Re: Getting rid of varbrace

Hi Sam,

> we need either a comment or a thorough purge.
> I favor the latter because I want the CLISP sources to _look_ like
> regular C.

Me too. Though I got the impression from our discussions last year
that a full purge - due to the massive commit this would cause - is
not wanted. That's why I went the `#define var` way.


> Note that the purge can be done with a perl one-liner.

Yes, but bear in mind that the purge is only the first step.  Removing
var causes problems.  In the second patch of the original mail I only
corrected those which caused compile errors for my configuration.
There are possibly many more of them when looking through all the
code.  Examples:

A declaration cannot immediately follow an label. There are LOTS of
those errors. Example:

label:
  var type name;

I solved this like so:

label:;
  var type name;


> No, please add braces instead.

This might work in some cases, but most of the affected code is highly
"unstructured". If I could easily add braces without rewritting it,
then there wouldn't be labels and gotos but if, while and friends.

I tried to find more instances of those with:

find src -type f -name '*.d' -exec \
  pcregrep -MHno \
    '[[:alnum:]]+:([^:;{]|\n)*?var([^;]|\n)*+;' '{}' ';' \
    2>/dev/null \
  | sed 's_^src/_\x0src/_g' \
  | grep -z -v -F ': */' \
  | less

But this shows many false positives...


Another issue is conflicting definitions of the same variable. Example:

var type1 foo;
// some code
var type2 foo;

This is solved by renaming one of them, if possible. In two places, I
manually added braces (like varbrace would've done), specifically in
src/intlog.d, function I_logcount_I, the variables with the name (as
written in the source code) "x16". The issue is that "x16" might be
"#define" to something else, and I didn't exactly understood why and
when.


> Please either make a separate patch or tell me where the
unused var is.

I attached a patch with the small change (including an entry in the
changelog, just to be sure ;) )


> I will let Bruno decide on this.

Bruno, what do you think?

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot

_______________________________________________
clisp-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/clisp-devel

# HG changeset patch
# User Daniel Jour <[email protected]>
# Date 1490300140 -3600
#      Thu Mar 23 21:15:40 2017 +0100
# Node ID 2bf1af7c25525b264405d361aa6759f1059f1438
# Parent  86e4c0dc1cfadf2ddb4a95c698d0105c1789c71f
remove unused variable "size" from src/array.d (make_bit_vector)

diff -r 86e4c0dc1cfa -r 2bf1af7c2552 src/ChangeLog
--- a/src/ChangeLog	Wed Mar 22 17:05:24 2017 -0400
+++ b/src/ChangeLog	Thu Mar 23 21:15:40 2017 +0100
@@ -1,3 +1,7 @@
+2017-03-23  Daniel Jour <[email protected]>
+
+	* src/array.d (make_bit_vector): Remove unused local variable "size".
+
 2017-03-22  Sam Steingold  <[email protected]>
 
 	* makemake.in (check-script): Add a test for -x raising an error.
diff -r 86e4c0dc1cfa -r 2bf1af7c2552 src/array.d
--- a/src/array.d	Wed Mar 22 17:05:24 2017 -0400
+++ b/src/array.d	Thu Mar 23 21:15:40 2017 +0100
@@ -5156,7 +5156,6 @@
 
 LISPFUNN(make_bit_vector,1)
 { /* (SYS::MAKE-BIT-VECTOR size) returns a Bit-Vector with size bits. */
-  var uintL size;
   if (!posfixnump(STACK_0)) {
    bad_size:
     /* STACK_0 = size, TYPE-ERROR slot DATUM */