In search of ... a flexible merge software.

[barney]

Ok, folk, I'm not certain this one exists, but I'll ask, just in case.

I'm using several duplicated software items on each machine.  These are mostly [PIM] databases and note-taking apps.  I need a way to keep them concurrent on each machine.  Right now, I use Syncless to replicate changes to a different data folder and to a USB key.  That works well for one (1) machine. 

However, when I'm using the same software on two (2) or more machines ... let's take MyInfo as an example.

MyInfo creates .mio files, with .mib backups.  The process mentioned works well.  But when I'm actively using MyInfo on two (2) or more machines, I need a way to merge the different files so that all data, from each machine, is preserved.

Let's say I've added a new article on machine A, and copied several Web pages on machine B.  On both machines, this was copied to daily.mio.  Now I need a way to merge the two (2) daily.mio files so that the resultant daily.mio contains all the data from both machines.  Might be able to do it with WinMerge, but this needs to be an automated process.  What I need is a kind of cross between WinMerge and Syncless.  Dunno that I've ever seen such a beastie.

Any ideas?

No, creating a separate file for each machine is impractical.  Daily.toshiba.mio, daily.gw.mio, and daily.gw2.mio could be created, but each would have to be loaded and updated on each machine - a maintenance nightmare I'd prefer not to encounter- but a bigger downside would be that now I'd have to search three (3) files for information that should ideally be contained in one (1).  In addition, I'd have to triple the MyInfo file loading on each machine, which would end up being a resource dilemma.

I looked at SyncBackSE, but I don't thimk it'll do this on a file level any more than Syncless does.  Syncless does directories, and while it's a bit more flexible, SyncBackSE does the same.  None of the backup software to which I have access seems to cover this, either.

[skwire]

Unless I'm completely mistaken, these types of files are almost certainly proprietary (to the software) binary files and, thus, are going to be extremely difficult, if not impossible, to merge together.

[Target]

Unless I'm completely mistaken, these types of files are almost certainly proprietary (to the software) binary files and, thus, are going to be extremely difficult, if not impossible, to merge together.

that's what I was thinking - perhaps it's time to explore another option (sounds like you need something browser based that runs across a network...)

[barney]

Unless I'm completely mistaken, these types of files are almost certainly proprietary (to the software) binary files and, thus, are going to be extremely difficult, if not impossible, to merge together.

that's what I was thinking - perhaps it's time to explore another option (sounds like you need something browser based that runs across a network...)

Yeah, I'm afraid you're both correct.  I had but faint hope of getting a positive result, but I had to try - there's some pretty sharp folk hangin' out here .

Problem with another option is that I'll prolly have to do it myself  , or pay someone to do it for me  .  Either way, it's gonna take more time than is presently available to give it  .  And I suspect it'll be definitely a non-trivial task <sigh />.  I wonder if it would even be marketable?

[4wd]

Perhaps if the software in question has a merge function, (or a boring old copy/paste), you could automate the process somewhat by using AutoHK/AutoIt to simulate keys/menus/etc.

Otherwise, choosing software that has a flat file database is your best option for what you want I would think.

[x16wda]

Does the software let you do an export to a text format, so you could massage the exports together and merge back in?

[barney]

There's a problem with either flat file or export to text.  Much of what I'm doing is to capture complete Web pages, including graphics.  Some of the captures are intrinsically worthless without those graphics.  And therein lies the dilemma.

If I'm capturing, for instance, a tutorial on a particular coding discipline, I need to be able to see the code examples in the page.  Oft times, that code is presented as a screen capture, typically a .jpg or .png file.  If I cannot see the code, the capture is futile.  That's the reason I'm using MyInfo - it has the most complete capture capability of any software I've used to date, and I've tried (and sometimes paid for) a bunch of 'em.  It also puts the URL of the capture at the bottom of the page, so I have attribution and citation references available.  The only problem I have with it, so far, is the inability to merge files.  I can do a copy/paste from one database to another, record by record, but I have to be able to recognize what needs to be moved - and it is a tedious process, very time-consuming  /.  That's why I was hoping against hope for a way of automating/simplifying that process.

[skwire]

Have you contacted the authors of MyInfo to see if they have any suggestions?

[barney]

Have you contacted the authors of MyInfo to see if they have any suggestions?

Awaiting a response.