forums.silverfrost.com

[steveDoyle]

HI Paul

I am getting some odd behaviour with the %WB control.
it reads and displays the test web page as expected. when i press the [OK] button the program crashes. the feedback message seem to refer to the button (%bt) but it works without issue if i comment out the %wb line
it crashes on win xp & 10
Steve

SUBROUTINE WEB_READ_TEST
IMPLICIT NONE
INCLUDE 'windows.inc'
CHARACTER*256 FPATH
INTEGER LF, LENGU

FPATH = 'http://epsassets.manchester.ac.uk/medialand/' //
+ 'cpi/vol1/software/cpi_ver_num.htm'
c actual length not LEN()
LF = LENGU(FPATH)
c lf= 76

IWIN = WINIO@('%ca[Web file read test]&')
IWIN = WINIO@('%60.20wb[' // FPATH(1:LF) // ']&')
IWIN = WINIO@('%ff&')
IWIN = WINIO@('%bt[OK]&')
IWIN = WINIO@('%ff')

END

Runtime error from program:c:\apps_cpi\bin\sprint.exe
Access Violation
The instruction at address 0e6a4523 attempted to read from location 00000000
0e6a3da1 constructed_button_procedure@16#70 [+0782]
7e418734 routine at address 7E418734 [+0000]
7e418816 routine at address 7E418816 [+0000]
7e428ea0 routine at address 7E428EA0 [+0000]
7e428eec routine at address 7E428EEC [+0000]
7c90e473 routine at address 7C90E473 [+0000]
7e4196c7 routine at address 7E4196C7 [+0000]
0e638706 __yield_program_control [+0221]


eax=ef011180 ebx=0000000f ecx=0e5fd1c4
edx=7c90e514 esi=0e6a3da1 edi=00000000
ebp=0e5fd260 esp=0e5fd1d4 IOPL=0
ds=0023 es=0023 fs=003b
gs=0000 cs=001b ss=0023
flgs=00010216 [NC EP NZ SN DN NV]

0e6a4523 mov ecx,[edi]
0e6a4525 cmp ecx,[ebp-0x20]
0e6a4528 jnle e6a4388

[PaulLaidler]

Steve

It works OK for me although I have had to make some minor changes. For example I changed LENGU to LENG and I changed to free format for simplicity.

%wb has been available for a number of years but it is quite possible that it won't work on XP. It should work on Windows 10.

[John-Silver]

Paul,
Steve wrote:
it crashes on win xp & 10

..., so your it shuld work on XP and Win 10 doesn't seem to be 100% valid.
have you tried on Win XP for example ?
_________________
''Computers (HAL and MARVIN excepted) are incredibly rigid. They question nothing. Especially input data.Human beings are incredibly trusting of computers and don't check input data. Together cocking up even the simplest calculation ... "

[steveDoyle]

HI

I initially suspected that it was related to the contents of the .HTM file but
on further investigation i found it works as expected (win XP) when it is a simple standalone code.

When exactly the same code is called from inside a complex application(s) it crashes. When it is called from a slightly less complex program it works as expected.

I guess my complex applications (stable for 20+ years) have shared a fundamental fault which is starting to reveal itself.

thanks

Steve