New features #751

SteveTheTechie · 2018-02-25T08:24:39Z

This closes #750, closes #721, closes #605, closes #513, closes #467, closes #454, closes #436, closes #232, closes #231, closes #169, closes #163, and closes #86

Many new features and improvements:

Expanded on iconSet concept to create linkInfo option to allow configuration of all header link attributes, including icons, text, and titles. I needed a more general approach to accommodate 2 new link possibilities: Collapse All and Expand All. (for the opt groups) The new header option usage provides the means to manage this increased flexibility. (Feature Request: Collapsable opt groups #232)
Menu height can now be dynamically resized via the jQuery UI resizable widget, if loaded and resizableMenu option set to true. (Update jquery.multiselect.js #436)
Fixed some issues w/ the menu height getting mangled when filtering was in used. (ISSUE-163 Make optgroups searchable and return its children as well. #605)
Modifed how header option is used. Can still set to false to hide header, but now it is also used to specify what header links are seen and in what order. (Feature Request: Collapsable opt groups #232)
The multiselect can now also be used as a fixed list box (always open; no pop-up; no button) via the listbox option. (Fixed menu (list box) variant #721)
New options groupsSelectable and groupsCollapsable control whether option groups can be selected and/or collapsed, respectively. (Feature Request: Collapsable opt groups #232)
Option groups can now be set to be collapsable. A new small button shows up at the left of each option group heading to collapse/expand the option group. (Feature Request: Collapsable opt groups #232)
New resync and value methods allow quickly updating the widget options checked state from the underlying native select without needing to recreate the options via a refresh(). (Resync method #750)
Can now use the up/down keys on closed single select to change the selected option if the button is focused. (Add keyboard event handlers for closed state in single select mode #86)
Fixed a previously unknown bug caused by binding menu events to the outer $menu container div instead of the $checkboxes container. This bug caused the menu events handler to also try to handle header events.
Added the ability to reference option groups by their label or index for use as a filter. Now usable programmatically in addOption, checkAll, uncheckAll, flipAll, collapseAll, and expandAll. (Feature Request: Ability to Select/Unselect All/disable programatically for an opt group #231)
Simplified the auto width determination for the _setMenuWidth() method by tweaking the related CSS rules. As a benefit, the _getScrollBarWidth() function has been eliminated.
Updated and simplified the _setMenuHeight() method to allow it to better work with filtering. (ISSUE-163 Make optgroups searchable and return its children as well. #605)
Updated the open() method so that the first checked option gets the focus when not using the filter. (Reopening 'multiple : false' should jump to current selection #169, No automatic scroll when selected option is not visible #467)
Re-wrote the filter widget--can now search option group headings and supports alternative filter rules. (multiselectfilter plugin ignores optgroup labels #163, Add the possibility to define own filter rules #454, ISSUE-163 Make optgroups searchable and return its children as well. #605)
Updated and added unit tests.
Updated demos and changed theme to Redmond--no more ugly construction orange.
Updated i18n files for new linkInfo option format.

**Closes #750, #721, #605, #467, #454, #436, #232, #163, #86** ### Many new features and improvements: 1. Expanded on iconSet concept to create linkInfo option to allow configuration of all header link attributes, including icons, text, and titles.s. 2. Menu height can now be dynamically resized via the jQuery UI resizable widget, if loaded and resizableMenu option set to true. 3. Fixed some issues w/ the menu height getting mangled when filtering was in used. 4. Modifed how header option is used. Can still set to false to hide header, but now it is also used to specify what header links are seen and in what order. 5. The multiselect can now also be used as a fixed list box (always open; no pop-up) via the listbox option. 6. New options groupsSelectable and groupsCollapsable control whether option groups can be selected and/or collapsed, respectively. 7. Option groups can now be set to be collapsable. A new small button shows up at the left of each option group heading to collapse/expand the option group. 8. New resync and value methods allow quickly updating the widget options checked state from the underlying native select. 9. Can now use the up/down keys on closed single select to change the selected option if the button is focussed. 10. Fixed a previously unknown bug caused by binding menu events to the outer $menu container div instead of the $checkboxes container. This bug caused the menu events handler to also try to handle header events. 11. Added the ability to reference option groups by their label or index for use as a filter. Now usable programmatically in addOption, checkAll, uncheckAll, flipAll, collapseAll, and expandAll. 12. Simplified the auto width determination for the _setMenuWidth() method by tweaking the related CSS rules. As a benefit, the _getScrollBarWidth() function has been eliminated. 13. Updated and simplified the _setMenuHeight() method to allow it to better work with filtering. 14. Re-wrote the filter widget--can now search option group headings and supports alternative filter rules. 15. Updated and added unit tests. 16. Updated demos and changed theme to Redmond--no more ugly construction orange. 17. Updated i18n files for new linkInfo option format.

