some useful features

Jungle Ide Home Forums Jungle Ide Feature requests some useful features

This topic contains 3 replies, has 2 voices, and was last updated by  Ziggy 4 years, 8 months ago.

Viewing 4 posts - 1 through 4 (of 4 total)
  • Author
  • #5704


    Working in the Jungle ide often find myself thinking that in the Jungle ide lack of such useful features of which I often use in other ide:
    – “column selection mode” – I think everything is clear, i.e. it’s menu item and hotkey for toggle the selection mode to the column mode and back.
    – “find usage” – is something like “jump to definition” only in reverse, i.e. find where the variable\constant\method\function\class is used in the code, of course, we can use the search, but it is a matter of convenience and time saving.



    speaking of search
    The search can be started from the context menu, but seach not bound to the context i.e. logical and convenient it would be if the cursor is set to a “any_variable”, and the start search\replacement through the context menu, then search\replacement field would be filled with this “any_variable”.



    well, seems I missed, column selection mode is executed with pressed alt button.



    If you select text and then open the find dialog (Ctrl+F) the selected text will be shown as the “find” text in the dialog.

    Also, column mode is enabled by pressing Alt on selection.

    Then, find all usages: This should be possible when we get a better support for Generics, Aliases and the EachIn operator. We’re working on this on a new parser version, but it’s hard to do! We would love to add some refactoring tools to Jungle Ide, based on an improved deterministic parser that understands most of the ambiguities Monkey language has (which are not few!)
    imagine this call:

    MyCall(identifier<param1, param2>(identifier2))

    How many parameters does it has? One single cast parameter, or two bools? As Monkey syntax is not context free, it makes it very complicated to get correct “find usages” always, unless a several passes parser is involved, but this usually means slow parsing on an IDE. So, until we get this kind of issues solved (mainly caused by the IMHO very bad idea of choosing < and > for generics), we can’t provide a 100% reliable find usages.
    That’s a reason why they did not choose < and > for generics when designing the D programming language, or the generics support for VB (to name 2). Java and other languages that deliberately inherit from the C++ templates syntax, are using this operators, so they have a single symbol for several tasks, which is, form a language design perspective, in my very personal opinion, a mistake. Good languages syntax is not context dependent. But, anyway, I guess we’re not here to theorize…

    We’ll get there hopefully soon, but no ETA yet as every change in the parser needs to be tested very well before going on production.

Viewing 4 posts - 1 through 4 (of 4 total)

You must be logged in to reply to this topic.