Please enable JavaScript to view this site.

A-Shell Development History

Navigation: Version 6.1/6.2, builds 1300-1428

1373 – 02 February 2014

Scroll Prev Top Next More

1373.9.1

XTREE bug fix: ListX feature wasn't loading any list items past the first null item.

1373.9.2

MBF2_STICKY bug fix: it was failing to recognize main window toolbars as off limit (thus allowing sticky controls to extend over them).

1373.9.3

AUI_EVENTWAIT bug fix: it was failing to convert incoming parentid value of 0 to the current auto parent setting.

1373.9.4

Click handling bug fix: The click handler was incorrectly treating the auto parent dialog as modal, even when it wasn't (and thus as ignoring clicks outside of it).

1373.9.5

Windows printing refinement: printing to a non-shareable port (LPTx or COMx) now internally blocks other instances of A-Shell from trying to print to the same port at the same time. (Previously this rare but possible situation was generating a Windows error 183.)

1373.9.6

TABX bug fix: the use of accelerator keys to select a new panel wasn't working reliably.

1373.9.7

MX_NOBUF, MX_FLUSHBUF bug fix: failing to pass a status parameter was leading to a memory fault. (Problem introduced in 1372.0)

1373.8

XTREE bug fix: Exitchars feature wasn't working (since 1365.0)

1373.7

XTREE bug fix: left arrow in empty editable tree was causing infinite loop.

1373.6

Eliminate bogus log file warning about a bad free request related to APEX.

1373.5

XTREE bug fix: eliminate bogus complaints about COLDEF after using XTROP_REPLACE.

1373.4

XTREE refinement: warning messages about configuration problems are now logged to ashlog.log as well as being displayed in message boxes.

1373.3.1

APEX bug fix: preview window size/position was not properly being saved and restored, particularly in multi-monitor environments.

1373.3.2

APEX bug fix: changing the printer or printer setup while previewing a file containing //XOFFSET or //YOFFSET commands other than at the top was causing the effect of the last offset command(s) to wrap around to the start of the file.

1373.3.3

Printing bug fix: a UNICODE/ASCII mismatch was causing the printer options displayed (and returned) from the printer selection dialog to be corrupted.

1373.2.1

XTREE bug fix: close a memory handle leak that was causing A-Shell to eventually run out of memory handles after heavy use. (Introduced in 1365.)

1373.2.2

Increase memory handle buffer from 384 to 512, mostly to allow for heavier usage of associative arrays, which use 1 handle each while in scope.

1373.1.1

APEX bug fix: search function was broken by the UNICODE conversion (1365.0)

1373.0.1

Language enhancement: (compiler edit 651) recognizes two new dot functions .ARGTYP and .ARGSIZ.

1373.0.2

MX_SHELLEX refinement: local mode (clisvr = "S" or "L") now supports the waitflag options and the exitstatus parameter. Previously, these only worked in the normal/default (client) mode where the operation is forwarded to ATE via the AG_SHLEXEC command. Also, both modes now support shell commands that require COM initialization.

1373.0.3

Compiler/language enhancement (compiler edit 652): Named Parameters in Functions and Procedures.

1373.0.4

COMPIL.LIT 1.1(126), COMPLP.LIT 1.1(126): update to support /P switch.

1373.0.5

Add new bit flag for MIAMEX, MX_COMPIL: &h02000000 /P, preprocess for named parameter support.

1373.0.6

Compiler/language enhancement (compiler edit 653): Named Parameters in Subroutines.

1373.0.7

ISAM 1.1 refinement: To avoid needlessly complaining about an invalid number of index levels in some IDX files created by ancient (particularly AMOS) versions of ISMBLD, any value less than three or greater than nine is now treated as three, which was originally the only choice. A warning is logged to the ashlog.log file, along with a recommendation that it be rebuilt, but then it goes on to work as a legacy-format IDX file. Previously, A-Shell complained and returned ERF 35 (smashed index) for any invalid value except 255, which was, until recently, the only legacy/invalid value we had seen in the field. Note that if the file really was corrupt, it will surely generate an ERF 35 in short order anyway, so the only downside to attempting to treat it as a legacy IDX is that you might miss the warning which used to appear on the screen.