Hi Thorsten :) On 09/01/2007, at 10:03 PM, Thorsten Bosbach wrote:
Clytie Siddall wrote:1. Which Linux build belongs to a machine running SuSe and using rpm packages? Is it Linux Intel or Linux x86? How do we distinguish between the two different Linux systems based on users' descriptions?From the TestTools' point of view it is just Linux
Yes, but what I mean is, against which build in qatrack do I log these results? There are LinuxIntel builds and Linux x86 builds. I don't know which of these builds corresponds to Andrés SuSe (RPM).
2. How do I manually check testtool errors? For example, the Linux results:You have to load the .res file with errors and/or warnings in the TestTool application (testtool.exe for win32/ testtool.bin for other systems / testtool on mac).The workflow for one test could be: start TestTool,open the .bas file of the test in the testtool e.g b_updt.bas (File -> Open),start the test (Program -> Run), directly stop the test (Program-> Cancel), resultfile was automatically loaded,You may try to reproduce the error manually, if there are enough comments, If you doublecklick on the coloured message, the sourcefile is loaded and points on the line, where the error occurred.Only reproduceable errors/warnings are 'worth' to think about: Now you have 2 choices to try to reproduce the error/warnings: 1. just run the test case with the error 2. if you yet don't know how to do this, run the whole test completelyIf it isn't reproduceable, means it doesn't show up anymore, forget about it.If it is reproduceable, it depends ....some messages already have an issue id - known errors - don't need to think about it.
OK. I'll have to find someone with a Linux system who's willing to follow these errors up. (I run OSX [Intel].)
... I would be curious about your warnings during initialization. ___Filename Errors Warnings Warnings_during_initialization ============================ ====== ======== ==============================b_updt.res 1 1 3 ch_updt.res 0 0 3 closeoffice.res 0 0 3 c_updt.res 1 31 3 d_updt.res 0 1 3 first.res 0 9 3 f_updt_help.res 0 0 3 f_updt_pkgmgr.res 0 0 3 f_updt_spadmin.res 0 0 3 f_updt_templates.res 1 4 3 f_updt_windowfuncs.res 0 1 3 gallery.res 2 7 3 i_updt_1.res 0 0 3 i_updt_2.res 4 4 3 ma_updt.res 5 11 3 m_updt.res 0 2 3 oasis_graphics_update.res 0 2 3 topten.res 0 8 3 w_updt.res 5 6 3 ww_updt.res 3 2 3 xforms_updt.res 0 1 3===================================================================== === ========Total 21 files 22 90 63 ___show quite a few errors for some scripts. How do I trace those errors to the actions creating them?3. What do we do for warnings about filter names?If there are currently no reference files for the language you are using, generate them. If the strings differ to the reference, think about which one is right.
I think André said he was looking into that. Please bear in mind that, due to a degenerative disease, I have very little concentration available, and in particular, I have great difficulty with any analytic tasks. Running testtool at all on my own system was a huge effort. I cannot "just" do the things you can do, anymore. I wish I could. ;)
4. What do errors citing "Invisible Window/Control" mean?I currently don't have an example... Depending on the context... A complete resultfile as attachment would help (directly to me)
I think André has supplied the link to the issue (72816) where we are tracking the VI release results.
I am concerned that the Linux build has so many errors, but I am relieved to be able to release the Windows build today. Two down (OSX Intel and Windows32), three to go (OSX PPC and both Linux builds). ;)
from Clytie (vi-VN, Vietnamese free-software translation team / nhóm Việt hóa phần mềm tự do)
http://groups-beta.google.com/group/vi-VN
PGP.sig
Description: This is a digitally signed message part
