**Test Suite Log**

[HaO] I want to document test suite failures when running the test suite here.
Others are invited to contribute.

The purpose of the page is to find any test suite differences when new versions are tested.


**Tcl**


***Windows Vista Pro 32 bit GER, MSVC6++, Platform SDK 2003SP1, Makfile.vc***

%|Test|8.4.20|8.5.14|8.5.15|8.6.0|8.6.1|%
&|filesystem-1.3||fail|fail|fail|fail|&
&|filesystem-1.4||fail|fail|fail|fail|&
&|filesystem-20.7|fail|||||&
&|filesystem-20.8|fail|||||&
&|socket-14.0||||fail(5)|fail(5)|&
&|winFCmd-6.1|fail|fail(1)|fail|fail(1)||&
&|winFCmd-6.9|fail|fail(1)|fail|fail(1)||&
&|winFCmd-6.13|fail|fail(1)|fail|fail(1)||&
&|winFCmd-9.3|fail|fail(1)|fail|fail(1)||&
&|winFCmd-12.6|fail|fail(1)|fail|||&
&|zlib-8.9||||fail(2)|fail|&
&|zlib-8.10||||fail(2)|fail|&
&|zlib-8.11||||fail(2)|fail|&
&|odbc-15.2||||fail(3)|fail(4)|&
&|odbc-15.3||||fail(3)|fail(4)|&
&|odbc-15.4||||fail(3)|fail(4)|&
&|odbc-15.5||||fail(3)|fail(4)|&
&|odbc-15.6||||fail(3)|fail(4)|&
&|odbc-20.1a||||fail(4)|fail(4)|&
&|odbc-21.2||||fail(4)|fail(4)|&
&|odbc-22.3||||fail|fail(4)|&
&|odbc-25.1a||||fail(4)|fail(4)|&
&|odbc-27.1a||||fail(4)|fail(4)|&
&|odbc-27.2b||||fail(4)|fail(4)|&

****Remarks****

   1. Fails typically the first test run, repeating will not fail
   1. Failed only once with privileged user
   1. Transaction issues, probably due to driver
   1. Message: database is locked (5)
   1. After discussion with [rmax], this is only on '''my''' machine and probably due to local configuration that getaddrinfo() does not return ipv4 addresses if there is a ipv6 address.

**Tk**

***Windows Vista Pro 32 bit GER, MSVC6++, Platform SDK 2003SP1, Makfile.vc***

%|Test|8.5.14|8.5.15|8.6.0|8.6.1|%
&|bind-15.23|||fail||&
&|bitmap-4.1|||fail||&
&|border-3.1|||fail||&
&|cursor-4.1|||fail||&
&|font-18.1|||fail||&
&|spinbox-21.1|fail|fail|fail|fail|&
&|text-19.16|||fail|fail|&
&|textDisp-19.11.20|||fail|fail|&
&|textDisp-19.11.23|||fail|fail|&
&|textDisp-29.2|fail|fail|fail|fail|&
&|textDisp-29.2.1|fail|fail|fail|fail|&
&|textDisp-29.2.2|fail|fail|fail|fail|&
&|textDisp-29.2.3|fail|fail|fail|fail|&
&|textDisp-29.2.4|fail|fail|fail|fail|&
&|textImage-3.2|fail|fail|fail|fail|&
&|textTag-15.1|fail|fail|||&
&|winClipboard-1.2||fail(2)||fail(1)|&
&|winClipboard-1.4||fail(2)||fail(1)|&
&|winClipboard-1.5||fail(2)||fail(1)|&
&|winClipboard-1.6||||fail(1)|&
&|winClipboard-2.2||fail(2)||fail(1)|&
&|winWm-7.3|||fail||&
&|wm-attributes-1.5.1||fail|fail||&
&|wm-attributes-1.5.2|||fail||&
&|wm-attributes-1.5.4|||fail||&

****Remarks****

   1. Win Clipboard tests feel quite unrelyable and fail with different errors. This time, the error message was 'TK SELECTION EXISTS'.
   1. As 1., but with error message 'couldn't open clipboard: Succesfully processed.' (translated from german).


**Test Logs**

I also archive (for me) the test log files build with:
======none
nmake -f Makefile.vc test OPTS=threads TESTFLAGS="-verbose \"body start error\"" > testlog_tcl8420rc0.txt 2>&1
======
Drop me an E-Mail if you want to get them.

<<discussion>>
[DKF]: FWIW, Tcl is clean for me in both "production" and "full debug" builds (where full debug is "configured with `--enable-symbols=all`) on OSX.
<<categories>> Development