Author: Raymond Toy
Author Date: 2022-12-05 01:58:34 UTC

Move some defvars from csimp.lisp to global.lisp

- Moved rns*, plogabs, %p%i, fourth%pi, half%pi, %pi2, and half%pi3.
- Removed special declarations no longer needed for these.

Author: Raymond Toy
Author Date: 2022-12-04 22:43:29 UTC

Move some defvars from suprv1.lisp to global.lisp

- Moved *refchkl*, *mdebug*, errcatch, lessorder,
  greatorder, *in-translate-file*, and *linelabel*.
- Removed special declarations no longer needed for these.

Author: Raymond Toy
Author Date: 2022-12-03 22:46:31 UTC

Remove no-longer needed calls to putprop

We moved these to defmvar, so these aren't needed anymore.

Author: Raymond Toy
Author Date: 2022-12-03 18:33:14 UTC

Set evfun property via defmvar in globals.lisp

In mlisp, the evfun property is set for some symbols defined in
globals.lisp. Instead, set this property via defmvar in globals.lisp,
and remove the variable from the list in mlisp.lisp.

Added docstrings too.

Author: Raymond Toy
Author Date: 2022-12-03 15:46:56 UTC

Defmvar supports setting properties

Update defmvar to support the option `:properties` which has the from
`((ind1 val1) (ind2 val2) ...)`, which sets the property `ind1` to the
value `val1` and so on for the variable.

Updated a couple of places in this file where there was a separate
putprop to define the property in the defmvar.

Author: Raymond Toy
Author Date: 2022-12-03 01:12:05 UTC

Move defmvars from inmis to globals

- Moved $listconstvars and $listdummyvars
- Removed from unnecessary declarations that are no longer needed.

Author: Raymond Toy
Author Date: 2022-11-25 17:18:16 UTC

Fix some typos and make asdf and defsystem closer

In maxima.asd split defmfun to its own module just like maxima.system
and update dependencies accordingly.

Fix a couple of typos in specifying the new "globals" module.

Author: Raymond Toy
Author Date: 2022-11-15 23:16:04 UTC

Rename ll and ul to *ll* and *ul*

Since ll and ul are declared to be special in defint.lisp, rename them
to *ll* and *ul* to make it clearer that these are special vars.

Author: Raymond Toy
Author Date: 2022-11-08 18:51:04 UTC

Revert code that wraps the options list neatly.

We'll do that in a separate change since it's not relevant to this
change.

Author: Raymond Toy
Author Date: 2022-10-29 14:47:13 UTC

Support makeinfo versions with 3 parts and simplify computation

Test versions of makeinfo have version numbers like 6.8.90. Update
configure to handle this so that we get a version of 6.8.90 instead of
8.90, which would be wrong.

We also simplify the tests for the version number by creating a
numerical value by multiply the major version by 100 and adding the
minor version. This assumes that the minor version is never larger
than 99. Then we can just compare integers. Thus 5.0 is version 500,
and that's easy to compare against 5.1 (version 501).

Author: Robert Dodier
Author Date: 2022-10-28 15:43:22 UTC

Minor cleanup issues in tests for lexical symbols.

 * src/lexical_symbols.lisp: new function 'unlexicalize',
   currently a convenience for testing, not sure if permanent

 * src/simp.lisp: special case for MDO and MDOIN (for-loops) in ALIKE1

 * tests/rtest2.mac: adjust tests and results to make some symbols
   global and unlexicalize others

Author: Raymond Toy
Author Date: 2022-10-16 00:42:13 UTC

Handle lisp args better for category program

The current code does
```
$lispprog $lispargs
```

But when lispargs is complicated, this doesn't actually pass the
desired lispargs correctly. For example for cmucl, we want to do
```
lispargs="-noinit -nositeinit -eval '(setf *gc-verbose* nil *default-external-format* :utf8)' -eval '(set-system-external-format :utf8 :utf8)' -load $catprogname -eval '(quit)'"
```

But the actual args are parsed in an unexpected way. This can be tested
via this simple program
```
#include <stdio.h>
int
main(int argc, char **argv)
{
        int i;

        for (i = 0; i < argc; i++) printf("argv[%d]='%s'\n", i, argv[i]);
        return 0;
}
```

