Help! Importing 5,000+ .rtf's from another outliner


First thanks so much for reading this to help! I have a bit of an importing nightmare, and if someone knows how to simplify things, it could save me literally weeks of work.

So here it is: Over a decade, I’ve organized my research on a Windows outliner much like Scrivener, named “Texnotes.” I have to move it to Scrivener now, and thankfully I can export the 5,000+ Texnotes articles (as rtf, html, etc.). I can also export the tree as a txt. However, what is the best approach to take these two things to ultimately create a mirror of my old Texnotes file? The way I’m doing manually will take several weeks of work.[1]

Thanks in advance!



P.S. I’m willing to buy Windows Scrivener if that will ease things at all.

[1] Right now, I’m doing it all manually. All my 5,000+ files are exported alphabetically as rtfs and in one folder. Then another file is open which shows me my outline of titles to those 5,000+ nodes. So what I’m doing is dragging the rtfs into Scrivener 1 by 1, and manually creating a new tree by referencing my old tree. Can you think of a more efficient way? Can I maybe import the 5000+ nodes with their current hierarchy into the tree pane of Scrivener? That alone would save me countless hours.

Can Texnotes export as .opml like many mind mapping apps can do? Scrivener can import .opml and maintain both the structure and the text

Hey Lunk! Thanks for the assist. These are the formats I can export in:

–Texnotes Note Format (TNP)
–Rich Tex Format (RTF)
–Plain Text (TXT)
–Unicode Text (TXT)
–Simple HTML (HTML)
–Accurate HTML with Stylesheets (HTML + CSS)

The other structure preserving option for import: Markdown. Scrivener preserves the outline specified using different headings, at least according to 10.1.4 in the manual.

So it looks like unless you can convince the developer of texnotes to support OPML or MMD, you will need to find a tool that could convert your source files first. Technically is shouldn’t be too challenging, but someone would need to be able to use Bash, Perl, Python, Ruby etc. to build the converter…

I described a technique for splitting with hierarchy in this thread, for programs that don’t support OPML (Word in that case). If you can export your outline with any degree of control, say inserting the title of an outline node into the output and putting hashes (#) in front of those lines depending on their depth ("# level one", “## level two”, etc.) then that might be a good option.

If you need formatting, the next best approach is File/Import/Import and Split… with the use of some kind of divider between sections. If it can export with something like “* * * *” between outline nodes then you can type that into the Import and Split pane and end up with 5,000 items on import. You’ll have to do the nesting yourself, but that’s not the hard part.

Worst case there is splitting. I wouldn’t waste time importing each file one by one when it is far more efficient to put your cursor between what should be two different files and hit Cmd-K, or select what should be the title of the next section, and use Opt-Cmd-K to split with that text as the title.