The option "--enable-64bit" is only necessary for some platforms:
https://core.tcl-lang.org/tcl/tktview/1230558fffffffffffff

When entier's arg had a mode of float or number, the translator has
attempted to check if the arg is coming from a sqrt call and
incorrectly assumed it could take the car of the translated arg.

(%i1) foo () := entier (2.0)$

(%i2) translate (foo);
<lisp error>

Even if the translated arg was a cons, the sqrt check is outdated and
wouldn't do anything useful.  The translations for entier and friends
need work, so just remove this bogus case.

No problems with the test suite, share test suite or rtest_translator.
New tests have been added to rtest_translator.

Applying patches created by Leo Butler and Tomio Arisaka, as noted in:
https://sourceforge.net/p/maxima/bugs/4424/

Also include the examples in the bug report in share/draw/rtest_draw.mac.

Fixes SF bug SF bug #4424: "draw2d doesnt accept fill_density in order to create opaque shapes"

about shadowing the Common Lisp functon MERGE.

Also moved the entries related to the test suite at the end of the
section.  I think it makes more sense this way since the section is
about bug reporting, so the bug reporting stuff should be
first.  (This is really only relevant if you're reading the section in
the manual in its entirety.   Help should only show exactly the
desired help item, of course.)

For each of the trig and hyperbolic functions, add a simple (common)
sentence that adds cross references to the variables that control
simplification: %piargs, %iargs, halfangles, triginverses, and
trigsign.

We use define macro, `xreftrigvars`, here so that all of the
references are consistent.

Gnuplot 6 is part of Ubuntu 24.04 (and probably other Linux
distributions) since several month and it seems, Maxima works
with Gnuplot 6 without problems. Did some tests here (on
Linux and also on Windows, of course) and found no severe
issues.

This commit introduces a workaround for bug #4416. A proper fix likely requires changes to the Taylor code.

    Workaround in tlimp:
        Modified `tlimp` to avoid applying tlimit on expressions containing `asin` or `acos`. Such expressions are instead handled
        by `limit1`, which computes the limit correctly for this case.

    Simplification of `tlimit-taylor`:
        Removed the call to `partial-logarc`. Eliminated the local definition of `taylor_simplify` to simplify the code.

    Append tests to rtest_limit_extra.mac

These changes fix `rtest_limit_extra.mac` test 256, so remove this as a known failure.

In letsimp, reorganize expressions (via NFORMAT) so that minus expressions are represented as ((MMINUS) FOO)
instead of ((MTIMES) -1 FOO) (the usual, simplified form).

Fixes SF bug #4417: "letsimp treats minus expressions as -1 times something"

mat_norm takes two args, not one.  The second arg should be 1, which
is what was used in the 5.47.0 manual.  (Not sure how this got
changed.)

Forgot closing "$" for `load("lapack")` and the matrix should be named
`M`, not `A` since the rest of the example references `M`.

Fix #4413

Test that transpose_b works, and add one more test that tranpose_a
works with all the other options.

rtest_unit_usersetunits.mac must be run in a clean session
(without the unit package being loaded already).

Both rtest_unit.mac and rtest_unit_usersetunits.mac should probably
be run by themselves (not with other test scripts) since the
unit package redefines some global functions.

We mostly want to test that the keyword arg `beta` works now, but we
might as well add a few more tests for dgemm.  The tests are based on
the matrices in the user doc for dgemm.

Also, when matching Maxima keyword args, call $verbify to convert any
nouns to verbs since the list of valid keywords are all verbs.

that results are the same as if typed in at the console input prompt,
since TOPLEVEL-MACSYMA-EVAL is applied to user inputs, and the unit package
implements its built-in conversion functions via TOPLEVEL-MACSYMA-EVAL.

There are fewer test case failures, with this commit, but there are still
a number of failures.

Add nice Mathjax equations for the minpack docs.  Clean up some of the
text as well to clarify things.

No problems with the test suite, share test suite or rtest_translator.
New tests have been added to rtest_translator.

Since TRANSLATE-WITH-FLONUM-OP has recently been extended to handle
numerical functions that take multiple args, we can now translate
atan2 in transf instead of having a separate simple translator in
troper.

See also commits 0c3b407a and d5489b33.

No functional changes.

No problems with the test suite, share test suite or rtest_translator.
New tests have been added to rtest_translator.

Also provide a description of the default value of tolerance.

Merge all the different calling sequences into one, which is more
accurate, and, I think, better.

Also provide a description of the default value of tolerance.

Merge all the different calling sequences into one, which is more
accurate, and, I think, better.

This extends TRANSLATE-WITH-FLONUM-OP to handle basic numerical
functions that take multiple args.  This will be used in upcoming
commits.

No functional changes.

No problems with the test suite, share test suite or rtest_translator.

No functional changes.

No problems with the test suite, share test suite or rtest_translator.

It was kind of hard to figure out the return value from the functions.
Make it a list of the items returned so it's a bit clearer what each
element is.

Also added templates for running the examples using the
update-examples script.

rtest_sign.mac: test case for SF bug #2910: "In maxima 5.35.1 powerseries fail if there complex root"
(already fixed by commit 0d2e681d).