Then
```
$ lispargs="-noinit -nositeinit -eval '(setf *gc-verbose* nil *default-external-format* :utf8)' -eval '(set-system-external-format :utf8 :utf8)' -load $catprogname -eval '(quit)'"
$ ./args $lispargs
argv[0]='./args'
argv[1]='-noinit'
argv[2]='-nositeinit'
argv[3]='-eval'
argv[4]=''(setf'
argv[5]='*gc-verbose*'
argv[6]='nil'
argv[7]='*default-external-format*'
argv[8]=':utf8)''
argv[9]='-eval'
argv[10]=''(set-system-external-format'
argv[11]=':utf8'
argv[12]=':utf8)''
argv[13]='-load'
argv[14]='-eval'
argv[15]=''(quit)''

```
Note that the `(setf *gc-verbose* nil)` is broken up into 3 separate
args which is not what we want. It should be just one arg, which is
the arg to `-eval`. What we really want is:
```
argv[0]='./args'
argv[1]='-noinit'
argv[2]='-nositeinit'
argv[3]='-eval'
argv[4]='(setf *gc-verbose* nil *default-external-format* :utf8)'
argv[5]='-eval'
argv[6]='(set-system-external-format :utf8 :utf8)'
argv[7]='-load'
argv[8]='-eval'
argv[9]='(quit)'
```

To fix this, we add a `runlisp` function to run the lisp with the
desired args. And instead of putting all the args into lispargs, we
just call runlisp after deciding which lisp is being used and just
append the desired args to the runlisp command.

Author: Raymond Toy
Author Date: 2022-10-10 19:59:15 UTC

Disable user init for abcl and acl

Also add comment that for each lisp, we want to disable loading of any
user init files.

Author: Wolfgang Dautermann
Author Date: 2022-10-01 11:33:38 UTC

Add quit() with an exit code to the german documentation.

Author: Raymond Toy
Author Date: 2022-09-20 23:52:28 UTC

Merge branch 'master' into rtoy-html-docs-for-langs

Author: Raymond Toy
Author Date: 2022-09-04 22:58:07 UTC

Ignore generated zeilberger.texi

Author: Raymond Toy
Author Date: 2022-09-02 15:02:48 UTC

Handle case for makeinfo version < 6

From Viktor Toth, via mailing list on Sep 2, 2022.

Author: Raymond Toy
Author Date: 2022-08-17 16:20:12 UTC

Use ptindex for the packages and new category macro for that

Because the packages are huge, we define a new index (pt) to hold the
packages. This requires using a new macro so we can find the right
links.

Author: Raymond Toy
Author Date: 2022-08-12 00:14:13 UTC

Update makeinfo version test to work with MacOS

The main problem was that on MacOS, expr didn't like the regexp
operator "+" in, say, "[0=9]+". This is easily fixed by using the
equivalent "[0-9][0-9]*".

We also added some more checks to make sure that makeinfo_version is
not empty and that makeinfo_major and makeinfo_minor are not empty
either. If any of these are empty, produce an error message.

Tested this on MacOS and verified that the warning messages are
produced (by using a faulty regexp for expr).

Author: Raymond Toy
Author Date: 2022-08-10 00:55:30 UTC

Fix typo and make sure --disable-build-docs works

First, fix typo: "{$makeinfo_old}" should be "${makeinfo_old}". That
was preventing us from signaling an error.

Second, make sure if --disable-build-docs is set, we don't try to test
for the makeinfo version. Do this by moving all of the code inside
the test for enable_build_docs being true.

Author: Raymond Toy
Author Date: 2022-08-07 15:41:15 UTC

Remove @math in the second arg in m4_math calls.

This is automatically done by m4_math now.

Author: Raymond Toy
Author Date: 2022-07-24 22:12:39 UTC

Update docs for output_format_for_help

Since describe_uses_html has been replaced by output_format_for_help,
update the docs to reflect this and describe what it does and the
values it can take.

Author: Raymond Toy
Author Date: 2022-07-18 00:12:44 UTC

Update intro to special functions

Instead of the example table there now, use an itemized list for each
subsection, and in each item, add a table of all the functions in the
subsection.

Not yet complete.

Author: Raymond Toy
Author Date: 2022-07-16 17:14:44 UTC

Add support for displaying localized html files

Update common-lang.mk to build maxima-index-html.lisp, including
installing and uninstalling it.

Update hdescribe to include the lang subdir when setting up $url_base.

I did a simple test by configuring with --enable-lang=pt. The set
LANG=pt_PT.UTF-8 (is that right?) and run maxima-local. Then "? sin"
brings up the html page for the sin function correctly.

