UR's Export Is Unreliable

[schferk]

After heavy editing - I didn't want to do a tort - I'm positive now:

UR ist faulty in its csv export. It does mix up item titles, be they direct neighbours or in the vicinity one to the other. The exported item list is NOT necessarily in the order of the items in the original tree, and thus, copying this list into an AO tree (or any other prog) and then doing a one-by-one copy and paste operation between the two progs will fill up a posssibly FALSE tree, taken not from the original tree, but from the export list (original content of item 200 becoming the content of item 202, and vice versa, e.g.).

All my programming efforts here are not at fault, neither is the Windows clipboard function, nor is AO or any of my editors : manual comparing of the items, one by one, of the raw, un-processed list exported by UR shows, here and there, or even many, discrepancies between the UR tree and the exported list.

The thread's title is correct then, and for the time being, I cannot see that it's some special chars in those title lines that trigger the mixing up.

Thus, intermediate solution, selecting the original tree, copying it into a second list, and let Beyond Compare compare for discrepancies. Then, wherever necessary, manual reordering of misplaced items in UR's faulty export list, needed to be processed by the script for the tab info.

Thus, people using UR's export (and perhaps also print???) functions for serious business, should check UR's output twice.

But now, let's have a good laugh together :

If you copy the tree directly from UR's original tree - which doesn't function well / always at first try... -, you can paste that clipboard content directly into AO or whatever, with all the leading tabs in the right places altogether. So, my formidable detour by UR's faulty export tree, then being (correctly) processed by my TSE script : 100 p.c. unnecessary.

All this unpaid extra work just for discovering that UR's Export Is Unreliable. Frightening.

But anything can be topped, then :

http://www.welt.de/vermischtes/artic...-steigern.html

[schferk]

RTF export is unreliable as well.

I'm positive now on this second bug but which occurs rarely, compared to the one described above.

We know that UR exports only those parts of the tree that are both expanded and selected; I prefer it otherwise but see the point in this way to do it.

But then, here and then, UR SKIPS some such and-expanded-and-selected parts of the tree, so that your export isn't sure to contain all you want to export (and this when you've got a normal, continuous selection, not as a possible side effect of discontinuous selection by control-mous-clicks, and not for possible monster tree exports, but some dozens of items to be exported will suffice - again, it's quite rare, but I'm positive).

If you export in .rtf format, be it for further processing / publishing out of MS Word or out of a dtp prog (InDesign or whatever), you risk here to leave out whole chapters, without ever knowing it except after publication when third parties speak of inconsistencies in your product.

Thus, anybody wishing to export UR data is well advised to also do it by external macro going back and forth, copying and pasting by clipboard, then to compare the results.

(I'm speaking here of precisely this export command: File - Export - Text items to a rtf doc - include item titles/insert page breaks/include item notes - Target file and Enter. And since there's a lot of confusion with the following command, "Item notes to a rtf doc", let me add that it's not that those skipped items only contained pics, and thus did not contain text, as one might think by (too) carefully reading those badly chosen/distinguished command denominations.)

[Nobodo]

I am having the same problem as OP on exporting.

I am exporting to HTML.

It looks like it completely ignores the way items have been moved around in the Data Explorer, and instead builds toc.html in the order items were created.

The images are of a hoisted section in UR itself, and the resulting toc.html on an html export I just did a few minutes ago.

[kinook]

It does look like the sort order is different, but I'm not seeing any missing info in my tests. Please ZIP and send a .urd file that demonstrates the problem. Thanks.
http://www.kinook.com/Forum/showthread.php?t=3038

[Nobodo]

I am not seeing any missing data, just incorrect sort order.

The sort order is important to me because it is a list of programming projects on my todo list, sorted in the order I am working on them. When I export them for business analysts to view, they think I'm working on things that I probably won't touch for weeks.

Thanks,
Mark.

[Nobodo]

More info -- it appears that the export sorts items as they were originally created. I change the order of items daily as priorities change, so it would be great if the export honored the changes.

Thanks,
Mark.