TCL通过3种不同的模式匹配删除重复的行

Question

我是编程TCL的初学者，想要从匹配

的输入文件中删除行

1. 完全相同的内容
2. 模式："ghi\/\njkl\/\nrccu1" -> "Point \.*: 10 Sinks" \.* （单词“Point”后面的数字和“[color.*”后面的内容可能不同 - 所有其他内容都需要完全匹配）
3. "mno\/\npqr\/\nrccu1" -> "stu\/\nvwx\/\nrccu1" [label = "depth: 1"] [color=salmon] [fontcolor=salmon] [style=solid]; （单词“Point”后面的数字和“[color.*”后面的内容可能不同 - 所有其他内容都需要完全匹配）

现在我有以下输入文件

"abc\/\ndef\/\nrccu1" [shape=octagon,color=red,style=filled];
"abc\/\ndef\/\nrccu1" [shape=octagon,color=red,style=filled];

"ghi\/\njkl\/\nrccu1" -> "Point 1: 10 Sinks" [color=salmon] [style=solid] [weight=8];
"123\/\n456\/\nrccu1" -> "Point 9: 10 Sinks" [color=grey] [style=solid] [weight=8];
"ghi\/\njkl\/\nrccu1" -> "Point 8: 10 Sinks" [color=grey] [style=solid] [weight=8];
"ghi\/\njkl\/\nrccu1" -> "Point 13: 20 Sinks" [color=grey] [style=solid] [weight=8];

"mno\/\npqr\/\nrccu1" -> "stu\/\nvwx\/\nrccu1" [label = "depth: 1"] [color=salmon] [fontcolor=salmon] [style=solid];
"mno\/\npqr\/\nrccu1" -> "stu\/\nvwx\/\nrccu1" [label = "depth: 4"] [color=salmon] [fontcolor=salmon] [style=solid];

"mno\/\npqr\/\nrccu1" -> "stu\/\nvwx\/\nrccu1" [label = "depth: 1"] [color=grey] [fontcolor=red] [style=solid];

输出文件应包含：

"abc\/\ndef\/\nrccu1" [shape=octagon,color=red,style=filled];

"ghi\/\njkl\/\nrccu1" -> "Point 1: 10 Sinks" [color=salmon] [style=solid] [weight=8];
"123\/\n456\/\nrccu1" -> "Point 9: 10 Sinks" [color=grey] [style=solid] [weight=8];
"ghi\/\njkl\/\nrccu1" -> "Point 13: 20 Sinks" [color=grey] [style=solid] [weight=8];

"mno\/\npqr\/\nrccu1" -> "stu\/\nvwx\/\nrccu1" [label = "depth: 1"] [color=salmon] [fontcolor=salmon] [style=solid];
"mno\/\npqr\/\nrccu1" -> "stu\/\nvwx\/\nrccu1" [label = "depth: 4"] [color=salmon] [fontcolor=salmon] [style=solid];

Answer 1

set in [open input.file r]
array set seen {}
while {[gets $in line] != -1} {
    # blank lines should be printed as-is
    if {[string length [string trim $line]] == 0} {
        puts $line
        continue
    }

    # create the "key" for this line
    regsub {\mPoint \d+} $line {} key
    regsub {\mcolor=\w+} $key {} key

    # print the line only if the key is unique
    if { ! [info exists seen($key)]} {
        puts $line
        set seen($key) true
    }
}
close $in