Author: Raymond Toy
Author Date: 2022-07-15 23:16:15 UTC

Update more info formulas to use @example instead of @math

For the formulas that used @math, use @example to make them
displayed. This looks better in the info docs instead of having the
formula left-justified. Kind of matches how the formulas are
displayed in html and TeX too.

Author: Raymond Toy
Author Date: 2022-07-14 16:03:42 UTC

Add formulas for the mean, var, etc for the remaining distributions

This completes all of the continuous distributions.

Also added a short note that the mean, variance, skewness, and
kurtosis are undefined for the Cauchy distribution. That explains
where there's no function to compute these.

Author: Raymond Toy
Author Date: 2022-07-08 20:00:00 UTC

Fix up @math{nc_t(n,ncp)}

This is typeset in a funny way in html and TeX, with a subscript t,
and italicized "nc".

This is now rewritten to look like the function nc_t, roman chars and
no subscript, similar to how \sin is typeset.

Author: Raymond Toy
Author Date: 2022-07-05 22:26:15 UTC

Convert A&S and DLMF references to links

Use the @urlaands and @urldlmf macros to convert the references to
links.

Author: Raymond Toy
Author Date: 2022-07-02 19:25:12 UTC

Clean up some math expressions in vectorsimp

The html versions of the examples in vectorsimp (expandcrosscross,
expandcurlcurl, expandlaplaciantodivgrad) were messed up. Clean these
up show they display correctly.

Author: Raymond Toy
Author Date: 2022-06-30 22:09:52 UTC

Adds some tests for airy hypergeometrics and bigfloat eval

Tests that the hypergeometric representation is right and that
bigfloat eval is close to float eval.

Author: Raymond Toy
Author Date: 2022-06-29 20:26:21 UTC

Bfloat eval for bessel_y and bessel_k when order is not an integer.

Use the hypergeometric representation to evaluate bessel_y and
bessel_k when the order is not an integer. The representation is not
valid in that case.

Author: Raymond Toy
Author Date: 2022-06-28 22:32:36 UTC

Enable debugging print when enabled for hdescribe

Author: Raymond Toy
Author Date: 2022-06-24 21:08:40 UTC

Add macros for producing A&S and DLMF links

The macros urlaands and urldlmf produce nice URL links that we can
click on in the HTML and pdf manual. The info manual has the links
shown so the user can just cut-n-paste them to a browser.

Modified a couple references to A&S (Special.texi) and DLMF
(nset.texi) to use the macros to see how it looks.

Author: Raymond Toy
Author Date: 2022-06-15 16:02:36 UTC

Always use mathjax in HTML docs

We now always use MathJax for HTML docs so that formulas look nice.

o Remove the --enable-mathax option from configure.ac
o build_html.sh.in sets HTML_MATH to mathjax so texinfo will use
  Mathjax.
o texi2html.init.in no longer needs @mathjax_script@. Texinfo takes
  care of this for us.

NOTE: The current HTML docs look kind of bad now because displayed
formulas aren't and all HTML formulas are wrapped in "$$". This needs
to be fixed before we can deploy this.

Author: Robert Dodier
Author Date: 2022-06-06 05:46:19 UTC

Reorganize English-language documentation so that it is in doc/info/en,
as with German, Spanish, etc. in doc/info/de, doc/info/es, etc.

Author: Raymond Toy
Author Date: 2022-06-07 16:37:24 UTC

Update comment on when simplification applies

Update comment to match implementation which only simplifies if s is
rational and a is the integer 1.

Author: Raymond Toy
Author Date: 2022-06-05 16:43:07 UTC

Remove nregex.lisp

Not used anymore.

Author: Raymond Toy
Author Date: 2022-06-02 20:04:12 UTC

Make defun-check error out if name starts with $

Defun-check should not be used for user-visible functions. Defmfun
should be used for that. Defun-check doesn't add the properties that
defmfun does.

Also add simple docstring to compiler-macro.

Author: Robert Dodier
Author Date: 2022-05-28 00:25:07 UTC

In PSLQ-INTEGER-RELATIONS, reorder loop so that termination test goes before iteration step.

This fixes the bug report to the mailing list, that guess_exact_value(0.1)
triggers a division by zero error. In that case and related cases,
the solution is found by the initialization step,
and no iteration should be performed.

Division by zero errors were also triggered by guess_exact_value(x)
for x = 0.125, 0.2, 0.25, float(1/15), and float(1/17) (with Clisp,
other Lisp implementations might or might not behave the same).

