# HG changeset patch # User Henry S. Thompson # Date 1494952132 -3600 # Node ID 9bb415e0adc9ce6d40d97f0c378db3a28d2a5649 # Parent 793496d2d165d2333954cdff5522e0fcb00ebcb6 try to fix error processin odd REUTER|IDN\!'.SPX' external ref diff -r 793496d2d165 -r 9bb415e0adc9 n2a.xsl --- a/n2a.xsl Tue May 16 17:27:19 2017 +0100 +++ b/n2a.xsl Tue May 16 17:28:52 2017 +0100 @@ -43,7 +43,7 @@ e:crnf($r)) else if (local-name($b)='v') then $b/text() - else e:shouldnt()"/> + else e:shouldnt('rnf',concat(local-name($b),' ',string($b)))"/> @@ -91,7 +91,7 @@ + else e:shouldnt('delta',local-name($c))"/> @@ -106,6 +106,8 @@ - shouldn't happen + + + shouldn't happen in : diff -r 793496d2d165 -r 9bb415e0adc9 notes.txt --- a/notes.txt Tue May 16 17:27:19 2017 +0100 +++ b/notes.txt Tue May 16 17:28:52 2017 +0100 @@ -224,3 +224,14 @@ ------------ @@ string identify, to say nothing of actual value, is lost -- fix? @@ row/column/both spans +------ +enron/kenneth_lay__19506 contains this formula: + + [1]!'.SPX' + +which crashes tokenise/rnf + +Changes intended to fix this fixed a bug (?) which wasn't properly +merging e.g. +3 -- no examples of larger numbers available to check +with... + diff -r 793496d2d165 -r 9bb415e0adc9 tokenise.xsl --- a/tokenise.xsl Tue May 16 17:27:19 2017 +0100 +++ b/tokenise.xsl Tue May 16 17:28:52 2017 +0100 @@ -1,7 +1,6 @@ - - + ("[^"]*")|(\{[^}]+})|(,)|([^=\-+*/();:,.$<>^!]+(?:\.[^=\-+*/();:,.$<>^!]+)*\()|([)])|(^=|\()|((?:(?:'[^']+')|(?:\[[0-9]+\][^!]*)|(?:[a-zA-Z_][a-zA-Z0-9._]*)!))|(\$?[A-Z]+\$?[0-9]+)|([a-zA-Z_\\][a-zA-Z0-9._]*)|(.) @@ -50,7 +49,8 @@ - tok1: ||| + tok1: |||| + + exp: :, + + + + + + + + + + + + + + + + + + + + + + - + + string($tokens[$i])), + $nrs,$stop)"/> - - - - amal: - + amal : + - - +