Hi,
here again a small patch to wv/text.c. First a small confession:
the 'case range' foul-up was mine. Realized too late it was a
gnu-ism :-(.
It still contains an off-by-one error, which is fixed here.
I also moved a comment block.
Also code to handle "soft-hyphens" (i.e. preferred word break
indicators that become hyphens only on line breaks, and
otherwise vanish; \- in LaTeX). It appears that the Word test
files I have use 0x2013 for this, but also for "en-dash" (the
official meaning), -- in LaTeX. Two completely different
symbols using the same code. Word renders them correctly and
differently. What does Word know that wv doesn't?
I don't understand this.
Martin
--
Martin Vermeer [EMAIL PROTECTED] [EMAIL PROTECTED]
-- Linux means: never having to delete your love letters
Index: text.c
===================================================================
RCS file: /cvsroot/wv/text.c,v
retrieving revision 1.42
diff -u -r1.42 text.c
--- text.c 2000/07/11 16:40:37 1.42
+++ text.c 2000/07/13 17:48:23
@@ -533,12 +533,28 @@
int wvConvertUnicodeToLaTeX(U16 char16)
{
- if( (char16 >= 0x9f) && (char16 <= 0xff) )
- {
- printf("%c", char16);
- return(1);
- }
+ /*
+ german and scandinavian characters, MV 1.7.2000
+ See man iso_8859_1
+
+ This requires the inputencoding latin1 package,
+ see latin1.def. Chars in range 160...255 are just
+ put through as these are legal iso-8859-1 symbols.
+ (see above)
+ Best way to do it until LaTeX is Unicode enabled
+ (Omega project).
+ -- MV 4.7.2000
+
+ We use a separate if-statement here ... the 'case range'
+ construct is gcc specific :-( -- MV 13/07/2000
+ */
+
+ if ( (char16 >= 0xa0) && (char16 <= 0xff) )
+ {
+ printf("%c", char16);
+ return(1);
+ }
switch(char16)
{
case 37:
@@ -551,7 +567,18 @@
case 31:
case 45:
case 0x2013:
- printf("--"); /* en-dash */
+ /*
+ soft-hyphen? Or en-dash? I find that making
+ this a soft-hyphen works very well, but makes
+ the occasional "hard" word-connection hyphen
+ (like the "-" in roller-coaster) disappear.
+ (Are these actually en-dashes? Dunno.)
+ How does MS Word distinguish between the 0x2013's
+ that signify soft hyphens and those that signify
+ word-connection hyphens? wvware should be able
+ to as well. -- MV 8.7.2000
+ */
+ printf("\\-");
return(1);
case 12:
case 13:
@@ -571,16 +598,6 @@
printf(">");
return(1);
- /*
- german and scandinavian characters, MV 1.7.2000
- See man iso_8859_1
-
- This requires the inputencoding latin1 package,
- see latin1.def. Chars in range 159...255 are just
- put through as these are legal iso-8859-1 symbols.
- -- MV 4.7.2000
- */
-
case 0x2019:
printf("'"); /* Right single quote, Win */
return(1);
@@ -657,7 +674,7 @@
case 31:
case 45:
case 0x2013:
- printf("--"); /* en-dash */
+ printf("-"); /* en-dash */
return(1);
case 12:
case 13:
