argument (and no pointer), it indeed does _not_ work.

a new thingy: <hansl> function void scheck (int a, string *s) printf "%s\n", s end function string hey = "ho" scheck(2, &hey) </hansl> produces: <output> gretl-Version 1.10.90cvs Aktuelle Sitzung: 2015-05-13 09:38 ? function void scheck (int a, string *s) > function void scheck (int a, string *s) Unrecognized data type 'string *' Fehler bei Skriptausführung: Stopp > function void scheck (int a, string *s) </output> Apart from the untranslated error message (could by my own fault)

apparently gretl doesn't accept strings to be passed in pointer form? Is this deliberate? AFAICS the docs with respect to pointers just talk about function arguments in general, not just scalars and matrices.

Am 13.05.2015 um 20:10 schrieb Allin Cottrell: > On Wed, 13 May 2015, Sven Schreiber wrote: > > >> apparently gretl doesn't accept strings to be passed in pointer form? >> Is this deliberate? AFAICS the docs with respect to pointers just >> talk >> about function arguments in general, not just scalars and matrices. > > What we say is (Guide, chapter 13), "Each of the type-specifiers, with > the exception of list and string, may be modified by prepending an > asterisk to the associated parameter name" to get a pointerized > version. > Right, fair enough. Note however that at the top of section 13.4 it says "Series, scalar, and matrix arguments to functions can be passed in two ways: “as they are”, or as pointers" Since I'm successfully passing around bundles in pointer-form (as also done in section 10.7 of the manual), the bundle type is missing in this enumeration I think. (It's because of these slight glitches that it's not always 100% clear to me whether the omission of string types in this list is intended or not.)

Am 14.05.2015 um 08:17 schrieb Riccardo (Jack) Lucchetti: > Il 2015-05-13 22:21 Sven Schreiber ha scritto: >> >> Since I'm successfully passing around bundles in pointer-form (as also >> done in section 10.7 of the manual), the bundle type is missing in this >> enumeration I think. >> >> (It's because of these slight glitches that it's not always 100% clear >> to me whether the omission of string types in this list is intended or >> not.) >> > > This is a fantastic job you're doing, Sven, thanks for that. Would it be > possible for you to send us directly appropriate patches to the LaTeX > source files? > Well thank you for your kind if not exaggerated words. In principle yes it would be possible, in practice I have to dig out the old emails to remind me again how to prepare the patches, and to check if it can be done easily on Windows or if I have to fire up a Linux machine for that.

Am 15.05.2015 um 00:58 schrieb Riccardo (Jack) Lucchetti: > > Your best bet is to download thge wold "doc" directory, possibly from > here: http://gretl.cvs.sourceforge.net/viewvc/gretl/gretl/doc/ > > and then edit the files in the "tex" subdir. I'm attaching a very small diff-file related to the file functions.tex in that subdir. Since it's been so long since my last diff-ing I hesitate to commit it myself, can you check the file (also the contents) and do it for me?