Re: ftoc bug report

In the past, this could occur if the user put "-header" into
their scan options, as the headers in the scan listing would
mess up the message parsing.  It is looking for message numbers
in the first line of the display.

There are now explicit "-noheader" flags to scan, but something
else could be messed up for that user.

>>>Alexander Zangerl said:

 > 
 > another one from the debianites: 
 > <URL:http://bugs.debian.org/293560>
 > 
 > the submitter gets a syntax error on opening the first pgp-signed
 > message:
 > 
 > --------------
 > syntax error in expression "int(1+1+(11-)*(101-1-2)/(101-))"
 >     while executing
 > "expr int($minlineno+1+($msgid-$minmsgid)*($maxlineno-$minlineno-2)/($max
msg
     id-$minmsgid))"
 >     (procedure "Ftoc_FindMsg" line 51)
 >     invoked from within
 > "Ftoc_FindMsg $msg"
 >     (procedure "Ftoc_FindMsgs" line 5)
 >     invoked from within
 > "Ftoc_FindMsgs $seqids"
 >     (procedure "Ftoc_ShowSequence" line 21)
 >     invoked from within
 > "Ftoc_ShowSequence $seq $msgids"
 >     (procedure "Ftoc_ShowSequences" line 16)
 >     invoked from within
 > "Ftoc_ShowSequences"
 >     (procedure "FolderChange" line 54)
 >     invoked from within
 > "FolderChange privat/az {Msg_Show cur}"
 >     invoked from within
 > "time [list  FolderChange $folder $msgShowProc"
 >     (procedure "Folder_Change" line 3)
 >     invoked from within
 > "Folder_Change privat/az"
 >     (command bound to event)
 > ------------------
 > 
 > looking at the code in question, it seems to me that $minmsgid 
 > might only be blank if the cache is severly borked.
 > 
 > two questions for the ones who know the innards of exmh better: 
 > *) would a update-scan-caches operation fix this issue?
 > *) or is my assumption re the cause of the problem wrong and something
 > else is broken?
 > 
 > i'm grateful for any infos...
 > az

--
Brent Welch
Software Architect, Panasas Inc
Delivering the premier storage system for scalable Linux clusters

www.panasas.com
[email protected]