Author: PeterPall
Author Date: 2022-05-23 05:29:48 UTC

Bumped the version number

Author: Raymond Toy
Author Date: 2022-05-04 01:56:09 UTC

Fix #3965: --list-avail fails for gcl

Basically just had to add "/" to the path, as we already do for clisp,
sbcl, ecl, and openmcl.

Tested this by building maxima and installing it in a temp directory
and running "maxima -l gcl --list-avail". This lists all of the lisps
I configured for running maxima.

Author: Raymond Toy
Author Date: 2022-04-27 16:13:12 UTC

Add more tests for zeta(-x)

Verify that float(zeta(-n/2)) is close to zeta(float(-n/2)) for
various values of odd n.

Author: Raymond Toy
Author Date: 2022-04-21 23:24:37 UTC

Fix #3967: elliptic_e(5*%pi/4,1) incorrect

elliptic_e(5*%pi/4,1) was returning 2-1/sqrt(2), but
elliptic_e(5*%pi/4,1),numer returned 2.707... or basically
2+1/sqrt(2).

This happens because we were simplifying elliptic_e(5*%pi/4,1) to
sin(5*%pi/4) = -1/sqrt(2). But we need to reduce the arg 5*%pi/4 by
%pi so we get sin(%pi/4) = 1/sqrt(2).

Add a test for this and that the symbolic and numerical values are
consistent.

Also needed to update the test for elliptic_e(x,1) to include the
reduction of the arg for the sin part.

Author: Raymond Toy
Author Date: 2022-04-19 16:40:26 UTC

Experiment to use reader macro to embed maxima expressions in lisp

We implement a reader macro (from Stavros) to allow embedding Maxima
expressions in Lisp code to make it easier to write.

So we can write

  (with-maxima-expr ""sin(?phi) + 2*round(?phi/%pi)*elliptic_ec(?m)")

instead of

(add (ftake '%sin phi)
     (mul 2
   (mul (ftake '%round (div phi '$%pi))
        (ftake '%elliptic_ec m))))

We add this macro to various places in ellipt.lisp for testing.

However some tests fail and this needs to be fixed before we can
generally allow this feature.

Author: PeterPall
Author Date: 2022-04-18 11:56:32 UTC

We need to package the lisp source file on "make dist".

Author: PeterPall
Author Date: 2022-04-18 08:52:06 UTC

Tried to repair the dist-info target

Author: Raymond Toy
Author Date: 2022-04-14 19:15:33 UTC

Remove commented out stuff for zeta

Clean up code, removing unneeded or commented-out stuff for zeta. The
def-simplifier for zeta is working.

Author: Raymond Toy
Author Date: 2022-04-13 22:30:13 UTC

Add tests for limit of li[n]

Author: Viktor T. Toth
Author Date: 2022-04-13 05:02:35 UTC

Updating configure.ac with new version number

Author: Raymond Toy
Author Date: 2022-03-10 22:35:36 UTC

Add test for li[3](-.862)

The imaginary part should be zero.

Author: Raymond Toy
Author Date: 2022-01-22 22:59:35 UTC

Fix a few grammatical issues in the comments.

Author: Raymond Toy
Author Date: 2022-01-06 19:31:16 UTC

Add a couple of tests for sech/csch/acsch.

Verifies that we don't overflow for the arguments where we used to
overflow unnecessarily.

Author: Robert Dodier
Author Date: 2021-11-19 03:58:05 UTC

Set version number to 5.45.9000 for testing content-dependency machinery.

Author: Raymond Toy
Author Date: 2021-11-14 17:57:49 UTC

Update tests for ecl-20.4.24

Author: Raymond Toy
Author Date: 2021-10-14 20:55:34 UTC

Convert bessel.lisp to use def-simplifier

Testsuite passes

Author: Raymond Toy
Author Date: 2021-10-14 20:05:17 UTC

Remove commented-out stuff

Author: Raymond Toy
Author Date: 2021-10-13 21:29:23 UTC

Remove commented-out stuff

Author: kjak
Author Date: 2021-10-09 20:23:17 UTC

translator: fix MQAPPLY array assignments when translate_fast_arrays:true

This is a very old Maxima bug that was not present in Macsyma.

In the translate_fast_arrays case, MASET did not handle the MQAPPLY
case. A lisp error would occur at runtime when attempting the
assignment.

(%i1) foo () := (a : make_array ('fixnum, 5), b() := a, b()[1] : 5)$

(%i2) translate_fast_arrays : true$

(%i3) translate (foo)$

(%i4) foo ();
<lisp error>

Now a special case for MQAPPLY has been added to MASET (similar to
the special case that exists in MAREF).