mlh758 · 2018-02-25T15:17:53Z

By the description this is a lot of welcome changes. I won't be able to look it over today but I'll try to get it merged early this week.

SteveTheTechie · 2018-02-25T19:56:14Z

@mlh758 Understood... I put a lot of time into this, so I would expect that you would need some extra time in to look it over. I think it is pretty complete, but I will also continue to look for where I may have missed things. I look forward to any questions you may have as well.

SteveTheTechie · 2018-02-26T00:00:29Z

@mlh758 My intention is that this PR will essentially be my last contribution for a Version 3 release--I am hopeful that you will feel comfortable doing the release. 😉

To that end, I would appreciate it if you would take some added time to give this a critical look, not just from whether the feature set, options, methods, naming, comments, etc. make sense, but if there is anything I am missing or could have done better.

There were a number of things that I specifically did and did not include in this:

Doing collapsable option groups "right" was a bit more involved than I anticipated. I needed to enable Collapse All and Expand All links functionality for the header, even though I knew those may not always be needed. That is what led me to a number of decisions around revamping the header functionality.
The jQuery UI Resizable widget is one of those widgets that I have previously alluded to that does not always work as advertised--it has given me problems in the past in other efforts. This is why I added an option for enabling its use for dynamic height resizing.
I did more with CSS rules this time around. This worked out really well and allowed me to eliminate some JavaScript.
I implemented your suggestion to create a helper function to find option groups by text or index. It worked out really well. 👍
I did not include RTL support... I do not really know how to do that well, but I am willing to work w/ someone on it in the future, as the gained knowledge could benefit me on my main web app project.
I did not do anything on the open jQuery Validate related issues... I am not familiar w/ jQuery Validate.
I did not do anything with the open ASP related issues... I am not familiar w/ ASP.
There are probably a few open issues that I could have checked w/ the new code, but I did not. I will leave that to you--hopefully, some of those can be closed as being outdated.
I did not do anything related to infinite scrolling, JSON data, or AJAX. I think that those and the Table View idea I posted as an issue should be done together as a future version update geared to collectively enhancing the widget to support "large data sets" since they all kind of relate to that in one way or another.
I did not do anything related to divorcing the code from jQuery UI. I believe that should be done in a future update. For some perspective, here is a widget that has jQuery as a dependency, but not jQuery UI (it used to depend on jQuery UI also): https://fullcalendar.io/docs/usage/ Doing the divorce is doable.

mlh758 · 2018-02-27T14:43:15Z

I opened the callbacks demo and tried clicking on a collapsed option group. Instead of selecting everything within it, or doing nothing, it seems to call uncheckAll and clear selections. I think it should behave as a normal option group click.

The filter also seems to no longer function in the dialog gist.

SteveTheTechie · 2018-02-27T18:29:20Z

I will take a look. When you say dialog gist, I am not sure what you are referring to.

mlh758 · 2018-02-27T19:08:24Z

src/jquery.multiselect.filter.js

-         $instance._oldToggleChecked(flag, group, true);
+      // is fired, only the currently displayed filtered inputs are checked if filter entered.
+      var instance = this.instance,
+            filter = this.$input[0];


I prefer having a var per declaration instead of combining them into a single statement.

mlh758 · 2018-02-27T19:09:57Z

src/jquery.multiselect.filter.js

-            $rows.eq(i).show();
-            return $inputs.get(i);
+      var term = this.$input[0].value.toLowerCase().replace(/^\s+|\s+$/g,''),
+            filterRule = this.options.filterRule || 'contains',


Same thing here about a var per variable.

mlh758 · 2018-02-27T19:18:13Z

src/jquery.multiselect.filter.js

+         // If not searching in groups then show all group headings in the results.
+         $checkboxes.find('.' + optgroupClass).removeClass(hiddenClass);
+      }
+      var filteredInputs = $checkboxes.children().map(function(x) {


I would prefer variable names more descriptive than x and y here.

