jcparkyn / nodexr Goto Github PK

This would replace the current modal dialog that appears after clicking the Edit button.
Syntax colouring would be disabled while editing to prevent the need for parsing the expression while typing.
A dialog would appear below the output to allow the user to cancel the edit (also by pressing Escape) or confirm (also by pressing Enter).

Currently the parser fails if the { is not followed by a digit.

This is a limitation of svg gradients; can be fixed by adjusting the vertical position of all inputs by a small (>1px) amount.

Clicking on a node input dot with a connected noodle should disconnect the noodle.

This happens whenever an input is initially hidden, and an input of the same type is initially visible in the same location.

Right now, this project is the top Google result for "Blazor node editor." And it appears to be pretty much the only Blazor node editor available... and its design is constrained to the highly specific domain of regex editing.

Would it be possible to separate out the node editor into its own package, and then have the regex editor be one thing implemented on top of that, so that people who want node editors for other purposes will be able to use it without it dragging a bunch of irrelevant regex-editing cruft into their projects?

These have not currently been implemented because they would not be properly parsed after using the 'Create Link' button, but they could potentially be very useful.

• Integer (with min/max number of digits and optional sign)
• Decimal (with optional sign and exponent)
• Separated (similar to Quantifier, but with a separator expression to go between each match)
• Optional (Alias for Quantifier - Zero or one)
• Define - for naming and reusing sub expressions. This could potentially be difficult to implement.

Expressions of the form /(...)...(...)/ are treated as a single group, so the quantifier does not add any additional grouping. This means that only the last group in the expression is quantified properly.

[a-zA-Z] -> Wildcard set to 'letters'
[0-9] -> Wildcard set to 'digits' (but this would create inconsistency with the parsed expression, so might be best avoided)
Variant of [ \t\r\n] -> Whitespace with corresponding options ticked.

Inversions of these should also be parsed.

Output is a-zA-Z
but should be [azA-Z]

The app should have an option to generate a node graph from a Regex that has been input by the user.

Hi, I'm writing a Trading application, I need a node-based logic editor, will this work for my needs, does it have a start node and end node, custom node example, etc.
this is screenshot of winForms.Can I replicate with these nodes just by writing custom nodes ?

Should demonstrate some of the possible functionality.

Caused by the @key used when rendering the output segments.

Add a node to support unicode classes and unicode blocks.
Output should be \p{...}.

The node should also have an 'invert' option, so the output is \P{...}.

This will give the option to repeat a character or set, similar to the Quantifier node. It will have the same options as the Quantifier node, but without "search type".

Would also affect move and delete operations.

Instead, the text should overflow with a scrollbar.

Give every node a previous input, so that nodes can be chained together for concatenation. This aims to provide a more intuitive alternative to the concatenate node and the multiple outputs of the output node.

Steps to reproduce:

1. Create one text node
2. Create another text node
3. Connect output of each node to input of another node

Current behavior: App hangs

Expected behavior: App doesn't connect the last node and shows user-friendly error

These should also be used by the 'share link' button (but potentially ignored if they are unchanged from the default).

There should be a dropdown at the top right of the UI to select the flags used by the Regex engine for search/replace.

Changes needed:

• Updated screenshot with the improved wildcard node
• Links to example expressions

I came here because I saw your old post on reddit and wanted to se how you solved the nodes in your solution, trying to teach myself how it works before I give it a go...
I tested one of my commonly used regex in your website and it came up with some problems...
Just wanted you to see how it looks...
regex is ^[A-Z]([A-Z0-9]\d{3}[A-Z0-9])(A\d{2}|A|D|E)$
basically it's a verification of sorts...
But it gave the wrong output... no worries for my part but I though you might want to know...

This will require a check to avoid duplicate @key parameters (using @key="null" instead for multiple occurrences of the same segment).

might be a dumb question but how do I delete a node?

If clicked, certain components of nodes can be marked as selected, with a white outline (see attached screenshot). Only the node itself should be selectable.

Some improvements to be made to the modal that shows when clicking the info button on a node:

• Make the close button white to improve visibility.
• Make the background lighter to improve visibility.
• Clean up css styling for custom styles.

Fails to parse ^((25[0-5]|(2[0-4]|1[0-9]|[1-9]|)[0-9])(\.(?!$)|$)){4}$ because of the |).

Empty OR options also cannot be created with the OR node, as they are removed automatically.

This would replace the current implementation where nodes are moved when they are dropped, but could potentially reduce performance while dragging.