On 6/12/2016 7:32 AM, Susan Shumway wrote:
I don't know about scripted access to KC search, but I can offer up
some Google search tricks for those of you that prefer that method
over the (still improving) KC search function. These are from a NaSPA
deck I helped present early last year...
Why not provide a REST API for KC searches? Even better provide a
bespoke API for looking up message IDs that sends a response in JSON lines.
IBM Knowledge Center allows you to use Boolean operators (AND/+, OR/|,
NOT/-, and so on) to create complex queries.
Example: "dynamic vipa" | dvipa
- Finds topics that contain the phrase dynamic VIPA OR the word
DVIPA anywhere in the text.
IBM Knowledge Center allows you to use search operators (intext,
inurl, intitle, and so on) to create complex queries that further
refine your search results.
Example: "dynamic vipa" | dvipa intitle:introduction | intitle:overview
- Finds topics that contain the phrase dynamic VIPA OR the word
DVIPA anywhere in the text AND that contain the word introduction or
the word overview in the title.
Each book in the z/OS library is assigned a unique identifier, which
is included in its URL. For example, book identifier e0zm100
corresponds to z/OS Migration. You can use the inurl search operator
and the book's identifier to narrow the search results to a specific
book. (The list of book identifiers is included in the presentation
but is WAY too long to include here. Let me know if you're interested
and I'll see if it's still online anywhere.)
Example: "dynamic vipa" | dvipa intitle:introduction |
intitle:overview inurl:halz002
- Finds topics in the "z/OS Communications Server: IP Configuration
Guide" book that contain the phrase dynamic VIPA OR the word DVIPA
anywhere in the text AND that contain the word introduction or the
word overview in the title.
Tip: Book identifiers are also helpful during basic searching. If you
hover over a topic title link in the search results, the displayed URL
will contain the identifier of the containing book.
On 12/04/16 6:33 PM, Paul Gilmartin wrote:
On Sun, 4 Dec 2016 10:31:13 -0800, Charles Mills wrote:
And as a side benefit, it does in fact work very nicely from an Android
device (and presumably also from an iOS device). (But unlike Doc
Buddy, not
off-line.)
And also, as I said in another thread, it works for "everything,"
not just
messages.
I wonder how it compares to Google "<search terms> site:ibm.com",
which I've been relying on? Thanks for the suggestion. An unfortunate
"search terms" for site:ibm.com is "program management" which gives
a basket of deplorable MBA-stuff but hardly any binder hits on the
first page.
Years ago, I wrote a script to drive LOOKAT into Lynx which I had
ported to MVS. Nothing needed on the desktop except tn3270.
And there's Chicago Soft's Kwik-E-Ref (whatever) for $.
And John McKown lately pondered interfacing curl to ISPF Edit.
I just tried it. It works, sorta, on simple things such as cbttape.org.
I'm not much moved to try it on KC.
http://www.anybrowser.org/campaign/
Where did we fall off the path!? I wonder whether Susan Shumway
has any thoughts on scripted access to KC search?
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