@mlh758 These are just index variables... what would you suggest as an alternative... index1 and index2?

groupIndex and optionIndex?

@mlh758 However, if there are no groups, then groupIndex would actually be the optionIndex.
(There would only be options.) Thus, this would be confusing for the no groups case.

Yeah that wouldn't be any good. We could probably reassess what it looks like after the filtering issue is worked out.

Yeah, leave it as x and y for now then.

mlh758 · 2018-02-27T19:53:42Z

src/jquery.multiselect.js

+      var linkInfo = ( this.linkInfo = $.extend(true, {}, linkDefaults, options.linkInfo || {}) );
+
+      // Helper function used below
+      function _linkHTML(linkTemplate, linkID) {


This should probably live as it's own somewhere else, it doesn't need to be assigned to the widget itself but it's not capturing anything in here in a closure and it's not a callback so it doesn't need to be a nested declaration.

@mlh758 If you look closely, it uses the linkInfo variable just above it. I originally had it outside of _create(), but I had to reference this.linkInfo. Since _create() is the only thing using it, I just decided to pull it inside of _create().

Would you prefer it be outside of _create()? Keep in mind it will still need to use the linkInfo stuff somehow.

_create is assigning it to this.linkInfo so the _linkHTML method would be able to access it that way. I would prefer to have it outside the _create method since it isn't a callback, and the data it's capturing in the closure is also accessible via this.

@mlh758 Ok, I will move it back outside _create() then.

mlh758 · 2018-02-27T20:00:42Z

src/jquery.multiselect.js

+         'title': 'Collapse'
+      },
+      'expand': {
+         'class': 'NOT USED',


Do these even need to exist if they're not used?

Mainly there for documenting so that people do not think they were missed and try to use them. I can remove them.

@mlh758 So do you want me to remove them then?
I could put the "NOT USED" bit in comments if that would be better.

I think you can just remove them, not really a need for a comment either. It seems like their lack of existence would be enough to indicate they aren't used.

mlh758 · 2018-02-27T20:29:43Z

src/jquery.multiselect.js

+					var prev, 
+                   current = null, 
+                   next = null;
+					self.$inputs.each( function() {


This should be a separate method, not a block dropped into the middle of they key handlers.

@mlh758 I struggled about what to do with this. It closes a PR, but it is quite a bit of code to just add specialized up/down key functionality for a closed single select. I could make a separate method, but it will only be used in the keydown handler.

The change itself is welcome, It's just a lot of extra code in them middle of an already large, awkwardly indented block. Putting it into a method like handleButtonKeyboardNav and just passing that in would be cleaner here.

@mlh758 ok will do

mlh758 · 2018-02-27T20:31:08Z

src/jquery.multiselect.js

      // optgroup label toggle support
-      this.$menu.on('click.multiselect', '.ui-multiselect-optgroup a', function(e) {
+      self.$checkboxes.on('click.multiselect', 'a', function(e) {


If this is now binding to checkboxes instead of menu, the method name should change accordingly.

mlh758 · 2018-02-27T20:34:09Z

src/jquery.multiselect.js

@@ -549,6 +700,34 @@
          checked: nodes.length ? nodes[0].checked : null
        });
      })
+      // collapse button
+      .on('click.multiselect', 'button', function(e) {


Maybe specify a class for the collapse button so it's more specific to that button instead of any button?

@mlh758 Ok. BTW, are you ok w/ a button vs a styled anchor?

Yeah I'm okay with it being a button.

mlh758 · 2018-02-27T20:41:30Z

src/jquery.multiselect.js

+
+      // Close each widget when clicking on any other element/anywhere else on the page,
+      // another widget instance, or when scrolling w/ the mouse wheel outside the menu button.
+      self.document.on('mousedown.' + self._namespaceID


These document listeners don't get destroyed when the widget is destroyed, and are created for each widget.

