Author: Fabian
Email:
Message:
> > Could it be that the indexer ignores my
"(.*?)" and replaces it with "(.*)" ?
And how can I avoid that and limit the section to one line
until the ";"?
>
>
> This is because regex has so called "hungry"
approach -
> they try to scan the longest possible substring.
> Everything between "SEVERITY']" and the very
last
> apostroph+semicolon pair matches your regex.
>
> Try to modify your expression this way:
>
> Section SEVERITY 2 64 "SEVERITY'] =
'([^']*)';" $1
Your solution works nearly, as we hava also unescaped ' in
the file. But with "SEVERITY'] = '([^;]*)';"
everything works fine. Btw: I know about the
"hungry" approach and the ? is normally used to
restrict the regexp to the shortest substring ...
nevertheless "(.*?)" doesn't work. Perhaps you
want to include that sometime in the future ;)
Now we have a new Section Problem. I declard the regex
"NOTES_TEXT'] = '([^;]*);" $1
It works. But we have multiple NOTES_TEXT in the file ...
and need all of them. Our current workaround is to declare
Section NOTES1 11 0 "[0]['NOTES_TEXT'] =
'([^;]*);" $1
Section NOTES2 12 0 "[1]['NOTES_TEXT'] =
'([^;]*);" $1
Section NOTES3 13 0 "[2]['NOTES_TEXT'] =
'([^;]*);" $1
[...]
But that's not elegant and might not catch all.
Do you have any regexp modifier like "/g" in your
regex engine to search for ALL occurences and return ALL of
them (not only $1 for the first one)?
And there is another thing. Our Indexer has ALWAYS exit
status 0 ... even if we get an Oracle-Error because we
didn't set ORACLE_HOME ...
Is this a problem of Oracle or of the indexer? We are
executing the indexer with an .sh script and the error
handling should stop after an error ... which it does only
with nonzero exit status - which never happens.
Reply: <http://www.mnogosearch.org/board/message.php?id=19044&g
t;
------------------------------------------------------------
---------
To unsubscribe, e-mail: general-unsubscribe mnogosearch.org
For additional commands, e-mail: general-helpmnogosearch.org
|