Hi Maurizio,
Could you report results using cpinfo and Clipper 5.2e (as opposed to
'd')?
Brgds,
Viktor
On 2009 Oct 7, at 14:13, Maurizio Faccio adinet wrote:
Hi
I've started the bug report.
//INFO report.
Clipper (R) 5.2d Intl. (Rev. 215) SPANISH - SPANISH DBFNTX Collation
CauseWay 386 DOS extender version 3.49 in use.
Active host is DPMI.
Virtual Memory Manager is not active.
I am using the clipper without using any NTX*.OBJ, the native
Spanish version of clipper
I will run the CPINFO and send you the resultant txt
The test that I have made was:
- create a database with this structure without indexes.
ASCII N 3 0
CLICHAR C 1 0
CLIUPPER C 1 0
CLILOWER C 1 0
CLIUPNRO N 3 0
CLILONRO N 3 0
HRBCHAR C 1 0
HRBUPPER C 1 0
HRBLOWER C 1 0
HRBUPNRO N 3 0
HRBLONRO N 3 0
I fill the database with ascii 1 to 255
Then I've wrote two programs that write the char, upper, lower char,
and the ascii number of each character, one in clipper and one in
harbour.
while !eof()
FIELD->clichar:=CHR(nronro)
FIELD->cliupper:=UPPER(CHR(nronro))
FIELD->clilower:=LOWER(CHR(nronro))
FIELD->cliupnro:=ASC(UPPER(CHR(nronro)))
FIELD->clilonro:=ASC(LOWER(CHR(nronro)))
DBSKIP()
ENDDO
Then I filter wich clipper and harbour UPPER and LOWER generates
different characters.
Thamk you
Viktor Szakáts escribió:
Hi Xavi,
Our reference compiler version is 5.2e. Results
for older versions/revisions don't matter for Harbour.
Since we also support 5.3b, we may also try to
replicate it as an extra (if possible to solve
in a reasonable way), so such tests are also
useful to see.
Brgds,
Viktor
On 2009 Oct 7, at 13:02, Xavi wrote:
Chen,
But the whole idea of Harbour' codepage ES850C was to supply a
driver that is
Clipper compatible (this is the "C" in the codepage name) to
allow such mutual
updates.
Maybe for don't have cpinfo.
Only by changing one version, the collation sequence is different .-
10/06/09 17:45:02 DOS 5.00 Clipper (R) 5.2d Intl. (x215)
(1994.03.25)
Character encoding: NTXSORT v1.21e x15 24/Jun/93
==================================================
character '˜' (151:A L ) is the same as upper and lower
number of upper and lower characters is different
Warning: irregular CP which needs special definition in Harbour
upper: "ABCDE∆FGHIJKLMN¥OPQRSTUVWXY˚Z∂⁄˙"
lower: "a bcde≠fghi¡jklmn¤o¢pqrstu˜vwxy™z∫≥‰"
==================================================
and may use any version of National Language Drivers .-
10/06/09 17:43:55 DOS 5.00 Clipper (R) 5.01
Character encoding: NATSORT v1.2i x14 19/Mar/93
==================================================
simple byte sorting !!!
character '˜' (151:A ) not defined as upper or lower
number of upper and lower characters is different
letters are not sorted continuously
Warning: irregular CP which needs special definition in Harbour
upper: "ABCDEFGHIJKLMNOPQRSTUVWXYZ⁄∂∆˙˚¥"
lower: "abcdefghijklmnopqrstuvwxyz™≠≥∫‰ ¡¢¤"
==================================================
Not mention of MDXSORT.
To allow such mutual updates we need a codepage for each.
What is the solution, change ES850C to a single specify version?
--
Xavi
_______________________________________________
Harbour mailing list
Harbour@harbour-project.org
http://lists.harbour-project.org/mailman/listinfo/harbour
_______________________________________________
Harbour mailing list
Harbour@harbour-project.org
http://lists.harbour-project.org/mailman/listinfo/harbour
<ES850C.DBF>_______________________________________________
Harbour mailing list
Harbour@harbour-project.org
http://lists.harbour-project.org/mailman/listinfo/harbour
_______________________________________________
Harbour mailing list
Harbour@harbour-project.org
http://lists.harbour-project.org/mailman/listinfo/harbour
