Hi All, I'm monitoring log files of testcases. The logfile always contains a last line where the status is reported (successful/NOT successful). I'm able to find these status reports with following commandline script: For /F "tokens=*" %%A in ('Dir /B /A:-D %1') Do ( type %%~fA | find "INFO - Testcase was successful!" >> D:\scripts\splitfiles\logs\temp\successful.txt For /F "tokens=*" %%A in ('Dir /B /A:-D %1') Do ( type %%~fA | find "INFO - Testcase was NOT successful!" >> D:\scripts\splitfiles\logs\temp\not_successful.txt But now my question: How can I find the last 2 ERROR lines in my logfile? Here is an example of my last logfile: 04 Aug 2008, 11:24:06 ERROR - Assertion failed param1: 1 matching "MercuryTest_21588" param2: No matches assertionType: CONTAINS 04 Aug 2008, 11:24:06 INFO - Condition was false 04 Aug 2008, 11:24:29 INFO - ***** Executing Step: Step 3 ***** 04 Aug 2008, 11:24:31 INFO - Loaded URL: http://test.wap.xxxxxx.com/xxxxxxxx... successfully 04 Aug 2008, 11:24:31 ERROR - Failed at Step 3: field 1 - 2 (MercuryTest_21588) not found(Control type: [link] Criteria: [.text] Criteria value: [1 - 2 (MercuryTest_21588)]) 04 Aug 2008, 11:24:31 ERROR - Failed at Step 3: field not found(Control type: [link] Criteria: [.text] Criteria value: [1 - 2 (MercuryTest_21588)]) 04 Aug 2008, 11:24:31 INFO - ***** Executing Step: Reset Promos ***** 04 Aug 2008, 11:24:31 INFO - start call of sequence: WAP:xxxxCommons.ResetAllPromotionAreas 04 Aug 2008, 11:24:31 INFO - start call of sequence: WAP:xxxxCommons.SchedulePromotion 04 Aug 2008, 11:24:33 INFO - Loaded URL: http://test.xxxxxxxx.xxxx.com/defau... successfully 04 Aug 2008, 11:25:04 INFO - Useragent set to IE6 04 Aug 2008, 11:25:04 INFO - Loaded standard security configuration 04 Aug 2008, 11:25:04 INFO - Testcase was NOT successful! I'm only interested in the last two ERROR lines, I want to pipe the contents of these to the "not_successfull.txt" but I cannot find how to achieve that? Any help will be greatly appreciated! Thanks in advance! Igor (This is also posted at Windows XP, but I figured out that it belongs in the Programming section, so apologies for the cross-post. I will be more careful next time!!)

DOS doesn't have a head/tail command so it is not that easy. You can do it programatically using vbscript and run the script file using cscript.

vbscript Set objFS = CreateObject("Scripting.FileSystemObject") Dim ArrStore() i=0 strFileName="test.txt" Set objFile = objFS.OpenTextFile(strFileName) Do Until objFile.AtEndOfLine strLine=objFile.ReadLine ArrLine = Split(strLine) If ArrLine(4) = "ERROR" Then ReDim Preserve ArrStore(i) ArrStore(i)=strLine i=i+1 End If Loop For j=UBound(ArrStore)-1 To UBound(ArrStore) WScript.Echo ArrStore(j) Next

Hi all, Thanks for your quick resposes! I have NO experience with VBScript AT ALL, but I have tried.. What I did: - I copied the code into a file and named the file: "C:\Scripts\LogFiles\ExtractErrorLines.vbs" - I created a text-file: "test.txt" in the same directory containing 3 lines where only the 3rd line contains the word "ERROR" Then I ran the VBScript. It generates an error though. Here the Error-message: Script: C:\Scripts\LogFiles\ExtractErrorLines.vbs Line: 16 Char: 1 Error: Subscript out of range: 'UBound' Code: 800A0009 Source: Microsoft VBScript runtime error I'm almost sure this can be done with plain commandline batchfile (prefferred solution), but I will be happy to use VBScript if I can get a working solution from it. Maybe someone can help? Thanks in advance! Igor

@echo off > not_successfull.txt setLocal EnableDelayedExpansion find "ERROR" < logfile > E for /f "tokens=* delims= " %%a in ('find /c "ERROR" ^< E') do ( set /a skip=%%a-2 ) for /f "tokens=* delims= " %%a in (E) do ( set /a N+=1 if !N! gtr !skip! echo %%a >> not_successfull.txt ) ===================================== If at first you don't succeed, you're about average. M2

@Mechanix2Go: EXACTLY!! ;-) Works like a charm! Thanks a lot!! Cheers, Igor

I hope somebody can help a bit more.. I have changed the @Mechanix2Go script to process a list of files: -------------------- @echo off > not_successfull.txt setLocal EnableDelayedExpansion For /F "tokens=1,2 delims=. " %%i in ('Dir /B /A:-D "C:\Scripts\Mercury\Logs\TE_*.log"') Do ( find "ERROR - " < "C:\Scripts\Mercury\Logs\%%i.%%j" > E ) for /f "tokens=* delims= " %%a in ('find /c "ERROR - " ^< E') do ( set /a skip=%%a-2 ) for /f "tokens=* delims= " %%a in (E) do ( set /a N+=1 if !N! gtr !skip! echo %%a >> not_successfull_%%i%%.txt ) del E /q/f ) -------------------- But now, two new problems occur: 1) The file-list of: Dir /B /A:-D "C:\Scripts\Mercury\Logs\TE_*.log" produces many files containing ip-addresses and several random spaces in the filename. So the %%i and %%J variables in the first 'For'-loop result in incorrect filenames when used as "%%i.%%j" Example: the filename TE_100.130.131.10_Test login link_20080807.log results in %%i= TE_100 and %%j=130 Is there a way to process the filenames containing dots and spaces? 2) I tried to construct a not_successfull.txt file for each logfile by using the %%i variable. But when using the not_successfull_%%i%%.txt as in my example above, the script only produces one file: not_successfull_logfile3.txt I would expect to end-up with files like: not_successfull_logfile1.txt not_successfull_logfile2.txt not_successfull_logfile3.txt Any help would be greatly appreciated (again)!! Thanks in advance, Igor