Josiah, I am glad you are continuing this conversation. I do not have the capacity to contribute as I would at present. So I thought I would dump some design patterns and ideas I had for this.
After the save to disk by the browser, a move rather than rename will be faster on the same disk. To do this the previous version in the selected folder needs to be moved to a backups folder first, and the oldest deleted. If all that takes place is renames and deletes, downloads to folders then backups will be fast, with the only full write to disk done by the browser. Now if we have document folders that contain only tiddlywiki html files we could use the fact that mywiki.html exists in these folder to look for mywiki.html in the browser download folder(s). Add a new wiki to this folder and it automatically joins those to be collected. each browser should have a default downloads folder thus our solution could use these first and only the exceptions configured. If a tiddlywiki is collected and removed from downloads immediately, it will always save as the same and original filename. No confusing versions or overwrite prompts. We need to cater for versions if the batch is not running for a while, but it can just iterate the versions the same way with moves. It would be possible to run a batch command in any folder that adds (appends) that folder to those to be scanned for in download folders. Even better would be to add this to the desktops file manager to get a r-click to select that folder. This iteration can also scan for subfolders. In windows at least a simple registry entry can associate .tw files with your default browser and clicking a tw file will open as usual in the browser. If this is done tw files could be placed in any folder with any files including other html files and you could even auto detect downloaded tw files and move them to a default folder. With .tw files you could select your documents folder and it and all its sub folders could be scanned to locate .tw files to check for in downloads, and to identify where they beling. Again reducing the need for configuration. A recent change to tiddlywiki allows you to set the download filename before you download the first time. You could do this to ensure it is unique and maybe even name it to a .tw file. Then it would automagicaly be managed by the local scripts. Whilst these solutions aim to be universal we do need a different one for each operating system, so will it ever be universal? A multiple os intaller could solve this. Having .tw files could make it possible to auto configure everything. Using the above should be compatible with timimi being installed because once it is running you just open the wikis and timimi handles the save and no download occurs. Timimi already works with .tw files. This process should work along with tiddlydesktop and tiddlyserver as well but we could consider some helper functions to make the transition simpler. Once again this can be improved by using .tw files because we can differentiate these from common or garden html files. Finally we could integrate it with a node install by pointing to the right folder. Regards Tony -- You received this message because you are subscribed to the Google Groups "TiddlyWiki" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/tiddlywiki. To view this discussion on the web visit https://groups.google.com/d/msgid/tiddlywiki/43b64d25-48c0-4dce-8ea7-5ae87e09a7c3%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
