Dynamic Migration of old websites to new with header name

[philjava]

Hi

How would it be possible to automatically migrate old to new websites under the following criterias with CQ5:

- The old websites can be accessed only by clicking each hyperlink in a html table individually. 

- The old websites have no common html code and only consist of random project id's in the adress bar. For Example: Some id's are 100,101 then others 1567).

- The only common parameter of each old website are the first 5 letters of each old website's header class.  For Example: <span class="heading0">Example for Header of old Websites</span>

So the question: Is this possible to achieve only inside core CQ5? And if not any hint on how to do it is welcome

Regards

Phil

[Jörg_Hoh]

Hi Phil,

For this you shouldn't do a 1:1 migration, because you'll miss one of the biggest assets of AEM for it: the hierarchy of content pages. Also I guess that it's quite hard to automatically map HTML into a number of predefined templates... The process of extracting the HTML and inserting it into templates isn't the hardest part then.

regards,

Jörg

[philjava]

Thanks for your answers. Why would I miss out the hirarchy of the content pages when I would do the automapping? I don't know if I would be able to do it, but I guess there should be a way to include the hirarchy of the content pages no?

By content pages hierarchy you mean the structure system on the left hand side right?

[Sham_HC]

IMO not a good practice to include one time migration code into cq instead run as standard alone application. You can make use of jsoup [1] to read old website & then write into cq.  You might find [2] also helpful depending on where your current website is deployed.    

[1]   http://jsoup.org/

[2]   http://helpx.adobe.com/experience-manager/using/migrating-data-cq.html

[philjava]

Thanks for your answer.

In case I successfully read out the old webpages by strings (letters of the header) with jsoup, is there a detailed guide on how to then write into cq automatically, so it createes the templates for each according website automatically?

[Sham_HC · Accepted solution]

PhilJava wrote...

Thanks for your answer.

In case I successfully read out the old webpages by strings (letters of the header) with jsoup, is there a detailed guide on how to then write into cq automatically, so it createes the templates for each according website automatically?

The article refereed earlier [2] should give an example of write. Other references are http://docs.adobe.com/docs/en/crx/2-3/developing/accessing_jcr_connectors.html  http://dev.day.com/docs/en/crx/current/developing/accessing_the_crx.html