The end condition in the binary search loop in the new getType() was

incorrect. I must have written that binary search idiom a hundred
times, so it's rather embarrassing that I can't _automatically_ get
it right! This was causing all kinds of characters to be classified
as ON when they should have been various other classes.

Also while I'm here, I've added another test case to utf8.txt (a
small piece of Arabic within a predominantly L->R line), and also
supplied a means to compile minibidi.c with -DTEST_GETTYPE to
produce a command-line character class lookup tool. (Not sure what
use that'll be _other_ than debugging this precise problem, but I
don't like to throw it away now I've written it :-)

[originally from svn r5016]

diff --git a/minibidi.c b/minibidi.c
index c13276d0065e0c78dcb66ffd55be8a94f117d5f9..d36dee07ec36483c5155504c68424cdff9e289cd 100644 (file)
--- a/minibidi.c
+++ b/minibidi.c
@@ -38,6 +38,14 @@
 #define OISL   0x80    /* Override is L */
 #define OISR   0x40    /* Override is R */
 
+/* For standalone compilation in a testing mode.
+ * Still depends on the PuTTY headers for snewn and sfree, but can avoid
+ * _linking_ with any other PuTTY code. */
+#ifdef TEST_GETTYPE
+#define safemalloc malloc
+#define safefree free
+#endif
+
 /* Shaping Helpers */
 #define STYPE(xh) ((((xh) >= SHAPE_FIRST) && ((xh) <= SHAPE_LAST)) ? \
 shapetypes[(xh)-SHAPE_FIRST].type : SU) /*))*/
@@ -848,7 +856,7 @@ unsigned char getType(int ch)
     i = -1;
     j = lenof(lookup);
 
-    while (j - i > 2) {
+    while (j - i > 1) {
        k = (i + j) / 2;
        if (ch < lookup[k].first)
            j = k;
@@ -1810,3 +1818,47 @@ void doMirror(wchar_t* ch)
        }
     }
 }
+
+#ifdef TEST_GETTYPE
+
+#include <stdio.h>
+#include <assert.h>
+
+int main(int argc, char **argv)
+{
+    static const struct { int type; char *name; } typetoname[] = {
+#define TYPETONAME(X) { X , #X }
+       TYPETONAME(L),
+       TYPETONAME(LRE),
+       TYPETONAME(LRO),
+       TYPETONAME(R),
+       TYPETONAME(AL),
+       TYPETONAME(RLE),
+       TYPETONAME(RLO),
+       TYPETONAME(PDF),
+       TYPETONAME(EN),
+       TYPETONAME(ES),
+       TYPETONAME(ET),
+       TYPETONAME(AN),
+       TYPETONAME(CS),
+       TYPETONAME(NSM),
+       TYPETONAME(BN),
+       TYPETONAME(B),
+       TYPETONAME(S),
+       TYPETONAME(WS),
+       TYPETONAME(ON),
+#undef TYPETONAME
+    };
+    int i;
+
+    for (i = 1; i < argc; i++) {
+       unsigned long chr = strtoul(argv[i], NULL, 0);
+       int type = getType(chr);
+       assert(typetoname[type].type == type);
+       printf("U+%04x: %s\n", chr, typetoname[type].name);
+    }
+
+    return 0;
+}
+
+#endif

diff --git a/testdata/utf8.txt b/testdata/utf8.txt
index 7ad058f6aa61864d70251729d7ceea0031e46336..a8a87f9b4ef07dfec1674dc42adf12c03f5c168a 100644 (file)
--- a/testdata/utf8.txt
+++ b/testdata/utf8.txt
@@ -18,3 +18,4 @@ Arabic and bidirectional text:
             (من مجمع الزوائد ومنبع الفوائد للهيثمي ، ج 1 ، ص  74-84)           
                عن \e[44mجرير\e[m \e[41mرضي\e[m الله عنه قال قال رسول الله صلى الله عليه
           وسلم: بني الاسلام على خمس شهادة ان لا اله الا الله واقام              
+Mixed LTR and RTL text: \e[44mجرير\e[m \e[41mرضي\e[m back to LTR.