If the widget is destroyed and a new one created, these leak and multiply.

@mlh758 Ok, I will have to address that in the destroy.

Looking at the following below... doesn't this.document.off(this._namespaceID); get rid of the document listeners for this widget's namespace?

destroy: function() { // remove classes + data $.Widget.prototype.destroy.call(this); // unbind events this.document.off(this._namespaceID); $(this.element[0].form).off(this._namespaceID); if (!this.options.listbox) { this.$button.remove(); } this.$menu.remove(); this.element.show(); return this; },

That's what it is supposed to be doing, and in master it appears to be (I just double checked). I wonder if another change along the way broke that?

It looks like I was removing the initial period from the namespace ID and it is needed for off().

mlh758 · 2018-02-27T20:44:52Z

src/jquery.multiselect.js


      // The following adds up item heights.  If the height sum exceeds the option height or if the number
      //   of item heights summed equal or exceed the native select size attribute, the loop is aborted.
      // If the loop is aborted, this means that the menu must be scrolled to see all the items.
-      $checkboxes.find('li,a').each( function() {
+      self.$checkboxes.find('li:not(.ui-multiselect-optgroup),a').filter(':visible').each( function() {


Why exclude the option groups here?

@mlh758 Because the option group li's contain a ul which contains all the child option li's. My concern was that I would be adding the height of the container li plus the heights of each child li. I do not want to be "double dipping" on the heights.

The anchor is really supposed to handle the height of the option group heading, but I do have a lingering question of whether the collapse button may be taller. The bottom line is that need a way to get the height of the optgroup headings without the child option li's. The current method is not perfect, but it seems to work well enough.

Got it, since find also goes into descendants. Makes sense.

@mlh758 If you have any thoughts on any better ways to account for the heights of the option group headings, I am all ears. As I noted, I do not think the current method is perfect.

I don't at this time, I just wanted to know why the change was made since it was so recently added.

@mlh758 The main potential problem here is if someone is using collapsable groups, then you also have the height of the collapse buttons to consider. If someone changes the CSS to make the buttons taller, then potentially the button heights should be examined instead of the anchor heights. Right now, I am not seeing much of a difference (I have not tried changing the button heights), but it has occurred to me that if the element being examined is an anchor, maybe I should look for a sibling button and use the Math.max height of the two (button and anchor).

mlh758 · 2018-02-27T21:00:00Z

This really should have been 10 different PRs, would have been a lot easier on both of us.

Especially with adding #721 in after I asked you not to. I would prefer that as a separate widget like the filter is a separate widget since it changes a lot of behavior.

mlh758 · 2018-02-27T21:25:30Z

The extra demos gist I made: https://gist.github.com/mlh758/b1b24931db7705441f9786bd68cf3569

One of them is for a dialog, we've used it before for testing/troubleshooting. That's where the speed test thing came from.

SteveTheTechie · 2018-02-28T06:43:16Z

I opened the callbacks demo and tried clicking on a collapsed option group. Instead of selecting everything within it, or doing nothing, it seems to call uncheckAll and clear selections. I think it should behave as a normal option group click.

Ok, I see that and I agree it is an issue.

However, there is really more to this. You have to consider what happens when the filter is in use, also. The filter was previously set up so that you could not search for optgroup labels, but it would always show all optgroup headings. If you clicked on a heading, you would only set the checks for the options that were visible / not filtered out.

Now, since I tried to enable searching for option group headings in the spirit of the PRs that requested it, if the optgroup searching option is enabled, it only shows optgroups for which its heading or one of its children contains the search text. If an optgroup heading matches, then all of its children are shown, also.

So... now there are 3 cases that can show / hide options... for the filter cases, clicking the option group heading is only supposed to change the visible options.

Consider the case of a filter in use + collapsed option group and someone clicks on the heading... The way I intended to work was that only visible optgroup options would get change on a heading click, regardless of how they got visible/hidden. I could change the behavior if a filter is in use vs not... your thoughts?

mlh758 · 2018-02-28T16:00:06Z

but it would always show all optgroup headings.