No problems with the test suite or share test suite. A new test has
been added to rtest_translator.

Author: Raymond Toy
Author Date: 2021-09-23 22:05:12 UTC

Print correct exponent if bfloat rounds up

As reported on the maxima list
(https://sourceforge.net/p/maxima/mailman/message/37353663/), some bfloats
aren't printed with the correct exponent if the number needs to be
rounded up and the rounding causes the exponent to change.

The solution here is to look at the number of digits that would be
printed. If it's greater than effective-printprec, then we must have
had to round the number up to the next power of 10, but we didn't
account for that in the exponent. So, when the number of digits is
not the expected length, increase the exponent by the number of excess
digits. We also need to remove the excess digits from the number so
that the printed result has the right number of digits (if bftrunc is
false).

A couple of simple tests are added to rtest3.mac.

Author: Raymond Toy
Author Date: 2021-08-26 23:05:22 UTC

Remove nil'ed out code.

Remove all the code that was replaced by def-simplifier.

Author: Raymond Toy
Author Date: 2021-08-24 22:48:52 UTC

Remove nil'ed out code.

Test suite still passes.

Author: Raymond Toy
Author Date: 2021-08-13 23:22:06 UTC

Comment out definitions of cons-exp

Comment out cons-exp but leave it for a bit until we're sure all uses
are gone. `git grep '(cons-exp'` shows that there are no longer any
occurrences.

Author: Raymond Toy
Author Date: 2021-08-08 22:42:54 UTC

Replace POW with POWER and remove POW.

We should use POWER everywhere instead of POW, since POW just
basically expands to POWER.

Testsuite runs fine with this change.

Author: Raymond Toy
Author Date: 2021-08-08 14:38:04 UTC

Need reversealias property for grind and plot labels

Grind and plot needs the reversealias property to be able to print out
the right thing. Without this, the result is the verb (%) form. We
really want the noun (%) form. This happens in msize-atom which will
strip the dollar sign off the noun form to produce the desired result.

Also added some comments on what the properties are need for.

Author: Raymond Toy
Author Date: 2021-08-04 16:06:27 UTC

Remove old simp functions

These were all replaced by def-simplifier, so we can remove the old
versions now.

Author: Raymond Toy
Author Date: 2021-08-02 20:07:29 UTC

Do arg count check first

The arg count check should be done right away instead of later.
There's no point in doing a simpcheck on args if the number of args is
incorrect.

Author: Viktor T. Toth
Author Date: 2021-06-24 04:01:00 UTC

Removed unnecessary sqrt from elliptic-ec.

Resolves GCL fail. Verified several test cases against prior versions with other LISPs.

(cherry picked from commit 1103b38c18c36523c910b506579f7ee5a726ca04)

Author: Robert Dodier
Author Date: 2021-06-16 06:39:02 UTC

Minor changes to Texinfo documentation for Gentran, in order to get it to build successfully.

Author: Raymond Toy
Author Date: 2021-05-08 15:21:28 UTC

First cut at implementing pejroot

This appears to work with real polynomial and real solutions. Need to
tweak the least-squares solver to work with complex numbers.

Author: Raymond Toy
Author Date: 2021-05-06 00:04:03 UTC

Fix typo.

Author: PeterPall
Author Date: 2021-04-03 07:43:27 UTC

wxMaxima's main branch now has a politically correct name

Author: PeterPall
Author Date: 2020-12-29 09:21:15 UTC

Windows installer: Updated the wxMaxima version

This resolves a small privacy issue: If the user still used the old
Internet Explorer 9 as help browser and opened the wxMaxima manual
the manual downloaded a HTML5 compatibility shim over the internet.

Debian deems downloading a file without asking the user first as a
privacy issue

(cherry picked from commit 2477f34f0fa126e17271ccde3d21a8eb0d20d1e5)

Author: Raymond Toy
Author Date: 2020-12-11 01:05:58 UTC

Merge branch 'master' into fix-incorrect-zeta-values

Author: Raymond Toy
Author Date: 2020-12-10 02:06:52 UTC

Add example where ierr = -1

Author: Raymond Toy
Author Date: 2020-10-22 00:11:15 UTC

Add link to fftpack5_fft from fft

Author: Raymond Toy
Author Date: 2020-10-14 00:36:48 UTC

Print docstring if exact match fails for documentation

When trying to describe something (via "? foo") and there's no exact
match, try to look up the docstring and return that. We print a note
that the exact match failed but there's a docstring. This is to let
the user know there is not such topic in the manual. But perhaps the
symbol has some lisp documentation attached, so it's useful if we can
display that.

