(All) #964
ddir/dprodir would display "datax3" rather than "data"
(All) #967
In the memo editor, using the up and down arrows, if the cursor
is in a column on the source line beyond the last column of the
destination row, the cursor may be placed in the wrong location.
(*nix only) #968
When running without the license manager under Unix/Linux, *report
and *cabe would not release the session upon exit.
(All) #969
TAB characters in help files were not properly handled.
(Sun) #1011
Files > 2gb cannot be accessed on SUN OS
(All) #975
Added the following flags to swapcpu:
-Q - Quiet mode. Supresses "file already in destination order"
messages.
-CN - Convert to native byte order.
-CF - Convert to foreign byte order.
(All) #976 swapcpu index already swapped crash segv
Running swapcpu on a file with an index already in destation byte
order could crash swapcpu.
(All) #992
The following scenario corrupts the index, possibly giving a DKNF
error in the future:
A new key is being inserted into the index. (Or, an existing key
is being modified, and the new key value is being inserted.) The
leaf node into which the ken would be inserted is full, as are
both of its sibling nodes. (Meaning a new leaf node gets created
by splitting the current node.) The parent node needs to be
updated to insert the new leaf node. The problem comes when the
parent node, and the parent's right-side sibling are full. (And
possibly the left-side sibling as well, in which case we need to
continue up to the grandparent node, where the same situation
occurs.)
(All) #1042
Blobfix doesn't fix the internal record number reference within
the blob header. This does no affect anything at runtime, but
does make examining the structure for debugging purposes more
difficult.
(All) #1047
If you have more than ome memo fieidl in a record, update one of
them, and then execute a "memo ... delete" on the other one, and
the other memo field never had any data in it, the blob file can
be corrupted.
(All) #1059
In *report, if you CALL a process from within @WBRK or @WGT, and
that processing table DECLAREs new variables, referencing those
variables can cause *report to crash/freeze.
(All) #1064
OPENDIR()'s return value was truncated to the last 3 digits, so
if more than 999 files are returned, the return value was wrong.
Note, however, that all files were returned.
(All) #966
If you have a lookup file with memo fields, and to a COPY to that
lookup, the memo fields won't be copied correctly.
(All) #970
If you have an array aliased to fields which includes a memo
field, executing "CLEAR arrayname" will crash filePro later.
(All) #1003
If you defined memo dummy fields or variables in a file which has
to "real" memo or blob fields, filePro would crash upon access to
those fields.
(All) #1012
If you have a filePro map with empty field names in the middle,
any reference-by-name of the first field after the blank field
name(s) will not work correctly.
(All) #1017
If you concatenate fields such that the combined length is greater
than 32,767 you would get an out-of-memory error.
(All) #1027
In *report, if the first assignment to a dummy field that is not
declared in automatic is in a CALLed table, then *report may crash
at a reference to a dummy field dome within @DONE processing.
(All) #1028
Under certain circumstances, using "@" in an edit cn crash filePro.
(All) #886
Attempting to use sockets without a socket license could crash
filePro rather than give an error.
(All) #963
In the meo editor, if the cursor is on the last line of a word-
wrapped line of text, and is on the first line within the memo
editor window, and the rest of the window is fil;led with
additional text, then placing the cursor at the end of the test
on that line and backspacing such that the now-shorter text
would re-wordwrap to take less lines, taking the cursor to the
text which has been re-wordwrapped to the previous line (ie: the
one above the top of the window), can cause filePro to crash.
(All) #1002
If the width of a listbox was at least 78, and dropshadows are
turned off, the contents of the listbox are shifted right 2 by
columns.
(All) #1048
DECLARE EXTERN in sort/select processing for a vairable which is
DECLAREd GLOBAL within automatic processing could cause a
"variable already declared" error after sort/select processing
was finished.