multiple keyword search - regular expression, posted by Johannes Liegl on Wed Aug 19 11:49:37 2009
|
Dear Sirs,
searching for keywords regular expression becomes very long. Are there any other type easy multiple keyword search supported. For example a keyword search like keyword1 & keyword2 results in different output than a keyword search like keyword2 & keyword1. A simple keyword search like "keyword1 & keyword2 and keyword2 &keyword1 should find all documents in a database containing both keywords resulting in the same output.
Thank you very much for you help in advance.
Best Regards
Johannes Liegl
|
Re: multiple keyword search - regular expression, posted by Stefan Ritt on Thu Sep 3 18:14:26 2009
|
Johannes Liegl wrote: |
Dear Sirs,
searching for keywords regular expression becomes very long. Are there any other type easy multiple keyword search supported. For example a keyword search like keyword1 & keyword2 results in different output than a keyword search like keyword2 & keyword1. A simple keyword search like "keyword1 & keyword2 and keyword2 &keyword1 should find all documents in a database containing both keywords resulting in the same output.
|
I had this request already several times. Indeed I myself am not very happy with regular expressions. If anybody knows of any other convenient search method, together with freely available C code, I'm more than happy to include this into elog. |
Re: multiple keyword search - regular expression, posted by Yoshio Imai on Fri Sep 4 13:59:19 2009
|
Maybe there is a simple way.
If I understand it correctly, the entries are accepted/rejected for inclusion in the search result list based on one call to regexec. In order to search for N keywords, you could replace this by a loop of N calls to regexec, and only accept the entry if all of them find a match. |
Re: multiple keyword search - regular expression, posted by Stefan Ritt on Fri Sep 4 14:07:31 2009
|
Yoshio Imai wrote: | Maybe there is a simple way.
If I understand it correctly, the entries are accepted/rejected for inclusion in the search result list based on one call to regexec. In order to search for N keywords, you could replace this by a loop of N calls to regexec, and only accept the entry if all of them find a match. |
That's right, but some people want something like
(NOT "House" and "Car") OR "Tree"
Using regex'es is a bit exotic for most users. So I hope there is some source code which implements a Google type search more or less 1:1. And Google does not use regex'es (is that right?). Here is the syntax for the Google search:
http://www.google.com/support/websearch/bin/answer.py?hl=en&answer=136861 |
Fix text prevents user from editing text during creation, instead of just edit, posted by Allen on Mon Aug 24 21:47:14 2009
|
When we set Fix text = 1, according to the syntax, this should prevent users from modifying the text field during an edit, but it looks like it is blocking access at both time of edit and creation, meaning you can never add anything to it. Is that the intended functionality? |
Re: Fix text prevents user from editing text during creation, instead of just edit, posted by Stefan Ritt on Fri Sep 4 09:48:55 2009
|
Allen wrote: |
When we set Fix text = 1, according to the syntax, this should prevent users from modifying the text field during an edit, but it looks like it is blocking access at both time of edit and creation, meaning you can never add anything to it. Is that the intended functionality?
|
Thanks for reporting that bug. I realized that this was not working at all since a long time. I fixed it in SVN revision 2245. |
Time Display Format, posted by james Goedsir on Tue Sep 1 04:15:57 2009
|
My Time field is displayed i a 12 Hour format I would like to change it to 24 Hours, can someone advised how it can be done |
Re: Time Display Format, posted by james Goedsir on Wed Sep 2 09:39:20 2009
|
james Goedsir wrote: |
My Time field is displayed i a 12 Hour format I would like to change it to 24 Hours, can someone advised how it can be done
|
Sorted no problem |
fckeditor update, posted by Arno Teunisse on Tue Aug 25 21:08:51 2009
|
Hello
Just a few fckeditor related questions. How do elog versions and fckeditor versions relate. ?
Can I just drop another version of the fckeditor over an other version? What things should I consider when doing so ?
thanks for you're time.
|
Re: fckeditor update, posted by Stefan Ritt on Mon Aug 31 11:22:20 2009
|
Arno Teunisse wrote: |
Hello
Just a few fckeditor related questions. How do elog versions and fckeditor versions relate. ?
Can I just drop another version of the fckeditor over an other version? What things should I consider when doing so ?
thanks for you're time.
|
The relation is not very "stong". In the past I updated between major version of fckeditor without chaning any elog code, so just give it a try. |
Multiple keyword search, posted by Yoichi Aso on Wed Jan 14 19:52:04 2009
|
Hi,
I have a question regarding how to perform a multiple-keyword search.