It only showed option group headings where some children were visible. This meant you could search for something and have results displayed where you just had a few options left under a couple of option groups. Clicking one of those option groups would select only the visible members of that option group.

I think re-ordering the way the filter is done in your code might make this easier, and your addition of the hidden class further simplifies the problem.

Apply the search to the options as normal, adding the hidden class to items that should be excluded.
If enabled, apply the search to the option groups. If an option group qualifies, remove the hidden class from all of its children.
Hide all option groups that have no visible children as before.

The downside is some potential redundant adding/removal of the hidden class on some options as the option group filtering will potentially undo some of the filtering applied in step 1, but I'd be okay with that to simplify the mechanism.

It looks like the collapse adds a class to the ul, which is separate from the hiding in the filter. It seems like their behavior should be able to overlap safely.

SteveTheTechie · 2018-02-28T17:50:29Z

@mlh758

This really should have been 10 different PRs, would have been a lot easier on both of us.

My enthusiasm for doing more PRs for this is fading at the moment (it will likely come back later in the future), so I needed to get as much done as I could before I feel the need to move on to other things.

Especially with adding #721 in after I asked you not to. I would prefer that as a separate widget like the filter is a separate widget since it changes a lot of behavior.

Let's just say that we have a difference of opinion on this. I am not all that sold on the separate widget approach--it adds unnecessary overhead and perpetuates the use of jQuery UI, which I think needs to be phased out in favor of another class based approach. Also, #721 only removes the button and fixes the menu position in the html. It is basically a subtractive approach, and I tend to believe that additional widgets work better for adding or replacing things, not taking them away. If do this as another widget, I would end up having to replace quite a few main widget methods for just some small changes. Lastly, I believe that the groupColumns option is also a change in behavior/look but that is part of the main widget and not a separate widget--so that does not need to be a separate widget, but this does?

mlh758 · 2018-02-28T17:52:16Z

groupColumns option is also a change in behavior/look but that is part of the main widget and not a separate widget--so that does not need to be a separate widget, but this does?

Touche, leave it in.

SteveTheTechie · 2018-02-28T19:17:07Z

@mlh758

The jQuery UI Resizable widget is one of those widgets that I have previously alluded to that does not always work as advertised--it has given me problems in the past in other efforts. This is why I added an option for enabling its use for dynamic height resizing.

Case in point... try running the Animations demo w/ resizable enabled.

mlh758 · 2018-03-02T21:14:06Z

Just a heads up, tomorrow I'll be leaving on a short vacation so I won't be responding much for the next 5 days.

SteveTheTechie · 2018-03-02T21:24:29Z

Ok thx I will probably post an update to this PR by Sunday.

Have a good vacation!

1. Make optgroup clicks function on the optgroups regardless of collapsed status. However, filtered status is still observed for excluding from clicks. 2. Fixed issue of filter not working in dialogs. Apparently, widget instance() method is no longer supported, per StackOverflow thread. Using .data('ech-multiselect') to fetch the instance now--this works w/ the dialog. 3. When filtering, now hide optgroup headings unless a match on children li's or optgroup heading (if searching optgroup headings option set) 4. Formatting tweaks. 5. Switch to using arrays for list items in options... e.g. header option to specify displayed links in desired order. 6. As IE10+ is IE minimum, switched to using array.indexOf() for array lookups and classList where appropriate. 7. Ditched NOT USED in linkDefaults object. 8. Made _linkHTML() an internal method function. 9. Implemented classes for the optgroup collapse button and the optgroup label anchor. 10. Fixed resync. find('option') not find('options') 11. Fixed destroy. '.' prefix on namespace is not optional for .off() (document listeners issue) 12. Moved single select button up/down code to separate method and simplified that code. 13. Changed bindMenuEvents to bindCheckboxEvents

SteveTheTechie · 2018-03-05T02:42:48Z

@mlh758 I know you are on vacation... Let me know if the changes I made will address the

SteveTheTechie · 2018-03-05T02:43:22Z

@mlh758 Oops closed this by mistake.

mlh758 · 2018-03-05T15:29:16Z

Apparently, widget instance() method is no longer supported, per StackOverflow thread. Using .data('ech-multiselect') to fetch the instance now--this works w/ the dialog.

