# Difference between revisions of "BlueprintVectorPaintBucketTool"

From Inkscape Wiki

(New page: (17:24:35) Juca: I was thinking about the paint bucket tool... (17:25:01) Juca: I dont like the white borders that are left when one fill a vector curve (17:25:48) Juca: I mean... the s...) |
|||

Line 1: | Line 1: | ||

− | (17:24:35) Juca: I was thinking about the paint bucket tool... | + | * (17:24:35) Juca: I was thinking about the paint bucket tool... |

− | + | * (17:25:01) Juca: I dont like the white borders that are left when one fill a vector curve | |

− | (17:25:01) Juca: I dont like the white borders that are left when one fill a vector curve | + | * (17:25:48) Juca: I mean... the spacing between the filled curve and the filling curve created by the paint bucket tool |

− | + | * (17:26:25) jegHegy_: it's there because stuff gets rasterized at the current zoom level | |

− | (17:25:48) Juca: I mean... the spacing between the filled curve and the filling curve created by the paint bucket tool | + | * (17:26:34) jegHegy_: including antialiasing |

− | + | * (17:26:55) ^-: [Johan] y | |

− | (17:26:25) jegHegy_: it's there because stuff gets rasterized at the current zoom level | + | * (17:26:56) Juca: I understand that the paint bucket vectorizes a pixmap of the drawing |

− | + | * (17:27:06) ^-: [Johan] what you can do is increase the stroke width of the output | |

− | (17:26:34) jegHegy_: including antialiasing | + | * (17:27:11) Juca: but there should be 2 algorithms for the paint bucket |

− | + | * (17:27:24) jegHegy_: dynamic offset helps smooth things out too | |

− | (17:26:55) ^-: [Johan] y | + | * (17:27:27) Juca: the current one |

− | + | * (17:27:30) Juca: and also another | |

− | (17:26:56) Juca: I understand that the paint bucket vectorizes a pixmap of the drawing | + | * (17:27:45) Juca: that would gets the actual vector curves |

− | + | * (17:28:27) ***jegHegy_ assumed the current way was kind of a shortcut until the math gets done for the vector stuff | |

− | (17:27:06) ^-: [Johan] what you can do is increase the stroke width of the output | + | * (17:28:37) Juca: the error comes from rounding errors in the process of rasterization and revectorization |

− | + | * (17:29:51) Juca: maybe this could be done in the future using vector effects feature (which is not implemented yet) | |

− | (17:27:11) Juca: but there should be 2 algorithms for the paint bucket | + | * (17:29:57) Juca: and which is SVG 1.2 |

− | + | * (17:31:12) Juca: but I was also wondering whether there is a way of using LPEs for that task too | |

− | (17:27:24) jegHegy_: dynamic offset helps smooth things out too | + | * (17:31:19) Juca: what do you think, johan? |

− | + | * (17:32:52) ^-: [Johan] hm | |

− | (17:27:27) Juca: the current one | + | * (17:33:12) ^-: [Johan] to get an exact solution (or nearly exact) is pretty hard in general i think |

− | + | * (17:33:35) ^-: [Johan] we can think about it though | |

− | (17:27:30) Juca: and also another | + | * (17:33:38) ^-: [Johan] it'd be nice |

− | + | * (17:33:46) ^-: [Johan] the current way would be a "quick" way | |

− | (17:27:45) Juca: that would gets the actual vector curves | + | * (17:34:00) ^-: [Johan] and the mathematical way slower, but more pleasing |

− | + | ||

− | (17:28:27) ***jegHegy_ assumed the current way was kind of a shortcut until the math gets done for the vector stuff | + | |

− | + | ||

− | (17:28:37) Juca: the error comes from rounding errors in the process of rasterization and revectorization | + | |

− | + | ||

− | (17:29:51) Juca: maybe this could be done in the future using vector effects feature (which is not implemented yet) | + | |

− | + | ||

− | (17:29:57) Juca: and which is SVG 1.2 | + | |

− | + | ||

− | (17:31:12) Juca: but I was also wondering whether there is a way of using LPEs for that task too | + | |

− | + | ||

− | (17:31:19) Juca: what do you think, johan? | + | |

− | + | ||

− | (17:32:52) ^-: [Johan] hm | + | |

− | + | ||

− | (17:33:12) ^-: [Johan] to get an exact solution (or nearly exact) is pretty hard in general i think | + | |

− | + | ||

− | (17:33:35) ^-: [Johan] we can think about it though | + | |

− | + | ||

− | (17:33:38) ^-: [Johan] it'd be nice | + | |

− | + | ||

− | (17:33:46) ^-: [Johan] the current way would be a "quick" way | + | |

− | + | ||

− | (17:34:00) ^-: [Johan] and the mathematical way slower, but more pleasing | + |

## Revision as of 20:47, 16 July 2008

- (17:24:35) Juca: I was thinking about the paint bucket tool...
- (17:25:01) Juca: I dont like the white borders that are left when one fill a vector curve
- (17:25:48) Juca: I mean... the spacing between the filled curve and the filling curve created by the paint bucket tool
- (17:26:25) jegHegy_: it's there because stuff gets rasterized at the current zoom level
- (17:26:34) jegHegy_: including antialiasing
- (17:26:55) ^-: [Johan] y
- (17:26:56) Juca: I understand that the paint bucket vectorizes a pixmap of the drawing
- (17:27:06) ^-: [Johan] what you can do is increase the stroke width of the output
- (17:27:11) Juca: but there should be 2 algorithms for the paint bucket
- (17:27:24) jegHegy_: dynamic offset helps smooth things out too
- (17:27:27) Juca: the current one
- (17:27:30) Juca: and also another
- (17:27:45) Juca: that would gets the actual vector curves
- (17:28:27) ***jegHegy_ assumed the current way was kind of a shortcut until the math gets done for the vector stuff
- (17:28:37) Juca: the error comes from rounding errors in the process of rasterization and revectorization
- (17:29:51) Juca: maybe this could be done in the future using vector effects feature (which is not implemented yet)
- (17:29:57) Juca: and which is SVG 1.2
- (17:31:12) Juca: but I was also wondering whether there is a way of using LPEs for that task too
- (17:31:19) Juca: what do you think, johan?
- (17:32:52) ^-: [Johan] hm
- (17:33:12) ^-: [Johan] to get an exact solution (or nearly exact) is pretty hard in general i think
- (17:33:35) ^-: [Johan] we can think about it though
- (17:33:38) ^-: [Johan] it'd be nice
- (17:33:46) ^-: [Johan] the current way would be a "quick" way
- (17:34:00) ^-: [Johan] and the mathematical way slower, but more pleasing