gschem feature request: An improved find dialog

Bug #698793 reported by KaiMartin
12
This bug affects 1 person
Affects Status Importance Assigned to Milestone
gEDA
In Progress
Wishlist
Edward Hennessy

Bug Description

The find dialog in gschem is rather spartan. It just jumps to the next found substring. As the size of projects increases, the need for more powerful searches grows.

Features I'd love to see implemented:
* Allow for globbing with "?" and "*", like the bash shell does.
* Optionally, parse the search string for regular expressions
* respond with "not found" and keep the dialog instead of a silent exit.
* highlight/select all the objects that contain a hit to the search.
* An option to include invisible attributes in search
* Give feedback, how many incidences of the search string were found. Explicitely give information, on which subsheet the search string was found.
* Optionally give a list of objects that contain the search string. A click to an item in the list, jumps to the corresponding object.
* Access to a history of searches done in the past.
* Start the dialog with the last search string.
* Replace search string individually or all at once.
* Replace the symbol that contains the search string.

---<(kaimartin)>---

Changed in geda:
assignee: nobody → Edward Hennessy (ehennes)
Revision history for this message
Edward Hennessy (ehennes) wrote :

Feature request partially implemented.

Commit: 89bfcc07846078567c9de83ae1cfd6e6d9ae0515

Provides:
* Explicitly give information, on which subsheet the search string was found.
* Optionally give a list of objects that contain the search string. A click to an item in the list, jumps to the corresponding object.

Revision history for this message
KaiMartin (kmk-familieknaak) wrote :

Hi Edward.
I am pleased to see my wishes (partially) come true!
Thank you for coding!

Looking forward to the new "docked" design of the UI.

---<)kaimartin(>----

Revision history for this message
Jason (whitewaterssoftwareinfo) wrote :

As per your request from the discussion on the geda-user mailing list I will repost the text of my suggestion:

> From: Jason White
> ...
> I would be tempted to offer an option to split the attribute and the
> value into different columns. That way if your search hits upon
> multiple types of attributes, you can sort by type with a single click
> on the column header.

> From: Edward Hennessy in reply to Jason White
> ...
> Thanks for the ideas. Sorting by column would be very useful.
>
> I’m not sure how to mix attributes and regular text objects in the result.
> But, I can see that sorting by the attribute value would be useful. Maybe
> it can split the columns only when all results are attributes?

Yes, that would be sufficient. What I envisioned is that there would always be an attribute and a value column, regardless of results. If a row in the results was not an attribute object, the the attribute field could simply say "Text Object", this way Text Objects can be sorted amongst different types of attributes.

As for the cases where no rows are Attributes, the above behavior would still be sufficient to allow the user to search. So I think it would be possible to get away with just have one "mode" of search. If a need arises then a new mode can be added in the future. I don't know anything about how this feature is implemented, but if its possible and not too much work, it would be nice if it was structured so that it would be easy for different search modules to be switched in and out. (Do I have anything specific in mind? No... Just a general though.)

Changed in geda:
status: New → In Progress
Revision history for this message
Edward Hennessy (ehennes) wrote :

Added both pattern and regular expression search capabilities to the find text operation.

Commit: 93bddf6b20c75326d357ef7df42f643bda5f8721

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Duplicates of this bug

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.