Missing texts and invalid characters

[Dev-iL]

Following this pull-request comment, please find attached two files (with renamed extensions), that can help reproduce/demonstrate the problem:

• a03584e36f15005c4c16b4def82aac5e.caption
• E25 - DSL construction.srt

Some noticeable problems:

1. The 1st subtitle (line 9 in the .caption; line 3 in the .srt) gets a line break at ].
2. The 2nd subtitle (ln 13 / ln 7) has a b=+ that shouldn't be there.
3. After subtitle 25 (103 / 109), the .srt file has some intermediate characters between the subtitles.
4. (...and more of the above).

[Dev-iL]

I have simplified the preparesrt logic somewhat, and now it seems to work properly. Before I submit a PR, here's my version:

        public string PrepareSrt()
        {
            const int METADATA_LINES = 7, CHARS_BEFORE_TIMESTAMP = 13, CHARS_AFTER_TIMESTAMP = 14;
            //read all file in memory
            string content = File.ReadAllText(filePath);

            // Discard the first lines, containing metadata used by Lynda desktop app to link subtitle to video:
            string output = RemoveFirstLines(content, METADATA_LINES);

            // Before every timestamp we have a constant amount of characters (starting by [NUL][SOH] and ending with a newline)
            output = Regex.Replace(output, @"\u0000\u0001[\s\S]{" + CHARS_BEFORE_TIMESTAMP + "}[\r\n]*", "");
            
            // After every timestamp we also have a constant amount of characters:
            output = Regex.Replace(output, @"(?<=\[\d\d:\d\d:\d\d\.\d\d\])[\s\S]{" + CHARS_AFTER_TIMESTAMP + "}", "");

            // Cleanup remaining non-UTF8 ASCII chars:            
            output = Regex.Replace(output, @"[^\u0020-\u007F \u000D\n\r]+", "");

            return output;
        }

[mdomnita]

Thanks, @Dev-iL . I stopped working on this because my Lynda free subscription ended a while ago and I don't get free subscription from my new workplace. Maybe I will make a new one on another email address and check out if it still works.