For example, when I want to find entries which contain two keywords "abc" and "def" in the body text but in an unknown order,
one way I came up with was to use a regular expression like this.
(abc|def).*(abc|def)
But this will also match entries containing two "abc" or two "def".
The following one will eliminate this problem.
(abc.*def)|(def.*abc)
But when I use more than two keywords, this type of regular expressions becomes very long (because I have to list all the permutations of the keywords) and it may not fit in the search text field (there seems to be a limit on the length of the search text).
Is there any way to allow multiple-keyword search easily ?
It would be nice if I can just enter two or more keywords separated by white spaces and elog finds entries containing all the keywords.
Thanks,
Yoichi |
Re: Multiple keyword search, posted by Johannes Liegl on Wed Aug 19 11:54:52 2009
|
Yoichi Aso wrote: | Hi,
I have a question regarding how to perform a multiple-keyword search.
For example, when I want to find entries which contain two keywords "abc" and "def" in the body text but in an unknown order,
one way I came up with was to use a regular expression like this.
(abc|def).*(abc|def)
But this will also match entries containing two "abc" or two "def".
The following one will eliminate this problem.
(abc.*def)|(def.*abc)
But when I use more than two keywords, this type of regular expressions becomes very long (because I have to list all the permutations of the keywords) and it may not fit in the search text field (there seems to be a limit on the length of the search text).
Is there any way to allow multiple-keyword search easily ?
It would be nice if I can just enter two or more keywords separated by white spaces and elog finds entries containing all the keywords.
Thanks,
Yoichi |
Dear Yoichi,
well we are facing the same problem and I would like to ask you if you have been able finding another method searching for keywords in documents?
I am looking forward to getting feedback. Thank you very much for your help in advance.
Best Regards
Johannes Liegl |
Logbook Parser, posted by Alan Grant on Tue Aug 11 00:20:11 2009
|
We are exploring whether it's possible/feasible to import ELog logbooks into a another database for special purposes (plotting/statisical, etc). Target database is TBD (perhaps Access).
Does anyone have or know of a logbook parser program? From cut/pasting into, for example, Excel, it does appear that the data fields are already line-feed delimited so offhand it would seem possible to parse if one really wanted to pursue it.
Regards,
- Alan |
Re: Logbook Parser, posted by Stefan Ritt on Tue Aug 11 08:29:23 2009
|
Alan Grant wrote: |
We are exploring whether it's possible/feasible to import ELog logbooks into a another database for special purposes (plotting/statisical, etc). Target database is TBD (perhaps Access).
Does anyone have or know of a logbook parser program? From cut/pasting into, for example, Excel, it does appear that the data fields are already line-feed delimited so offhand it would seem possible to parse if one really wanted to pursue it.
Regards,
- Alan
|
You can export to CSV (comma-separated-values) if you go to "Find" and then click on "Export: CSV". These fiels you ran read right into Excel or other spreadsheet programs for further analysis. |
Re: Logbook Parser, posted by Steve Williamson on Tue Aug 11 13:02:22 2009
|
Stefan Ritt wrote: |
Alan Grant wrote: |
We are exploring whether it's possible/feasible to import ELog logbooks into a another database for special purposes (plotting/statisical, etc). Target database is TBD (perhaps Access).
Does anyone have or know of a logbook parser program? From cut/pasting into, for example, Excel, it does appear that the data fields are already line-feed delimited so offhand it would seem possible to parse if one really wanted to pursue it.
Regards,
- Alan
|
You can export to CSV (comma-separated-values) if you go to "Find" and then click on "Export: CSV". These fiels you ran read right into Excel or other spreadsheet programs for further analysis.
|
excuse my butting in ... I've found the exports useful in the past - however, is is possible to run the export from a script in order to produce reports? Utilities like wget won't work as the export process doesn't return the data as html.
regards
Steve
|
Re: Logbook Parser, posted by Stefan Ritt on Tue Aug 11 13:25:48 2009
|
Steve Williamson wrote: |
excuse my butting in ... I've found the exports useful in the past - however, is is possible to run the export from a script in order to produce reports? Utilities like wget won't work as the export process doesn't return the data as html.
|
That's not true. wget does work. Try that one:
wget --no-check-certificate -O export.csv https://midas.psi.ch/elogs/linux+demo/?mode=CSV1
actaully wget doesn't care if the return is HTML or a GIF image or anything else, it just saves it into the output file. |
Re: Logbook Parser, posted by Steve Williamson on Wed Aug 12 14:40:52 2009
|
Stefan Ritt wrote: |
Steve Williamson wrote: |
excuse my butting in ... I've found the exports useful in the past - however, is is possible to run the export from a script in order to produce reports? Utilities like wget won't work as the export process doesn't return the data as html.
