#6001: [with patch, needs work] tear out docstrings in the notebook into a new
window
-------------------------+--------------------------------------------------
 Reporter:  mpatel       |       Owner:  boothby
     Type:  enhancement  |      Status:  new    
 Priority:  minor        |   Milestone:         
Component:  notebook     |    Keywords:         
-------------------------+--------------------------------------------------

Comment(by mpatel):

 I've tested the new, possibly improved [attachment:introspect_v2.js
 `introspect.js`] on Fedora 9 Linux in Firefox 3.0.10, Opera 9.64, and a Qt
 4.5.0 !WebKit-based "browser" demo.

 I'm not sure about how well the latter, which I found in
 `/usr/lib64/qt4/demos/browser,` approximates Safari.  It returns
 {{{
 "Mozilla/5.0 (X11; U; Linux; en-US) AppleWebKit/527+ (KHTML, like Gecko,
 Safari/419.3)  demobrowser/0.1"
 }}}
 when I enter `navigator.userAgent` in the !JavaScript console (check the
 "Tools" menu or right-click on any element, then click on "Inspect").  To
 get "tab" introspection working, I just added `'sl':keyboard_saf` near the
 end of `keyboards.py.`  Perhaps this is related to ticket #4046.

 Anyway, it seems all the browsers have their own way of doing just about
 everything, and no major UI library handles pop-ups seamlessly across
 browsers.  We may not (cannot?) get identical behavior on all major
 browsers.

 Some possible tests, besides the obvious:

  * Check whether multiple tear-outs from a worksheet go to a common
 window.
  * Ditto, from multiple worksheets.
  * Ditto, after reloading a worksheet(s).
  * Ditto, after navigating to another domain (e.g., `google.com`) in a
 tear-out window(s) (see [http://en.wikipedia.org/wiki/Same_origin_policy
 same origin policy]).
  * Check the behavior when tear-outs go to an ''unsized'' window and the
 browser is set to open tabs vs. set to open windows.
  * Ditto for ''sized'' windows.

 For the latter, see the `window.open()` lines in `introspect.js.`  In
 Firefox, I still get a blank tab when I force new windows to open in tabs
 and use unsized pop-ups.  I have got this to work with a somewhat
 different approach, but I haven't yet reconciled it with what's in v2.

 Drag-and-drop sometimes behaves oddly in Firefox, adding copious spurious
 placeholders to the list.  I've noticed this after I navigate to a
 different domain in a tear-out window, forcing the script to open a new
 window.  I think this is a jQuery [UI] bug, since it goes away when I
 substitute the latest versions.  This and other quirks may be a good
 reason to upgrade both.  Worst case: Disable some tear-out features for
 certain browsers.

 Some ideas, though my patience with browser "programming" has all but
 vanished:

  * Do something server-side instead or in addition.
  * Add a top-level toolbar to the tear-out window, e.g., to toggle, close,
 resize all docstrings.
  * Add a simple "cell" input that calls
 `window.opener.evaluate_cell_inspection().`
  * Cookies, preferably chocolate.

 Some ideas related to #5653:

  * Collect anonymous statistics on docstring access: compute correlations,
 do Bayesian predictions, rank the greatest (and worst) docstring writers
 of '''''all time'''''.
  * Generate a large graph of all docstrings and their cross-links.  Drive
 the graph theorists wild.  Actually, I think it's rather sparse at the
 moment.  Not an [http://en.wikipedia.org/wiki/Expander_graph expander]?
  * Upgrade Sphinx and test the Qt help builder, though it may
 [http://groups.google.com/group/sphinx-
 dev/browse_thread/thread/64dd226d9b4cec36/6fa3e0d251dae768#6fa3e0d251dae768
 need work].  Try running Qt's `assistant` (`assistant-qt4` on Fedora 9)
 for a taste.
  * Add "PDF" to the toolbar, with maybe a rude surprise for PDF plug-in
 users.
  * One-click copy-pasting of examples for the truly lazy.

-- 
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/6001#comment:6>
Sage <http://sagemath.org/>
Sage - Open Source Mathematical Software: Building the Car Instead of 
Reinventing the Wheel

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"sage-trac" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/sage-trac?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to