You can test this by doing "load(hompack)", and then "?
hompack_fixpdf". There's no manual entry for hompack_fixpdf, but
there is a docstring for this function.

Author: Raymond Toy
Author Date: 2020-10-03 23:16:02 UTC

Support hompack

Author: Raymond Toy
Author Date: 2020-09-29 01:23:35 UTC

Update to MathJax 3

Author: Raymond Toy
Author Date: 2020-09-16 23:35:50 UTC

Define $nonnegintegerp with defmfun for better error checking

Author: Robert Dodier
Author Date: 2020-08-28 18:34:15 UTC

Merge /u/metayan/maxima/ branch ch-5.43 into branch-5_43

https://sourceforge.net/p/maxima/code/merge-requests/12/

Author: Raymond Toy
Author Date: 2020-04-14 23:32:22 UTC

Clean up cobyla interface

* Remove old implementation code
* Add docstring that was originally part of fmin_cobyla.mac
* Indent new code a bit better so the lines aren't so long

Author: Raymond Toy
Author Date: 2020-04-11 07:00:29 UTC

Convert run_testsuite to use defmfun.

Straighforward change to rename run-testsuite as $run_testsuite using
defmfun. Get rid of the original $run_testsuite which basically just
called run-testsuite anyway.

Author: Raymond Toy
Author Date: 2020-02-16 17:42:19 UTC

Fix typos in new tests

Author: Raymond Toy
Author Date: 2020-02-15 00:52:30 UTC

Remove special case for order k/2

This can be handled by the case of rational args and the results are
the same.

Author: Raymond Toy
Author Date: 2020-02-08 21:49:42 UTC

Update test to match new output form for gamma_incomplete

Author: Raymond Toy
Author Date: 2020-02-03 06:38:08 UTC

Add test from the mailing list

Author: Raymond Toy
Author Date: 2019-12-14 05:28:07 UTC

Merge branch 'master' into update-gentran

Author: Fabrizio Caruso
Author Date: 2019-12-04 07:52:37 UTC

Winding number

Author: Raymond Toy
Author Date: 2019-09-03 18:44:21 UTC

One more change from Tony Mc

Author: Raymond Toy
Author Date: 2019-08-12 01:50:44 UTC

Remove old commented out rule

Rule for building xmaxima.html updated to depend on manual.css.

Author: PeterPall
Author Date: 2019-06-20 17:44:02 UTC

A try to make GCL work: I suspect the dockerfile somehow doesn't use the ANSI version.

Author: Raymond Toy
Author Date: 2019-05-19 16:49:18 UTC

Use the new @subentry for subentries

We previously used a hack to get multi-level indices. Texinfo 6.6dev
now supports this via @subentry, so have the macros expand out to use
@subntry.

Author: Raymond Toy
Author Date: 2019-05-06 18:25:53 UTC

Merge branch 'master' into new-gentran

Author: PeterPall
Author Date: 2019-05-04 07:36:23 UTC

Windows Installer: Automatically find the newest wxMaxima release

Pro:
 - The maxima release no more needs to know the maxima version
   number
 - All updating wxMaxima in the Windows Installer requires from
   now on is building a new Windows Installer.

Contra:
 - Often all version numbers being fixed is an requirement, too.

As my private linux box only contains 4 Gigabytes of RAM and
8 Gigabytes are necessary for building an Windows Installer this
commit is entirely untested, unfortunately.

Author: PeterPall
Author Date: 2019-03-26 18:22:10 UTC

Created functions that return a search path including $load_directory

Author: PeterPall
Author Date: 2019-03-15 06:27:39 UTC

Tried to introduce a way to create a raw gnuplot programm.

Author: Elias Martenson
Author Date: 2019-03-02 15:33:33 UTC

Change calls to maxima-intern to use plain intern

Author: PeterPall
Author Date: 2019-02-24 14:38:42 UTC

Make maxima be built as a standalone application.

Author: PeterPall
Author Date: 2019-02-10 20:38:18 UTC

Now it seems to build a package.

Author: Wolfgang Dautermann
Author Date: 2019-01-25 15:48:06 UTC

Update included wxmaxima in the Windows installer.

(already included in the Windows installer, I released yesterday.
The version/md5-checksum did not match for wxmaxima 19.01.1 too...)