|
That's not true. wget does work. Try that one:
wget --no-check-certificate -O export.csv https://midas.psi.ch/elogs/linux+demo/?mode=CSV1
actaully wget doesn't care if the return is HTML or a GIF image or anything else, it just saves it into the output file.
|
you're right, of course, on all counts!
when I was testing wget/elog to try to automate an extract I was getting a lot of stuff like:
/Change_Log/587">Software Only</a></td><td class="92^M^H<88>^\ÿ^Y"ÿ"><a href="../Change_Log/587">23416</a></td><td class="92^M^H<88>^\ÿ^Y"Ã<a href="../Change_Log/587">New</a></td><td class="92^M^H<88>^\ÿ^Y"ÿ"><a href="../Change_Log/587">Awaited</a></td>
but I must have been getting something wrong, using your command line as an example it works perfectly! Thanks again for elog!! |
Re: Logbook Parser, posted by Alan Grant on Tue Aug 11 16:25:28 2009
|
Steve Williamson wrote: |
Stefan Ritt wrote: |
Alan Grant wrote: |
We are exploring whether it's possible/feasible to import ELog logbooks into a another database for special purposes (plotting/statisical, etc). Target database is TBD (perhaps Access).
Does anyone have or know of a logbook parser program? From cut/pasting into, for example, Excel, it does appear that the data fields are already line-feed delimited so offhand it would seem possible to parse if one really wanted to pursue it.
Regards,
- Alan
|
You can export to CSV (comma-separated-values) if you go to "Find" and then click on "Export: CSV". These fiels you ran read right into Excel or other spreadsheet programs for further analysis.
|
excuse my butting in ... I've found the exports useful in the past - however, is is possible to run the export from a script in order to produce reports? Utilities like wget won't work as the export process doesn't return the data as html.
regards
Steve
|
Steve, just a word of thanks for "butting in" ... my next thought was how could I schedule an export to feed the other database so it wouldn't have to be done manually each day. Your question took care of that for me! :)
Good community. Thanks. |
Comment on: Alphabetize Quick Option filter, posted by David Pilgram on Mon Aug 10 21:07:15 2009
|
(For some reason I could not add this in Dennis's thread.)
I like this new feature, BUT
I happen to have two Options: Options System, and Options Status.
System are a very few items, whereas Status has a long list, which, like Dennis's example, can be added to.
Keeping the latter in alpha order is great, but it's a shame that the cost is that Options System are also
sorted alphabetically, whereas it has a natural order which it would be preferable to keep - for example (and
this is made up)
Options System: 3.1, NT, 2000, XP, Vista
where the natural order here is chronological.
Perhaps the configuration file option could be more specific, for example
Sort attribute Options Status = 1
which would then NOT sort Options System. If both are needed to be sorted, both should be specified, or back to
the original syntax which defaults to sort *all* Options. |
Re: Comment on: Alphabetize Quick Option filter, posted by David Pilgram on Mon Aug 10 21:19:50 2009
|
I've just noticed that it has also sorted another Option, which are selected as radio buttons. Again, this is a
list which has a natural - again, in this case, chronological - order.
Because of this, I'm going to have to turn off this feature as it is on my system. I hope something can be sorted
on this.
> (For some reason I could not add this in Dennis's thread.)
>
> I like this new feature, BUT
>
> I happen to have two Options: Options System, and Options Status.
>
> System are a very few items, whereas Status has a long list, which, like Dennis's example, can be added to.
> Keeping the latter in alpha order is great, but it's a shame that the cost is that Options System are also
> sorted alphabetically, whereas it has a natural order which it would be preferable to keep - for example (and
> this is made up)
>
> Options System: 3.1, NT, 2000, XP, Vista
>
> where the natural order here is chronological.
>
> Perhaps the configuration file option could be more specific, for example
>
> Sort attribute Options Status = 1
>
> which would then NOT sort Options System. If both are needed to be sorted, both should be specified, or back to
> the original syntax which defaults to sort *all* Options. |
Re: Comment on: Alphabetize Quick Option filter, posted by Stefan Ritt on Tue Aug 11 08:38:56 2009
|
Ok, that makes sense, so I changed it to
Sort Attribute Options Status = 1
as you suggested.
> (For some reason I could not add this in Dennis's thread.)
>
> I like this new feature, BUT
>
> I happen to have two Options: Options System, and Options Status.
