Fetch Livelink html-source

[fmlab]

Hey guys,

I have to evaluate Livelink contents, in the current case Workflows. Unfortunately the LAPI seems to be too stiff for that.

Somehow getting a specific html-source into another lightweight-engine (eg. PHP) I could achieve the evaluation very flexible withouth heavy maintenance.

Now I am looking for a tool that will fetch my desired source (depending on a parameter), and somehow returns it to my other routine, or simply saves the source-code to my filesystem.

Can anyone help here?

 

[ggriffit]

can you provide some more information on what you are trying to do?

Greg Griffiths
Livelink Certified Developer & ECM Global Star Champion 2005 & 2006

http://www.greggriffiths.org/livelink/

 

[fmlab]

I need to evaluate an unknown number of workflows (as in Personal > Workflows in Livelink "LES").
These workflows contain many attributes, which I could fetch via PCRE.

So basically I need to write some kind of crawler, that reads the main page of workflows, and follows into every workflow beneath. In these workflows I have to follow the links that lead to the attribute-pages, which I can then read with PCRE.

So again, it comes down to fetching the html-source of a Livelink page, as I can reach it via my browser.

 

[appnair]

What do you mean main page of workflows? of what workflow do you consider main perhaps the workflow volume is what you are after.I am not sure reading with perl on a browsed stream is going to be any efficient perhaps you have your reasons for doing that.The attributes you are after could be sourced with SQL queries are they workflow attributes or standard library/document attributes.

Well, if I called the wrong number, why did you answer the phone?
James Thurber, New Yorker cartoon caption, June 5, 1937
Certified OT Developer and probably certfiable,Livelink ECM Champion 2008

http://www.linkedin.com/in/appunair

 

[fmlab]

With main I mean the page that lists the workflows that you have rights to view/edit on.

In my scenario performance comes way behind flexibility, and since neither me nor my coworkers are real programmers I see a to-be-compiled solution (such as JAVA) way to stiff and unflexible via the original LAPI.

When the whole evaluate-the-workflow-attribs thing started we were in touch with our Livelink in-house support, but their knowledge ended with offering a customized report. Here the problem was that the attributes were stored along with "foreign" attributes in generic tables, which we won`t get access to.

That`s why I see the only way to make this smoothly is PHP, as I know what I am doing there.

Do you see a way of fetching that html-source?

 

[ggriffit]

am sure this can be done with a few simple SQL statements or even some LAPI which would be much better and more robust and future proof than trying to process each page for links via a crawler type thing.

Greg Griffiths
Livelink Certified Developer & ECM Global Star Champion 2005 & 2006

http://www.greggriffiths.org/livelink/

 

[fmlab]

As I`ve already said I won`t be granted SQL access, thus I need a solution that is based on my user-rights and workstation-features.

 

[ggriffit]

You had mentioned a LiveReport earlier in the thread, that was what I was re-mentioning, it should be simple enough to generate a LiveReport to pull through the Category Data on a workflow.

Alternatively an good developer should be able to create a simple LAPI or OScript program to do this for you.

Greg Griffiths
Livelink Certified Developer & ECM Global Star Champion 2005 & 2006

http://www.greggriffiths.org/livelink/

 

[fmlab]

Thanks for that.
Any further suggestions regarding the fetch-html-source topic?

 

[appnair]

well I see that you are kind of up against a wall in a organization that does not want to fix what the performance problem is but seriously as Greg says a better attempt at solving the root cause should be attempted.Having said that If personal.workflows is taking a long time to load perhaps a simple workflow status object to a workflow that you and your colleaugues are interested could be created or maybe the PERL approach is better form a usability side.I am not familar with using perl on the workflow pages so perhaps some others may chip in.

Well, if I called the wrong number, why did you answer the phone?
James Thurber, New Yorker cartoon caption, June 5, 1937
Certified OT Developer and probably certfiable,Livelink ECM Champion 2008

http://www.linkedin.com/in/appunair

 

[fmlab]

I am working in a large corp with a big "S" from Germany. There are a lot of obstacles alike, and there is no way I can get root-access :-(

I`ve also been fighting to try LAPI, but I cannot get over the routines complaint about a missing cert. No idea how to get that. Our LL support suggested to use the SSL cert, which did not work and does not make any sense here.

I will go crazy soon!

Any further suggestions?

 

[appnair]

H'mmm I don't want to add fuel to your fire.If you can access the livelink web gui and look under personal.workflows ,using lapi also you can do that.You will need help from your company LL administartion as to getting ports,connection cgi etc and I think you are having trouble getting anything from them.LAPI just means that things that you can perform with your userid/password albeit you don't want to do a bunch of clicks within the livelink web UI.It DOES not require any more permissions than what you currently have.Obviously since lapi is a client side code interfacing with livelink Enterprise Server you will need the lapi client libraries as well.All of this is posted just to disple some of the myths against livelink pls don't misunderstand me.

Well, if I called the wrong number, why did you answer the phone?
James Thurber, New Yorker cartoon caption, June 5, 1937
Certified OT Developer and probably certfiable,Livelink ECM Champion 2008

http://www.linkedin.com/in/appunair

 

[fmlab]

You are right with most of what you`ve said. We suspect that our instance of LL requires client-side certificates though. We have already tried only using CGI- and user-information for a session, but as we found out nothing happened here.
But then again: I am not really a programmer. That is why I am looking for a way to just get the html-source, which I could evaluate with a other language than JAVA or C#.