Test the Analysis Pipeline with an Indexed Document
In this example, we index the PDF guides of the Exalead CloudView documentation. We change the Author name
displayed in the hit content of search results. Instead of ‘EXALEAD R&D’, we want to
display ‘EXALEAD’.
You can add a new connector that includes the Exalead CloudView functional and reference product documentation. For this
example, we keep most of the default parameters.
Open the Administration Console: http://<HOSTNAME>:<BASEPORT+1>/admin
Go to Index > Connectors and click Add
connector.
Complete the fields as follows:
For Name, type Functional_guides
For Creation mode, select
new.
For Type, select Files.
Click Accept.
The Functional_guides connector page is created.
Note:
When creating connectors, always use intuitive names as by default, they appear as
navigation facet in
the Mashup UI.
On this connector Configuration page, go to the
Filesystem paths section and enter the path:
for UNIX platforms, /<INSTALLDIR>/docs/pdf
for Windows platforms,
<INSTALLDIR>\docs\pdf
Click Apply.
Go to the Home page and under the connectors list, click
Scan next to the Functional_guides
connector.
This triggers the indexing.
Go to the Mashup UI: http://<HOSTNAME>:<BASEPORT>/mashup-ui and enter #all in the search field.
Search results are displayed.
Note:
for each hit the Author meta displays Exalead R&D.
Hit with Author Meta Displaying Exalead R&D
Define the New Author
Name
Once PDFs are indexed, the first thing you need to define is the new Author
name to be displayed.
In the Administration Console, go to Index > Data Processing > Edit
> Pipeline name (e.g. ap0) > Document Processors.
In Processor types, search for the Replace
Values processor and add it above the
rename_extracted_author processor.
Click the icon and rename this new processor:
rename_exalead.
Define the following parameters:
Input from: extracted_author
String to replace: EXALEAD
R&D
Replacement string: EXALEAD
Click Apply.
Test the Document Meta Processing
You can process a document to see how the modification is taken into
account.
In the Administration Console, go to Index > Data Processing >
Test.
In Select an indexed document, choose:
your connector’s name in Source:
Functional_guides
the document to be processed in Document URI:
CloudView_GetStartedGuide_EN_R2015x.pdf
Select your analysis pipeline (by default ap0).
Click Process.
The list of generated metas and annotations is displayed on the right. You can see
that the Author
has been properly renamed ‘EXALEAD’. Metas have all been newly generated, that is why
they are displayed in green.
Now you can take a closer look at what happened in the analysis pipeline.
Display the Analysis Pipeline Details
Look for the extracted_author meta using the Filter
metas search filter on the right.
Select the analysis pipeline name (for example ap0) to expand
it.
Only document and semantic processors using the extracted_author
meta are active. You can see the two processors involved when renaming and extracting
the Author.
Notice the green icon before the rename_extracted_author
processor. It means that the condition defined for this processor is met.
Select the rename_exalead processor.
You can see that the extracted_author meta is ‘EXALEAD’. This meta
has been modified, that is why it is displayed in blue.
Select the rename_extracted_author processor.
You can see that the extracted_author meta has been removed. The
final name is still ‘EXALEAD’.
Go to the Home page and click Clear
documents from the Functional_guides
connector.
Click Scan.
Once the scan is complete, go back to the Mashup UI page and refresh the view.
icon and rename this new processor:
