Multi-value output buffers (those including | character), other than 'tracks', are not properly handled when number of values doesn't match number of files on which web source is executed.

Consider this web source:

Input:
CODE
!web
!sources
!framework
!has
!some
!bugs

Script:
CODE
[ParserScriptAlbum]=...
outputto "field"
do
movechar 1
sayrest
say "|"
moveline 1 1
while "!"

Output: (6 values)
CODE
output["field"]= "web|sources|framework|has|some|bugs|"

The following screenshots present different scenarios according to number of files on which web source was executed:

Three files

Number of files is smaller than number of values. Extra values are discarded as if they don't exist! User is unaware of how many values were really "collected" by script.

Six files

Number of files matches number of values. Script works as expected.

Ten files

Number of files is greater than number of values. Extra files are populated with content of raw output buffer!