Welp, that means this is my fault. Can you link that Stack Overflow thread? instance was added in newer jQuery UI so it not being supported would be very strange. The issue should be that my dialog demo doesn't pull in a new enough jQuery UI.

mlh758 · 2018-03-05T15:31:54Z

src/jquery.multiselect.filter.js

@@ -258,7 +256,7 @@
    */
    destroy: function() {
      $.Widget.prototype.destroy.call(this);
-      this.$input.val('').trigger("keyup");
+      this.$input.val('').trigger("keyup").off('keydown input search');


Good call here.

mlh758 · 2018-03-05T15:36:06Z

src/jquery.multiselect.js

-            if (!$button.hasClass('ui-state-disabled')) {
-              $button.addClass('ui-state-focus');
+            if (!this.classList.contains('ui-state-disabled')) {
+              this.classList.add('ui-state-focus');


Cool, I didn't know this was available in IE10.

Per https://developer.mozilla.org/en-US/docs/Web/API/Element/classList , "Basic Support" is available... Note that I am not using anything fancy w/ it.

mlh758 · 2018-03-05T15:40:47Z

Code wise I'm happy. My only hangup is wanting to figure out what happened with the instance thing, although since that was the only new jQuery UI feature in place, it's probably best to leave it out anyway. No reason to force people to upgrade for one method.

I haven't had a chance to pull this down and test it, but I don't see any more code changes I want.

SteveTheTechie · 2018-03-05T16:42:32Z

Unfortunately, my computer crashed last night that I have been doing these updates on. Laptop screen is toast. It is an old laptop, so I am not sure what I am going to do at the moment. (typing this on work computer... cannot use it for updates)

All I can say on the instance thing was I kept getting an error of "no such method 'instance' in multiselect instance" in multiselectfilter.js line## in the dialog gist ... so I had to figure out a different approach and I did some research on StackOverflow to find the data() approach. I do not have the thread readily at hand, unfortunately. I also saw a comment on the jQuery UI forum that the instance method was buggy. Either way, I thought maybe using the data approach may be more reliable.

One thing to look at is that I think I put in something like this.element.multiselect().data('ech-multiselect'), but in hindsight, I do not think the multiselect() part should be needed. (just use this.element.data('ech-multiselect'))

mlh758 · 2018-03-05T16:49:43Z

Yup, that was caused by me pulling in an old version of jQuery UI for that demo. The filter thing was my fault. I'll leave your change in though since there's no reason to force an upgrade over a single method that is only convenience. If I find any issues in testing I'll figure out something else to do to fix things up but this will probably be fine to merge at this point.

SteveTheTechie · 2018-03-05T16:56:06Z

Here are the relevant posts, although I guess not needed now as problem solved:
https://stackoverflow.com/questions/30365669/jquery-autocomplete-custom-data-error-no-such-method-instance-for-autocomplete
https://forum.jquery.com/topic/issue-with-jquery-ui-selectmenu-no-such-method-instance

SteveTheTechie · 2018-03-05T17:18:37Z

src/jquery.multiselect.js

+      selectedListSeparator: ', ',        // (string) This allows customization of the list separator.  Use ',<br/>' to make the button grow vertically showing 1 selection per line.
+      maxSelected: null,                  // (integer | null)  If selected count > maxSelected, then message is displayed, and new selection is undone.
+      openEffect: null,                   // (array) An array containing menu opening effect information.
+      closeEffect: null,                  // (array) An array containing menu closing effect information.


@mlh758 I changed these option names (show/hide -> openEffect/closeEffect) to be more clear about what they actually do. However, I also tend to think that the widget may be relying on deprecated jQuery UI API signatures for these, as current docs mention the use of objects for animations. Therefore, these need to be re-looked at to see if they should be updated to reflect the current API. I will post an issue for this.

This reverts commit 1815e57.

SteveTheTechie · 2018-03-08T18:15:23Z

@mlh758 Are you still testing this? Are there any further changes needed?

mlh758 · 2018-03-08T18:31:20Z

I only got back into town yesterday evening. I haven't had a chance to pull it down yet.