>
> System are a very few items, whereas Status has a long list, which, like Dennis's example, can be added to.
> Keeping the latter in alpha order is great, but it's a shame that the cost is that Options System are also
> sorted alphabetically, whereas it has a natural order which it would be preferable to keep - for example (and
> this is made up)
>
> Options System: 3.1, NT, 2000, XP, Vista
>
> where the natural order here is chronological.
>
> Perhaps the configuration file option could be more specific, for example
>
> Sort attribute Options Status = 1
>
> which would then NOT sort Options System. If both are needed to be sorted, both should be specified, or back to
> the original syntax which defaults to sort *all* Options. |
Re: Comment on: Alphabetize Quick Option filter, posted by David Pilgram on Tue Aug 11 10:07:08 2009
|
Thanks Stefan! Works great.
> Ok, that makes sense, so I changed it to
>
> Sort Attribute Options Status = 1
>
> as you suggested.
>
> > (For some reason I could not add this in Dennis's thread.)
> >
> > I like this new feature, BUT
> >
> > I happen to have two Options: Options System, and Options Status.
> >
> > System are a very few items, whereas Status has a long list, which, like Dennis's example, can be added to.
> > Keeping the latter in alpha order is great, but it's a shame that the cost is that Options System are also
> > sorted alphabetically, whereas it has a natural order which it would be preferable to keep - for example (and
> > this is made up)
> >
> > Options System: 3.1, NT, 2000, XP, Vista
> >
> > where the natural order here is chronological.
> >
> > Perhaps the configuration file option could be more specific, for example
> >
> > Sort attribute Options Status = 1
> >
> > which would then NOT sort Options System. If both are needed to be sorted, both should be specified, or back to
> > the original syntax which defaults to sort *all* Options. |
Re: Comment on: Alphabetize Quick Option filter, posted by Dennis Seitz on Tue Aug 11 17:46:33 2009
|
Yes, many thanks, Stefan, from me, too! It's really great that you respond so quickly to requests and suggestions.
And thanks to David for the fine tuning, great suggestion.
Dennis
> Thanks Stefan! Works great.
>
> > Ok, that makes sense, so I changed it to
> >
> > Sort Attribute Options Status = 1
> >
> > as you suggested.
> >
> > > (For some reason I could not add this in Dennis's thread.)
> > >
> > > I like this new feature, BUT
> > >
> > > I happen to have two Options: Options System, and Options Status.
> > >
> > > System are a very few items, whereas Status has a long list, which, like Dennis's example, can be added to.
> > > Keeping the latter in alpha order is great, but it's a shame that the cost is that Options System are also
> > > sorted alphabetically, whereas it has a natural order which it would be preferable to keep - for example (and
> > > this is made up)
> > >
> > > Options System: 3.1, NT, 2000, XP, Vista
> > >
> > > where the natural order here is chronological.
> > >
> > > Perhaps the configuration file option could be more specific, for example
> > >
> > > Sort attribute Options Status = 1
> > >
> > > which would then NOT sort Options System. If both are needed to be sorted, both should be specified, or back to
> > > the original syntax which defaults to sort *all* Options. |
List Option, posted by Alan Grant on Tue Aug 11 08:10:21 2009
|
Hello Stefan.
Currently this is defined as a maximum of 100 literals in the cfg file. I would like to see the option to reference an external text file as input for this.
As a side question, I would also like to increase the max to a greater value, for example, even 5000. I assume I can change the source (I recall var was something like "List_Option_Max") and see if that would still work, but would you know offhand if that would cause a problem anywhere else?
Regards,
Alan
(PS: Just getting started with ELog. Please excuse if these questions sound newbie. I also searched the Forum first but haven't found any answers to them yet.) |
Re: List Option, posted by Stefan Ritt on Tue Aug 11 08:33:32 2009
|
Alan Grant wrote: |
Currently this is defined as a maximum of 100 literals in the cfg file. I would like to see the option to reference an external text file as input for this.
|
I will put this on the wish list.
Alan Grant wrote: |
As a side question, I would also like to increase the max to a greater value, for example, even 5000. I assume I can change the source (I recall var was something like "List_Option_Max") and see if that would still work, but would you know offhand if that would cause a problem anywhere else?
|
I limited this to 100 entries because it will be hard to handle it. Imagine a drop-down list box with 5000 entries. It would fill your complete screen and you still won't see all 5000 entries. In that case it might be better to use a free text field and enter the attribute value as free text.
You can increase MAX_N_LIST in elogd.c, but at some point you will get a stack overflow and elogd will just crash.
- Stefan
|
|