02859b32806a590d0dd38b3819c0c5756ffebbd1
tdreszer
  Wed Oct 26 16:01:10 2011 -0700
knocked off another couple of minor bugs in js.
diff --git src/hg/js/subCfg.js src/hg/js/subCfg.js
index 623840c..83dfae4 100644
--- src/hg/js/subCfg.js
+++ src/hg/js/subCfg.js
@@ -5,30 +5,58 @@
 // opened.  Composite and view level controls (parents) when updated override related
 // subtrack controls (children).  Subtrack controls, when updated overide parent controls
 // for the one subtrack.  Controls wil get class 'changed' added when changes are made.
 // When the form is submitted, all controls not marked as "changed" will be unnamed and will
 // therefore not make it into the cart.
 
 // Definitions as used here:
 // obj: an input or select style html control which may be marked as "changed"
 //   parentObj: composite or view level obj which has subtrack level childObjs associated
 //   childObj: subtrack level obj that has composite and or view level parentObjs
 // cfg: subtrack level embedded dialog which can be opened or closed (hidden) and isn't
 //      populated till first opened.  Can also be a viewCfg and maybe a compositeCfg
 // populate: act of filling a subtrack cfg with controls
 // fauxVis: fake control for subtrack visDD, which will be replaced with true vis when clicked.
 
+// TESTED:
+// OK  subCBs enable/disable subVis and subCfg
+// OK  subCBs set compVis to full
+// OK  subVis faux works
+// OK  subVis ajaxed on click
+// OK  subVis change is persistent (as seen in: hgTracks image, rightClick, return to hgTrackUi)
+// OK  subVis overridden by viewVis
+// OK  subVis overridden by compVis
+// BUG subVis set compVis to full (when compVis set to hide, but subVis already populated, setting subVis to non-hide DOES NOT set compVis to full)
+// OK  subCfg wrench disabled
+// OK  subCfg wrench enable/disable by subCB
+// OK  subCfg wrench enable/disable by subVis
+// OK  subCfg populates (vias ajax) on click
+// OK  subCfg settings persist (as seen in hgTracks rightClick)
+// OK  subCfg open closes mdbArrow display
+// OK  subCfg closed by mdbArrow open
+// OK  subCfgs all closed on a table sort
+// OK  subCfg overridden by viewCfg
+//     STRANGE: clicking wrench to open does not leave viewCfg so viewCfg change isn't (yet) seen.
+// OK  subCfg overridden by compCfg
+// OK  matrix clicks enable/disable sub
+// OK  matrix clicks set compVis to full
+// OK  matrix [+][-] enable/disable subs
+// OK  matrix [+] set compVis to full
+// OK  viewVis changes set compVis to full
+// OK  subVis correct in hgTrackUi, hgTracks image, rightClick memnu, rightClick poCfg
+// -   cart limited to minimum changed
+
 // TODO:
 // - SOLVED: checkboxes: working with name = boolshad.{name}   FIXME: multishad?
 // - SOLVED: filterBy,filterComp working: they rely upon a unique id, and id with '.' screwed it all up.  So replaced '.' with '_-'
 // - SOLVED: OpenChromSynth: subtrack filterby needs to be updated by composite filterBy.
 // - SOLVED: check subtrack to enable/disable fauxVis and wrench
 // - SOLVED: matCB should effect subCb including enable/disable fauxVis and wrench
 // - SOLVED: composite/view vis should effect subVis and enable/disable fauxVis and wrench
 // - SOLVED: inside out: changing subtrack vis should affect subCB and matCB
 // - SOLVED: Loosing checked tracks!!!  Setting vis clears checkboxes?
 // - TESTED: hui.c #ifdef SUBTRACK_CFG should switch full functionality on/off
 // - SOLVED: Make "disabled" subCB clickable!
 // - SOLVED: PROBLEM is fauxDisabled.  SOLUTION is convert fauxDisabled to true disabled on form.submit()
 // - DECIDED: When parent vis makes subs hidden, should they go to unchecked?   No, disabled!
 // - DECIDED: Should user be able to click on disabled vis to check the CB?  No, not important.
 // - DECIDED: Make vis changes "reshape" composite!  NOTE: Do we want to do this???   I am against this as too disruptive.  We may want to end reshaping in CGIs as well!