SteveTheTechie · 2018-03-08T18:34:39Z

oops sorry 😊

mlh758 · 2018-03-09T15:51:38Z

Everything looks good. Going to merge now.

mlh758 · 2018-03-09T16:09:13Z

I published version 3 pre-release with minified files attached so people can start messing with it.

SteveTheTechie added 4 commits February 25, 2018 00:23

File mix-up

cb06423

File mix-up

31e5da1

Fix formatting

bc711d2

SteveTheTechie mentioned this pull request Feb 26, 2018

I have a number of updates... #717

Closed

9 tasks

SteveTheTechie added 2 commits February 25, 2018 22:13

Omitted a break statement

5286ae1

No array.indexOf() & simplify btn keypress handler

7f7f3e5

mlh758 requested changes Feb 27, 2018

View reviewed changes

SteveTheTechie closed this Mar 5, 2018

SteveTheTechie reopened this Mar 5, 2018

mlh758 reviewed Mar 5, 2018

View reviewed changes

mlh758 approved these changes Mar 5, 2018

View reviewed changes

SteveTheTechie commented Mar 5, 2018

View reviewed changes

SteveTheTechie added 2 commits March 5, 2018 15:35

Ooops, missed some link classes

1815e57

Revert "Ooops, missed some link classes"

07e04d7

This reverts commit 1815e57.

mlh758 merged commit 5d42519 into ehynds:version3 Mar 9, 2018

New features #751

New features #751

Conversation

SteveTheTechie commented Feb 25, 2018 • edited Loading

Many new features and improvements:

mlh758 commented Feb 25, 2018

SteveTheTechie commented Feb 25, 2018

SteveTheTechie commented Feb 26, 2018 • edited Loading

There were a number of things that I specifically did and did not include in this:

mlh758 commented Feb 27, 2018

SteveTheTechie commented Feb 27, 2018

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

SteveTheTechie Feb 28, 2018 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

SteveTheTechie Feb 28, 2018 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

SteveTheTechie Feb 28, 2018 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

SteveTheTechie Feb 28, 2018 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

SteveTheTechie Mar 2, 2018 • edited Loading

Choose a reason for hiding this comment

mlh758 commented Feb 27, 2018

mlh758 commented Feb 27, 2018

SteveTheTechie commented Feb 28, 2018

mlh758 commented Feb 28, 2018 • edited Loading

SteveTheTechie commented Feb 28, 2018

mlh758 commented Feb 28, 2018 • edited Loading

SteveTheTechie commented Feb 28, 2018

mlh758 commented Mar 2, 2018

SteveTheTechie commented Mar 2, 2018 • edited Loading

SteveTheTechie commented Mar 5, 2018

SteveTheTechie commented Mar 5, 2018

mlh758 commented Mar 5, 2018 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

mlh758 commented Mar 5, 2018

SteveTheTechie commented Mar 5, 2018 • edited Loading

mlh758 commented Mar 5, 2018

SteveTheTechie commented Mar 5, 2018

SteveTheTechie Mar 5, 2018 • edited Loading

Choose a reason for hiding this comment

SteveTheTechie commented Mar 8, 2018

mlh758 commented Mar 8, 2018

SteveTheTechie commented Mar 8, 2018

mlh758 commented Mar 9, 2018

mlh758 commented Mar 9, 2018

SteveTheTechie commented Feb 25, 2018 •

edited

Loading

SteveTheTechie commented Feb 26, 2018 •

edited

Loading

SteveTheTechie Feb 28, 2018 •

edited

Loading

SteveTheTechie Feb 28, 2018 •

edited

Loading

SteveTheTechie Feb 28, 2018 •

edited

Loading

SteveTheTechie Feb 28, 2018 •

edited

Loading

SteveTheTechie Mar 2, 2018 •

edited

Loading

mlh758 commented Feb 28, 2018 •

edited

Loading

mlh758 commented Feb 28, 2018 •

edited

Loading

SteveTheTechie commented Mar 2, 2018 •

edited

Loading

mlh758 commented Mar 5, 2018 •

edited

Loading

SteveTheTechie commented Mar 5, 2018 •

edited

Loading

SteveTheTechie Mar 5, 2018 •

edited

Loading